Biogeographical diversity of microbialcommunities in arcto-alpine plants

Manoj Kumar

Biogeographical diversity of m

icrobial comm

unities in arcto-alpine plants

M

anoj Kum

ar

Invitation

You are invited to the publicdefense of my PhD thesis

Biogeographical diversity ofmicrobial communities in

arcto-alpine plants

Manoj Kumar

University of GroningenAcademy Buliding

Broerstraat 5, Groningen

Monday, 12.12.2016at 12:45

Paranymphs

Mukil Maruthamuthum.marutha.muthu@rug.nl

Stephanie Grizardstephanie.grizard@univ-lyon.fr

14255_Manoj_Kumar_Cover.indd 1 08-11-16 19:22

Biogeographical Diversity of Plant Associated Microbes in Arcto-Alpine Plants

   

The research reported in this thesis was carried out at the Microbial Ecology group, which is part of the Genomics Research in Ecology and Evolution in Nature (GREEN) research group, Groningen Institute for Evolutionary Life Sciences (GELIFES), University of Groningen (The Netherlands) and Ecology and Evolutionary Biology section of Department of Biological and Environmental Science, University of Jyväskylä, Finland. The research was supported by The Netherlands Organisation for Scientific Research (NWO, 821.01.005) awarded to J.D. van Elsas, Academy of Finland research grant (259180) and Finnish cultural foundation Lapland regional fund awarded for R. Nissinen. The research was furthermore supported by Finnish cultural foundation grant and Short Term Scientific Mission (STSM) grant (COST Action FA1103) awarded to Manoj Kumar. Layout design: Manoj Kumar Cover design: Manoj Kumar and Riitta Nissinen Bookmark Design: Bala Subramaniam ISBN: 978-94-6299-492-8 (printed version) ISBN: 978-94-6299-493-5 (electronic version)

Biogeographical diversity of plant associated microbes in arcto-alpine plants

Proefschrift

ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen

op gezag van de rector magnificus prof. dr. E. Sterken

en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op

maandag 12 December 2016 om 12:45 uur

door Manoj Kumar Gopala Krishnan

geboren op 23 March 1986

te Coimbatore, India

Promotor Prof. dr. J.D. van Elsas Copromotor Dr. R.M. Nissinen Assessment Committee Prof. dr. J.T.M. Elzenga Prof. dr. M.G.A. van der Heijden Prof. dr. G. Berg

“Every th ing i s everywhere ; but the env ironment se l e c t s” Martinus Wilhelm Beijerinck and Lourens Baas Becking

Contents Chapter 1 1

General introduction Chapter 2 23

Plants impact structure and function of bacterial communities in Arctic soils  

Chapter 3 49

Plants assemble species specific bacterial communities from common core taxa in three arcto-alpine climate zones

Chapter 4 79

Fungal communities associated with arcto-alpine plants are strongly shaped by regional effects

Chapter 5 109

Strong regionality and dominance of anaerobic bacterial taxa characterize diazotrophic bacterial communities of the arctic plant species Oxyria digyna and Saxifraga oppositifolia

Chapter 6 129

Microbial diversity in arcto-alpine plants: a synthesis

Summary/Uittreksel 145

Acknowledgements 151

Author affiliations 155

!

!

Chapter 1

Manoj Kumar

General introduction

!2

Cha

pter

1

!!

3

Gen

eral

Int

rodu

ctio

n

Unlike plants and animals, microorganisms are assumed to be cosmopolitan and so they can be found globally across habitats like soils, sediments, lakes and the sea (Hanson et al., 2012). With respect to this, a range of studies have reported endemism in bacteria and fungi, where some taxa are restricted to a distinct habitat (Allen et al., 1995; Lindström and Langenheder, 2012; Oakley et al., 2010). This has been mainly found for, e.g., some extreme environments like hot springs (Bahl et al., 2011; Papke et al., 2003; Takacs-Vesbach et al., 2008; Whitaker et al., 2003). Microbial life in extreme conditions, such as the extremes of temperature, is indeed challenging. With respect to low temperatures, polar regions appear to act as an effective filter that limits colonization of the habitat (Brinkmeyer and Knittel, 2003; Cowan et al., 2014; Cox et al., 2016). Due to their large population sizes, immense phenotypic and genotypic variation, diminutive extinction and high dispersal rates, microbes follow very particular population structuring c.q. biological diversification patterns (Gerstein and Moore, 2011; Kirk et al., 2004). Thus, Fierer and Lennon (2011) suggested that the physiology, life history and ecology of most microbial taxa – even numerically dominant ones - are still understudied and thus remain poorly understood. This is especially true for cold regions such as the Arctic and alpine regions.

Furthermore, the interactions of microbes with plants in arctic and also alpine regions constitute a key ecological phenomenon that has been unexplored. Endophytic bacteria and fungi are ubiquitous in plants (Berg et al., 2014; Elvira-Recuenco and Vuurde, 2011; Hallmann et al., 1997), but the assembly, structure and function of endosphere communities have so far mainly been studied in agricultural plant species or in model plant systems in artificial settings (Badri et al., 2009; Bulgarelli et al., 2013; Compant et al., 2010; Zhang et al., 2006). The factors governing the assembly of the microbiota associated with perennial wild plants in the low temperature and low nutrient soils, such as prevalent in the Arctic, might differ greatly from those of well-fertilized crop plants, as discussed below. Origin and evolution of arctic plants During the Pleistocene (ca 2.5-0.01 million years before present, a period of repeated glaciations), the circumpolar flora only consisted of about 1,500 species of vascular plants. Plants that survived the advancing and retreating glacial fronts from the Arctic and Alpine regions mixed and migrated back and forth in this period, forming the bulk of the arcto-alpine flora we have today (Quinn, 2008b). Alsos et al. (2007) illustrated that most plant species in the high-Arctic (Spitsbergen) are indeed cold-adapted, being dispersed from surrounding continents on several occasions. Through phylogenetic analyses of

!4

Cha

pter

1

plastid DNA, Wang et al. (2016) then illustrated the origin and circum-arctic distribution of the pioneer arctic plant Oxyria digyna, which originated from the Qinghai-Tibet plateau. They postulated that several arctic plants indeed originated in the high mountains of Asia and later spread to the rest of the world. Open landscapes, strong winds and extensive snow (and ice) covers possibly aided in the dispersal of these plants that were or became well adapted to such conditions. The arctic pioneer plant Dryas octopetala was found to move along retreating glaciers during the late quaternary (between the last interglacial and glacial period) across Europe, as evidenced by studying plant fossils (Birks, 2008). This general pattern has also been found for other arctic and alpine plants (Quinn, 2008b). Plant adaptive responses to cold environments Arcto-alpine plants have to withstand stressful environments that are characterized by low temperature, low nutrient availability, severe drought, short growth season (3-4 months) and extremes of atmospheric and soil moisture levels (Billings and Mooney, 1968; Kume et al., 1999; Quinn, 2008b). All of these factors limit plant growth. To overcome the stress posed by these varied factors, plants adapt by various strategies. First, a general reduction of plant size, as clearly observed from the tenet that plant size decreases with increasing altitude/latitude in the Northern hemisphere. Plants in extreme polar desert habitats only measure about 5-10 cm (Quinn, 2008a). Second, plants in these habitats reveal stress-adaptive growth forms like chamaephytes (plants that hold regenerating buds above the soil level), hemicryptophytes (hold buds at the surface), rosettes (prevent frost damage), cushion forms (maintain temperature raised inside the plants), prostrate shrubs and tussocks (Körner, 2003; Larcher et al., 2010). Furthermore, the allocation of nutrients among different plant parts and their physiological hardening has been observed to prevent frost damage (Margesin et al., 2007; Smallwood and Bowles, 2002). Third, plants secrete the red anthocyanin, especially during the beginning and end of the growth season and convert light to heat when temperatures are low. These anthocyanins are masked by chlorophyll during the summer season. During winter, a majority of arctic plants also synthesize “antifreeze” compounds to protect the tissues from freezing (Quinn, 2008c). Fourth, plants also strictly ‘select’ their microhabitat for growth. Thus Silene acaulis, a cushion-forming plant, is usually observed growing at exposed ridges. Finally, reproductive strategies - like high germination rates, early flowering and ability of seeds to survive over winter - along with a dominantly vegetative propagation modus, a low optimal temperature for photosynthesis and a strategic allocation of captured sunlight are other key adaptation strategies of arcto-alpine plants.

!!

5

Gen

eral

Int

rodu

ctio

n

Why study arcto-alpine plant-microbe interactions? Approximately 5.7 X 106 km2 of the surface of Earth are covered by Arctic regions (Derry and Staddon, 1999), and the arctic habitat is predicted to experience major shifts in vegetation, with potential extinctions of flora and fauna (IPCC 2007). Sturm et al. (2001) reported a progressive (past) increase over time in the dominance of certain shrubs in the Arctic, by comparing photographs of Arctic regions taken with 50-year intervals. The fact that these shrubs displayed a positive response to warming indicated global warming as the cause.

The overall rise in temperature due to global warming is troublesome for the arctic region, as soil defrosting may result in as-yet unpredictable changes in carbon release and storage patterns. Thus, the natural ecosystem in the Arctic may also show potentially undesirable changes in soil nutrient mineralization by the local microbial communities (Zinger et al., 2009). Major questions are on the availability of carbon from thawing permafrost soils, the decomposition rates, the availability of nitrogen, phosphorus and sulfur, and on whether an ecological niche is created on which the emerging plants will rely for their growth and survival (Kirk et al., 2004). A detailed study of microbial functions in arctic/alpine soils is thus needed (Derry and Staddon, 1999).

It is unknown to what extent the aforementioned plant adaptations have affected their interactions with plant-associated microbes. However, it may be assumed that varying interactions, with bearing on plant fitness, may have ensued. For instance, the release of organic compounds by plant roots likely assists in spurring the abundance and diversity of bacteria in the rhizosphere, as compared to those in the corresponding soils (Derry and Staddon, 1999; Tam et al., 2001). In spite of the fact that we currently dispose of a large body of knowledge on the strategies used by temperate-climate plant species to direct, deal with, and potentially make use of, their associated microbiomes, those used by arctic/alpine plants are still largely unknown. Hence, it is important to study the interactions of local microbiomes with these pioneer plant species and their role in the establishment of plants in the new habitats. Hereunder, I briefly discuss plant-microbe interactions from a theoretical perspective, with special emphasis on those specific for the arctic/alpine regions. Plant–microbe interactions Microorganisms (bacteria and fungi) have been living in the Earth’s surface layers for over several billions of years. Ever since the (later) emergence of land plants, both beneficial and detrimental interactions with the resident

!6

Cha

pter

1

microbiomes have emerged (Zhang et al., 2006). In theoretical terms, the plant-microbe interactions are differentiated into the following classes:

Symbiosis – as an example, nodulation or mycorrhization of plant parts by bacteria (in particular rhizobia) or fungi

Commensal (epiphytic or endophytic) presence – plant tissues can carry bacterial or fungal cells at their surface or within, without any apparent harm

Pathogenicity – particular bacteria and/or fungi can infest plants and cause plant disease Of crucial importance is the fact that specialized fungi interact with

plants by forming symbiotic relationships named “mycorrhizae”. About 90% of all vascular plant species, including forest trees, crops, grass and tundra plants can form different symbiotic relationships with fungi; i.e., ectomycorrhizal (EM), ericoid mycorrhizal (ECM) and arbuscular mycorrhizal (AM) (Bonfante and Genre, 2010). The ability to interact with fungi is considered to be important for plants in most, if not all, of the ecosystems (Kytöviita 2005). Nutrient acquisition, solubilization of nutrients and minerals, stress tolerance, N and P mobilization and resistance towards plant pathogens are significant benefits provided by mycorrhizae to plants (Finlay, 2007, 2008). Likewise, host plants are necessary for the growth and reproduction of many of the mycorrhiza-forming fungal species (Bonfante and Genre, 2010).

In addition to these associations with clear ecological outcomes, soil bacteria and fungi can also interact with plants yielding less well defined effects, which can be of slightly deleterious, neutral or beneficial nature (Van Elsas et al., 2007).

What connects all of the above interactions is the fact that any successful interaction establishes in phases. To these, a prior phase of recognition and attraction of the microbial partner is primordial. This first interactive phase commonly takes place in the rhizosphere, i.e. the soil surrounding the roots of plants. The microbial communities in soil are attracted to root-exuded compounds, i.e. sugars, amino acids, organic acids, sterols, proteins and/or secondary metabolites. Using such molecules as chemoattractants, they can thus colonize the rhizosphere and rhizoplane (Badri et al., 2009; Compant et al., 2010; Sørensen and Sessitsch, 2007). Some bacteria and fungi from the rhizosphere can subsequently further colonize and invade plant tissues and get established as endophytes. Thus, plants can be thought of selectively attracting and repelling microbes first into their rhizospheres and then placing strong filters on the ones that are admitted to become endophytic.

Although it is attractive to state that the resultant plant-associated microbes can support the colonization of plants in soils that pose harsh

!!

7

Gen

eral

Int

rodu

ctio

n

conditions, there is actually only scarce evidence for this tenet when it comes to arctic/alpine plants. Granted, such effects have been revealed in soils with low nutrients, pH or high metal toxicity (Bradley et al., 1981; Mapelli et al., 2011; Sousa et al., 2014). Furthermore, there is extensive knowledge pointing to the fact that specific micro-organisms can provide resistance to plants against plant pathogens, either by outcompeting these or by antagonizing them, and also assist plants in their tolerance to conditions of drought (Miniaci et al. 2007; Hardoim et al. 2008; Compant et al. 2010; Zhang et al. 2006; Sessitsch & Howieson 2002). However, again, we lack knowledge on all of these facets of plant microbiomes for arctic/alpine soils.

We currently know that virtually all plants on Earth harbor microbiomes, that are often partially rhizospheric, partially endophytic (living in the endosphere) (Tan and Zou, 2001). Thus, microbial cells have been found to dwell in the roots, stems, flowers, fruits, tubers, bark and even seeds of diverse plants (Hardoim et al., 2012; Müller et al., 2015; Zhang et al., 2006). Moreover, plants may even ‘select’ their own microbiome from soil via roots secreting signaling molecules (Yergeau et al., 2007). Duineveld et al. (1998), using a direct molecular method, first revealed that plants influence the microbiome of their rhizosphere. Then, Smalla et al. (2001) indicated that different plants attract a different microbiome for the same soil. Moreover, Bragazza et al. (2012) showed that plants [in peat lands] can affect the functioning of the soil microbial communities. These microbial communities differed based upon the plant species, plant physiology and even between different plant parts. Thus, a vision has developed of plants creating specific ecological niches that, consequently, select specific microbial species (Garbeva et al., 2004; Yergeau et al., 2007).

As discussed above, the interactions of plants with microorganisms may give benefits of different kinds to the plant. Moreover, these plant-microbial associations were also observed in perennial trees, herbaceous plants and arctic tundra (Nissinen et al., 2012). Thus, we currently face a situation of emerging evidence for a strong role of microbiomes in the ecology of a range of temperate climate plants, versus a general paucity of knowledge on the diversity, community composition and role of microbiomes in arctic and alpine plants. Colonization of land by arctic/alpine plants - an invisible help Arcto-alpine pioneer plants can grow in extremely challenging conditions, for instance in cold rocky areas along receding glaciers. Usually, the soil/moraine forefields of such areas are low in sources of C and N, and they are initially vegetation-free (Miniaci et al., 2007). A logical hypothesis is that local

!8

Cha

pter

1

microbiomes can offer (ecophysiological) help to pioneer plant species. Indeed, a study on bio-weathering of recently deglaciated moraine soils at the Midre Lovénbreen glacier in Svalbard in High Arctic illustrated the role of bacteria in soil formation (Mapelli et al., 2011). Bacteria with high phylogenetic diversity able to drive key biogeochemical processes were shown to play key roles in C and N inputs in these soils. This microbial nutrient deposition contributes to soil fertility, thus assisting early ecosystem development (Mapelli et al., 2011; Nemergut et al., 2007). There is a feedback mechanism as well, as Miniaci et al. (2007), from their studies in alpine glacier forefields, concluded that vegetation patches in the glacier forefield increased microbial activities and biomass. This clearly affected the plants, as a possible payback. However, we know very little about the ecological importance of this “invisible help” and need more detailed studies to understand the role of these early colonizers in ecosystem development. In the upcoming years, freshly deglaciated soils may become hot spots for colonization by plants due to rising local temperatures. Plant–microbe interactions may then play pivotal roles in these events. Several plant factors, like stress tolerance, uptake and processing of nitrogen and phosphorus, survival, early colonization, dispersal by seeds, slow growth rate and survival may be affected or driven by plant-associated microbiomes. To understand the key processes, we need to gather fundamental knowledge on the microbial communities in such habitats and their effects on pioneer plant species. The microbial diversity in arcto-alpine biomes Bacteria - Several accounts (censi) of bacterial communities from different arctic habitats, i.e. permafrost soils (Bakermans et al., 2014; Gittel et al., 2014; Yergeau et al., 2010), cryoconite holes (Cameron et al., 2012; Edwards et al., 2011, 2013) and tundra soils (Männistö et al., 2007; Neufeld and Mohn, 2005; Schmidt and Bölter, 2002; Wallenstein et al., 2007), have been provided in recent years. The similarities between the catalogues of 16S rRNA gene sequences in soils from Finland and in Alaskan tundra soils have supported the idea of a similar circumpolar distribution of bacteria (Männistö et al., 2007). Likewise, Nissinen et al. (2012) reported that endophytic bacterial communities from Kilpisjärvi, Finland, are host-plant-specific when observed at a high taxonomic level. A tight association of specific Sphingomonas spp. with O. digyna and Diapensia lapponica was observed. In contrast, a study of Antarctic rhizosphere soil communities of Deschampsia antarctica and Colobanthus quitensis revealed the occurrence of similar bacterial community structures and no influence of plant species (Teixeira et al., 2010). However, in the latter case, the bacterial communities in the respective bulk soils were found to be different from those of the rhizosphere. The authors hypothesized that, even though

!!

9

Gen

eral

Int

rodu

ctio

n

vegetation influences the microbial communities, other factors such as differential moisture content or freeze-thaw cycles might have exerted more powerful effects (Teixeira et al., 2010). Diazotrophic bacteria - Nitrogen (N) is considered to be a key limiting factor for both microbial and plant growth in the Arctic tundra and hence the diversity of diazotrophic communities from these (low nitrogen) soils has been relatively well studied. Wallenstein et al. (2009) observed overall increases and decreases in microbial enzyme activities with the presence and absence of soil nitrogen, respectively. Subsequently, a series of N-addition experiments by Sistla et al. (2012) revealed that microbial growth, enzyme synthesis and carbon allocation in tussock tundra soils during the arctic summer are N-limited. A study of a chronosequence along the forefield of the Damma glacier by Brankatschk et al. (2011) reported a very low N content in soils from 10 to 70 years with scarce vegetation. Furthermore, they observed that potential N-fixation activity was undetectable, and denitrification and nitrification rates were extremely low. The lowest nifH gene copy numbers were detected from newly exposed 10-year old soils. In contrast, a densely plant-covered 120-year old soil showed both higher enzyme activities and higher copy numbers of the nifH, nirK and nirS genes. This suggested that plants had a strong bearing on the microbial associations and an established increase of nitrification and denitrification activities. A study by Nordin et al. (2004) reported that less than 1% of the experimentally added N was recovered from plants, whereas about 49% was recovered from the soil microbial biomass. The study explains that, even though plants are capable of directly taking up added N (NH4

+, NO3-,

amino acids), they compete poorly with microbes. Hence it is clear that microbes constitute an integral part of plant growth in the N limited tundra soils. Fungi - It is known that the majority of plants in tropical and temperate regions form mycorrhizal relationships in order to enhance their capacities to obtain N and P compounds. In line with this, it has been estimated that 61-86% of the plant nitrogen in the arctic tundra is furnished, from bound nitrogen, by ECM and EM fungal communities, a value similar to that in temperate climate regions (Hobbie and Hobbie, 2006). An early report by Högberg (1997) already concluded that soil nitrogen uptake by non-mycorrhizal plants is less than that by mycorrhizal plants. However, the possible dominance of non-mycorrhizal or low-mycorrhizal flora in the Arctic environment is puzzling considering the importance of mycorrhizal symbioses in comparable (nutrient poor) temperate-climate habitats (Kytöviita 2005). Furthermore, very little is known about the role of mycorrhizae in the soil C balance of the tundra ecosystem, in spite of the fact that it contains ca. 11% of world’s C pool (Ludley and Robinson, 2008). The precise roles of mycorrhizae

!10

Cha

pter

1

in the arctic tundra have not been extensively studied and there is very little information about the types of symbioses and fungal adaptations to Arctic conditions. The diversity of root-associated fungi in the arctic tundra has been investigated in a pioneer study by Bjorbækmo et al. (2010). They found that a total of 137 fungal OTUs, mostly representing ECM genera such as Cenococcum, Cortinarius, Hebeloma, Inocybe and Tomentella were present in the root systems of D. octopetala.

With respect to the soil microbiomes, Schadt et al. (2003) reported seasonal variation in the fungal biomasses of tundra soils, with a maximum in fungal activity during late winter under the snow. However, in case of AM fungi, Kytöviita & Ruotsalainen (2007) suggested that the activity decreases with lower temperature and host plants receive thus less benefit from the symbiosis. Also, a decrease in temperature was the single most important factor that limited mycorrhizal growth and colonization rates in the Arctic (Ruotsalainen & Kytöviita 2004; Kytöviita 2005). In a different study by the same group, an increase in temperature from 12 to 17˚C benefited the low-arctic host plants Potentilla crantzii and Rananculus acris with increased mycorrhizal (AM) association, providing more nutrients (especially phosphorus) (Kytöviita & Ruotsalainen 2007). It was hypothesized that the rising temperature improves mycorrhizal performance, in turn providing more nutrients to the host and thus supporting the persistence of cold-climate meadow plants. Factors such as a change in temperature, soil moisture, freezing-thawing, decomposition of organic polymers and phenolic compounds, and availability of soil carbon, were behind these fungal community shifts. However, the precise effects of each of these factors on the microbial shifts were unclear. Sampling regions Mayrhofen, Austria (European Alps) The region above the tree-line over the snow-covered mountains (altitude ca. 3000m above sea level) is termed the alpine habitat. The European Alps (43.25-48.25˚N, 4.25-16.25˚E) constitute a long mountain range in the south of Europe, with the highest peaks reaching 4400 to 4800m above sea level (Casty et al., 2005). The relatively short, snow-free growing season spans 3-4 months. The regional climate is characterized by low temperatures (i.e. annual mean of 2-4˚C) and high precipitation rates by mainly snow. Its scarce vegetation reflects the alpine and arctic tundra (Nemergut et al., 2005). Kilpisjärvi, Finland (Low-Arctic) Kilpisjärvi is located in the northernmost part of Finland, at the borders of Finland, Sweden, and Norway (68˚N), 400 km north of Arctic Circle. Its flora

!!

11

Gen

eral

Int

rodu

ctio

n

is dominated by mountain birch forest in the valleys and by tundra plant species at higher elevations. The annual mean temperature is about -2.2°C, with a plant growing season of ca. 90-100 days in the valleys. Kilpisjärvi is one of the coldest places in continental Europe (Männistö et al., 2007). Ny-Alesund, Svalbard (High-Arctic) Spitsbergen (Norway) is part of an isolated archipelago located in the high Arctic at 74-81°N; most of the land cover is dominated by glaciers and permafrost layers, and the mean annual temperature is -4°C. The soil temperatures have been reported to be below zero for more than 250 days per year ranging from -6C to -25C (Coulson & Hodkinson 1995). Any biological activity is restricted to less than 10% of the total land mass coupled with about 3 months of plant growing season. The Svalbard flora is dominated by tree-less tundra and arcto-alpine plant species (Reigstad et al., 2011). Plant species studied Oxyria digyna and Saxifraga oppositifolia are among the key pioneer species in the arcto-alpine soils, growing in low-nutrient tundra soils (usually along the rocky patches or snow-melt areas), and were early colonizers of arctic and alpine regions after the ice age. They grow in similar habitats and thrive in tree-less rocky/fell/tundra habitats, but they differ in their mycorrhizal associations and physiological characteristics. Our sampling sites from all the three regions were located in the rocky areas with small soil pockets that held the target plants. O. digyna (mountain sorrel) is a pioneer plant species and is common in the arctic tundra, with distribution from the high Arctic (80°N. Spitsbergen, Norway) to temperate and alpine habitats. O. digyna usually grows in cold, moist habitats and typically colonizes arctic and alpine stream banks, tallus/scree slopes, snow patches and rocky slopes, primarily those facing north. It is also a typical pioneer species in glacier forelands (Robbins and Matthews, 2009) and has been estimated to be one of the first plants to colonize terrain left barren by retreating ice, for instance at the end of the last ice age. O. digyna is a non-mycorrhizal plant species, but it is very successful in colonizing novel habitats and establishing there, in spite of the limited dispersal of its winged fruits (Marr et al., 2008). It is not just a transient pioneer species, but establishes well in novel sites and forms part of the permanent flora. S. opposi t i fo l ia , a semi-woody cushion plant, is a long-lived perennial herb with opposite scale-like leaves, attractive purple-colored flowers and a wide circumpolar distribution. The distribution spans from Spitsbergen, Norway, along Scandinavian Lapland all the way up to 83 °N in Greenland, the Siberian

!12

Cha

pter

1

and Canadian Arctic to the alpine regions of Europe, Asia and North America. The habitat range even extends to the polar desert and several sites in Devon Island, Canada. The leaves can tolerate extreme water stress (up to -55 bar in lab) (Teeri, 1973). S. oppositifolia is one of the first plant species to colonize soil after glacier retreat. It has also been selected for coordinated observations within the International Tundra Experiment (Kume et al., 1999). It is an ectomycorrizal species and its dispersal and reproduction is mainly by seeds. Being one of the earliest flowering species (period of 8-9 days after snow melt) among the arctic tundra makes it a strong contender to colonize newly exposed unvegetated soils (Stenström and Molau, 1992). Objective of this thesis Given the fact that the diversity and factors shaping community structures of plant-associated microbes in the Arctic are poorly known compared to those in temperate climate regions, I decided to make an inventory of the factors (biotic and abiotic) that determine the composition and diversity of soil- and plant-associated bacterial and fungal communities in selected arcto-alpine soils. A key question was how and to what extent pioneer plant species in these habitats make use of their associated microbiomes, and to what extent they select these. Furthermore, I also addressed the diversity of diazotrophic bacterial communities associated with the same plants in the N limited arctic habitats.

To achieve these goals, I characterized the plant-associated microbial diversities in two pioneer arctic plants with differing mycorrhizal association and the soils associated with these plants. The selected plants, O. digyna (non-mycorrhizal) and S. oppositifolia (EM), are among the early colonizers of the harsh soil habitats and prefer similar growth conditions (Figure 1). The sampling took place in selected low-arctic (Kilpisjärvi, Finland), high-arctic (Ny-Ålesund, Svalbard, Norway) and alpine (Innsbruck, Austria) environments. I associated the local microbial community structures with differences in location, climate, soil properties and plant species (Figure 1). Furthermore, I addressed the putative role of these microbes in the nitrogen cycle in order to shed light on the putative role in plant development. Hypotheses

Bacterial community structures and functioning in arctic and alpine soils are highly influenced by geographic region, vegetation, pH and soil chemical parameters.

Plant-associated endosphere and rhizosphere microbial communities are selected from bulk soil communities.

!!

13

Gen

eral

Int

rodu

ctio

n

Endophytic bacterial diversity is primarily dependent on plant species, indicating a strong selection by the plants.

Plants select species-specific communities from the soil microbiota in a context (biogeographical zone)-dependent manner. The abundance and diversity of bacterial communities do not correlate with latitude or climate, but community structure does.

Nitrogen acquisition by plants in arctic soils is driven strongly by plant-associated bacteria, and, conversely, plants strongly shape the nitrogen cycling communities.

!Figure 1 Flow chart of our hypothesis of drivers of the soil and plant-associated microbiomes in selected arcto-alpine habitats: Mayrhofen (Alps), Kilpisjärvi (low-Arctic) and Ny-Ålesund (high-Arctic)

!Outline of the thesis The aforementioned hypotheses were tested and answered in more detail in the following chapters. The influence of geographic region on bulk soil microbiomes was assessed in chapters 2, 3, 4, 5, the influence of plant species on rhizosphere communities in chapters 2, 3, 4, 5 and the impact of plant species on endosphere communities in chapters 3, 4 and 5.

Specifically, a brief overview of our current understanding of the plant-associated bacterial and fungal communities in arcto-alpine regions is discussed in the introductory chapter 1. Microbiomes may help in initial seed germination, root growth (by producing growth hormones), settlement and adherence of the roots, N fixation and P solubilization. In addition, the pioneer plants may establish proper conditions for the microbes to survive and settle as well and provide an environmental filter for microbial establishment in return for the nutrients received from them. In chapter 2, I used microbial community fingerprintings to investigate the impact of plants or soil physico-chemical properties on the diversity and community structure of bacterial (and diazotroph) communities from Kilpisjärvi, Finland (low-arctic) and

!14

Cha

pter

1

Longyearbyen, Svalbard (high-arctic). In addition, Biolog Gen III plates were used to determine the functional diversities of both soil and rhizosphere bacterial communities from the same sites.

In order to understand the key microbial players that determine the impact of geographical region and plant species on the diversity of bacterial and fungal communities, I then used next generation sequencing (NGS) methods for the following chapters. The samples for NGS analyses (chapters 3, 4, 5) were collected from three distinct geographical regions: Mayrhofen, Innsbruck (alpine), Kilpisjärvi, Finland (low-arctic) and Ny-Ålesund, Svalbard (high-arctic). On the basis of NGS, chapter 3 addresses the phylogenetic diversity and community structure of the bacterial communities from bulk soil, rhizosphere soil and the endosphere of O. digyna and S. oppositifolia. We investigated the key bacterial OTUs that impact the host plant species per geographical region. In addition to determining the impact of region and host plant specificity, we also dig deeper in order to pinpoint the “core” endosphere communities associated with O. digyna and S. oppositifolia from all three sampling regions. We also hypothesize that “core” endospheres will help in the establishment of the plants in these habitats and play key roles in acquiring N in these low nutrient soils. The diversity and community structures of the fungal communities associated with bulk soil and rhizosphere and endosphere of both O. digyna and S. oppositifolia are reported in the chapter 4. We also analyzed the impact of soil physico-chemical properties (soil pH, SOM and available P) on the bacterial and fungal bulk soil communities in chapters 3 and 4 respectively. With nifH gene as a proxy for diazotrophs, a new nested primer system was developed and used to study the diversity and community structure of potential nitrogen fixing bacterial (PNFB) communities in chapter 5. In addition, the impact of geographic region and host-specificity of the PNFB communities was studied and we found anoxic PNFB enriched in the Arctic regions. In chapter 6, I finally summarize the new knowledge on the relative impact of geographic region and plant species on the plant-associated microbial communities in the (nutrient poor) arctic and alpine regions. Considering the scarce knowledge we have about plant-microbe interactions in the selected arcto-alpine plants, I make some concluding remarks and considerations for the future research in this exciting and ecologically important research area. References

Allen, E. B., Allen, M. F., Helm, D. J., Trappe, J. M., Molina, R., and Rincon, E. (1995). Patterns and regulation of mycorrhizal plant and fungal diversity. Plant and Soil 170, 47–62. doi:10.1007/BF02183054.

Alsos, I. G., Eidesen, P. B., Ehrich, D., Skrede, I., Westergaard, K., Jacobsen, G. H., et al.

!!

15

Gen

eral

Int

rodu

ctio

n

(2007). Frequent long-distance plant colonization in the changing Arctic. Science (New York, N.Y.) 316, 1606–9. doi: 10.1126/ science.1139178.

Badri, D. V, Weir, T. L., van der Lelie, D., and Vivanco, J. M. (2009). Rhizosphere chemical dialogues: plant-microbe interactions. Current opinion in biotechnology 20, 642–50. doi:10.1016/j.copbio.2009.09.014.

Bahl, J., Lau, M. C. Y., Smith, G. J. D., Vijaykrishna, D., Cary, S. C., Lacap, D. C., et al. (2011). Ancient origins determine global biogeography of hot and cold desert cyanobacteria. Nature communications 2, 163. doi: 10.1038/ncomms1167.

Bakermans, C., Skidmore, M. L., Douglas, S., and McKay, C. P. (2014). Molecular characterization of bacteria from permafrost of the Taylor Valley, Antarctica. FEMS microbiology ecology, 1–16. doi:10.1111/1574-6941.12310.

Berg, G., Grube, M., Schloter, M., and Smalla, K. (2014). Unraveling the plant microbiome: looking back and future perspectives. Frontiers in Microbiology 5, 1–7. doi:10.3389/fmicb.2014.00148.

Billings, W., and Mooney, H. (1968). The ecology of arctic and alpine plants. Biological Reviews 43, 481–529.

Birks, H. H. (2008). The Late-Quaternary history of arctic and alpine plants. Plant Ecology & Diversity 1, 135–146. doi:10.1080/17550870802328652.

Bjorbækmo, M. F. M., Carlsen, T., Brysting, A., Vrålstad, T., Høiland, K., Ugland, K. I., et al. (2010). High diversity of root associated fungi in both alpine and arctic Dryas octopetala. BMC plant biology 10, 244. doi:10.1186/1471-2229-10-244.

Bonfante, P., and Genre, A. (2010). Mechanisms underlying beneficial plant-fungus interactions in mycorrhizal symbiosis. Nature communications 1, 48. doi:10.1038/ ncomms1046.

Bradley, R., Burt, A. J., and Read, D. J. (1981). Mycorrhizal infection and resistance to heavy metal toxicity in Calluna vulgaris. Nature 292, 335–337. doi:10.1038/ 292335a0.

Bragazza, L., Parisod, J., Buttler, A., and Bardgett, R. D. (2012). Biogeochemical plant–soil microbe feedback in response to climate warming in peatlands. Nature Climate Change 3, 1–5. doi:10.1038/nclimate1781.

Brankatschk, R., Töwe, S., Kleineidam, K., Schloter, M., and Zeyer, J. (2011). Abundances and potential activities of nitrogen cycling microbial communities along a chronosequence of a glacier forefield. The ISME journal 5, 1025–37. doi:10.1038/ ismej.2010.184.

Brinkmeyer, R., and Knittel, K. (2003). Diversity and Structure of Bacterial Communities in Arctic versus Antarctic Pack Ice. Applied and environmental microbiology 69, 6610–6619. doi:10.1128/AEM.69.11.6610.

Bulgarelli, D., Schlaeppi, K., Spaepen, S., Ver Loren van Themaat, E., and Schulze-Lefert, P. (2013). Structure and Functions of the Bacterial Microbiota of Plants. Annual review of plant biology. doi:10.1146/annurev-arplant-050312-120106.

Cameron, K. a, Hodson, A. J., and Osborn, a M. (2012). Structure and diversity of bacterial, eukaryotic and archaeal communities in glacial cryoconite holes from the Arctic and the Antarctic. FEMS microbiology ecology 82, 254–67. doi:10.1111/j.1574-6941.2011.01277.x.

Casty, C., Wanner, H., Luterbacher, J., Esper, J., and Böhm, R. (2005). Temperature and precipitation variability in the European Alps since 1500. International Journal of Climatology 25, 1855–1880. doi:10.1002/joc.1216.

Compant, S., Clément, C., and Sessitsch, A. (2010). Plant growth-promoting bacteria in the rhizo- and endosphere of plants: Their role, colonization, mechanisms involved and prospects for utilization. Soil Biology and Biochemistry 42, 669–678.

!16

Cha

pter

1

doi:10.1016/j.soilbio.2009.11.024. Coulson, S., and Hodkinson, I. (1995). Thermal environments of Arctic soil organisms

during winter. Arctic and Alpine Research 27, 364–370. doi:10.2307/ 1552029. Cowan, D. a, Makhalanyane, T. P., Dennis, P. G., and Hopkins, D. W. (2014). Microbial

ecology and biogeochemistry of continental Antarctic soils. Frontiers in microbiology 5, 154. doi:10.3389/fmicb.2014.00154.

Cox, F., Newsham, K. K., Bol, R., Dungait, J. A. J., and Robinson, C. H. (2016). Not poles apart: Antarctic soil fungal communities show similarities to those of the distant Arctic. Ecology Letters 19, 528–536. doi: 10.1111/ele.12587.

Derry, A., and Staddon, W. (1999). Functional diversity and community structure of micro-organisms in three arctic soils as determined by sole-carbon-source-utilization. Biodiversity and Conservation 8, 205–221.

Duineveld, B., Rosado, A., van Elsas, J. D., and van Veen, J. A. (1998). Analysis of the dynamics of bacterial communities in the rhizosphere of the chrysanthemum via denaturing gradient gel electrophoresis and substrate utilization. Applied and Environmental Microbiology 64, 4950–4957.

Edwards, A., Anesio, A. M., Rassner, S. M., Sattler, B., Hubbard, B., Perkins, W. T., et al. (2011). Possible interactions between bacterial diversity, microbial activity and supraglacial hydrology of cryoconite holes in Svalbard. The ISME journal 5, 150–60. doi:10.1038/ismej.2010.100.

Edwards, A., Douglas, B., Anesio, A. M., Rassner, S. M., Irvine-Fynn, T. D. L., Sattler, B., et al. (2013). A distinctive fungal community inhabiting cryoconite holes on glaciers in Svalbard. Fungal Ecology 6, 168–176. doi:10.1016/j.funeco.2012.11.001.

Van Elsas, J. D., Tam, L., Finlay, R. D., Killham, K., and Trevors, J. T. (2007). “Microbial Interactions in Soil,” in Modern Soil Microbiology, eds. J. D. van Elsas, J. K. Jansson, and J. T. Trevors (CRC Press), 177–210.

Elvira-Recuenco, M., and Vuurde, J. W. L. van (2011). Natural incidence of endophytic bacteria in pea cultivars under field conditions. Canadian Journal of Microbiology.

Fierer, N., and Lennon, J. T. (2011). The generation and maintenance of diversity in microbial communities. American journal of botany 98, 439–48. doi:10.3732/ajb. 1000498.

Finlay, R. D. (2007). “The Fungi in Soil,” in Modern Soil Microbiology, eds. J. D. van Elsas, J. K. Jansson, and J. T. Trevors (CRC Press), 107–146.

Finlay, R. D. (2008). Ecological aspects of mycorrhizal symbiosis: with special emphasis on the functional diversity of interactions involving the extraradical mycelium. Journal of experimental botany 59, 1115–26. doi:10.1093/jxb/ern059.

Garbeva, P., van Veen, J. a, and van Elsas, J. D. (2004). Microbial diversity in soil: selection microbial populations by plant and soil type and implications for disease suppressiveness. Annual review of phytopathology 42, 243–70. doi:10.1146/annurev. phyto.42.012604.135455.

Gerstein, A. C., and Moore, J.-S. (2011). Small is the new big: assessing the population structure of microorganisms. Molecular Ecology 20, 4385–4387. doi:10.1111/j.1365-294X.2011.05263.x.

Gittel, A., Bárta, J., Kohoutová, I., Mikutta, R., Owens, S., Gilbert, J., et al. (2014). Distinct microbial communities associated with buried soils in the Siberian tundra. The ISME journal 8, 841–53. doi:10.1038/ismej.2013.219.

Hallmann, J., Quadt-Hallmann, A., Mahaffee, W. F., and Kloepper, J. W. (1997). Bacterial endophytes in agricultural crops. Canadian Journal of Microbiology 43, 895–914. doi:http://dx.doi.org/10.1139/m97-131.

Hanson, C. a., Fuhrman, J. a., Horner-Devine, M. C., and Martiny, J. B. H. (2012). Beyond biogeographic patterns: processes shaping the microbial landscape. Nature Reviews

!!

17

Gen

eral

Int

rodu

ctio

n

Microbiology 10, 1–10. doi:10.1038/nrmicro2795. Hardoim, P. R., Hardoim, C. C. P., van Overbeek, L. S., and van Elsas, J. D. (2012).

Dynamics of seed-borne rice endophytes on early plant growth stages. PloS one 7, e30438. doi:10.1371/journal.pone.0030438.

Hardoim, P. R., van Overbeek, L. S., and Elsas, J. D. Van (2008). Properties of bacterial endophytes and their proposed role in plant growth. Trends in microbiology 16, 463–71. doi:10.1016/j.tim.2008.07.008.

Hobbie, J. E., and Hobbie, E. A. (2006). 15 N in Symbiotic Fungi and Plants Estimates Nitrogen and Carbon Flux Rates in Arctic Tundra. Ecology 87, 816–822. doi:10.1890/0012-9658(2006)87[816:NISFAP]2.0.CO;2.

Högberg, P. (1997). Tansley Review No. 95 15N natural abundance in soil-plant systems. New Phytologist 137, 179–203.

Kirk, J. L., Beaudette, L. a, Hart, M., Moutoglis, P., Klironomos, J. N., Lee, H., et al. (2004). Methods of studying soil microbial diversity. Journal of microbiological methods 58, 169–88. doi:10.1016/j.mimet.2004.04.006.

Körner, C. (2003). “The Alpine Life Forms,” in Arctic Plant Life: Functional Plant Ecology of High Mountain Ecosystems (Springer), 9–20.

Kume, A., Nakatsubo, T., Bekku, Y., and Masuzawa, T. (1999). Ecological significance of different growth forms of purple saxifrage, Saxifraga oppositifolia L., in the High Arctic, Ny-Ålesund, Svalbard. Arctic, Antarctic, and Alpine Research 31, 27–33.

Kytöviita, M.-M. (2005). Asymmetric symbiont adaptation to Arctic conditions could explain why high Arctic plants are non-mycorrhizal. FEMS microbiology ecology 53, 27–32. doi:10.1016/j.femsec.2004.09.014.

Kytöviita, M.-M., and Ruotsalainen, A. L. (2007). Mycorrhizal benefit in two low arctic herbs increases with increasing temperature. American journal of botany 94, 1309–1315. doi:10.3732/ajb.94.8.1309.

Larcher, W., Kainmüller, C., and Wagner, J. (2010). Survival types of high mountain plants under extreme temperatures. Flora - Morphology, Distribution, Functional Ecology of Plants 205, 3–18. doi:10.1016/j.flora.2008.12.005.

Lindström, E. S., and Langenheder, S. (2012). Local and regional factors influencing bacterial community assembly. Environmental Microbiology Reports 4, 1–9. doi:10.1111/ j.1758-2229.2011.00257.x.

Ludley, K. E., and Robinson, C. H. (2008). “Decomposer” Basidiomycota in Arctic and Antarctic ecosystems. Soil Biology and Biochemistry 40, 11–29. doi:10.1016/j.soilbio.2007.07.023.

Männistö, M. K., Tiirola, M., and Häggblom, M. M. (2007). Bacterial communities in Arctic fjelds of Finnish Lapland are stable but highly pH-dependent. FEMS Microbiology Ecology 59, 452–465. doi:10.1111/j.1574-6941.2006.00232.x.

Mapelli, F., Marasco, R., Rizzi, A., Baldi, F., Ventura, S., Daffonchio, D., et al. (2011). Bacterial communities involved in soil formation and plant establishment triggered by pyrite bioweathering on arctic moraines. Microbial ecology 61, 438–47. doi:10.1007/s00248-010-9758-7.

Margesin, R., Neuner, G., and Storey, K. B. (2007). Cold-loving microbes, plants, and animals--fundamental and applied aspects. Die Naturwissenschaften 94, 77–99. doi:10.1007/s00114-006-0162-6.

Marr, K. L., Allen, G. a., and Hebda, R. J. (2008). Refugia in the Cordilleran ice sheet of western North America: chloroplast DNA diversity in the Arctic-alpine plant Oxyria digyna. Journal of Biogeography 35, 1323–1334. doi:10.1111/j.1365-2699.2007 .01879.x.

Miniaci, C., Bunge, M., Duc, L., Edwards, I., Bürgmann, H., and Zeyer, J. (2007). Effects

!18

Cha

pter

1

of pioneering plants on microbial structures and functions in a glacier forefield. Biology and Fertility of Soils 44, 289–297. doi:10.1007/s00374-007-0203-0.

Müller, H., Berg, C., Landa, B. B., Auerbach, A., Moissl-Eichinger, C., and Berg, G. (2015). Plant genotype-specific archaeal and bacterial endophytes but similar Bacillus antagonists colonize Mediterranean olive trees. Frontiers in Microbiology 6, 138. doi:10.3389/fmicb.2015.00138.

Nemergut, D. R., Anderson, S. P., Cleveland, C. C., Martin, A. P., Miller, A. E., Seimon, A., et al. (2007). Microbial community succession in an unvegetated, recently deglaciated soil. Microbial ecology 53, 110–22. doi:10.1007/s00248-006-9144-7.

Nemergut, D. R., Costello, E. K., Meyer, A. F., Pescador, M. Y., Weintraub, M. N., and Schmidt, S. K. (2005). Structure and function of alpine and arctic soil microbial communities. Research in microbiology 156, 775–84. doi:10.1016/j.resmic.2005.03.004.

Neufeld, J., and Mohn, W. (2005). Unexpectedly high bacterial diversity in arctic tundra relative to boreal forest soils, revealed by serial analysis of ribosomal sequence tags. Applied and environmental microbiology 71, 5710–5718. doi:10.1128/AEM. 71.10.5710.

Nissinen, R. M., Männistö, M. K., and van Elsas, J. D. (2012). Endophytic bacterial communities in three arctic plants from low arctic fell tundra are cold-adapted and host-plant specific. FEMS microbiology ecology 82, 510–22. doi:10.1111/j.1574-6941.2012.01464.x.

Nordin, A., Schmidt, I., and Shaver, G. (2004). Nitrogen uptake by arctic soil microbes and plants in relation to soil nitrogen supply. Ecology 85, 955–962. doi: 10.1890/03-0084.

Oakley, B. B., Carbonero, F., van der Gast, C. J., Hawkins, R. J., and Purdy, K. J. (2010). Evolutionary divergence and biogeography of sympatric niche-differentiated bacterial populations. The ISME journal 4, 488–97. doi: 10.1038/ismej.2009.146.

Papke, R. T., Ramsing, N. B., Bateson, M. M., and Ward, D. M. (2003). Geographical isolation in hot spring cyanobacteria. Environmental Microbiology 5, 650–659. doi: 10.1046/j.1462-2920.2003.00460.x.

Quinn, A. J. (2008a). “Arctic and Antarctic tundra,” in Arctic and Alpine Biomes, ed. L. S. Woodward (Greenwood Press), 33–82.

Quinn, A. J. (2008b). “Introduction to Arctic and Alpine Biomes,” in Arctic and Alpine Biomes, ed. L. S. Woodward (Greenwood Press), 1–32.

Quinn, A. J. (2008c). “Mid-Latitude Alpine Tundra,” in Arctic and Alpine Biomes, ed. L. S. Woodward (Greenwood Press), 83–156.

Reigstad, L. J., Jorgensen, S. L., Lauritzen, S.-E., Schleper, C., and Urich, T. (2011). Sulfur-oxidizing chemolithotrophic proteobacteria dominate the microbiota in high arctic thermal springs on Svalbard. Astrobiology 11, 665–78. doi:10.1089/ ast.2010.0551.

Robbins, J., and Matthews, J. (2009). Pioneer vegetation on glacier forelands in southern Norway: emerging communities? Journal of Vegetation Science 20, 889–902. doi:10.1111/j.1654-1103.2009.01090.x.

Ruotsalainen, A. L., and Kytöviita, M.-M. (2004). Mycorrhiza does not alter low temperature impact on Gnaphalium norvegicum. Oecologia 140, 226–33. doi:10.1007/ s00442-004-1586-3.

Schadt, C. W., Martin, A. P., Lipson, D. a, and Schmidt, S. K. (2003). Seasonal dynamics of previously unknown fungal lineages in tundra soils. Science (New York, N.Y.) 301, 1359–61. doi:10.1126/science.1086940.

Schmidt, N., and Bölter, M. (2002). Fungal and bacterial biomass in tundra soils along an arctic transect from Taimyr Peninsula, central Siberia. Polar Biology 25, 871–877. doi:10.1007/s00300-002-0422-7.

Sessitsch, A., Howieson, J., Perret, X., Antoun, H., and Martinez-Romero, E. (2002). Advances in Rhizobium research. Critical Reviews in Plant Sciences 21, 323–378.

!!

19

Gen

eral

Int

rodu

ctio

n

doi:10.1080/0735-260291044278. Sistla, S. a., Asao, S., and Schimel, J. P. (2012). Detecting microbial N-limitation in

tussock tundra soil: Implications for Arctic soil organic carbon cycling. Soil Biology and Biochemistry 55, 78–84. doi:10.1016/j.soilbio. 2012.06.010.

Smalla, K., Wieland, G., Buchner, A., Zock, A., Parzy, J., Kaiser, S., et al. (2001). Bulk and rhizosphere soil bacterial communities studied by denaturing gradient gel electrophoresis: plant-dependent enrichment and seasonal shifts revealed. Applied and Environmental Microbiology 67, 4742–4751. doi:10.1128/AEM.67.10.4742.

Smallwood, M., and Bowles, D. J. (2002). Plants in a cold climate. Philosophical transactions of the Royal Society of London. Series B, Biological sciences 357, 831–47. doi:10.1098/rstb.2002.1073.

Sørensen, J., and Sessitsch, A. (2007). “Plant-associated bacteria lifestyle and molecular interactions,” in Modern Soil Microbiology, eds. J. D. Van Elsas, J. T. Trevors, and J. K. Jansson (CRC Press), 211–236.

Sousa, N. R., Ramos, M. a., Marques, A. P. G. C., and Castro, P. M. L. (2014). A genotype dependent-response to cadmium contamination in soil is displayed by Pinus pinaster in symbiosis with different mycorrhizal fungi. Applied Soil Ecology 76, 7–13. doi:10.1016/j.apsoil.2013.12.005.

Stenström, M., and Molau, U. (1992). Reproductive ecology of Saxifraga oppositifolia: phenology, mating system, and reproductive success. Arctic and Alpine Research 24, 337–343. doi:10.2307/1551289.

Sturm, M., Racine, C., Tape, K., Cronin, T. W., Caldwell, R. L., and Marshall, J. (2001). Increasing shrub abundance in the Arctic. Nature 411, 546–547.

Takacs-Vesbach, C., Mitchell, K., Jackson-Weaver, O., and Reysenbach, A. L. (2008). Volcanic calderas delineate biogeographic provinces among Yellowstone thermophiles. Environmental Microbiology 10, 1681–1689. doi:10.1111/j.1462-2920. 2008.01584.x.

Tam, L., Derry, A., Kevan, P., and Trevors, J. (2001). Functional diversity and community structure of microorganisms in rhizosphere and non-rhizosphere Canadian arctic soils. Biodiversity and Conservation 10, 1933–1947.

Tan, R. X., and Zou, W. X. (2001). Endophytes: a rich source of functional metabolites. Natural product reports 18, 448–59.

Teeri, J. (1973). Polar desert adaptations of a high arctic plant species. Science 179, 496–497. doi:10.1126/science.179.4072.496.

Teixeira, L. C. R. S., Peixoto, R. S., Cury, J. C., Sul, W. J., Pellizari, V. H., Tiedje, J., et al. (2010). Bacterial diversity in rhizosphere soil from Antarctic vascular plants of Admiralty Bay, maritime Antarctica. The ISME journal 4, 989–1001. doi:10.1038/ ismej.2010.35.

Wallenstein, M. D., Mcmahon, S. K., and Schimel, J. P. (2009). Seasonal variation in enzyme activities and temperature sensitivities in Arctic tundra soils. Global Change Biology 15, 1631–1639. doi:10.1111/j.1365-2486.2008 .01819.x.

Wallenstein, M. D., McMahon, S., and Schimel, J. (2007). Bacterial and fungal community structure in Arctic tundra tussock and shrub soils. FEMS microbiology ecology 59, 428–35. doi:10.1111/j.1574-6941.2006.00260.x.

Wang, Q., Liu, J., Allen, G. A., Ma, Y., Yue, W., Marr, K. L., et al. (2016). Arctic plant origins and early formation of circumarctic distributions: a case study of the mountain sorrel, Oxyria digyna. New Phytologist 209, 343–353. doi:10.1111/nph.13568.

Whitaker, R. J., Grogan, D. W., and Taylor, J. W. (2003). Geographic Barriers Isolate Endemic Populations of Hyperthermophilic Archaea. Science 301, 2002–2004. doi:10.1126/science.1086909.

Yergeau, E., Hogues, H., Whyte, L. G., and Greer, C. W. (2010). The functional potential

!20

Cha

pter

1

of high Arctic permafrost revealed by metagenomic sequencing, qPCR and microarray analyses. The ISME journal 4, 1206–14. doi:10.1038/ismej.2010.41.

Yergeau, E., Newsham, K. K., Pearce, D. A., and Kowalchuk, G. A. (2007). Patterns of bacterial diversity across a range of Antarctic terrestrial habitats. Environmental microbiology 9, 2670–82. doi:10.1111/j.1462-2920.2007.01379.x.

Zhang, H. W., Song, Y. C., and Tan, R. X. (2006). Biology and chemistry of endophytes. Natural product reports 23, 753–71. doi:10.1039/b609472b.

Zinger, L., Shahnavaz, B., Baptist, F., Geremia, R. a, and Choler, P. (2009). Microbial diversity in alpine tundra soils correlates with snow cover dynamics. The ISME journal 3, 850–9. doi:10.1038/ismej.2009.20.

! !

!!

21

Gen

eral

Int

rodu

ctio

n

! !

 22

 

!

!

Chapter 2

Manoj Kumar, Minna K Männistö, Jan Dirk van Elsas, Riitta M. Nissinen

Plants impact structure and function of bacterial communities in Arctic soils!

Plant and Soil, February 2016, DOI: 10.1007/s11104-015-2702-3 !

!

!24

Cha

pter

2

Abstract Background and aims Microorganisms are prime drivers of ecosystem functions in the Arctic, and they are essential for vegetation succession. However, very little is known about the phylogenetic and functional diversities of the bacterial communities associated with arctic plants, especially in low-organic- matter soils. Here, we studied the diversity and community structure of the total and diazotrophic bacterial communities in the rhizospheres of two arctic pioneer plant species, Oxyria digyna and Saxifraga oppositifolia, as well as the phenotypic profiles of these communities. Methods Plant and soil samples were harvested from four sampling sites located in the low (Finnish) or high (Norwegian) Arctic. Bacterial diversities and community structures were determined with 16S rRNA and nifH gene-targeted PCR and T-RFLP profiling. The phenotypic (functional and antibiotic resistance) profiles of the soil bacterial communities were determined with BIOLOG Gen III plates. Results and conclusions Location and soil type, as well as plant species were shown to significantly influence the bacterial community structures. This was evident from the clustering of the communities per sampling location and per plant species in ordination plots. Geographical location as well as, plants also influenced the functional profiles of the soil bacterial communities. The antibiotic resistance profiles were associated with the location the communities were derived from, whereas the abilities to tolerate oxidative stress and specific antibiotics were associated with the rhizosphere communities Keywords Rhizosphere bacterial diversity; Arctic plants; Functional diversity; T-RFLP; BIOLOG Gen III, D-serine

!

!25!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

Introduction The diversity and community structure of soil bacteria are influenced by a range of biotic and abiotic factors. Several studies have reported pH to be a major factor influencing community structures across soil habitats (Männistö et al. 2007; Fierer and Jackson, 2006; Lauber et al. 2009). Moreover, factors such as soil moisture regime (Brockett et al. 2012), C:N ratio and phosphate level (Kuramae et al. 2012), the presence of plants (Knelman et al. 2012; Duineveld et al. 1998; 2001; Garbeva et al. 2004; Kuramae et al. 2012; Urbanova et al. 2015) and the input of plant-derived exudates (Duineveld et al, 1998; 2001; Fierer & Ladau 2012; Pii et al. 2015) have also been shown to influence soil bacterial communities. Such factors are in turn related to the more ‘overall’ factors geographic location (Yergeau et al. 2007), elevation (Singh et al. 2012) and snow cover and its dynamics (Schimel et al. 2004; Zinger et al. 2009).

Ever since the emergence of terrestrial plants, relationships (both beneficial and detrimental ones) have developed between plants and microbes, and plant-associated microbes have differentiated into symbionts, endophytes, pathogens and parasites (Zhang et al. 2006). Among the plant-inhabiting microorganisms, bacteria play major roles as colonizers of the endosphere as well as root surface (the rhizoplane). They also colonize the soil surrounding plant roots, i.e. the rhizosphere, which is considered to be a hot spot for plant-microbe interactions. Progressively-increasing evidence has revealed that plant-associated bacteria play major roles in nutrient uptake and resistance to abiotic stress factors, like drought and cold, in plants (Sessitsch et al. 2002; Zhang et al. 2006; Miniaci et al. 2007; Hardoim et al. 2008; Compant et al. 2010).

Plant roots release organic compounds such as sugars, polysaccharides, amino acids, fatty acids, sterols, proteins and secondary metabolites in their exudates (Badri et al. 2009; Cesco et al. 2010; Chaparro et al. 2014; Pii et al. 2015b). Such compounds can have strong impact on the diversity and community composition of soil microbes. Furthermore, plants have been shown to be able to select, to a considerable extent, their own microbial community from the soil by secreting unique signaling molecules (Duineveld et al. 2001; Smalla et al. 2001; Yergeau et al. 2007), thus impacting local soil nutrient cycling. However, very little is known of the plant-associated bacterial communities in arctic climate zones, in spite of the fact that climate change is predicted to have a significant impact on the arctic ecosystem, in terms of the plant communities and the extent of plant coverage (Walker and Wahren 2006).

Plants in the Arctic are challenged with several abiotic stress factors, i.e. cold winters and short, cool summers, drought, anoxia and high

!

!26

Cha

pter

2

fluctuations of solar radiation (Kume et al. 1999; Körner 2003). However, plants living in arctic systems are well adapted to their environments, as during the short summers, they have growth rates comparable to temperate plants (Chapin 1983; Chapin et al. 1987; Smallwood and Bowles 2002; Margesin et al. 2007). Rather than by low temperatures directly, growth rates of arctic plants may be limited by a restricted availability of nutrients due to slow mineralization processes. Available nitrogen is probably the major limiting factor for plant growth in Arctic soils (Schimel et al. 2004; Wallenstein et al. 2009), and bacterial communities involved in nitrogen fixing and cycling are likely to have a significant impact on plant growth.

In the light of the paucity of information on plant-microbe interactions in the Arctic, we focused on the impact of two key pioneering arctic plant species, Oxyria digyna and Saxifraga oppositifolia, on the soil bacterial communities. Both selected plants are found in relatively poorly characterized low-organic-matter soils in the Arctic (Robbins and Matthews 2009), are perennial (although O. digyna is suggested to be relatively short-lived) and propagated by seed. O. digyna is non-mycorrhizal, whereas S. oppositifolia is reported to be mainly ectomycorrhizal (Marr et al. 2008, Stenström and Molau, 1992). A better understanding of the microbiota that is associated with these plants is expected to shed light on the key (microbial) processes in arctic pioneer species. To enable a broad view of such microbiota, we analyzed the systems from two angles, i.e. (1) ‘overall’ based on community DNA (using direct molecular tools), and (2) function-based, using community-level phenotypic arrays. We hypothesized that in these low organic matter soils, the presence of plant, but also plant species have a strong influence on the rhizosphere bacterial community structure (16S, nifH) and function (metabolic profiles). Materials and Methods Study sites and sample collection Samples were collected from four sites in two different climatic zones/location in the Arctic: three sites in the Kilpisjärvi region (northwestern Lapland, Finland - low Arctic, 69 °N) and one site in Longyearbyen (Svalbard, Norway; high Arctic, 74-81°N). See Figure 1.

The three Kilpisjärvi sites Saana, Jehkas and Salluoaivi (abbreviated hereafter as S, J and Sa, respectively - See Table 1) were characterized by scattered arcto-alpine vegetation (about 800 m.a.s.l..). At the three sites, the annual mean temperature was about -2.2°C, and the plant growing season was about 90-100 days. O. digyna and (in J and Sa) S. oppositifolia were the main plant species, other dominating plant species were mosses (Polytrichum sp. and

!

!27!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

other), lichens, Bistorta vivipara and Saxifraga species. In S and J Ranunculus glacialis was abundant. The sampling site at Svalbard was located close to the Longyearbreen glacier (abbreviated hereafter as LG) near Longyearbyen. The area is dominated by glaciers and permafrost with a mean annual temperature of -4°C, and growing season of 70-90 days. The vegetation was also characterized as tundra, with scattered pioneer plants. In addition to O. digyna and S. oppositifolia, LG site was dominated by lichens and mosses. B. vivipara and Saxifraga sp. were also present.

Figure 1 Map showing the geographic location of the sampling sites in Longyearbyen, (Svalbard, Norway) and Kilpisjärvi (Finland) and a detailed view of the three sampling sites in Kilpisjärvi.

Six replicate plants per species plus six bulk soil samples were collected

from each site in July, 2011. The two target plant species, O. digyna and S. oppositifolia, were collected as whole plants with adhering soil [from site S, only O. digyna was harvested]. Soil and plant samples were taken to the lab, stored at 4°C and processed within 48-72 hours of sampling. To obtain rhizosphere soil, loose soil was first removed by careful shaking, after which tightly-adhering soil was sampled from the root surfaces of each plant with sterile small blunt end forceps avoiding any damage to roots. To preserve material, three 250 mg aliquots of bulk and rhizosphere soil from each sample (replicate) were snap-frozen with liquid N2 and stored at -80°C. The bulk soil samples were collected approximately 5 meters from plants with a sterile spade 15-20 cm below surface, and contained no visible plant roots.

!

!28

Cha

pter

2

Table 1 Background data of sampling sites and habitats

Location (climatic zone)

Site name (abbreviation)

Latitude Elevation Habitat Vegetation cover

Kilpisjärvi (low arctic) Saana (S) N 69.05 890 m.a.s.l.

Boulder/scree slope, close to melting ice patch

30-50%

Kilpisjärvi (low arctic) Jehkas (J) N 69.08 925 m.a.s.l.

Boulder/scree slope, close to melt-water stream

<30%

Kilpisjärvi (low arctic) Salluoaivi (Sa) N 69.13 861 m.a.s.l.

Small valley, ice patch, close to melt-water pond

30-50%

Longyearbyen (high arctic)

Longyearbreen Glacier (LG)

N 78.20 n.a.

Glacier forefront, scree dominated valley

<30%

Soil analysis Soil pH-H2O was measured as 1:5 (v:v) fresh soil to water ratio. Soil slurry was shaken briefly, incubated at room temperature for 2 h, and pH was measured using a pH meter (model 220, Denver instruments). The organic matter content of the soil was determined based on loss upon ignition, by heating pre-dried soil at 600°C for 4 h. Soil microbial biomass was measured using the substrate-induced respiration (SIR) method of Anderson and Domsch (1978). Carbon-based respiration was induced by adding 20 mg of glucose in a mixture of 1 g of soil and 2ml demineralized water and was shaken at 150 rpm at room temperature for 2.5 h. The CO2 released was measured using gas chromotography and the microbial biomass estimated according to Anderson and Domsch (1978).

The soluble ammonium (NH4+-N) and nitrate (NO3

--N) levels were measured with the San++ automated wet chemistry analyzer (Skalar, Breda, the Netherlands) according to manufacturers instructions: 12.5 g of fresh soil were suspended in 30 ml of 1M KCl. The mixture was shaken overnight, filtered using nitrogen free filter paper (Macherey-Nagel # 321) and the filtrate used for further analysis. The analysis followed the standard method for 2-100 mg/kg total N (Cat# 461-218 issue, Skalar, Breda, the Netherlands).

Soluble phosphate was analyzed spectrophometrically by using a slightly modified protocol of Strickland and Parsons (1968). 1.2 cm3 of dry soil was mixed in 2 ml of demineralized water and incubated at 20°C for 22 hrs. Then, 60 ml of demineralized water was added to the suspension, shaken at 20°C for 1 h and filtered using paraffin wax filter papers (Schleicher and Schuell # 512). Five ml of filtrate were mixed with 5 ml of reagent mixture (5N sulphuric acid, 4% ( w/v) molybdate solution, 1.75% (w/v) ascorbic acid solution, 0.275% (w/v) potassium antimonyl tartarate solution, Vortex-mixed

!

!29!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

and incubated at room temperature for 20 min. The absorbance value of samples and standards were measured at 882 nm using a spectrophotometer (GENESYS 5, Spectronic Instruments) and the soluble phosphate was calculated (See supplementary text file S1 for formulae). Soil DNA extraction and PCR Soil DNA was extracted using the MoBio Power soil kit (MoBio, Carlsbad, CA USA) according to the manufacturer’s instructions with slight modifications: 0.5 g of soil sample was used instead of 0.25 g because of low microbial counts and the samples were homogenized for 30 sec with a bead beater (Fast Prep, MP biomedicals). Approximately 5-10 ng of DNA was used for PCR amplification with universal 16S rRNA eubacterial primers to determine the total bacterial diversity. FAM labeled 27F (AGAGTTTGATCMTGGCTCAG) and unlabeled 1492R (TACGGYTACCTTGTTACGACTT) primers were used for the amplification (Lane et al. 1991). Each 30 !l PCR reaction contained 5-10 ng of sample DNA, 1X PCR buffer, 1 mg/ml of BSA, 0.2 mM dNTP’s, 0.3 !M of each primer and 100 U/ml Fermentas DreamTaq DNA Polymerase (Thermo Scientific). A BioRad thermocycler was used for the amplifications and conditions were: 2 min denaturation at 94°C followed by 30 cycles of denaturing, annealing, and extension at 94°C for 45 s, 55°C for 1 min and 72°C for 2 min, respectively. Final extension was carried out at 72°C for 12 min. All PCR amplified products were checked on 1% agarose gels for the presence of the desired bands, of about 1450 bps.

FAM-labeled primers POL-F (TGCCAYCCSAARGCBGACTC) and POL-R (ATSGCCATCATYTCNCC) were used to analyse the diversity of nitrogen fixing bacterial communities (Poly et al. 2001). PCR was carried out in a reaction volume of 35 !l using approximately 50-60 ng of sample DNA, 1X PCR buffer, 1 mg/ml of BSA, 0.25 mMdNTP’s, 0.25 !M of each primers and 25 U/ml FermentasDreamTaq DNA Polymerase (Thermo Scientific). A thermocycler (Bio-Rad) was used for the PCR amplifications, with conditions: a 4 min denaturation step at 94°C followed by 35 cycles of denaturing, annealing and extension at 94°C for 1 min, 54°C for 1 min and 72°C for 2 min respectively. The final extension was carried out at 72°C for 10 min, after which the reaction mix was kept at 8°C.

The DNA fragments were purified using innuPREP PCR pure 96 Kit (Analytik Jena, Jena) based on the manufacturer’s instructions and centrifuged at 2,325 x g (4-15 C, Sigma centrifuge).

!

!30

Cha

pter

2

DNA restriction and T-RFLP analysis Amplified 16S and nifH PCR fragments were digested using MspI and MnlI restriction enzymes respectively. 20 !l restriction mix contains, approximately 150-200 ng (8-10 !l) of amplified fragments, 1 !g of t-RNA, 2 !l of restriction buffer, 1 !l of MspI/MnlI enzyme and MilliQ water. Samples were incubated at 37°C for overnight.

The samples were precipitated using ethanol/EDTA/sodium acetate mixture. For each 20 !l of restriction mix 4 !l of EDTA/Na-acetate mixture (Equal volume of 0.125 M EDTA and 3 M Na-acetate) and 50 !l of 100% ethanol was added and incubated at room temperature for 15 min. The samples were centrifuged (Eppendorf Centrifuge) at 2,900 g at 6°C for 1 hr, after which the supernatant was removed by turning the plate and spinning it briefly at low speed (100g for 30 s). Then, the pellet was washed with 100 !l of 70% ice-cold ethanol and centrifuged again at 2900 g at 6°C for 20 mins. The supernatant was again removed as above, and the pellet was dried at 37°C for 10 mins. Then the pellet was dissolved in 8 !l of MilliQ water and 1-2 !l of sample was used for T-RFLP analysis with Genescan LIZ600 as standard and using ABIPRISM 377 automatic sequencer from Fragment analysis service, University of Turku, Finland.

Community phenotypic profiling A protocol modified from Trevor et al. (2010) was used for the extraction of microbial cells from both bulk and rhizosphere soils. 20 g of bulk (2:3, w/v), and 10 g of rhizosphere (1:2 w/v) soils were suspended in sterile 0.1% sodium pyrophosphate (pH 7.0) and shaken at 140 rpm for 1 h at 10°C. The resulting suspensions were centrifuged at 1,200 g for 10 min at 4°C and the supernatants (10 – 13 ml for bulk soils, 7-8 ml for rhizosphere soils) transferred into fresh sterile tubes. The washing step was repeated with fresh sterile 0.1% sodium pyrophosphate and the supernatants were then combined. The extracts (25 ml bulk soil, 15 ml rhizosphere soil) were then centrifuged at 2,000 g for 10 min at 4°C, after which supernatants were transferred to new sterile centrifuge tubes and centrifuged at 7,100 g for 30 min at 4°C to pellet the cells. The pellets were resuspended in 1.7 ml of 0.1 M potassium buffer, transferred to sterile 2 ml tubes, and centrifuged (860 g for 3 min) to remove any remaining soil debris. The supernatants were transferred to sterile tubes, centrifuged at 10,400 g for 5 min at 4°C. The final pelleted cells were then used for phenotypic profiling. Here, the protocol of Derry et al (1999) was followed.

Phenotypic profiling of the bulk and rhizosphere soil microbial communities was done with GenIII MicroPlates (BIOLOG, CA, USA),

!

!31!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

containing 71 carbon source utilization assays and 23 chemical sensitivity assays (list of substrates at http://www.biolog.com/pdf/milit/00P 185rA GEN III MicroPlate IFU Mar2008.pdf). The pelleted cells were washed with 0.1M potassium phosphate buffer twice and resuspended in inoculation fluid IF-A (BIOLOG, CA, USA). Three replicate plates were incubated per sample. 100 !l of cell suspension was inoculated per well, and incubated at 10°C in a plastic box to prevent drying during prolonged incubation. The soil communities from LG site showed lower metabolic activities than those from all Kilpisjärvi soils, with little utilization of most carbon sources during the first week of incubation. To exclude false negatives, we therefore extended incubation to up to 30 days, at which point most of the samples showed clear activity, as indicated by color development, and used the final absorbance values for comparative analyses. The color change was measured using a microplate reader at 600 nm (Versamax Microplate reader, Molecular devices). Samples with no color change in positive control were rejected.

Data analysis The sizes of T-RFLP fragments of both 16S rRNA and nifH samples obtained from the ABIPRISM automatic sequencer were determined using Peak Scanner software (Applied Biosystems). Only peaks ranging from 50 to 600 bp (16S rRNA genes) and 40- 360 bp (nifH) were chosen for analysis. The T-RF’s were aligned and excess noise was filtered out based on peak area using on-line software T-REX (http://trex.biohpc.org/). Samples with less than a total of 30 TRF’s for 16s rRNA and 20 TRF's for nifH were removed. Minimal of three replicates per sample group were analyzed, with the exception of nifH profiles of LG and Saana bulk soil samples, where only 2 replicates were used. The final data were normalized as described by Dunbar et al (2001). Normalized T-RFPL peaks were used as a proxy for bacterial community members in analyzes of community structures. Community structures were analyzed using PRIMER (PRIMER-E Ltd). Bray-Curtis dissimilarity analysis was used to quantify community compositional differences. Non-metric multidimensional (NMDS) scaling (Kruskal stress formula = 1 and minimum stress =0.01) was used to compare the community structures of total and nitrogen-fixing bacterial communities based on T-RFLP profiles. Hierarchical cluster analysis of the T-RFLP profiles was performed with CLUSTER5 using the complete linkage clustering mode (Clarke 1993). The effect of sampling site and vegetation on the T-RFLP profiles was tested using two-way crossed analysis of similarities (ANOSIM) and factors contributing to the differences evaluated using similarity percentage (SIMPER) analysis included in the PRIMER software (Clarke 1993).

!

!32

Cha

pter

2

The diversity indices for both bacterial and diazotrophs were analyzed by calculating species richness and H’ Shannon diversity indices using Univariate Diversity Indices (DIVERSE, Primer 6). The obtained species richness and Shannon values were used for determining differences in the diversity using two-way ANOVA and clustered box-plots from IBM SPSS Statistics viewer.

For the analysis of phenotypic profiles of the communities (BIOLOG GenIII MicroPlate data), the absorbance values from the last reading from which background values were subtracted (negative control) were used for the analyses. In order to get comparable values, each sample’s absorbance was divided by that sample plate’s average absorbance as described by Campbell et al. (1997) and Lucas et al. (2013). Differences in the profiles were analyzed with Primer 6.0 package with Permanova+ package (Clarke and Gorley 2006). Bray-Curtis dissimilarity analysis was used to quantify differences in metabolic profiles of the communities using the normalized, square root-transformed absorbance data. Furthermore, ANOSIM was used to determine the statistical significance between the communities, and SIMPER was used to identify the variables contributing to the dissimilarities (Clarke 1993).

Differences in physico-chemical properties between bulk and rhizosphere soils and between different sites were determined using Independent-Samples T test (IBM SPSS statistics viewer) The samples were grouped and tested based on site or soil type. Results Soil characteristics The physicochemical properties of the LG soil (Longyearbyen) differed from those of the S, Sa and J soils (Kilpisjärvi). Both the bulk and rhizosphere soils from the latter sampling sites were significantly (P < 0.001) more acidic (pH range 5.45-6.12) than the corresponding LG soil samples (pH range 6.14-6.53) (Table 2). No significant differences in pH were detected across the soils from the three Kilpisjärvi sites.The LG bulk soil had a significantly (P<0.01) higher OM content (5.09%) than the three Kilpisjärvi bulk soils (1.11-2.22%, average 1.6%; Table 2). A clear impact of the presence of plants was detectable, as all rhizosphere soils had a higher OM content and were more acidic as compared to the corresponding bulk soils (Table 2). No differences were detectable, with respect to pH and OM content, between the rhizosphere soils of O. digyna and S. oppositifolia.

The soil nitrogen contents were different between the LG (Longyearbyen) and the three Kilpisjärvi soils. The LG bulk soils contained

!

!33!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

2.22-2.47 mg soluble NH4+-N /kg soil, and the Kilpisjärvi soils <0.44 mg

NH4+-N/kg soil. Whereas the LG rhizosphere soils had similar soluble NH4

+-N levels to the bulk soils, those (both O. digyna and S. oppositifolia) in Kilpisjärvi had tenfold higher NH4

+-N concentrations (4.05-5.86 NH4+-N kg-1 soil, Table

2) than the corresponding bulk soils. In sharp contrast to the foregoing, the LG soils had significantly lower

concentrations of soluble NO3--N (0.11-0.22 NO3--N /kg soil) than the Kilpisjärvi soils (1.11-91.0 NO3--N /kg soil) (P<0.05; Table 2). Moreover, the rhizosphere soils in all three Kilpisjärvi sites, but not in the Longyearbyen site, contained 5-60 times higher NO3--N concentrations than the corresponding bulk soils.

The levels of soluble phosphate were low (i.e. 1.11 – 3.17 g/ L soil) in all soils and there was no significant differences between soil types or sampling sites. The microbial biomass levels were between 0.02 and 0.04% of the total organic matter in all bulk soils measured.

Bacterial diversity and community structure First, the T-RFLP based structures of the bacterial communities in all soil samples were used as the basis for the determination of the Shannon diversity indices, thus describing the ‘richness’ and ‘evenness’ of the dominant broad bacterial groups. Analyses of these indices revealed no significant differences between sampling sites or geographical location (Kilpisjärvi or Longyearbyen). No statistically significant differences were detected between the bulk soil and S. oppositifolia rhizosphere soils (P>0.05; Figure 2a). However, such indices were significantly higher for the O. digyna rhizosphere communities than for the bulk soil ones (P<0.02; two-way ANOVA) at each sampling site (Figure

Table 2 Mean values of physio-chemical parameters of bulk and rhizosphere soils (O.d. = Oxyria digyna), (S.o. =Saxifraga oppositifolia) from the sampling sites. Standard deviation (SD) is in parenthesis. ND = Not determined

Sites

Soil type pH (SD) Organic matter % of dry soil

Microbial biomass % of organic matter

Soluble NO3--N mg N kg-1 soil

Soluble NH4+-N mg N kg-1 soil

Soluble PO4 g L-1 soil

Saana Bulk 6.12 (0.25) 1.71 (0.71) 0.04 (0.01) 2.40 (1.74) 0.48 (0.17) 1.52 (0.58)

Saana O.d. rhizosphere 5.54 (0.23) 7.02 ND 19.53 (1.48) 5.42 (5.4) 1.32 (0)

Jehkas Bulk 6.21 (0.18) 1.65 (0.5) 0.03 (0.001) 1.11 (0.69) 0.33 (0.11) 1.56 (0.3)

Jehkas O.d. rhizosphere 5.45 (0.07) 21.2 ND 61.0 (9.58) 5.86 (0.37) 2.96 (0)

Jehkas S.o. rhizosphere 5.70 (0.12) 23.7 ND 41.5 (27.1) 4.82 (1.14) 3.17 (0.29)

Salluoaivi Bulk 6.06 (0.43) 2.22 (0.54) 0.03 (0.01) 1.16 (1.09) 0.44 (0.1) 2.31 (0.32)

Salluoaivi O.d. rhizosphere 5.98 (0.06) 16.0 (0.06) ND 30.1 (10.8) 5.40 (0.91) 2.04 (0.39)

Salluoaivi S.o. rhizosphere 5.92 (0.12) 17.0 (0.04) ND 7.88 (4.06) 4.05 (0.51) 1.11 (0.41)

Longyearbreen Bulk 6.53 (0.73) 5.09 (0.37) 0.02 (.001) 0.11 (0.07) 2.23 (0.38) 2.09 (0.46)

Longyearbreen O.d. rhizosphere 6.14 (0.11) ND ND 0.44 (0.25) 2.47 (0.13) ND

Longyearbreen S.o. rhizosphere 6.41 (0.11) 10.3 ND 0.32 (0.33) 2.22 (0.08) 2.14 (1.16)

!

!

!34

Cha

pter

2

2a). These differences were mainly driven by the higher richness values in the O. digyna rhizosphere soils (Supplemental Figure S1). In contrast to the diversity indices, the T-RFLP based bacterial community structures differed significantly per geographical location, i.e. Longyearbyen versus Kilpisjärvi (one-way ANOSIM; P<0.001, R = 0.79), with those at the three Kilpisjärvi sites being similar to each other. NMDS plots confirmed the differences between the two geographical locations, as clear clusterings of the bacterial patterns were found according to geographical location (Figure 3).

Figure 2 Box plot of Shannon diversity indices of (A) 16s rRNA based bacterial community structures and (B) nitrogen fixing bacterial communities based on nifH gene calculated using DIVERSE (PRIMER 6) with sampling sites on the category axis. Figure legends: Soil type: B - Bulk soil sample, O.d - O. digyna rhizosphere soil sample, S.o - S. oppositifolia rhizosphere soil sample. Sampling sites: S - Saana, J - Jehkas, Sa - Salluoaivi, LG - Longyearbreen Glacier.

Very interestingly, the presence of a plant as well as the plant species had significant impacts on the bacterial community structures. This was evident from the observation that (1) the bacterial communities in the rhizosphere soils of O. digyna and S. oppositifolia were significantly different from those in the corresponding bulk soils (P <0.01, R = 0.32; one-way ANOSIM), and (2) these were also different from each other. These observations were corroborated by the NMDS ordination, where the samples clustered according to plant species, rather than sampling site, within each geographical location (Figure 3a). Thus, plant roots were significant drivers of the structures of their local bacterial communities.!

!

!35!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

Figure 3 Non-metric Multi-Dimensional Scaling (nMDS) plots of bacterial community structure profiles analysed based on T-RFLP profiles of 16S rRNA gene (Msp I digestion) amplicons (3a) and nifH gene (Mnl I digestion) amplicons (3b). Similarity clustering was based on Bray Curtis dissimilarity matrixes and was performed with complete linkage method in PRIMER 6. Figure legends: Soil type: B - Bulk Soil sample, O.d - O. digyna rhizosphere soil sample, S.o - S. oppositifolia rhizosphere soil. Sampling sites: S - Saana, J - Jehkas, Sa - Salluoaivi, LG - Longyearbreen Glacier

Diazotroph diversity and community structure Given the potential importance of diazotrophs, we assessed these across the soil samples by nifH gene targeted PCR and subsequent T-RFLP analysis. The diversity of the diazotrophic communities at site S, as reflected by the Shannon indices, was significantly higher (P<0.001; two-way ANOVA) than that at the other three sites (Figure 2b). Both species richness and evenness influenced this difference (data not shown). Moreover, the Shannon indices of the S. oppositifolia rhizosphere soils were significantly higher than those of the O. digyna rhizosphere soils (P<0.005, two-way ANOVA) (Figure 2b).

Concerning the diazotroph community structures assessed by T-RFLP, the communities in the LG and S sites each differed clearly from those of sites

!

!36

Cha

pter

2

J and Sa, which were highly similar to each other. Within the sites, the diazotroph communities in the rhizosphere soils clustered along plant species in the NMDS plot, indicating that plant species was a major driver of the root-associated diazotroph communities (Figure 3b). Phenotypic profiles of bacterial communities Parallel to the molecular community fingerprints, we produced phenotypic fingerprints of the microbial communities from the rhizosphere and bulk soil samples of all sampling sites. Phenotypic profiles of the LG soil communities were significantly different from those of the Kilpisjärvi sites S, J and Sa (R = 0.418, p<0.03; one way ANOSIM). These differential responses were confirmed by principal coordinate analysis (PCoA), which revealed the existence of two main clusters on the first axis, one encompassing the majority of the S, Sa and J derived profiles, and the other one the LG derived ones, except one (Figure 4a). This clustering was reminiscent of the clustering based on the T-RFLP based community fingerprints (Figures 3a and 3b). SIMPER analysis revealed that the differences in the community phenotypes between the LG and the Kilpisjärvi sampling sites S, J and Sa were mainly driven by relatively higher tolerances to the antibiotics vancomycin, rifamycin, lincomycin and troleandomycin, as well as acidity (pH 5.0) in the Kilpisjärvi communities.

In contrast, there was a trend towards a more efficient utilization of sugars and organic acids by the LG soil communities. The main substrates that were preferentially utilized by communities from different sites and soil types are listed in the Supplementary Table S1.

Within each geographical location, the phenotypic fingerprints of the rhizosphere soil communities differed significantly (R=0.305, p<0.04 in two-way ANOSIM) from those of the bulk soil communities, as indicated by the tight clustering of most of the rhizosphere soil samples in the PCoA (Figure 4b).

This clustering of rhizosphere soils was mainly driven by their quite effective utilization of D-serine and D-arabitol, and by their higher tolerance to (potassium) tellurite, guanidine-HCl and fusidic acid, as well as to (sodium) bromate, surfactant Niaproof 4 and troleandomycin (Figure 4b). The main substrates utilized by the bulk and rhizosphere communities from all sampling sites are listed in Supplementary Table S1. No statistically significant differences were found between the phenotypic profiles of the bacterial communities from the two plant species. Furthermore, except for the clearly different LG site, no effect of sampling site was detectable (using ordination or statistical testing).

!

!37!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

Figure 4 (A) PCoA ordination of metabolic fingerprints of soil microbial communities from four sampling sites in Longyearbyen and Kilpisjärvi based on Bray Curtis similarity matrix. The vectors indicate the major substrates (correlation cut-off 0.8) driving the separation between the communities. (B) PCoA ordination of metabolic fingerprints of rhizosphere and bulk soil microbial communities from three sampling sites in Kilpisjärvi based on Bray Curtis similarity matrix. The vectors indicate the major substrates (correlation cut-off 0.8) driving the separation between the communities. Figure legends : Soil type: B - Bulk Soil sample, O.d - O. digyna rhizosphere soil sample, S.o - S. oppositifolia rhizosphere soil. Sampling sites: S - Saana, J - Jehkas, Sa - Salluoaivi, LG – Longyearbreen Glacier.

Discussion On the basis of the 16S rRNA and nifH gene based T-RFLP community profiles, as well as the functional diversity measurements, we here showed that the soil bacterial communities from the Longyearbyen sampling site were substantially different from those of the Kilpisjärvi soils. Whether such differences are attributable to sampling site, geographical (climate) or edaphic factors, is a matter of debate. We found significant variation in the soil

!

!38

Cha

pter

2

physicochemical characteristics between the Kilpisjärvi sampling sites S, Sa and J, and the Longyearbyen LG site, in particular with respect to the prevailing pH and the soil OM content. This higher OM content may be attributed to the long-term deposition of organic compounds by melt water from the Longyearbreen Glacier, coupled to the low decomposition rates in the high Arctic (Tockner et al. 2002). The fact that the NH4

+-N levels were higher in the Longyearbyen bulk soils than in the Kilpisjärvi ones, with the opposite being true for the NO3

--N levels, pointed to a lower nitrification rate in the high compared to the low Arctic, as reported by Alves et al. (2013). In general, arctic soils can have relatively low NO3

--N contents (Chu et al. 2010), which is consistent with the values in the bulk soil samples studied by us. In contrast, we detected significantly elevated NH4

+-N as well as NO3--N levels (next to

higher OM content; Table 2) in the Kilpisjärvi rhizosphere soils. These raised N levels may be a result of an enhanced release of organic matter from the plant, indicating higher degradation and nutrient cycling in these otherwise nutrient poor soils, as also reported by others (Tscherko et al. 2005; Deiglmayr et al. 2006; Brankatschk et al. 2011).

Remarkably, within each geographical location (Longyearbyen and Kilpisjärvi), the factor ‘plant’, as well as plant species (O. digyna versus S. oppositifolia) were drivers of the bacterial community structures. This was true for the 16S rRNA gene based (covering all bacteriota) as well as the nifH-based analyses (covering the diazotrophs). The O. digyna and S. oppositifolia plants in this study were sampled close to each other (20-100 cm apart), and were thus subjected to the same climatic and, presumably, soil conditions. Potentially, differences in litter and root exudates produced, in addition to mycorrhizal status, may have indicated the occurrence of unique, plant species-specific bacterial communities. Rhizosphere bacterial communities have been previously shown to be plant-specific in agricultural soils (Berg and Smalla, 2009; Hardoim et al. 2011; Smalla et al. 2001; Westover et al. 1997). In the Arctic, Ferrera-Rodríguez et al. (2013) detected plant species-specific bacterial communities in the rhizospheres of six plant species, including O. digyna, and the endophytic communities of three plant species sampled in Kilpisjärvi have previously been shown to be plant species-specific (Nissinen et al. 2012). The impact of plant exudates and litter is supposed to be particularly strong in the low-to-moderate nutrient content soils targeted in this study. In Antarctica, Teixeira et al. (2010) reported no link between plant species and rhizosphere community structure in their study addressing the two antarctic vascular plant species Deschampsia antarctica and Colobanthus quitensis, whereas another study detected the impact of vegetation on soil microbial communities also in Antarctica (Yergeau et al. 2007).

!

!39!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

The LG sampling site differed from the three other sites (S, Sa and J) with regard to bacterial community structures (Figure 3a, Figure 3b), and also in soil pH, mineral N content and soil organic content, all known to be important in structuring soil bacterial communities. In particular, pH has been suggested to be a main factor determining diversity and structure of soil microbial communities. (Chu et al. 2010; Fierer and Jackson 2006; Griffiths et al. 2011; Lauber et al. 2009; Männistö et al. 2007), with an overriding effect at moderate acidity (pH < 5.1). Thus the different pH levels between Kilpisjärvi and Longyearbyen soils may have driven the bacterial communities differently between the two locations. Corroborating this conclusion was the finding that the LG site soil communities were more sensitive to pH 5.0 than those from the Kilpisjärvi sampling sites in phenotypic profiling (Figure 4a, Supplemental Table S1).

Phenotypic arrays have been used in soil functional analyses for over 20 years now, and the method, although revealing aspects of the functional potential, has caveats as well. Such caveats revolve around the fact that potential rather than actual activities are measured, and that such activities most often come about as the result of the activation of only a small subpopulation of the whole community (Smalla et al. 1998). Moreover, any method used to release bacterial cells from soil particles inevitably adds another factor of uncertainty (bias) to this method. Thus, data from phenotypic arrays need to be interpreted with utmost care. Our data appeared to confirm that the communities from the three Kilpisjärvi sampling sites S, J and Sa grouped away from those from the Longyearbyen sampling site LG (Figure 4a). This was mainly driven by significantly higher antibiotic tolerances in the soil communities from the Kilpisjärvi sampling sites (Figure 4a). Further, the phenotypic profiles of O. digyna and S. oppositfolia rhizosphere communities clustered separately from those of bulk soil communities. This latter finding is consistent with data from Tam et al. (2001), who also found that the communities from rhizosphere and bulk soil samples from several plant species grouped separately in ordinations of phenotype profiles. Bragazza et al. (2012) and Grayston et al. (1998) also showed that plants can affect the functioning of soil microbial communities. Thus, plants create specific ecological niches in soil that select a particular combination of bacterial species (Garbeva et al. 2004; Yergeau et al. 2007) and functional groups.

The tolerances to fusidic acid, troleandomycin and to Niaproof 4 were higher in the rhizosphere than in the bulk soils across all sampling sites (data in supplemental Table S1). This finding might reflect a selection for a higher ability to withstand these compounds in the plant root vicinity in competitive situations. Many rhizobacteria secrete antibiotics and surfactants, that act

!

!40

Cha

pter

2

against other bacteria and fungi, and so a selection for tolerance to such compounds likely occurs in the rhizosphere (Debois et al. 2014). The higher antibiotic tolerance in the O. digyna versus the S. oppositifolia rhizosphere might reflect the differences in the mycorrhizal status of these plant species: as a non-mycorrhizal species, O. digyna would benefit from rhizosphere bacterial communities with ability to efficiently compete with possibly harmful soil fungi, whereas ectomycorrhizal S. oppositifolia would mainly rely on its fungal partner.

Resistance to (potassium) tellurite was clearly higher in the rhizosphere soils than in the corresponding bulk soils (Figure 4b). Tellurite, an oxyanion of tellurium, is highly toxic to most life forms, as it damages cells via generation of reactive oxygen species (ROS) and disturbance of the thiol redox balance (Arenas et al. 2014). Resistance to tellurite is rare in nature, and is found mostly in corynebacteria (Summers and Jacoby 1977; Taylor 1999), but it has also been found in strains of Firmicutes and Proteobacteria isolated from Antarctica and other extreme environents (Arenas et al. 2014). Such tellurite-resistant strains commonly tolerate oxidative stress. Thus, the relatively high tellurite resistance in the rhizosphere could reflect the exposure of such rhizosphere communities to ROS, the primary plant defense reaction towards microbes. The rhizosphere communities also showed an increased tolerance to the strong electron acceptor bromate, to the chaotropic agent guanidine-HCl and to the surfactant Niaproof 4, in comparison to bulk soil communities. These all could reflect adjustments to conditions that reign in the vicinity of active plant roots.

The rhizosphere soils further clustered by their effective utilization of D-serine and D-arabitol (Figure 4b, supplemental table S1). Plant root exudates typically contain sugars, organic acids and sugar alcohols, which can be plant-specific (Hardoim et al 2008; Compant et al. 2010; Haichar et al. 2008). We hypothesized that D-arabitol and D-serine might shape the rhizosphere communities of O. digyna and S. oppositifolia by creating a specific nutritional niche. However, alternative explanations are also possible. In particular D-serine is a rare amino acid which is not commonly metabolized by (soil) bacteria, and is inhibitory to the growth of most plants (Vranova et al. 2011). However, it has been reported to be present in substantial quantities in a few plant species, including rice (Gogami et al. 2009). Fungal pathogens with high virulence towards rice are able to utilize D-serine, suggesting nutritional adaptation to the host in these fungi (Van Buyten and Höfte 2013). Thus, we could speculate, that the ability to metabolize D-serine could be a strong advantage for rhizosphere bacteria of D-serine-harboring plants.

!

!41!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

The current study broadens our understanding of influence of pioneer plant species on community structure and function of bacterial communities in low organic matter arctic soils. We clearly identified the structuring of soil bacteria by these two plants, but also by the geographical site. The antibiotic resistance profiles were associated with the location the communities were derived from, whereas the abilities to tolerate oxidative stress and specific antibiotics were associated with the rhizosphere communities. Acknowledgements We thank NWO for financing this study with grant OND1344812.We wish to thank Koen Frans Verweij as well as the Kilpisjärvi Biological Station of the University of Helsinki for assistance in sampling, Joana Falcao Salles for helpful suggestions concerning the statistical analyses and Nelly Eck for her assistance in soil chemical analyses. We thank the anonymous reviewers for their comments, questions and suggestions, which greatly helped to improve our manuscript. References

Alves RJE, Wanek W, Zappe A, Richter A, Svenning MM, Schleper C, Urich T (2013) Nitrification rates in Arctic soils are associated with functionally distinct populations of ammonia-oxidizing archaea. The ISME journal 1–12.

Anderson J, Domsch K (1978) A Physiological method for the quantitative measurement of microbial biomass in soils. Soil biology and biochemistry 10:215–221.

Arenas FA, Pugin B, Henríquez NA, Arenas-Salinas MA, Díaz-Vásquez WA, Pozo MF, Muñoz CM, Chasteen TG, Pérez-Donoso JM, Vásquez CC (2014) Isolation, identification and characterization of highly tellurite-resistant, tellurite-reducing bacteria from Antarctica. Polar Science 8:40–52.

Badri D V, Weir TL, van der Lelie D, Vivanco JM (2009) Rhizosphere chemical dialogues: plant-microbe interactions. Current opinion in biotechnology 20:642–50.

Berg G, Smalla K (2009) Plant species and soil type cooperatively shape the structure and function of microbial communities in the rhizosphere. FEMS microbiology ecology 68:1–13.

Bragazza L, Parisod J, Buttler A, Bardgett RD (2012) Biogeochemical plant–soil microbe feedback in response to climate warming in peatlands. Nature Climate Change 3:1–5.

Brankatschk R, Töwe S, Kleineidam K, Schloter M, Zeyer J (2011) Abundances and potential activities of nitrogen cycling microbial communities along a chronosequence of a glacier forefield. The ISME journal 5:1025–37.

Brockett BFT, Prescott CE, Grayston SJ (2012) Soil moisture is the major factor influencing microbial community structure and enzyme activities across seven biogeoclimatic zones in western Canada. Soil Biology and Biochemistry 44:9–20.

Campbell CD, Grayston SJ, Hirst DJ (1997) Use of rhizosphere carbon sources in sole carbon source tests to discriminate soil microbial communities. Journal of Microbiological Methods 30:33–41.

!

!42

Cha

pter

2

Cesco S, Neumann G, Tomasi N, Pinton R, Weisskopf L (2010) Release of plant-borne flavonoids into the rhizosphere and their role in plant nutrition. Plant and Soil 329:1–25.

Chaparro JM, Badri D V, Vivanco JM (2014) Rhizosphere microbiome assemblage is affected by plant development. The ISME journal 8:790–803.

Chapin F (1983) Direct and indirect effects of temperature on arctic plants. Polar Biology 2:47–52.

Chapin F, Bloom A, Field C, Waring R (1987) Plant responses to multiple environmental factors. Bioscience 37:49–57.

Chu H, Fierer N, Lauber CL, Caporaso JG, Knight R, Grogan P(2010) Soil bacterial diversity in the Arctic is not fundamentally different from that found in other biomes. Environmental microbiology 12:2998–3006.

Clarke K (1993) Non-parametric multivariate analyses of changes in community structure. Australian journal of ecology 18:117–143.

Clarke KR, Gorley RN (2006) PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth. Compant S, Clément C, Sessitsch A (2010) Plant growth-promoting bacteria in the rhizo-

and endosphere of plants: Their role, colonization, mechanisms involved and prospects for utilization. Soil Biology and Biochemistry 42:669–678.

Debois D, Jourdan E, Smargiasso N, Thonart P, De Pauw E, Ongena M (2014) Spatiotemporal Monitoring of the Antibiome Secreted by Bacillus Biofilms on Plant Roots Using MALDI Mass Spectrometry Imaging. Analytical Chemistry 86:4431–4438.

Deiglmayr K, Philippot L, Tscherko D, Kandeler E (2006) Microbial succession of nitrate-reducing bacteria in the rhizosphere of Poa alpina across a glacier foreland in the Central Alps. Environmental microbiology 8:1600–12.

Derry A, Staddon W, Kevan P and Trevors J (1999) Functional diversity and community structure of micro-organisms in three arctic soils as determined by sole-carbon-source-utilization. Biodiversity and Conservation 8, 205-221

Duineveld B, Kowalchuk G, Keijzer A, van Elsas JD, van Veen JA (2001) Analysis of Bacterial communities in the rhizosphere of chrysanthemum via denaturing gradient gel electrophoresis of PCR-amplified 16S rRNA as well as DNA fragments coding. Applied and Environmental Microbiology 67:172–178.

Duineveld B, Rosado A, van Elsas JD, van Veen JA (1998) Analysis of the dynamics of bacterial communities in the rhizosphere of the chrysanthemum via denaturing gradient gel electrophoresis and substrate utilization. Applied and Environmental Microbiology 64:4950–4957.

Dunbar J, Ticknor L, Kuske C (2001) Phylogenetic specificity and reproducibility and new method for analysis of terminal restriction fragment profiles of 16S rRNA genes from bacterial communities. Applied and Environmental Microbiology 67:190–197.

Ferrera-Rodríguez O, Greer CW, Juck D, Consaul LL, Romero EM, Whyte LG (2013) Hydrocarbon-degrading potential of microbial communities from Arctic plants. Journal of applied microbiology 114:71–83.

Fierer N, Jackson RB (2006) The diversity and biogeography of soil bacterial communities. Proceedings of the National Academy of Sciences of the United States of America 103:626–31.

Fierer N, Ladau J (2012) Predicting microbial distributions in space and time. Nature methods 9:549–51.

Garbeva P, van Veen J a, van Elsas JD (2004) Microbial diversity in soil: selection microbial populations by plant and soil type and implications for disease suppressiveness. Annual review of phytopathology 42:243–70.

!

!43!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

Gogami Y, Ito K, Kamitani Y, Matsushima Y, Oikawa T (2009) Occurrence of D-serine in rice and characterization of rice serine racemase. Phytochemistry 70:380–7.

Grayston SJ, Wang S, Campbell CD, Edwards AC (1998) Selective influence of plant species on microbial diversity in the rhizosphere. Soil Biology and Biochemistry 30:369–378.

Griffiths RI, Thomson BC, James P, Bell T, Bailey M, Whiteley AS (2011) The bacterial biogeography of British soils. Environmental microbiology 13:1642–54.

Haichar FEZ, Marol C, Berge O, Rangel-Castro JI, Prosser JI, Balesdent J, Heulin T, Achouak W, (2008) Plant host habitat and root exudates shape soil bacterial community structure. The ISME journal 2:1221–30.

Hardoim PR, Andreote FD, Reinhold-Hurek B, Sessitsch A, van Overbeek LS, van Elsas JD (2011) Rice root-associated bacteria: insights into community structures across 10 cultivars. FEMS microbiology ecology 77:154–64.

Hardoim PR, van Overbeek LS, van Elsas JD (2008) Properties of bacterial endophytes and their proposed role in plant growth. Trends in microbiology 16:463–71.

Knelman JE, Legg TM, O’Neill SP, Washenberger CL, González A, Cleveland CC, Nemergut DR (2012) Bacterial community structure and function change in association with colonizer plants during early primary succession in a glacier forefield. Soil Biology and Biochemistry 46:172–180.

Körner C (2003) Alpine Plant Life - Functional Plant Ecology of High Mountain Ecosystems. Springer-Verlag Berlin Heidelberg.

Kume A, Nakatsubo T, Bekku Y, Masuzawa T (1999) Ecological significance of different growth forms of purple saxifrage, Saxifraga oppositifolia L., in the High Arctic, Ny-Ålesund, Svalbard. Arctic, Antarctic, and Alpine Research 31:27–33.

Kuramae EE, Yergeau E, Wong LC, Pijl AS, van Veen JA, Kowalchuk GA (2012) Soil characteristics more strongly influence soil bacterial communities than land-use type. FEMS microbiology ecology 79:12–24.

Lane DS, Stackebrandt E, Goodfellow M (1991) 16S/23S rRNA sequencing. In: Stackebrandt E And Goodfellow M (ed) Nucleic Acid Techniques in Bacterial Systematics. NY, John Wiley and Sons, pp 115–175

Lauber CL, Hamady M, Knight R, Fierer N (2009) Pyrosequencing-based assessment of soil pH as a predictor of soil bacterial community structure at the continental scale. Applied and Environmental Microbiology 75:5111–20.

Lucas JA, García-Villaraco A, Ramos B, García-Cristobal J, Algar E, Gutierrez-Mañero J (2013) Structural and functional study in the rhizosphere of Oryza sativa L. plants growing under biotic and abiotic stress. Journal of applied microbiology 115:218–35.

Männistö MK, Tiirola M, Häggblom MM (2007) Bacterial communities in Arctic fjelds of Finnish Lapland are stable but highly pH-dependent. FEMS Microbiology Ecology 59:452–465.

Margesin R, Neuner G, Storey KB (2007) Cold-loving microbes, plants, and animals--fundamental and applied aspects. Die Naturwissenschaften 94:77–99.

Marr KL, Allen GA, Hebda RJ (2008) Refugia in the Cordilleran ice sheet of western North America: chloroplast DNA diversity in the Arctic-alpine plant Oxyria digyna. Journal of Biogeography 35:1323–1334.

Miniaci C, Bunge M, Duc L, Edwards I, Bürgmann H, Zeyer J (2007) Effects of pioneering plants on microbial structures and functions in a glacier forefield. Biology and Fertility of Soils 44:289–297.

!

!44

Cha

pter

2

Nissinen RM, Männistö MK, van Elsas JD (2012) Endophytic bacterial communities in three arctic plants from low arctic fell tundra are cold-adapted and host-plant specific. FEMS microbiology ecology 82:510–22.

Pii Y, Mimmo T, Tomasi N, Terzano R, Cesco S, Crecchio C (2015a) Microbial interactions in the rhizosphere: beneficial influences of plant growth-promoting rhizobacteria on nutrient acquisition process. A review. Biology and Fertility of Soils 403–415.

Pii Y, Penn A, Terzano R, Crecchio C, Mimmo T, Cesco S (2015b) Plant-microorganism-soil interactions influence the Fe availability in the rhizosphere of cucumber plants. Plant Physiology and Biochemistry 87:45–52.

Poly F, Monrozier LJ, Bally R (2001) Improvement in the RFLP procedure for studying the diversity of nifH genes in communities of nitrogen fixers in soil. Research in Microbiology 152:95–103.

Robbins J, Matthews J (2009) Pioneer vegetation on glacier forelands in southern Norway: emerging communities? Journal of Vegetation Science 20:889–902.

Schimel JP, Bilbrough C, Welker JM (2004) Increased snow depth affects microbial activity and nitrogen mineralization in two Arctic tundra communities. Soil Biology and Biochemistry 36:217–227.

Sessitsch A, Howieson J, Perret X, Antoun H, Martinez-Romero E (2002) Advances in Rhizobium research. Critical Reviews in Plant Sciences 21:323–378.

Singh D, Takahashi K, Kim M, Chun J, Adams JA (2012) A hump-backed trend in bacterial diversity with elevation on Mount Fuji, Japan. Microbial ecology 63:429–37.

Smalla K, Wachtendorf UTE, Heuer H, Liu WT, Forney L (1998) Analysis of BIOLOG GN Substrate Utilization Patterns by Microbial Communities. Applied and Environmental Microbiology 64:1220–1225.

Smalla K, Wieland G, Buchner A, Zock A, Parzy J, Kaiser S, Roskot N, Heuer H, Berg G (2001) Bulk and rhizosphere soil bacterial communities studied by denaturing gradient gel electrophoresis: plant-dependent enrichment and seasonal shifts revealed. Applied and Environmental Microbiology 67:4742–4751.

Smallwood M, Bowles DJ (2002) Plants in a cold climate. Philosophical transactions of the Royal Society of London Series B, Biological sciences 357:831–47.

Stenström M, Molau U (1992) Reproductive ecology of Saxifraga oppositifolia: phenology, mating system, and reproductive success. Arctic and Alpine Research 24:337–343.

Summers AO, Jacoby GA (1977) Plasmid-determined resistance to tellurium compounds. Journal of bacteriology 129:276–81.

Tam L, Derry A, Kevan P, Trevors J (2001) Functional diversity and community structure of microorganisms in rhizosphere and non-rhizosphere Canadian arctic soils. Biodiversity and Conservation 10:1933–1947.

Taylor DE (1999) Bacterial tellurite resistance. Trends in microbiology 7:111–5. Teixeira LCRS, Peixoto RS, Cury JC, Sul WJ, Pellizari VH, Tiedje J, Rosado AS (2010)

Bacterial diversity in rhizosphere soil from Antarctic vascular plants of Admiralty Bay, maritime Antarctica. The ISME journal 4:989–1001.

Tockner K, Malard F, Uehlinger U, Ward J (2002) Nutrients and organic matter in a glacial river-floodplain system (Val Roseg, Switzerland). Limnology and Oceanography 47:266–277.

Tscherko D, Hammesfahr U, Zeltner G, et al (2005) Plant succession and rhizosphere microbial communities in a recently deglaciated alpine terrain. Basic and Applied Ecology 6:367–383.

!

!45!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

Urbanová M, "najdr J, Baldrian P (2015) Composition of fungal and bacterial communities in forest litter and soil is largely determined by dominant trees. Soil Biology and Biochemistry 84:53–64.

Van Buyten E, Höfte M (2013) Pythium species from rice roots differ in virulence, host colonization and nutritional profile. BMC plant biology 13:203.

Vranova V, Zahradnickova H, Janous D, Skene KR, Matharu AS, Rejsek K, Formanek P (2011) The significance of D-amino acids in soil, fate and utilization by microbes and plants: review and identification of knowledge gaps. Plant and Soil 354:21–39.

Walker M, Wahren C (2006) Plant community responses to experimental warming across the tundra biome. Proceedings of the National Academy of Sciences of the United States of America 103:1342–1346.

Wallenstein MD, Mcmahon SK, Schimel JP (2009) Seasonal variation in enzyme activities and temperature sensitivities in Arctic tundra soils. Global Change Biology 15:1631–1639. doi: 10.1111/j.1365-2486.2008.01819.x

Westover KM, Kennedy AC, Kelley SE (1997) Patterns of rhizosphere microbial community structure associated with co-occurring plant species. Journal of Ecology 85:863–873.

Yergeau E, Newsham KK, Pearce DA, Kowalchuk GA (2007) Patterns of bacterial diversity across a range of Antarctic terrestrial habitats. Environmental microbiology 9:2670–82.

Zhang HW, Song YC, Tan RX (2006) Biology and chemistry of endophytes. Natural product reports 23:753–71. doi: 10.1039/b609472b

Zinger L, Shahnavaz B, Baptist F, Geremia RA, Choler P (2009) Microbial diversity in alpine tundra soils correlates with snow cover dynamics. The ISME journal 3:850–9.

!

!46

Cha

pter

2

Supplementary Text 1: Formulae Formulae for calculation of NH4

+-N, PO4, and NO3--N measurements

S o l u b l e NH4

+-N and NO3--N m e a s u r e m e n t s

((Nitrate/Ammonium ppm-Blank ppm)*(Amount of 1M KCl + (Soil weight*Soil Moisture content)) * Dilution) / (Soil weight*Soil Moisture content) = Soluble Nitrate mg N/kg soil S o l u b l e p h o s p h a t e m e a s u r e m e n t s (O.D sample – O.D blank)* Factor * 1000 = Soluble Phosphate / 1 cm3 dry soil Factor is obtained from Calibration l ine using standard solution measurements.

Supplementary Figure S1. Box plot of species richness of 16s rRNA based bacterial community structure calculated using DIVERSE (PRIMER 6) with sampling sites on the category axis. Figure legends: Soil type: B - Bulk soil sample, O.d - O. digyna rhizosphere soil sample, S.o - S. oppositifolia rhizosphere soil sample. Sampling sites: S - Saana, J - Jehkas, Sa - Salluoaivi, LG - Longyearbreen Glacier.

!

!47!

Arc

tic P

lant

s Im

pact

Str

uctu

re a

nd F

unct

ion

of B

acte

ria

Supplementary Table S1: Vegetation list from the sampling sites

Saana Jehkas Sallooaivi Longyearbyen Higher plants Arabis alpinum 0 1 0 0 Carex sp. 1 0 0 0 Cassiope tetragone 1 1 0 0 Cerastium arcticum 0 0 0 1 Dryas octopetella 0 0 1 0 Harrimanella hypnoides (Cassiope hypnoides) 1 1 1 0 Juncus sp. 1 0 0 0 Luxula confuza 0 0 0 1 Oxyria digyna 1 1 1 1 Papaver dahlianum 0 0 0 1 Poaceae spp. 1 0 1 0 Polygonym viveparum (Bistorta vivepara) 1 1 1 1 Ranunculus glacialis 1 0 0 0 Ranunculus nivalis 1 1 1 0 Ranunculus pygmea 1 1 0 0 Salix herbecea 0 1 1 0 Salix polaris 0 0 1 1 Saxifraga cernua 0 0 0 1 Saxifraga cespitosa 0 0 0 1 Saxifraga hieracifolia 0 0 0 1 Saxifraga oppositifolia 1 1 1 1 Saxifraga tenuis 0 1 1 0 Sibbaldia procumbens 0 1 0 0 Siline aucalis 0 1 1 0 Stellaria longipes 0 0 0 1 Lichen Cetraria sp. 1 0 1 0 Stereocaulon sp. 1 0 1 0 Rhizocarpum geographicum 1 1 0 1 Cladonia sp. 1 1 1 0 Moss Polytrichum sp. 1 1 0 1

1 = Presence, 0 = Absence.

 

 48

 

 

 

 

 

 

   

!

!

!!

Chapter 3

Plants assemble species specific bacterial communities from common core taxa in three arcto-alpine climate zones

Manoj Kumar, Günter Brader, Angela Sessitsch, Anita Mäki, Jan Dirk van Elsas, Riitta M. Nissinen

!50

Cha

pter

3

Abstract Evidence for the pivotal role of plant-associated bacteria to plant health and productivity has accumulated rapidly in the last years. However, several key questions remain unanswered, among which is the impact of climate zones on the plant-associated microbiota. Importantly, plant-bacterial interactions in wild plants and in arcto-alpine biomes have not been well studied as compared to those in tropical and temperate biomes. We hypothesized that the bacterial communities associated with pioneer plants in the former regions have major roles in plant health support, and this would be reflected in the formation of climate and host plant specific endophytic communities. We also hypothesized that these endophytes would have a tight association with their respective host plants. To examine the drivers of the plant-associated bacterial communities, we compared the latter in the perennial plants Oxyria digyna and Saxifraga oppositifolia - naturally growing in three arcto-alpine regions [Mayrhofen, Austria - alpine, Kilpisjärvi, Finland - low Arctic and Ny-Ålesund, Svalbard -highArctic] - with those in the corresponding bulk soils. As expected, the bulk soil bacterial communities in the three regions were significantly different. The relative abundances of Proteobacteria decreased progressively from the alpine to the high-arctic soils, whereas those of Actinobacteria increased. The candidate division AD3 and Acidobacteria abounded in the low Arctic.

With respect to the endophytic bacterial communities, plant species and geographic region were the major determinants of the community structures. The plants in the alpine region had higher relative abundances of Proteobacteria, as compared to the plants from the low- and high-arctic regions which were dominated by Firmicutes. A highly conserved shared set of omnipresent bacterial taxa was found to occur in the two plant species, here denoted as the ‘tight’ core bacteriome. Burkholderiales, Actinomycetales and Rhizobiales were common members of this core, and these taxa were the main contributors to the differences in the endophytic bacterial community structures across compartments as well as regions. We postulate that this tight core is due to selection by the two plants, concomitant with a highly efficient adaptation of such “core bacteriome” to the plant-associated habitats. Keywords: endophytes, arcto-alpine plant, bacteria, biogeographical diversity, core bacteriome, Oxyria digyna, Saxifraga oppositifolia.

!!

51

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

Introduction Among the terrestrial environments on Earth, arctic and alpine ecosystems cover about 8% of the global land area, which is more than the area covered by tropical forests (Chapin and Körner, 1996). The arctic and alpine ecosystems have the least biologically usable heat and the lowest diversity of plants (Billings and Mooney, 1968). The plants in these systems are well adapted to cold and short growing seasons and low-nutrient soils. The typical plant species occurring in both biomes are collectively referred to as arcto-alpine vegetation. These plants are important in these soils, as they constitute the prime settlers that are at the basis of a living ecosystem. It has been hypothesized that the local microbiota plays an important role in the ecological success of these pioneering plants (Borin et al., 2010; Mapelli et al., 2011). While arctic and alpine biomes share many characteristics shaping vegetation, including short and cool growing seasons, cold winters and soils with low levels of nutrients, there are also distinct differences: the alpine tundra is characterized by high annual and diurnal temperature fluctuation and high solar intensity in the summer and, in general, well-drained soils. The vegetation in the Arctic, on the other hand, experiences long polar nights in winter versus 24-hour daylight during the growing season, and soils are typically water-logged due to underlying permafrost (Körner, 2003). These differences have led to climatic ecotypes within arcto-alpine vegetations, where growth morphology and phenology of the same plant species in different biomes reflect adaptation to distinct climates.

Endophytic bacteria are ubiquitous across both cultivated and wild plants, and have been shown to contribute to all major aspects of plant life, including regulation of growth and development, nutrient acquisition and protection from biotic and abiotic stressors (reviewed in Hardoim et al., 2008, 2015). Studies conducted mainly in agricultural or in model plant systems have in recent years offered a rapidly growing insight into the assembly, structure and function of the endophytic communities in plants (Compant et al., 2010; Zhang et al., 2006). Soil type and plant species or genotype are both known to shape the rhizosphere (Berg and Smalla, 2009; Garbeva et al., 2004) and root endosphere communities (Bulgarelli et al., 2013). Rhizosphere soil is considered to be the main source of endophytes (Bulgarelli et al., 2013), but vertical transmission via seeds have also been reported (Puente et al., 2009; Hardoim et al., 2012). However, the factors governing the plant-associated microbiota of perennial wild plants in demanding habitats and low-nutrient soils may differ from those of model or well-fertilized crop plants. For plants in the low-arctic fell tundra, we have previously shown that plant species, rather than sampling site, determines the structure of the endophytic (Nissinen et al., 2012) and rhizospheric (Kumar et al., 2016) microbial communities.

!52

Cha

pter

3

Most bacterial species are considered to be cosmopolitan, as they have been found across biogeographic regions inhabiting diverse habitats like soils, sediments, lakes and the sea (Hanson et al., 2012). Interestingly, the bacterial diversity levels in Arctic soils has been shown not to differ from that of other biomes (Chu et al., 2010). However, endemism is also observed in bacteria, with some taxa reportedly being restricted to distinct geographical regions (Cho, 2000; Oakley et al., 2010).

The main goal of this study was to investigate the factors shaping the bacterial communities associated with two plant species in several sites representing three geographic regions and climates, from the high Arctic to the Alps. Our target plant species, Oxyria digyna and Saxifraga oppositifolia, are arcto-alpine plant species with wide distribution from the high Arctic to the mid-latitude alpine tundra. Both species are typical pioneer species that are capable of efficiently colonizing low-nutrient tundra soils. We focused on the endophytic bacteria, and also examined the bacterial communities in the relevant rhizosphere and bulk soil samples.

We hypothesized that (1) geographic region, related to climate zone, determines the diversity and community structure of soil bacterial communities in the selected habitats, and (2) plants strongly shape the plant-associated microbial communities, resulting in plant species specific bacterial communities, regardless of the geographic region. We also hypothesized that (3) the plant-associated bacteria are tightly associated with their hosts, as indicated by their persistent presence (core microbiome) and their lifestyle (enrichment of these bacteria in the plant endophytic compartment).

To achieve our aims, we used community DNA based amplicon sequencing targeting the 16S rRNA gene region and subsequent analyses. Materials and Methods Sampling locations and study sites Plant and soil samples were collected from eight sampling sites in three distinct regions representing different climate zones; Ny-Ålesund, high Arctic (3 sampling sites), Kilpisjärvi, low Arctic (3 sampling sites) and Mayrhofen, European Alps (2 sampling sites) (Figure 1). Kilpisjärvi is at the northwestern Finland and located along the Fenno-Scandinavian border. Its flora is dominated by mountain birch forest in the valleys and by fell tundra at higher elevations. The annual mean temperature is about -2.2°C with plant growth season of ca. 90-100 days. Ny-Ålesund (Svalbard, Norway) is located on an isolated archipelago in the high Arctic; the land cover is dominated by glaciers and permafrost layers, and the mean annual temperature is -4°C. The soil temperatures have been reported to be below zero for more than 250 days per

!!

53

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

year ranging from -6°C to -25°C (Coulson and Hodkinson, 1995). Most biological activity is restricted to less than 10% of the total land mass coupled with about three months of plant growing season. The sampling location in the Mayrhofen is located above the tree line south of Mayrhofen over the snow-covered mountains (altitude ca. 2400 m above sea level) in the alpine tundra of the European Alps.

Figure 1 Map of Europe depicting our three sampling locations Mayrhofen from Austrian Alps, Kilpisjärvi from low-arctic Finnish Lapland and Ny-Ålesund from high-arctic Svalbard archipelago

!Sample collection and processing 12 replicates of bulk soil samples (5-15 cm depth) and six samples of both O. digyna and S. oppositifolia (as whole plants with adhering rhizosphere soils) were collected from all sites, except in site “Saana” (Kilpisjärvi) where only O.digyna plants were sampled and site “Cliff” (Mayrhofen) where we sampled only 6 bulk soil samples. Sampling was performed during summer 2012. All harvested plants were flowering at the time of the sampling. Rhizosphere and bulk soil samples were processed and stored as specified by Kumar et al. (2016). After removing rhizosphere soils, plant roots were thoroughly washed with water and surface sterilized by immersing the plant material into 3% sodium hypochlorite for 3 minutes and then subsequently in sterile double distilled water (3 x 90 s). 80-100 mg of root samples were weighed, snap frozen with liquid nitrogen and stored at -80˚C for further DNA analysis.

Soil pH and soil organic matter (SOM) content were measured as described in Kumar et al. (2016), while available phosphorous (P) was measured based on Bray No 1 extraction method (Bray and Kurtz, 1945). All

!54

Cha

pter

3

the soil chemical analyses were performed in duplicates (2 technical replicates) per sample, and with 4-8 biological replicates per site and sample type (Table 1). DNA isolation Microbial DNA from soil samples were extracted following manufacturer’s instruction using MoBio Power soil kit (MoBio, Carlsbad, CA USA). For soil samples 0.5 g of soil was used instead of 0.25 g because of low microbial counts in our soils (data not shown). For isolation of endophyte samples, Invisorb Spin Plant Mini Kit (STRATEC Biomedical AG, Germany) was used in order to ensure prolonged stability of endophytic DNA in the plant derived samples. An additional homogenization step of bead beating at 45 s was followed for the endophyte DNA extraction from plant samples. 16S rRNA gene library generation and sequencing After isolating DNA from all six plant replicates from both plant species, we selected the four best replicates (samples with best DNA yields and amplification performance) per sampling site for further analyses. Rhizosphere samples from the same plant samples were also selected along with eight bulk soil samples. 16S rRNA gene was amplified using primers 799f/1492r (Chelius and Triplett, 2001) and M13-1062f/1390r in a nested approach. The nested primers targeting the V6-V8 regions of 16s rRNA gene enable elimination of plant chloroplast 16S rRNA gene amplicons as well as separation of endophyte amplicons from plant microchondrial amplicons by size fractionation (799f-1492r, Chelius and Triplett (2001)) and produce an amplicon with high phylogenetic coverage and optimal size for IonTorrent sequencing (1062f-1390r). Primers 1062f (Ghyselinck et al., 2013)and 1390r (Zheng et al., 1996) were tagged with M13 sequences to enable sample barcoding as described below and in Mäki et al. (2016). Both reactions had 1 !l of sample DNA, 1x PCR buffer, 1 mg/ml of BSA, 0.2 mM dNTP’s, 0.3 !M of each primer and 1250 U/ml GoTaq DNA Polymerase (Promega, WI USA) in a 30!l reaction volume. 5-10 and 25-30 ng of soil and endophyte DNA, respectively, was used in the first PCR, and 1:10 diluted amplicons from the first PCR were used as a template for the second run. Amplifications for both PCR reactions were performed as follows: 3 mins denaturation at 95°C followed by 35 cycles of denaturing, annealing, and extension at 95°C for 45 secs, 54°C for 45 secs and 72°C for 1 min, respectively. Final extension was carried out at 72°C for 5 mins. Four (rhizo- and endosphere) or eight (bulk soil) samples technically best samples (good DNA yield, good PCR amplification) were included in the next generation sequencing library construction.

!!

55

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

Sequence libraries were prepared by running a third PCR to attach the M-13 barcode system developed by Mäki et al. (2016). Amplicons from second PCR were diluted 1:5 and re-amplified using barcode attached M13 system as forward primer and 1390r-P1 with adaptor A as a reverse primer. PCR mix and conditions were similar as described above, with an exception of only using 8 cycles for amplification. Amplified libraries were purified using Agencourt AMPure XP PCR purification system (Beckman Coulter, CA USA). Purified samples were quantified with Qubit Fluorometer (Invitrogen, MA USA) and an equivalent DNA quantity of each sample was pooled together. The pooled samples were then size fractionated (size selection range of 350-550 bp) using Pippin Prep (Sage Science, MA USA) 2% Agarose gel cassette (Marker B) following the manufacturer’s protocol. Size fractioned libraries were sequenced using Ion 314 chip kit V2 BC on Ion Torrent PGM (Life Technologies, CA USA) in Biocenter Oulu, Finland. Bioinformatics and statistical analysis The raw sequence reads were processed using QIIME (Caporaso et al., 2010) and UPARSE (Edgar, 2013) based on a 16S rRNA gene data analysis pipeline developed by Pylro et al. (2014) with slight modifications in quality filtering. Sequences were trimmed by removing sequences with low quality reads (Q score <25) and shorter base pair (<150) length. Furthermore, all the raw reads were trimmed (200 bp), clustered and aligned at 97% identity using USEARCH algorithm (Edgar, 2010). UCLUST algorithm along with Greengenes database (DeSantis et al., 2006) was used to assign taxonomies at 97% identity to the individual OTUs. In total, 426,135 high-quality reads (1468 reads - 5331 reads per sample) were clustered into 985 OTUs. For alpha diversity analysis all the samples were rarefied (subsampled) to 1400 reads per sample. Shannon index and species richness were obtained using Univariate Diversity Indices (DIVERSE, PRIMER 6 (PRIMER-E Ltd)). The differences in diversity indexes between the soil samples and their correlation with soil physico-chemical properties were determined using two-way ANOVA and Pearson correlation (SPSS Statistics, IBM). The significance of the differences between the soil samples were tested by Games-Howell post-hoc tests (two-way ANOVA).

To normalize the data for community structure and other analyses all the samples with more than the median reads were rarefied to the median (2780 reads), while the samples with less reads were used as such, as described in deCárcer et al. (2011). In addition, all the singletons and OTUs with less than 50 reads were removed before processing. The influence of sampling site, geographic region, plant compartment and plant species on bacterial

!56

Cha

pter

3

community structures, based on Bray-Curtis distance matrixes of square root transformed abundance data, were analysed using permutational manova (PERMANOVA) and visualized by PCoA ordinations at the OTU level. Taxonomic groups (phyla or OTU) with strongest impact on differences between community structures were identified with SIMPER (Similarity Percentages - species contributions), all performed with PRIMER 6 software package with PERMANOVA+ add-on (primer-e.com).

All the Ternary plots were made by calculating the mean relative abundances of OTUs per geographic region/compartment and with the function ‘ternaryplot’ ‘vcd’ (Meyer et al., 2015) from the R package. All other graphs (bar and scatter plots) were constructed using the R package ‘graphics’. Picking endosphere core OTUs The highly conserved OTUs (core OTUs) were manually picked by selecting OTUs that were constantly observed (present in at least 3 out of 4 replicates per site) in the endosphere of either O. digyna and S. oppositifolia or both. To determine the distribution of core OTUs reads across different compartments, the averaged read count per compartment was calculated for each OTU, the averages were summed up and presented as the relative distribution of each of the 13 core OTUs per compartment.

Figure 2 Venn diagrams of shared OTUs (number of reads of respective OTUs) across 3 bioclimatic regions (A) and compartments (B)

Results A total of 426,135 quality-filtered sequence reads was retrieved from the total 174 samples in our sample set, representing the endospheres and rhizospheres of the two plant species and the associated bulk soil samples from the three geographic regions (Figure 1). The sequences were separated into 985 OTUs (defined at the 97% cut-off level) and subjected to downstream analyses. Of

!!

57

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

these, 933 OTUs were present in at least one of the samples from each of the three regions, 43 in two regions and nine were restricted to one region only. 778 of the 985 OTUs were found in all compartments (bulk soil, rhizosphere soil or endosphere), 190 in two compartments and 17 were compartment-specific (Figure 2). Soil characteristics are different across three arcto-alpine regions Table 1 lists the soil characteristics in the three geographic regions: Mayrhofen (alpine), Kilpisjärvi (low-arctic) and Ny-Ålesund (high-arctic) (Figure 1). The Ny-Ålesund [bulk] soils had significantly higher pH (two-way ANOVA, p<0.05) and soil organic matter (SOM) values (two-way ANOVA, p<0.01), and significantly lower levels of available phosphorus (two-way ANOVA, p<0.05) than the Kilpisjärvi and Mayrhofen soils. The Kilpisjärvi soils had the lowest average pH values, but there were no significant differences in the other physico-chemical properties between the Kilpisjärvi and Mayrhofen bulk soils. Table 1 Soil physico-chemical properties

Region Sampling Site [] SOM (%) pH Available Phosphorous mg/kg

Mayrhofen Alps (A) [8] 0.01 (0.002) 7.03 (0.9) 1.84 (1.1) Cliff (C) [4] 0.02 (0.002) 4.60 (0.1) 1.48 (0.4) Average 0.01 (0.002) 5.81 (0.5) 1.66 (0.8)

Kilpisjärvi Jehkas New (JN) [8] 0.02 (0.002) 5.55 (0.2) 1.31 (0.4) Jehkas Old (JO) [8] 0.02 (0.008) 6.36 (0.4) 0.76 (0.4) Saana (S) [8] 0.02 (0.01) 5.49 (0.6) 2.45 (1.5) Average 0.02 (0.01) 5.80 (0.5) 1.51 (0.8)

Ny- Ålesund Knudsenheia (K) [8] 0.03 (0.01) 7.4 (0.9) 0.83 (0.5) Midtre Lovénbreen (M) [8]

0.03 (0.03) 6.4 (1.2) 0.63 (0.1)

Red River (RR) [8] 0.04 (0.01) 7.78 (0.5) 0.34 (0.1) Average 0.04 (0.02) 7.20 (0.9) 0.60 (0.3)

( ) – Standard deviation values [] – number of biological replicates/sampling site

Geographical region and soil properties impact the diversity of the bulk soil, but not of the rhizosphere or endosphere bacterial communities The species richness (SR) and !-diversity (Shannon index, SI) values of the bulk soil bacterial communities differed between the geographic regions. The Kilpisjärvi bulk soils (SR=33.95, SI=4.21) had significantly lower richness

!58

Cha

pter

3

(two-way ANOVA, p<0.01) and diversity (two-way ANOVA, p<0.01) values than the Mayrhofen (SR=41.04, SI=4.61) and Ny-Ålesund bulk soils (SR=42.55, SI=4.92) (Figure 3a and 3b). In contrast, there were no significant differences in the richness or diversity levels of the rhizosphere soil or endosphere samples between the regions (Figure 3a and 3b).

There was a significant positive relationship of both SR and SI with soil pH (pearson correlation [2-tailed], SR p<0.001, SI p<0.001,) and a negative one with the levels of available phosphorus (P) (pearson correlation [2-tailed], SR p<0.014, SI p<0.001). There was a significant positive correlation between SOM and SI, but not between SOM and SR (Figure 3c).

Figure 3 Estimated Shannon diversity index (A) and species richness (B) of bulk soil, rhizosphere soil and endophytic bacterial communities from three climatic regions Mayrhofen, Kilpisjärvi and Ny-Ålesund. (C) Scatter plots of bulk soil communities explaining the correlation (Pearson correlation) between Shannon diversity and species richness with soil-physico chemical properties from three climatic regions.

!

!!

59

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

Bacterial community structures from different regions differ at the phylum level Collectively, OTUs from all our samples fell into 21 bacterial phyla. Eight of these, i.e. Proteobacteria, Actinobacteria, Acidobacteria, candidate division AD3, Bacteroidetes, Firmicutes, Chloroflexi and Gemmatimonadetes, were prominent making up collectively making up about 97% of the total microbiome. The remaining 13 phyla were present at less than 1% relative abundance each (Figure 4b). Proteobacteria was the most abundant phylum (relative abundance of 48%), with !- (19%) and "-Proteobacteria (19%) being the most dominant classes.

Figure 4 (A) Ternary plot of distribution of OTUs across three climatic regions. Each circle represents one OTU, and the size, color and position of the circle represent its relative abundance, bacterial phylum and affiliation of the OTU with the different regions, respectively. (B-E) Average relative abundances of bacterial phyla distributed across different bioclimatic regions. (B) All samples, (C) Bulk soil samples, (D) Rhizosphere soil samples, (E) Endosphere samples

Proteobacteria were relatively more abundant in Mayrhofen (average relative abundance 57%; Figure 4b) than in the arctic regions (46% in Kilpisjärvi and 43% in Ny-Ålesund). The phylum Acidobacteria and the

!60

Cha

pter

3

candidate division AD3 were observed in higher average relative abundances in Kilpisjärvi than in the other two regions (Figure 4b). The AD3 candidate division had reduced diversity (Figure 4a, 4b), with a single abundant OTU (OTU 10) dominating the Kilpisjärvi bulk soil samples, representing about 25% of the total bulk soil community. The Ny-Ålesund samples were enriched with Actinobacteria (average relative abundances KJ=14%, IN=14%, NA=23%) (Figure 4a, 4b). Table 2 SIMPER analysis based on Bray-Curtis dissimilarity indexes at phylum level identifying the major phyla driving the dissimilarities between different regions or compartments.

SIMPER analyses confirmed that Proteobacteria in Mayrhofen, Acidobacteria and AD3 in Kilpisjärvi and Actinobacteria in Ny-Ålesund were the main phyla contributing to the overall dissimilarities between the regions in all our samples (Table 2). This trend was also consistent in the communities across the different compartments (bulk soil, rhizosphere soil, endosphere) (Figure 4c, 4d, 4e), with the exception of AD3, which was present at very low abundances in the endosphere (<0,9% ) in all three regions (Table 2). Additionally, the relative abundances of Firmicutes in the endosphere samples increased with increasing latitude (Figure 4e), being lowest in Mayrhofen and highest in Ny-Ålesund.

!

Bacterial phyla level dissimilarity index - Bioclimatic regions Average dissimilarity = 34.74 Austria Kilpisjarvi Phylum Average

Abundance Average Abundance

Contribution to Dissimilarity %

Proteobacteria 56.93 46.14 27.77 Acidobacteria 7.68 13.74 17.35 AD3 4.48 10.04 15.66 Actinobacteria 13.64 13.88 13.36 Firmicutes 2.36 4.06 7.17 Bacteroidetes 5.82 4.09 5.72 Gemmatimonadetes 2.5 2.69 3.9 Average dissimilarity = 32.69 Austria Ny-Alesund Phylum Average

Abundance Average Abundance

Contribution to Dissimilarity %

Proteobacteria 56.93 43.3 25.91 Actinobacteria 13.64 23.38 18.14 Acidobacteria 7.68 6.18 12.7 Firmicutes 2.36 5.48 9.62 AD3 4.48 2.23 8.19 Bacteroidetes 5.82 6.99 7.48 Chloroflexi 2.17 4.83 5.79 Gemmatimonadetes 2.5 3.68 5 Average dissimilarity = 37.43 Kilpisjarvi Ny-Alesund Phylum Average

Abundance Average Abundance

Contribution to Dissimilarity %

Proteobacteria 46.14 43.3 19.47 Actinobacteria 13.88 23.38 18.25 Acidobacteria 13.74 6.18 16.32 AD3 10.04 2.23 13.64 Firmicutes 4.06 5.48 10.04 Bacteroidetes 4.09 6.99 6.92 Chloroflexi 1.99 4.83 5.16 Gemmatimonadetes 2.69 3.68 4.52

!

Bacterial phyla level dissimilarity index – Compartment Average dissimilarity = 31.05 Bulk Soil Rhizosphere Phylum Average

Abundance Average Abundance

Contribution to Dissimilarity %

AD3 13.59 2.89 20.44 Proteobacteria 36.92 47.47 19.58 Actinobacteria 17.7 19.71 18.99 Acidobacteria 13.84 11.77 18.92 Gemmatimonadetes 5.88 2.88 5.97 Bacteroidetes 3.33 5.47 4.78 Chloroflexi 4.39 4.38 4.46 Average dissimilarity = 46.76 Bulk Soil Endosphere Phylum Average

Abundance Average Abundance

Contribution to Dissimilarity %

Proteobacteria 36.92 58.06 24.91 AD3 13.59 0.55 14.37 Acidobacteria 13.84 2.16 13.47 Actinobacteria 17.7 15.45 13.32 Firmicutes 0.56 11.81 12.16 Bacteroidetes 3.33 8.16 6.69 Gemmatimonadetes 5.88 0.48 5.8 Chloroflexi 4.39 0.76 4.12 Average dissimilarity = 34.93 Rhizosphere Endosphere Phylum Average

Abundance Average Abundance

Contribution to Dissimilarity %

Proteobacteria 47.47 58.06 23.81 Firmicutes 0.29 11.81 16.5 Acidobacteria 11.77 2.16 15.09 Actinobacteria 19.71 15.45 14.71 Bacteroidetes 5.47 8.16 8.24 Chloroflexi 4.38 0.76 5.51 AD3 2.89 0.55 4.44 Gemmatimonadetes 2.88 0.48 3.52

!

!!

61

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

Firmicutes, Proteobacteria and Bacteroidetes dominate endosphere communities SIMPER analysis indicated that OTUs affiliated with the Firmicutes were strongly enriched in all endosphere-derived sequence data sets, whereas these were virtually absent from the bulk and rhizosphere soil ones. In contrast, candidate division AD3, Chloroflexi, Gemmatimonadetes and Acidobacteria were present in very low abundances in the endosphere communities in all three geographic regions (<4% of total relative abundance collectively) (Figure 6, Table 2). The relative abundance of Proteobacteria and Bacteroidetes increased progressively from bulk to rhizosphere soil to endosphere, with a concomitant decrease in that of candidate division AD3, Gemmatimonadetes and Chloroflexi (Table 2, Figure 6b). This trend was similar in all three geographic regions. The divergence of endosphere communities from soil communities was also evident at the subphylum level. For example, OTUs representing the actinobacterial class Thermoleophilia were abundant in the bulk and

Figure 5 Principal Coordinate Analysis (PCoA) plots of bacterial communities from bulk soils, rhizosphere soils and endospheres of O. digyna and S. oppositifolia from three climatic regions Mayrhofen, Kilpisjärvi and Ny-Ålesund. (A) All samples, (B) Bulk soils and Rhizosphere soils, (C) Endospheres from O. digyna and S. oppositifolia.The symbol colors correspond to compartment (A and B) or plant species (C) and the shapes of the symbols correspond to the geographic regions. All ordinations are based on Bray-Curtis distance matrixes.

!62

Cha

pter

3

rhizosphere soil communities, whereas these were rare in the endosphere communities. The latter were dominated by the class Actinobacteria. Compartment impacts bacterial diversity and community structures more than geographic region or sampling site The species richness and diversity values of the endophytic communities, analyzed at the OTU level, were significantly lower than those of the bulk or rhizosphere soil communities (Figure 3a, b). The rhizosphere soils had the highest SR and SI values, with the differences between rhizosphere and bulk soils not being significant (two-way ANOVA , p>0.05, Figure 3a, b). However, we observed no such difference between the plant species, as O. digyna and S. oppositifolia had similar SR and SI indices in the rhizo- as well as the endosphere communities.

Figure 6 (A) Ternary plot of all OTUs plotted based on the compartment specificity. Each circle represents one OTU. The size, color and position of each OTU represents it relative abundance, bacterial phyla and contribution of the OTU to the nearby compartments respectively. (B) Distribution of average relative abundance of selected major bacterial phyla from all three bioclimatic regions across the compartments.

Also, the community structures of the endosphere bacterial communities differed clearly from those of the bulk and rhizosphere soil ones across all three regions, as demonstrated by PCoA (Figure 5a). A separate analysis of the soil-derived samples revealed that the bulk soil communities diverged from the rhizosphere soil ones in the three regions (Figure 5b). This was supported by PERMANOVA, where compartment was identified as a significant and strong driver of the differences between bacterial community structures (Pseudo-F=30,962, P<0,001, Table 3). In addition to compartment, sampling site (pseudo-F=4,0646, P<0,001) and region (Pseudo-F=6,5495, P<0,001) both had significant effects

!!

63

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

on the bacterial community structures (Table 3), although these factors both had less impact than compartment(Table 3). Table 3 PERMANOVA analyses of factors impacting differences between community structures of bacteria from different climatic regions, sampling sites, compartments or plant species.

PERMANOVA, Single Factor Analysis

Factor df SS MS Pseudo-F p-value(perm) All Samples

Compartment 2 1.02E+05 51,132 30.96 0.001

Site 7 56,283 8,040 4.064 0.001

Region 2 27,369 13,685 6.549 0.001 Bulk Soil

Site 7 43,653 6,236 7.671 0.001

Region 2 21,221 10,610 9.150 0.001 Rhizosphere Soil

Site 7 28,835 4,119 5.073 0.001

Region 2 12,352 6,176 5.996 0.001 Endosphere

Site 7 29,679 4,239 2.142 0.001

Region 2 11,842 5,921 2.788 0.001 PERMANOVA, Two-Factor Analysis

Rhizosphere Soil

Region 2 11,466 5,732 5.886 0.001 Plant species 1 2,910 2,910 2.988 0.007

Region X Plant species 2 2,983 1,491 1.531 0.067

Residuals 54 52,599 974

Endosphere Region 2 11,659 5,829 2.968 0.001

Plant species 1 7,922 7,922 4.033 0.001

Region X Plant species 2 6,383 3,191 1.625 0.003 Residuals 52 1,02E+05 1,964

PERMANOVA performed on each of the compartments separately revealed that region and sampling site had the greatest influence on the structure of bulk soil communities (Pseudo-F=9,1503, P<0,001 and Pseudo-F=7,6707, P<0,001, respectively) with their influence decreasing for the rhizosphere (Pseudo-F=5,9962, P=<0,001 and Pseudo-F=5,0728, P=<0,001, respectively) and endosphere (Pseudo-F=2,7877, P=<0,001 and Pseudo-F=2,1418, P<0,001, respectively) communities (Table 3). Interestingly, region shaped community structures more than sampling site in all compartments, indicating impact of climate or geographic distance (Table 3). Thus, in further

!64

Cha

pter

3

analyses, we focused on comparing communities from different regions and plant species. Plant species and region both impact endosphere bacterial community structures PERMANOVA identified both plant species and geographic region as significant drivers of the community structures of the rhizosphere soil communities (PERMANOVA p<0.01), but region (Pseudo-F= 5,8857) had more impact on the differences than plant species (Pseudo-F=2,9879) (Table 3). In contrast, while plant species, region and their interaction all had significant impact on endosphere community structures (PERMANOVA, P<0,01), plant species had stronger impact on the differences between the communities (Pseudo-F=4,0332) than region or interaction between these factors (Pseudo-F=2,9678 and Pseudo-F=1,6249, respectively) (Table 3).

The endophytic communities from all three regions, being relatively similar to each other, tended to diverge based on plant species (O. digyna or S. oppositifolia) on the first two axes in the PCoA ordination (Figure 5c), while we did not observe any plant species specific clustering in the PCoA of the corresponding rhizosphere communities (data not shown).

Thus, we found partial support for our hypothesis, that plant species strongly shape the plant-associated bacterial communities, as plant species emerged as the major (albeit not the only) significant determinant of the endophytic bacterial community structures over multiple sites and several regions (climate zones). In contrast, this was not true for the rhizosphere communities.

Figure 7 (A) Venn diagram of common shared OTUs and plant species specific OTUs (number of reads of the respective OTUs) between O. digyna and S. oppositifolia from all the endosphere samples. Average relative abundance of endophytic bacterial communities associated with O. digyna and S. oppositifolia endosphere samples at different taxonomical level. (B) bacterial class, (c) bacterial order. Only selected major bacterial orders and classes were classified and shown.

!!

65

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

Differences in endosphere bacterial community structures between the two plant species are explained by differential acquisition of shared bacterial taxa Remarkably, the majority of the endosphere bacterial taxa was present in both plant species, but in different relative abundances. A total of 841 OTUs was found in the endosphere samples, comprising 152,050 reads. A vast majority, i.e. 612 OTUs (149,422 reads, 98.3% of all endosphere reads), was shared between the two plant species, and many of these OTUs were consistently enriched along plant species (Figure 7a). For example, OTUs representing diverse Burkholderiales were enriched in the O. digyna samples (13,325 reads vs 4,396 reads in S. oppositifolia), while OTUs in the order Clostridiales (6,064 reads in S. oppositifolia vs 3,222 reads in O. digyna) were relatively more abundant in S. oppositifolia across the three climate zones (Figure 7c). SIMPER analysis revealed an enhanced accumulation of several OTUs of the Comamonadaceae, Oxalobacteraceae (Burkholderiales) and Bradyrhizobiaceae in O. digyna, versus higher abundances of OTUs of the Clostridiales, along with Actinobacteria, Myxococcales and Saprospirales in S. oppositifolia (Table 4). In addition to the shared bacterial taxa, 162 OTUs (1,749 reads, 1.1% of the total endosphere reads) and 57 OTUs (879 reads, 0.6%) were observed only in O. digyna and S. oppositifolia, respectively (Figure 6a). Thirteen bacterial taxa are highly conserved in the O. digyna and S. oppositifolia endospheres in all three regions, constituting a major portion of the communities We then searched for bacterial taxa that were highly conserved (‘tight’ core) in the O. digyna or S. oppositifolia endospheres using as a criterion OTUs that were present in at least three out of four endosphere samples per plant species across all sampling sites and all regions. Thirteen such OTUs were found, of which five, representing Bradyrhizobium (2 OTUs), Rhodoplanes (!-Proteobacteria), Janthinobacterium ("-Proteobacteria) and Planococcaceae (Firmicutes), were highly consistently present in both plant species (Table 5). Additionally, eight OTUs were consistently present in just one of the plant species. Thus O. digyna specific core OTUs belonged to Comamonadaceae ("-Proteobacteria) and Enterobacteriaceae (#-Proteobacteria), whereas S. oppositifolia core OTUs belonged to Micromonosporaceae, Micrococcaceae (Actinobacteria), Bradyrhizobiaceae (!-Proteobacteria) and unidentified "-Proteobacteria (Table 5). Collectively, these (highly conserved) core OTUs accounted for 38% of the total reads in the endosphere communities. Significantly, eleven of these core OTUs (all, except OTUs 171 and 429, Table 5) were among the main drivers of the divergence of the endosphere communities of the two plant species (Table 4). They also explained the differences between the endosphere and the soil bacterial communities, and those between the endosphere communities in different geographical regions (Table 4).

!66

Cha

pter

3

Most core taxa are enriched in the plant-associated compartments In order to test our hypothesis that the core endophytes are also specifically enriched in the plant endosphere compartment (rather than being ubiquitously present in all samples and compartments), we inspected the distribution of the core OTUs’ reads across different compartments (endosphere, rhizosphere soil and bulk soil) by calculating the average reads of individual OTUs per compartment. Of the 13 core OTUs, 10 OTUs were mainly present in the endosphere (over 50% of reads in the endosphere samples), and but only 5 OTUs (OTUs 4, 5, 15, 32 and 33) were clearly endosphere enriched. However, 11 out of 13 core OTUs were predominantly present in the plant associated compartments, as over 80% of their reads were detected in the endo- or rhizosphere (Figure 8).

Figure 8 Relative distribution of core OTUs’ reads across different compartments.The graph is based on average read count of each OTU in different compartments.

Thus, even though we did find a core set of OTUs that were

consistently present in our endosphere samples, we found no support to our hypothesis, that strictly endophytic lifestyle would be required for tight association with plant hosts. Discussion Factors shaping the bacterial diversity levels in soils across three climatic regions In this study, we examined the bacterial communities in three regions spanning over 3000 km in distance, i.e. Mayrhofen (alpine), Kilpisjärvi (low-arctic) and Ny-Ålesund (high-arctic). In these three regions, the climatic conditions are

!!

67

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

Tab

le 4

20 k

ey O

TU

s de

term

inin

g th

e co

mpo

sitio

n of

end

ophy

te c

omm

uniti

es i

n O

. di

gyna

and

S.

oppo

sitifo

lia f

rom

thr

ee r

egio

ns.

Num

eric

al v

alue

s in

dica

te %

con

trib

utio

n of

the

resp

ectiv

e O

TU

s in

det

erm

ing

the

diff

eren

ce in

com

mun

ity c

ompo

sitio

n be

twee

n en

dosp

here

and

rhi

zosp

here

, O. d

igyna

an

d S.

opp

ositi

folia

and

ave

rage

bet

wee

n th

e th

ree

bioc

limat

ic r

egio

ns r

espe

ctiv

ely.

* -

ind

icat

e O

TU

s (t

op 2

0 O

TU

s fr

om S

IMPE

R a

naly

sis)

whi

ch

stro

ngly

con

trib

uted

to

the

diff

eren

ces

in c

omm

unity

str

uctu

res

of e

ndos

pher

e ba

cter

ia b

etw

een

the

com

part

men

ts,

plan

t sp

ecie

s an

d ge

ogra

phic

re

gion

s. O

TU

s w

hich

are

als

o pa

rt o

f tig

htly

ass

ocia

ted

OT

Us

wer

e hi

ghlig

hted

by

bold

lett

ers

in th

e O

TU

# c

olum

n.

OT

U #

E

ndo

vs

Rhi

zo

Plan

t sp

ecie

s R

egio

n Ph

yla

Cla

ss

Ord

er

Fam

ily

Gen

us

Spec

ies

OT

U_2

1 0.

97*

1.71

* 1.

69*

Firm

icutes

C

lostri

dia

Clos

tridi

ales

C

lostri

diac

eae

Clos

tridi

um

O

TU

_5

0.95

* 1.

01*

1.04

* Fi

rmicu

tes

Bacil

li Ba

cilla

les

Plan

ococ

cacea

e

O

TU

_2

0.93

* 1.

73*

1.87

* Pr

oteo

bacte

ria !-

prot

eoba

cteria

R

hizo

bial

es

Brad

yrhi

zobi

acea

e Br

adyr

hizo

bium

OT

U_3

0.

81*

1.59

* 1.

68*

Prot

eoba

cteria

"-

prot

eoba

cteria

M

yxoc

occa

les

Hal

iang

iacea

e

O

TU

_15

0.73

* 1.

28*

1.20

* A

ctino

bacte

ria

Acti

noba

cteria

A

ctino

myc

etales

M

icroc

occa

ceae

Koc

uria

OT

U_8

0.

71*

1.52

* 1.

79*

Prot

eoba

cteria

#-

prot

eoba

cteria

Bu

rkho

lder

iales

O

xalob

acter

acea

e Ja

nthi

noba

cteriu

m

livid

um

OT

U_3

3 0.

65*

0.99

* 0.

97*

Acti

noba

cteria

A

ctino

bacte

ria

Acti

nom

yceta

les

Micr

omon

ospo

racea

e

O

TU

_4

0.62

* 1.

43*

1.24

* Pr

oteo

bacte

ria #

-pro

teoba

cteria

Bu

rkho

lder

iales

C

omam

onad

acea

e

O

TU

_706

0.

62*

1.43

* 1.

25*

Bacte

roid

etes

Sapr

ospi

rae

Sapr

ospi

rales

C

hitin

opha

gacea

e

O

TU

_13

0.53

* 1.

12*

1.24

* Pr

oteo

bacte

ria #

-pro

teoba

cteria

Bu

rkho

lder

iales

C

omam

onad

acea

e M

ethyli

bium

OT

U_8

4 0.

5*

0.99

* 1.

17*

Prot

eoba

cteria

#

-pro

teoba

cteria

Bu

rkho

lder

iales

C

omam

onad

acea

e L

imno

habi

tans

OT

U_4

8 0.

4 0.

69*

0.78

* Pr

oteo

bacte

ria $-p

roteo

bacte

ria

Pseu

dom

onad

ales

Ps

eudo

mon

adac

eae

Pseu

dom

onas

OT

U_1

6 0.

37

0.65

* 0.

63

Prot

eoba

cteria

!

-pro

teoba

cteria

R

hizo

bial

es

Hyp

hom

icrob

iacea

e R

hodo

plan

es

O

TU

_22

0.37

0.

74*

0.79

* Pr

oteo

bacte

ria #

-pro

teoba

cteria

Bu

rkho

lder

iales

C

omam

onad

acea

e

O

TU

_36

0.37

0.

92*

0.72

* A

ctino

bacte

ria

Acti

noba

cteria

A

ctino

myc

etales

OT

U_2

3 0.

29

0.71

* 0.

59

Prot

eoba

cteria

$ -

prot

eoba

cteria

X

anth

omon

adal

es

Sino

bacte

racea

e St

eroi

doba

cter

O

TU

_26

0.28

0.

64*

0.62

Pr

oteo

bacte

ria

Un_

Prot

eoba

cteria

O

TU

_41

0.27

0.

56*

0.51

Pr

oteo

bacte

ria $ -

prot

eoba

cteria

X

anth

omon

adal

es

Sino

bacte

racea

e

O

TU

_37

0.26

0.

71*

0.54

Pr

oteo

bacte

ria !

-pro

teoba

cteria

R

hizo

bial

es

Hyp

hom

icrob

iacea

e

O

TU

_83

0.26

0.

59*

0.47

A

ctino

bacte

ria

Acid

imicr

obiia

A

cidim

icrob

iales

!68

Cha

pter

3

Table 5 Highly conserved core OTUs of O. digyna and S. oppositifolia endospheres. OTUs present in minimum of three samples out of four in all sampling sites in all regions per plant species are included.

OTU # Phyla Class Order Family Genus

Core OTUs of both plant species

OTU_16 Proteobacteria !-proteobacteria Rhizobiales Hyphomicrobiaceae Rhodoplanes

OTU_2 Proteobacteria !-proteobacteria Rhizobiales Bradyrhizobiaceae Bradyrhizobium

OTU_429 Proteobacteria !-proteobacteria Rhizobiales Bradyrhizobiaceae

OTU_5 Firmicutes Bacilli Bacillales Planococcaceae

OTU_8 Proteobacteria "-proteobacteria Burkholderiales Oxalobacteraceae Janthinobacterium

Additional core OTUs of S. oppositifolia

OTU_15 Actinobacteria Actinobacteria Actinomycetales Micrococcaceae Kocuria

OTU_171 Proteobacteria ! -proteobacteria Rhizobiales Bradyrhizobiaceae

OTU_33 Actinobacteria Actinobacteria Actinomycetales Micromonosporaceae

OTU_7 Proteobacteria " -proteobacteria Ellin6067 Un_Ellin6067

Additional core OTUs of O. digyna

OTU_13 Proteobacteria "-proteobacteria Burkholderiales Comamonadaceae Methylibium

OTU_32 Proteobacteria !-proteobacteria Enterobacteriales Enterobacteriaceae

OTU_35 Proteobacteria "-proteobacteria Burkholderiales Comamonadaceae

OTU_4 Proteobacteria "-proteobacteria Burkholderiales Comamonadaceae

clearly different.The highest bacterial species richness and diversity values in the bulk soils were found in the Ny-Ålesund samples, which was consistent with data by Chu et al. (2011) and Neufeld and Mohn (2005) who also detected highest bacterial diversities in the high northern latitutes. However, our data stand in contrast to those from Yergeau et al. (2007), who reported decreasing bacterial diversity in Antarctic soils with increasing latitude towards the south pole. We found a clear positive correlation of bacterial diversity with soil pH and SOM, and a negative correlation with the level of available P, agreeing with several studies that put forth soil pH as a major driver of bacterial diversity (Fierer and Jackson, 2006; Fierer and Lennon, 2011; Lauber et al., 2009; Rousk et al., 2010; Shi et al., 2015). Soil nutritional status and available P have also been shown to significantly impact bacterial diversity (Siciliano et al., 2014).

With respect to compartment, the endosphere bacterial communities were significantly less diverse than those in the corresponding soils. However, in contrast with studies from other soils ("nceo#lu et al., 2011; Kowalchuk et al., 2002; Smalla et al., 2001), where rhizosphere soil communities have been reported to be less diverse and rich than those in bulk soils, we observed highest richness and diversity values in the rhizosphere soils. This was similar to findings in our previous study from the Kilpisjärvi site, where the

!!

69

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

rhizosphere samples had highest richness and diversity (Kumar et al., 2016). Miniaci et al. (2007) and Coleman-Derr et al. (2016) studying low-SOM glacier forefield or desert soils, respectively, also observed higher bacterial diversity and richness values in the rhizospheres than in the corresponding bulk soils. Further, Yergeau et al. (2007) found that soil bacterial diversities in Antarctic non-vegetated fell-field soils decreased with increasing (southern) latitude whereas bacterial diversity of soils from vegetated sites did not decrease with latitude. This suggests that a plant-incited “protective or nutritional” effect on bacterial communities becomes increasingly more important in soils with low available nutrients. OTUs that determine the divergence of the soil bacteriomes across three regions In this study, we detected very few endemic bacterial OTUs, as the great majority of the bacterial taxa were found in all three, geographically widespread, regions. However, these taxa were present in very different relative abundances, leading to region-driven community structures. Roughly, proteobacterial taxa decreased and Gram-positive ones increased towards the north, with Acidobacteria and candidate division AD3 being enriched in the Kilpisjärvi samples. This clear progressive change in bacterial community structures hints at specific effects of the shifting local conditions on the aforementioned taxa. Thus, habitat filtering rather than [long-distance] dispersal impacts the bacterial community compositions across the three cold climate sites.

The dominance of Proteobacteria in the bulk soil samples from Mayrhofen was consistent with findings by Margesin et al. (2009) in alpine soils. Moreover, corroborating earlier studies (Männistö et al., 2007, 2013), the high abundance of Acidobacteria was likely linked to the low pH in the Kilpisjärvi soils (Chu et al., 2010; Griffiths et al., 2011). Also, the high abundance of candidate division AD3 in Kilpisjärvi (Figure 6b) was consistent with similar findings for the Mitchell peninsula in Antarctica (Ji et al., 2015) and low-nutrient sandy soils (Zhou et al., 2003). However, earlier studies by Männistö et al. (2007, 2013) have not detected candidate division AD3 in high-SOM Kilpisjärvi soils. We here assume that the candidate division AD3 members that were found are well adapted to low SOM/ low nutrient soils. Compartment is the primary driver of bacterial community structures A striking observation was that both O. digyna and S. oppositifolia sampled in any of the three regions exhibited similar endosphere bacterial communities. We previously observed compartmental influence between bulk and rhizosphere

!70

Cha

pter

3

soils of O. digyna and S. oppositifolia (Kumar et al., 2016), and so extend this to the endospheres that were addressed in the current study. Clearly, even though the bulk soil bacterial communities were influenced by region and sampling site, which may relate to soil edaphic factors, plant endospheres shared similar bacterial endophytes across the three regions. This points to a strong and specific filtering effect of the two pioneering plants that were studied, allowing similar bacteria to colonize plants from the widely divergent soils in different regions.

As a token of the plant-incited filtering effect, members of the Proteobacteria, Actinobacteria, Bacteroidetes and Firmicutes dominated the endosphere bacterial communities. Several other studies, performed with both agricultural and wild plants, also reported these four phyla to be dominant in several endosphere, with Proteobacteria being the most dominant (Coleman-Derr et al., 2016; Santoyo et al., 2016; Zhao et al., 2016). Other taxa, including Acidobacteria, candidate division AD3, Chloroflexi and Gemmatimonadetes, were virtually absent from the endosphere. A general underrepresentation of Acidobacteria in the endosphere has also been observed in other systems (Coleman-Derr et al., 2016; Edwards et al., 2015; Zarraonaindia et al., 2015). The enrichment of Firmicutes in endosphere samples in this study was mainly ascribed to the raised abundance of OTUs belonging to the Clostridia (in particular OTU 21; genus Clostridium). Possibly, such organisms might have been selected for their capacities to fix nitrogen in the cold and often water-logged soils (Rosenblueth and Martínez-Romero, 2006) in the permafrost-impacted Arctic sites. Interestingly, nifH genes of Clostridia spp. have been reported to be frequent in soil samples fromCanadian high Arctic (Deslippe and Egger, 2006). Similarly, in our study plants were sampled at early growing season, when plants started flowering, and snow was still melting in most sampling sites. A small set of highly conserved OTUs shapes the endosphere bacterial communities in two arcto-alpine plant species O. digyna and S. oppositifolia, the target plant species in this study, are both perennial herbs with similar habitat requirements, producing tap root systems of similar size and depth; the plants often grow at close proximity to each other. The plants, however, are taxonomically quite distant: O. digyna is member of the Polygonaceae of the order Caryophyllales, whereas S. oppositifolia belongs to the order Saxifragales, which is estimated to have diverged from other core Eudicots 114-124 MYA (Soltis et al., 2000; Wikström et al., 2001). The plants also have differing mycorrhizal associations: O. digyna is non-mycorrhizal, whereas S. oppositifolia is endomycorrhizal, which is likely to have strong impact on its nutrient acquisition efficiency.

!!

71

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

Despite these differences, the endophytic communities of these two plants were strikingly similar. While we found a significant effect of plant species on community structure of endophytic bacterial communities (Table 3, Figure 5c), the plants shared a core microbiome, dominated by Burkholderiales, Actinomycetales and Rhizobiales, across three different arcto-alpine climatic regions. Of these, Actinomycetales and Burkholderiales have been reported as components of the core root microbiome of A. thaliana (Schlaeppi et al., 2014). Rhizobiales are known plant symbionts with nitrogen fixing abilities, while Burkholderiales are well known for their biodegradative capacities and antagonistic properties towards multiple soil-borne fungal pathogens (Benítez and McSpadden Gardener, 2009; Chebotar et al., 2015). The core microbiome OTUs representing Burkholderiales, especially Comamonadaceae and Oxalobacteraceae, were relatively more abundant in O. digyna. We have repeatedly isolated bacteria from O. digyna vegetative tissues with very high sequence homologies to the above core OTUs (Nissinen et al., 2012; unpublished). Further, we have isolated or detected (clone libraries) bacteria in O. digyna seeds with 100% (16S rRNA gene based) identity to six of the core OTUs (OTUs 2 and 16 representing Rhizobiales, OTUs 8, 13 and 35 (Burkholderiales) and OTU 15 (Actinomycetales) (unpublished data). Core OTUs related to similar strains from seeds were highly enriched (Figure 8) in the endosphere or rhizosphere soils. Part of these core organisms could thus be seed-transmitted and colonize the rhizo-and endosphere of developing seedlings, as previously described by Puente et al. (2009) in desert cacti, indicating the potential importance of such seed transmitted endophytes in pioneer plants. Horizontal transmission of a set of endophytes has also been observed by Hardoim et al. (2012) and Johnston-Monje and Raizada (2011).

In addition, these core OTUs were among the primary drivers of region, compartment or host-plant species differences among the bacterial communities. The higher relative abundance of Clostridia in Ny-Ålesund and Rhizobia in Mayrhofen in the endosphere communities is one such example, as discussed above.

In summary, we here report that, on the basis of data obtained with two plant species, host plant-specific endophytic communities can be acquired despite a distance of over 3000 km and differences in climate and chemistry between soils. These plant species-specific assemblages are formed from a shared core set of bacteria, most of which are strongly enriched in the endosphere. We surmised that selection by these perennial plants plays a role, possibly concomitant with a highly efficient adaptation and fitness of these bacteria in the plant environment. Some of these core OTUs could even be seed-inherited, explaining their tight association with the host plant. Very closely related endophytic taxa have previously been found to be shared by

!72

Cha

pter

3

plants from other cold climates (Carrell and Frank, 2015; Nissinen et al., 2012; Poosakkannu et al., 2015), indicating efficient establishment of specific bacteria in arcto-alpine plants. Conflict of interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Author contributions Study was conceptualized and designed by RN and MK. Field work was performed by MK, RN and GB. Sample processing was done by MK and RN. Supporting soil analysis was done by MK while library preparation for sequence analysis was done by MK with assistance of AM. Bioinformatics analysis was performed by MK and the data analysis was done by MK and RN. Manuscript draft was prepared by MK, RN and JE and revisions was done by MK, RN, AM, GB, AS and JE. Final version for the submission was prepared by MK and RN. Funding This research was funded by NWO project 821.01.005 (for JE), Finnish cultural foundation Lapland regional fund (for RN) and by Finnish Academy, project 259180 (for RN) Acknowledgements Authors acknowledge Maarten Loonen and other members of Netherlands Arctic Station, Ny-Ålesund, Svalbard and Kilpisjärvi Biological Station of the University of Helsinki for assistance in sampling. We also thank Andrés López-Sépulcre and Jenni Niku for their valuable suggestions with statistical analysis and Jolanda Brons for her assistance in soil physico-chemical analysis.

References

Badri, D. V, Weir, T. L., van der Lelie, D., and Vivanco, J. M. (2009). Rhizosphere chemical dialogues: plant-microbe interactions. Current opinion in biotechnology 20, 642–50. doi:10.1016/j.copbio.2009.09.014.

Benítez, M. S., and McSpadden Gardener, B. B. (2009). Linking sequence to function in soil bacteria: Sequence-directed isolation of novel bacteria contributing to soilborne plant disease suppression. Applied and Environmental Microbiology 75, 915–924. doi:10.1128/AEM. 01296-08.

!!

73

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

Berg, G., and Smalla, K. (2009). Plant species and soil type cooperatively shape the structure and function of microbial communities in the rhizosphere. FEMS microbiology ecology 68, 1–13. doi:10.1111/j.1574-6941.2009.00654.x.

Billings, W., and Mooney, H. (1968). The ecology of arctic and alpine plants. Biological Reviews 43, 481–529.

Borin, S., Ventura, S., Tambone, F., Mapelli, F., Schubotz, F., Brusetti, L., et al. (2010). Rock weathering creates oases of life in a high Arctic desert. Environmental Microbiology 12, 293–303. doi:10.1111/j.1462-2920.2009.02059.x.

Bray, R. H., and Kurtz, L. T. (1945). Determination of Total, Organic, and Available Forms of Phosphorus in Soils. Soil Science 59, 39–46. doi:10.1097/00010694-194501000-00006.

Bulgarelli, D., Schlaeppi, K., Spaepen, S., Ver Loren van Themaat, E., and Schulze-Lefert, P. (2013). Structure and Functions of the Bacterial Microbiota of Plants. Annual review of plant biology. doi:10.1146/annurev-arplant-050312-120106.

Caporaso, J. G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F. D., Costello, E. K., et al. (2010). QIIME allows analysis of high-throughput community sequencing data. Nature methods 7, 335–6. doi:10.1038/nmeth.f.303.

Carrell, A. A., and Frank, A. C. (2015). Bacterial endophyte communities in the foliage of coast redwood and giant sequoia. Frontiers in Microbiology 6, 1008. doi:10.3389/fmicb. 2015.01008.

Chapin, S. F., and Körner, C. (1996). “Arctic and Alpine Biodiversity: Its Patterns, Causes and Ecosystem consequences,” in Functional Roles of Biodiversity: A Global perspective, 7–32.

Chebotar, V. K., Malfanova, N. V., Shcherbakov, A. V., Ahtemova, G. A., Borisov, A. Y., Lugtenberg, B., et al. (2015). Endophytic bacteria in microbial preparations that improve plant development. Applied Biochemistry and Microbiology 51, 271–277. doi:10.1134/ S0003683815030059.

Chelius, M. K., and Triplett, E. W. (2001). The Diversity of Archaea and Bacteria in Association with the Roots of Zea mays L. Microbial ecology 41, 252–263. doi:10.1007/ s002480000087.

Cho, J. (2000). Biogeography and Degree of Endemicity of Fluorescent Pseudomonas Strains in Soil. 66, 5448–5456. doi:10.1128/AEM.66.12.5448-5456.2000.

Chu, H., Fierer, N., Lauber, C. L., Caporaso, J. G., Knight, R., and Grogan, P. (2010). Soil bacterial diversity in the Arctic is not fundamentally different from that found in other biomes. Environmental microbiology 12, 2998–3006. doi:10.1111/j.1462-2920.2010.02277.x.

Chu, H., Neufeld, J. D., Walker, V. K., and Grogan, P. (2011). The Influence of Vegetation Type on the Dominant Soil Bacteria, Archaea, and Fungi in a Low Arctic Tundra Landscape. Soil Science Society of America Journal 75, 1756. doi:10.2136/sssaj2011.0057.

Coleman-Derr, D., Desgarennes, D., Fonseca-Garcia, C., Gross, S., Clingenpeel, S., Woyke, T., et al. (2016). Plant compartment and biogeography affect microbiome composition in cultivated and native Agave species. New Phytologist 209, 798–811. doi:10.1111/nph.13697.

Compant, S., Clément, C., and Sessitsch, A. (2010). Plant growth-promoting bacteria in the rhizo- and endosphere of plants: Their role, colonization, mechanisms involved and prospects for utilization. Soil Biology and Biochemistry 42, 669–678. doi:10.1016/j.soilbio. 2009.11.024.

Coulson, S., and Hodkinson, I. (1995). Thermal environments of Arctic soil organisms during winter. Arctic and Alpine Research 27, 364–370. doi:10.2307/1552029.

deCárcer, D. A., Denman, S. E., McSweeney, C., and Morrison, M. (2011). Evaluation of subsampling-based normalization strategies for tagged high-throughput sequencing data

!74

Cha

pter

3

sets from gut microbiomes. Applied and Environmental Microbiology 77, 8795–8798. doi: 10.1128/AEM.05491-11.

DeSantis, T. Z., Hugenholtz, P., Larsen, N., Rojas, M., Brodie, E. L., Keller, K., et al. (2006). Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Applied and environmental microbiology 72, 5069–72. doi:10.1128/AEM. 03006-05.

Deslippe, J. R., and Egger, K. N. (2006). Molecular diversity of nifH genes from bacteria associated with high arctic dwarf shrubs. Microbial Ecology 51, 516–525. doi:10.1007/s00248-006-9070-8.

Edgar, R. C. (2010). Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461. doi:10.1093/bioinformatics/btq461.

Edgar, R. C. (2013). UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nature Methods 10, 996–998. doi:10.1038/nmeth.2604.

Edwards, J., Johnson, C., Santos-Medellín, C., Lurie, E., Podishetty, N. K., Bhatnagar, S., et al. (2015). Structure, variation, and assembly of the root-associated microbiomes of rice. Proceedings of the National Academy of Sciences of the United States of America 112, E911–20. doi:10.1073/pnas.1414592112.

Fierer, N., and Jackson, R. B. (2006). The diversity and biogeography of soil bacterial communities. Proceedings of the National Academy of Sciences of the United States of America 103, 626–31. doi:10.1073/pnas.0507535103.

Fierer, N., and Lennon, J. T. (2011). The generation and maintenance of diversity in microbial communities. American journal of botany 98, 439–48. doi:10.3732/ajb.1000498.

Garbeva, P., van Veen, J. A., and van Elsas, J. D. (2004). MICROBIAL DIVERSITY IN SOIL: Selection of Microbial Populations by Plant and Soil Type and Implications for Disease Suppressiveness. Annual Review of Phytopathology 42, 243–270. doi:10.1146/annurev. phyto.42.012604.135455.

Ghyselinck, J., Pfeiffer, S., Heylen, K., Sessitsch, A., De Vos, P., Larsen, P., et al. (2013). The Effect of Primer Choice and Short Read Sequences on the Outcome of 16S rRNA Gene Based Diversity Studies. PLoS ONE 8, e71360. doi:10.1371/journal.pone.0071360.

Griffiths, R. I., Thomson, B. C., James, P., Bell, T., Bailey, M., and Whiteley, A. S. (2011). The bacterial biogeography of British soils. Environmental microbiology 13, 1642–54. doi:10.1111/j.1462-2920.2011.02480.x.

Hanson, C. a., Fuhrman, J. a., Horner-Devine, M. C., and Martiny, J. B. H. (2012). Beyond biogeographic patterns: processes shaping the microbial landscape. Nature Reviews Microbiology 10, 1–10. doi:10.1038/nrmicro2795.

Hardoim, P. R., Hardoim, C. C. P., van Overbeek, L. S., and van Elsas, J. D. (2012). Dynamics of seed-borne rice endophytes on early plant growth stages. PloS one 7, e30438. doi:10.1371/journal.pone.0030438.

Hardoim, P. R., van Overbeek, L. S., Berg, G., Pirttilä, A. M., Compant, S., Campisano, A., et al. (2015). The Hidden World within Plants: Ecological and Evolutionary Considerations for Defining Functioning of Microbial Endophytes. Microbiology and Molecular Biology Reviews 79, 293–320. doi:10.1128/MMBR.00050-14.

Hardoim, P. R., van Overbeek, L. S., and Elsas, J. D. Van (2008). Properties of bacterial endophytes and their proposed role in plant growth. Trends in microbiology 16, 463–71. doi:10.1016/j.tim.2008.07.008.

"nceo#lu, Ö., Al-Soud, W. A., Salles, J. F., Semenov, A. V., and van Elsas, J. D. (2011). Comparative Analysis of Bacterial Communities in a Potato Field as Determined by Pyrosequencing. PLoS ONE 6, e23321. doi:10.1371/journal.pone.0023321.

Ji, M., van Dorst, J., Bissett, A., Brown, M. V., Palmer, A. S., Snape, I., et al. (2015). Microbial diversity at Mitchell Peninsula, Eastern Antarctica: a potential biodiversity

!!

75

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

“hotspot.” Polar Biology 39, 237–249. doi:10.1007/s00300-015-1776-y. Johnston-Monje, D., and Raizada, M. N. (2011). Conservation and Diversity of Seed

Associated Endophytes in Zea across Boundaries of Evolution, Ethnography and Ecology. PLoS ONE 6, e20396. doi:10.1371/journal.pone.0020396.

Körner, C. (2003). Alpine Plant Life - Functional Plant Ecology of High Mountain Ecosystems. Second Edi. Springer.

Kowalchuk, G. A., Buma, D. S., de Boer, W., Klinkhamer, P. G. L., and van Veen, J. A. (2002). Effects of above-ground plant species composition and diversity on the diversity of soil-borne microorganisms. International Journal of General and Molecular Microbiology 81, 509–520. doi:10.1023/A:1020565523615.

Kumar, M., Männistö, M. K., van Elsas, J. D., and Nissinen, R. M. (2016). Plants impact structure and function of bacterial communities in Arctic soils. Plant and Soil 399, 319–332. doi:10.1007/s11104-015-2702-3.

Lauber, C. L., Hamady, M., Knight, R., and Fierer, N. (2009). Pyrosequencing-based assessment of soil pH as a predictor of soil bacterial community structure at the continental scale. Applied and Environmental Microbiology 75, 5111–20. doi:10.1128/AEM. 00335-09.

Mäki, A., Rissanen, J. A., and Tiirola, M. (2016). A practical method for barcoding and size-trimming PCR templates for amplicon sequencing. Biotechniques 60, 88–90. doi:10.2144/000114380.

Männistö, M. K., Kurhela, E., Tiirola, M., and Häggblom, M. M. (2013). Acidobacteria dominate the active bacterial communities of Arctic tundra with widely divergent winter-time snow accumulation and soil temperatures. FEMS microbiology ecology 84, 47–59. doi:10.1111/1574-6941.12035.

Männistö, M. K., Tiirola, M., and Häggblom, M. M. (2007). Bacterial communities in Arctic fjelds of Finnish Lapland are stable but highly pH-dependent. FEMS Microbiology Ecology 59, 452–465. doi:10.1111/j.1574-6941.2006.00232.x.

Mapelli, F., Marasco, R., Rizzi, A., Baldi, F., Ventura, S., Daffonchio, D., et al. (2011). Bacterial communities involved in soil formation and plant establishment triggered by pyrite bioweathering on arctic moraines. Microbial ecology 61, 438–47. doi:10.1007/s00248-010-9758-7.

Margesin, R., Jud, M., Tscherko, D., and Schinner, F. (2009). Microbial communities and activities in alpine and subalpine soils. FEMS Microbiology Ecology 67, 208–218. doi:10.1111/j.1574-6941.2008.00620.x.

Meyer, D., Zeileis, A., and Hornik, K. (2015). vcd: Visualizing Categorical Data. R package version. 1.4–1. Available at: https://cran.r-project.org/web/packages/vcd/citation.html.

Miniaci, C., Bunge, M., Duc, L., Edwards, I., Bürgmann, H., and Zeyer, J. (2007). Effects of pioneering plants on microbial structures and functions in a glacier forefield. Biology and Fertility of Soils 44, 289–297. doi:10.1007/s00374-007-0203-0.

Neufeld, J., and Mohn, W. (2005). Unexpectedly high bacterial diversity in arctic tundra relative to boreal forest soils, revealed by serial analysis of ribosomal sequence tags. Applied and environmental microbiology 71, 5710–5718. doi:10.1128/AEM.71.10.5710.

Nissinen, R. M., Männistö, M. K., and van Elsas, J. D. (2012). Endophytic bacterial communities in three arctic plants from low arctic fell tundra are cold-adapted and host-plant specific. FEMS microbiology ecology 82, 510–22. doi:10.1111/j.1574-6941.2012.01464.x.

Oakley, B. B., Carbonero, F., van der Gast, C. J., Hawkins, R. J., and Purdy, K. J. (2010). Evolutionary divergence and biogeography of sympatric niche-differentiated bacterial populations. The ISME journal 4, 488–97. doi:10.1038/ismej.2009.146.

Poosakkannu, A., Nissinen, R., and Kytöviita, M. M. (2015). Culturable endophytic

!76

Cha

pter

3

microbial communities in the circumpolar grass, Deschampsia flexuosa in a sub-Arctic inland primary succession are habitat and growth stage specific. Environmental Microbiology Reports 7, 111–122. doi:10.1111/1758-2229.12195.

Puente, M. E., Li, C. Y., and Bashan, Y. (2009). Rock-degrading endophytic bacteria in cacti. Environmental and Experimental Botany 66, 389–401. doi:10.1016/j.envexpbot. 2009.04.010.

Pylro, V. S., Roesch, L. F. W., Morais, D. K., Clark, I. M., Hirsch, P. R., and Tótola, M. R. (2014). Data Analysis for 16S Microbial Profiling from Different Benchtop Sequencing Platforms. Journal of microbiological methods 107, 30–37. doi:10.1016/j.mimet.2014.08.018.

Rosenblueth, M., and Martínez-Romero, E. (2006). Bacterial endophytes and their interactions with hosts. Molecular plant-microbe interactions$: MPMI 19, 827–37. doi:10.1094/MPMI-19-0827.

Rousk, J., Bååth, E., Brookes, P. C., Lauber, C. L., Lozupone, C., Caporaso, J. G., et al. (2010). Soil bacterial and fungal communities across a pH gradient in an arable soil. The ISME journal 4, 1340–51. doi:10.1038/ismej.2010.58.

Santoyo, G., Moreno-Hagelsieb, G., del Carmen Orozco-Mosqueda, M., and Glick, B. R. (2016). Plant growth-promoting bacterial endophytes. Microbiological Research 183, 92–99. doi:10.1016/j.micres.2015.11.008.

Schlaeppi, K., Dombrowski, N., Oter, R. G., Ver Loren van Themaat, E., and Schulze-Lefert, P. (2014). Quantitative divergence of the bacterial root microbiota in Arabidopsis thaliana relatives. Proceedings of the National Academy of Sciences of the United States of America 111, 585–92. doi:10.1073/pnas.1321597111.

Shi, Y., Xiang, X., Shen, C., Chu, H., Neufeld, J. D., Walker, V. K., et al. (2015). Vegetation-Associated Impacts on Arctic Tundra Bacterial and Microeukaryotic Communities. Applied and Environmental Microbiology 81, 492–501. doi:10.1128/AEM.03229-14.

Siciliano, S. D., Palmer, A. S., Winsley, T., Lamb, E., Bissett, A., Brown, M. V., et al. (2014). Soil fertility is associated with fungal and bacterial richness, whereas pH is associated with community composition in polar soil microbial communities. Soil Biology and Biochemistry 78, 10–20. doi:10.1016/j.soilbio.2014.07.005.

Smalla, K., Wieland, G., Buchner, A., Zock, A., Parzy, J., Roskot, N., et al. (2001). Bulk and Rhizosphere Soil Bacterial Communities Studied by Denaturing Gradient Gel Electrophoresis$: Plant-Dependent Enrichment and Seasonal Shifts Revealed. Applied and environmental microbiology 67, 4742–4751. doi:10.1128/AEM.67.10.4742.

Soltis, D. E., Soltis, P. S., Chase, M. W., Mort, M. E., Albach, D. C., Zanis, M., et al. (2000). Angiosperm phylogeny inferred from 18S rDNA, rbcL, and atpB sequences. Botanical Journal of the Linnean Society 133, 381–461. doi:10.1006/bojl.2000.0380.

Wikström, N., Savolainen, V., and Chase, M. W. (2001). Evolution of the angiosperms: calibrating the family tree. Proceedings. Biological sciences / The Royal Society 268, 2211–20. doi:10.1098/rspb.2001.1782.

Yergeau, E., Newsham, K. K., Pearce, D. A., and Kowalchuk, G. A. (2007). Patterns of bacterial diversity across a range of Antarctic terrestrial habitats. Environmental microbiology 9, 2670–82. doi:10.1111/j.1462-2920.2007.01379.x.

Zarraonaindia, I., Owens, S. M., Weisenhorn, P., West, K., Hampton-Marcell, J., Lax, S., et al. (2015). The soil microbiome influences grapevine-associated microbiota. mBio 6, 1–10. doi:10.1128/mBio.02527-14.

Zhang, H. W., Song, Y. C., and Tan, R. X. (2006). Biology and chemistry of endophytes. Natural product reports 23, 753–71. doi:10.1039/b609472b.

Zhao, S., Zhou, N., Zhao, Z.-Y., Zhang, K., and Tian, C.-Y. (2016). High-Throughput Sequencing Analysis of the Endophytic Bacterial Diversity and Dynamics in Roots of the

!!

77

Bac

teria

l com

mun

ities

in a

rcto

-alp

ine

plan

ts.

Halophyte Salicornia europaea. Current Microbiology 72, 557–562. doi:10.1007/s00284-016-0990-3.

Zheng, D., Alm, E. W., Stahl, D. A., and Raskin, L. (1996). Characterization of universal small-subunit rRNA hybridization probes for quantitative molecular microbial ecology studies. Applied and environmental microbiology 62, 4504–13.

Zhou, J., Xia, B., Huang, H., Treves, D. S., Hauser, L. J., Mural, R. J., et al. (2003). Bacterial phylogenetic diversity and a novel candidate division of two humid region, sandy surface soils. Soil Biology and Biochemistry 35, 915–924. doi:10.1016/S0038-0717(03)00124-X.

 

 78

!

!

!!

Chapter 4

Fungal communities associated with arcto-alpine plants are strongly shaped by regional effects

Manoj Kumar, Jenni Niku, Livio Antonielli, Günter Brader, Angela Sessitsch, Sara Taskinen, Jan Dirk van Elsas, Riitta M. Nissinen

Submitted to Frontiers in Microbiology

!80

Cha

pter

4

Abstract The diversity of plant-associated fungal communities in arcto-alpine pioneer plants remains elusive when compared to that of temperate climate plants. Here, we investigated the diversity of plant root associated fungal communities from two pioneer arcto-alpine plants, Oxyria digyna and Saxifraga oppositifolia, from three distinct geographic regions. DNA was extracted from bulk and rhizosphere soils and from the plant roots, after which the ITS2 region was amplified and amplicons were sequenced in order to profile the fungal communities. Diversities as well as community structures of the fungal communities were primarily impacted by the factor compartment (bulk soil, rhizosphere soil, endosphere). The highest diversities were detected in the rhizosphere soils followed by bulk soils and endosphere samples. Endosphere fungal communities exhibited community structures that were clearly distinct from those of rhizosphere or bulk soil. We detected no impact of geographic region on the diversity values of the soil fungal communities. However, remarkably, the diversities of the endosphere communities of both plant species decreased with increasing latitude. Rhizosphere and endosphere fungal community structures were influenced by geographic region, with low host plant specificity. Soil pH clearly determined the community structures of the bulk soil fungal communities. The fungal genera Mycenella, Cadophora and Varicosporium dominated the endosphere, whereas Geomyces pannorum was predominantly present in the soils. Keywords: Fungal diversity, arcto-alpine plants, Geomyces pannorum, endophytes, Mycenella, Cadophoram, Varicosporium

!!

81

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

Introduction The diversity and ecological functioning of fungal communities in most habitats remain poorly studied when compared to those of bacterial communities (Anderson & Cairney, 2004). In particular, the diversity of fungi in arctic and alpine biomes have been largely unexplored (Robinson, 2001). A recent study on Antarctic soil fungal communities (comparing with previous studies of Arctic soil fungi) revealed a bipolar distribution of the fungal communities in Antarctica, i.e. these communities shared a high number of organisms with those of high northern latitude soils (Cox et al., 2016). Thus, strong parallel habitat filtering was hypothesized to occur. Moreover, the fungi colonizing such extreme environments may be strongly dispersed, e.g. by migrating birds (Marshall, 1998), and this points to an ability to colonize and survive in the (low-nutrient) polar soils.

A surprising richness and diversity of plant root-associated fungal communities has been found in the first pioneering studies in these habitats, as compared to the generally low diversity of plants (Blaalid et al., 2012; Botnen et al., 2014; Mundra et al., 2015; Yao et al., 2013, Bjorbækmo et al., 2010). We ignore to what extent such fungi fall in the class of mycorrhizal fungi, or just occur as root inhabitants without a noticeable mycorrhizal feature. Nevertheless and irrespective of their relative abundances, in particular ecto-mycorrhizal (EM) and ericoid mycorrhizal (ECM) fungal communities have been estimated to furnish about 61-86% of the plant nitrogen in the arctic tundra, a value similar to that estimated for temperate climate regions. In contrast, there is no conclusive evidence that arbuscular mycorrhizal (AM) fungi in the Arctic supply their symbiotic partners with any nitrogen (Hobbie and Hobbie, 2006).

Unfortunately, there are seemingly contrasting reports when it comes to the levels of mycorrhization of plants in the Arctic. Thus, Kytöviita (2005) suggested that, in contrast to the high mycorrhization rate of plants from tropical or temperate soils, non-mycorrhizal plants dominate Arctic soils. Koske (1987) and Kytöviita and Ruotsalainen (2007) also reported that most plants in this region do not avidly form mycorrhizal symbioses. Moreover, these authors claimed that the diversity of AM fungi tends to decrease with increasing latitude. In particular, herbaceous species like Saxifraga oppositifolia, Pedicularis capitata and Solidago virgaurea (which are mainly arbuscular mycorrhizal [AM] in temperate regions) have been reported to form EM symbiosis in the high Arctic (Kytöviita, 2005; Pietikäinen et al., 2007). It was hypothesized that the low temperatures, short growth seasons and/or soil disturbance by frost thawing might be at the basis of the apparent low abundance of mycorrhized plants in the Arctic. However, we still face a paucity of knowledge as to which

!82

Cha

pter

4

plant naturally associate with mycorrhizal fungi in the arcto-alpine regions and which plant do not posses this avidity.

Furthermore, there are so far only few studies that investigated the host plant specificity of endophytic fungi in arcto-alpine plants (Blaalid et al., 2014; Fisher et al., 1995; Higgins et al., 2007; Lv et al., 2010; Zhang et al., 2015). In four studies, EM or ECM fungi from several arctic and alpine plants showed no significant host specificity (Ryberg et al., 2009, 2011; Timling et al., 2012; Walker et al., 2011). In another more specific study, Botnen et al. (2014) also reported low host specificity of root-associated fungal communities from the arcto-alpine plant species Dryas octopetala, Salix polaris and Bistorta vivipara. However, rhizosphere AM fungal communities were found to be influenced by plant host (Becklin et al., 2012). In contrast to the foregoing four studies, Higgins et al. (2007) and Zhang et al. (2015) reported that endophytic fungal communities of tundra plants from the Canadian and Ny-Ålesund (Svalbard) Arctic regions were host plant specific. The conflicting observations of the degree of mycorrhization at high northern latitudes and the controversial claims of host-plant specificity in the high Arctic require further study, in particular with respect to the presumed importance of fungal endophytes in nutrient poor habitats.

In this study, we used next generation sequencing of soil and plant microbial community DNA to study the diversity and community structure of the respective fungal communities. We selected two pioneer plant species with differing mycorrhizal relationships, Oxyria digyna and S. oppositifolia. The former plant species has been reported to be non-mycorrhizal (Väre et al., 1992), whereas the latter one forms EM associations (Gardes and Dahlberg, 1996). Three distinct geographic regions were selected for this study: Mayrhofen in the European Alps (alpine climate), Kilpisjärvi in Finnish Lapland (low arctic climate) and Ny-Ålesund in the Svalbard archipelago (high arctic climate) (Figure 1). We hypothesized that the bulk soil fungal communities are mainly influenced by geographic region and/or by soil physico-chemical properties, while the plant-associated ones are mainly driven by plant species due to their distinct mycorrhizal association. We also hypothesized that the diversity of the soil- and plant-associated fungal communities will decrease with increasing latitude. To the best of our knowledge, this is the first comprehensive comparative study of plant-associated fungal communities in High Arctic and Alpine climate zones.

Materials and Methods Sampling locations and sample collection Samples were collected from three distinct bioclimatic regions: Innsbruck, Austria (Alpine), Kilpisjärvi, Finland (low-arctic) and Ny-Ålesund, Svalbard

!!

83

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

(high-arctic). Kilpisjärvi (S, JO, JN) and Ny-Ålesund (K, M, RR) had three sampling sites each, while we had only two sites from Mayrhofen (A, C) region (Figure 1, Table 1). From each site, six replicates of top (1-5 cm depth) and bottom (15-20 cm depth) soil samples were collected along with six samples of both O. digyna and S. oppositifolia (with adhering rhizosphere soils). Both plant species O. digyna and S. oppositifolia have circumpolar distribution and are one of the early colonizers on glacier retreat soils. O. digyna has been reported to be non-mycorrhizal while S. oppositifolia is ecto/endo-mycorrhizal. Bulk and rhizosphere soils were processed and stored as specified by Kumar et al. (2016b). Plant roots were thoroughly washed after removing rhizosphere soil, and surface-sterilized by immersion in 3% sodium hypochlorite (3 min) and then in sterile double-distilled water (3 x 90 sec). After that, 80-100 mg of root samples were stored at -80˚C for DNA extraction. Soil physicochemical analyses (soil pH, soil organic matter (SOM) and available phosphorus (P)) were performed as described in Kumar et al. (2016a).

Figure 1 Map with information on three distinct sampling locations: One from the Alps (Mayrhofen, Austrian Alps) and one each from low- (Kilpisjärvi, Finnish Lapland) and high-Arctic (Ny-Ålesund, Svalbard archipelago

Fungal ITS gene library generation and sequencing Total community DNA was isolated from both soil and plant samples using MoBio Power soil kit (MoBio, Carlsbad, CA USA) and Invisorb Spin plant Midi Kit (STRATEC Biomedical AG, Germany), respectively with minor modifications to manufacturers protocol as detailed in Kumar et al., 2016a. Fungal internal transcribed spacer (ITS) gene libraries were amplified using fITS7 and ITS4 (Ihrmark et al., 2012) primer sets. 30 !l reaction mixture

!84

Cha

pter

4

contained 1 !l of template DNA, 1x PCR buffer, 0.2 mM dNTP’s, 0.3 !M of each primer and 3000 U/ml DreamTaq DNA Polymerase (Thermo Scientific, MA, USA). 1 !l of template had 5-10 and 25-30 ng of soil and endophyte DNA respectively. The amplification procedure consisted of 5 mins denaturation at 95°C followed by 35 cycles of denaturing, annealing, and extension at 95°C for 30 s, 55°C for 30 secs and 72°C for 1 min, respectively. Final extension was carried out at 72°C for 7 mins. The sequence libraries were prepared by attaching M-13 barcode system to the amplicons (Mäki et al., 2016). Amplicons from first PCR were diluted 1:10 and re-amplified using barcode attached M13 system as forward primer and ITS4-P1 with adaptor A as a reverse primer. PCR mix and conditions were similar as described above, with an exception of only using 8 cycles for amplification. Amplified products were purified with Agencourt AMPure XP PCR purification system (Beckman Coulter, CA, USA) and quantified using Qubit dsDNA HS assay kit and Qubit fluorometer (Invitrogen, MA, USA). The quantified samples from each sample were then pooled together and were sequenced in house using Ion 314 chip kit v2 BC on Ion Torrent PGM (Life Technologies, MA, USA). Table 1 Overview of the sampling sites, geographic location with latitudes along with soil physico-chemical properties and details of plant species sampled from respective site.

Geographic Region Sampling site Soil pH SOM** %

Available phosphorus mg/kg

Latitude

Plant species sampled*

Mayrhofen (Alpine)

Alps (A) 0.01 (0.002) 7.03 (0.9) 1.84 (1.1) 47.18 OD, SO Cliff (C) 0.02 (0.002) 4.60 (0.1) 1.48 (0.4) 47.18 OD, SO

Kilpisjärvi (low-Arctic)

Jehkas New (JN) 0.02 (0.002) 5.55 (0.2) 1.31 (0.4) 69.08 OD, SO Jehkas Old (JO) 0.02 (0.008) 6.36 (0.4) 0.76 (0.4) 69.08 OD, SO Saana (S) 0.02 (0.01) 5.49 (0.6) 2.45 (1.5) 69.05 OD

Ny-Ålesund (high-Arctic)

Knudsenheia (K) 0.03 (0.01) 7.4 (0.9) 0.83 (0.5) 78.90 OD, SO Midtre Lovénbreen (M)

0.03 (0.03) 6.4 (1.2) 0.63 (0.1) 78.90 OD, SO

Red River (RR) 0.04 (0.01) 7.78 (0.5) 0.34 (0.1) 78.93 OD, SO ( ) – Standard deviation values * OD – O. digyna, SO – S. oppositifolia **SOM: soil organic matter

Sequence data processing and bioinformatic analysis Raw sequence data from Ion torrent seqencing were processed using QIIME 1.8.0 (Caporaso et al., 2010) software and UPARSE (Edgar, 2013). After removal of primers and barcodes reads with quality score above 25 and 200 bp were selected for further processing. In addition, the fungi sequences from ITS1 region were extracted with ITSx (Bengtsson-Palme et al., 2013), screened for chimeric sequences (UNITE 7.0) and aligned at 97% identity using USEARCH algorithm (Edgar, 2010). UCLUST algorithm along with UNITE 7.0 database was used to assign taxonomies at 97% identity to the individual

!!

85

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

OTUs. For alpha diversity determination all the samples were rarefied (subsampled) to 540 reads per sample and for community structure and other analysis all the samples with more than the median reads were rarefied to the median (3700 reads), while the samples with less then 3700 reads were used as such (deCárcer et al., 2011). Statistical analyses The ordination plots were produced using a model-based ordination method introduced in Hui et al. (2015) and Warton et al. (2015) implemented in R software. The method fits a generalized linear latent variable model with two latent variables to the data and then produces an ordination plot based on the predicted bivariate latent variables. Here we assumed a negative binomial distribution for the OTUs due to the highly over dispersed data. Shannon diversity index and species richness were calculated using Univariate Diversity Indices (DIVERSE, PRIMER 6 (PRIMER-E Ltd)) and significance values were calculated using IBM SPSS. PRIMER 6 software was further used to determine the similarities between the samples (ANOSIM), to access the similarity percentage of dominating OTUs or bacteria at different taxonomic level (SIMPER).

Ternary plots were produced by calculating the mean relative abundances of OTUs from rhizosphere and endosphere compartments and plotted with the function ‘ternaryplot’ ‘vcd’ (Meyer et al., 2015) from the R package. Picking endosphere core OTUs The highly conserved OTUs (core OTUs) were manually picked by selecting OTUs that were constantly observed (present in at least 2 out of 3 replicates per site) in the endosphere of either O. digyna and S. oppositifolia or both. Results In total, we analyzed 138 samples from bulk and rhizosphere soils, as well as from the endosphere, and obtained 434,630 high quality reads. The reads were grouped into OTUs at the 97% level of similarity, after which OTUs that contained less than 30 reads were removed (less than 0.02% of the sequences). This quality trimming resulted in 422,132 sequences, which were subsequently grouped into 978 OTUs and used in further analyses. Among these 978 OTUs, 325 were region-specific, another 321 were shared between two of the three regions and the remaining 332 were present across all the regions (Figure 2a). With respect to the compartments (bulk soil, rhizosphere soil or endosphere),

!86

Cha

pter

4

soil (bulk and rhizosphere) fungi covered a large proportion of the diversity (616 of the 978 OTUs), with 126 OTUs being specific to bulk soils, 76 OTUs to the rhizosphere soils and 414 OTUs shared between bulk and rhizosphere compartments. In contrast, we observed only two OTUs (OTUs 3 and 1676), from the phylum Ascomycota, to be specific to the endosphere. The great majority of the OTUs associated with the endosphere of one or both plant species were also found in the other compartments (282 OTUs) (Figure 2b).

Figure 2 Venn diagram depicting the shared OTUs and number of sequences of the respective OTUs across different (a) geographic regions and (b) sample compartments

Taxonomic distribution of the reads Ascomycota was the dominant phylum across all samples, comprising about 65% of the total sequences (272,749 sequences, 567 OTUs). Among the Ascomycota, Leotiomycetes (143,953 sequences, 217 OTUs) stood out as the dominant order. The second dominant phylum was Basidiomycota (102,143 sequences, 254 OTUs), comprising 24% of the sequences. Here, the order Agaricomycetes (94,465 sequences, 216 OTUs) dominated. The remaining sequences (11% of the total) were from the Zygomycota (26,862 sequences, 33 OTUs), the Chytridiomycota (1,413 sequences, 8 OTUs), the Glomeromycota (377 sequences, one OTU) and unidentified fungi (18,588 sequences, 115 OTUs).

Remarkably, five OTUs were dominant in an overall sense, as they covered 15 % of the total sequences. Of these, four (OTUs 1,8,9 and 58) belonged to the Ascomycota and one (OTU 6) to the Basidiomycota. Four of these five OTUs were highly abundant in the endosphere (48,507 of 52,174 sequences), i.e. Phialocephala (OTU 9), Cadophora (OTU 8), and Varicosporium (OTU 1) from the order Helotiales, and Mycenella (Mycenella trachyspora, OTU 6) from the Agaricales. The remaining OTU from Geomyces spp. (Geomyces pannorum, OTU 58) of the class Leotiomycetes was prevalently observed in the soil (12,955 of 12,976 sequences) (Table S1).

!!

87

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

Rhizosphere or endosphere fungal community structures are not shaped by host plant species The species richness (SR) (13.40) and Shannon diversity (SD) (3.42) values of the O. digyna rhizosphere samples were significantly (p<0.01) higher than those of S. oppositifolia (SR = 10.92, SD = 2.90) rhizosphere samples, whereas we observed no such difference in the respective endosphere diversity and richness values (Table 2). However, the fungal community structures of both the rhizospheres and endospheres of the two plant species showed no host plant specificity based on the generalized linear latent variable models (GLLVM) that we applied (Figure 3a, 3b). Given that the host plant species did not have any statistically significant impact on fungal community structures, we decided in the following to consider the plant-associated communities together.

Figure 3 Generalized linear latent variable model plots of plant-associated fungal communities from three different geographic regions based on compartment (a) endosphere samples (b) rhizosphere samples and (c) rhizosphere and endosphere combined. In addition, OTUs which determine the dissimilarity between the geographic regions based on SIMPER data are marked in figure (a) and (b).

Compartment is the primary determinant of fungal diversity and community structure Two-way ANOVA (SPSS) analysis showed that both fungal richness (p<0.001) and diversity (p<0.001) were significantly different between the three compartments (Table 2). Across all regions, the rhizosphere communities revealed the highest richness and diversity levels, followed by the bulk soils and the endospheres (Table 2). We observed no significant differences in the richness or diversity levels of the total fungal communities across the three geographic regions (p=0.58) (Table 2). When each compartment was analyzed separately, no trends in the richness or diversity in the bulk or rhizosphere soil samples between the geographic regions were detectable, while, surprisingly,

!4 !2 0 2 4

!4!2

02

4

(a)

OTU_10

OTU_14

OTU_16

OTU_1609

OTU_2

OTU_23

OTU_24

OTU_3OTU_38

OTU_4

OTU_5

OTU_7

OTU_9

RegionKilpisjärviMayrhofenNy!Ålesund

Plant SpeciesO. digynaS. oppositifolia

!4 !2 0 2 4

!4!2

02

4(b)

lvR

2[,2

]

OTU_108

OTU_118

OTU_187

OTU_2OTU_22 OTU_293

OTU_30

OTU_40

OTU_44

OTU_461

OTU_62

OTU_81

!88

Cha

pter

4

we observed a significant decrease in the diversity values (p<0.001) of the endophytic fungal communities from Mayrhofen to Ny-Ålesund. Even though the Kilpisjärvi fungal communities were not significantly different from either the Mayrhofen or Ny-Ålesund communities, we observed a decrease in diversity of the endophytic fungal communities with increasing latitude. Consistent with the foregoing, we also observed a steady (albeit not significant) decrease in the richness values of the endophytic (but not soil) fungi with increasing latitude (Table 2). Table 2 Mean values of species richness, Pielou’s evenness and Shannon diversity index calculated using DIVERSE (PRIMER 6). In the analyses where significant differences were found, the sample groups that were not significantly different are denoted with the same letter. Significance was determined with Tukey’s post-hoc analysis (IBM SPSS).

Factors Species richness Pielou’s evenness Shannon index All Regions Mayrhofen 9.47 0.69 2.81 Kilpisjärvi 8.92 0.68 2.71 Ny-Ålesund 8.48 0.66 2.60 Compartment Bulk soils 10.62a 0.73a 3.08a Rhizosphere soils 12.29b 0.73a 3.18a Endosphere 3.54c 0.57b 1.77b Host specificity by Compartment Rhizosphere O. digyna 13.40a 0.77a 3.42a S. oppositifolia 10.92b 0.68b 2.90b Endosphere O. digyna 3.31 0.60 1.84 S. oppositifolia 4.36 0.54 1.78 Compartment by Region Bulk Soils Mayrhofen 10.54 0.75 3.15 Kilpisjärvi 11.10 0.75 3.21 Ny-Ålesund 10.18 0.69 2.91 Rhizosphere Soils Mayrhofen 12.91 0.73 3.24 Kilpisjärvi 11.84 0.70 3.01 Ny-Ålesund 11.97 0.75 3.25 Endosphere Mayrhofen 5.01 0.62 2.09a Kilpisjärvi 3.39 0.58 1.78ab Ny-Ålesund 3.29 0.53 1.63b

The analyses of the fungal community structures indicated that these were also primarily influenced by compartment, in particular the endosphere compartment. The endosphere communities grouped separately from the bulk and rhizosphere soil ones across all three geographic regions in the GLLVM ordination plots. In contrast, there was no conspicuous difference between the bulk and rhizosphere soil fungal communities (Figure 4). We detected only a

!!

89

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

weak influence of geographic region on the community structures when the fungal communities from all compartments were analyzed together (Figure 4).

Figure 4 Generalized linear latent variable model plots of fungal communities from all the samples across three geographic regions. The ordination plot based on the full data set is given with different symbols and colors corresponding to different regions and compartments, respectively.

Figure 5 Generalized linear latent variable model plots of bulk soil fungal communities with pH as a co-variate and the sites were colored according to corresponding soil parameter values.

Soil pH and geographic region influence fungal community structures We then analyzed the possible impact of local soil parameters on the bulk soil fungal community structures. The analyses revealed that these were mainly

!!

!

!

!

!

!

!

!

!!!! !

!

!!

!

!

!!

!

!

!

!

!

!!

!!

!

!

!

!

!

!

!3 !2 !1 0 1 2 3

!3!2

!10

12

3!

RegionKilpisjärviMayrhofenNy!Ålesund

CompartmentBottom soilTop soilRhizosphereEndosphere

!3 !2 !1 0 1 2 3

!3!2

!10

12

3 RegionKilpisjärviMayrhofenNy!Ålesund

56

78

pH

1

!90

Cha

pter

4

influenced by soil pH (Figure 5), while the levels of soil organic matter (SOM) and available P had no influence (data not shown). In contrast, as evidenced by the GLLVM based analyses, the rhizosphere fungal communities were grouped based on the geographic region, expectedly with no host plant specificity (Figure 3b). This was confirmed using two-way crossed analyses (ANOSIM) with geographic region and host plant species as factors: geographic region was the primary determinant of the rhizosphere fungal community structures (R = 0.71). Also the endosphere fungal communities were separated based on geographic region, although the impact of region was weaker than on the rhizosphere soil communities (Figure 3a). Corroborating this, ANOSIM analyses confirmed that geographic region (R = 0.38) was a stronger driver of the endophytic fungal community structure than plant species (R =0.11).

Figure 6 Histogram with relative abundance of major bulk soil fungal classes from (a) Ascomycota and (b) Basidiomycota with differences in soil pH.

!

!!

91

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

Major members of the fungal communities across the compartments Bulk soil communities Leotiomycetes (42,786 sequences, 190 OTUs), Eurotiomycetes (16,348 sequences, 93 OTUs) and Dothideomycetes (8,528 sequences, 67 OTUs) from the phylum Ascomycota, and Agaricomycetes (40,556 sequences, 169 OTUs) from the phylum Basidiomycota, were the most abundant fungal classes observed across the bulk soil communities. Among the Ascomycota classes, Dothideomycetes (Davidiellaceae, Phaeosphaeriaceae, Pseudoeurotiaceae and Sporormiaceae) were enriched in soils with near-neutral pH (6-7) from Kilpisjärvi, while Leotiomycetes (Geoglossaceae, Helotiaceae, Hyaloscyphaeceae) preferred the moderately to slightly acidic soils (4-6) of Mayrhofen and Eurotiomycetes (Verrucariaceae) were predominantly present in the alkaline soils (pH 8-9; Figure 6a) in Ny-Ålesund. Likewise, in the Basidiomycota, the orders Filobasidiaceae and Thelebolaceae were predominantly present in soils with pH below 7 (Mayrhofen and Kilpisjärvi), whereas Strophariaceae (from the same phylum) preferred the more alkaline soils from Ny-Ålesund (Figure 6b). Rhizosphere soil communities Similar to the bulk soils, Agaricomycetes (33,736 sequences, 173 OTUs) and Leotiomycetes (25,027 sequences, 179 OTUs) were abundantly present in the rhizosphere soils. This was followed by Eurotiomycetes (12,104 sequences, 88 OTUs). Moreover, Agaricomycetes from the phylum Basidiomycota were highly enriched in the two arctic regions. In particular, the Kilpisjärvi communities were enriched with the genera Tomentella and Clavaria, whereas the genera Inocybe and Cortinarius were abundant in the Ny-Ålesund communities. Also Mortierella spp. (phylum Zygomycota) was predominantly observed in the arctic regions. In contrast the alpine region was enriched with the phylum Ascomycota. The classes Verrucariaceae from Eurotiomycetes and the genera Tetracladium and Acicuseptoria, of the Leotiomycetes and Dothideomycetes, respectively, were dominant in the Mayrhofen rhizosphere soils (Figure 7a). Endosphere communities Leotiomycetes (76,140 sequences, 217 OTUs) represented about 52 % of the total endosphere-derived sequences. This was followed by Agaricomycetes (20,173 sequences, 216 OTUs) and Dothideomycetes (19,912 sequences, 67 OTUs). Even though most of the endosphere OTUs were shared between O. digyna and S. oppositifolia with no strong host specificity, we did observe some endemism with respect to the geographic region. Leotiomycetes were particularly enriched in

!92

Cha

pter

4

the two arctic sites. They were represented by the genera Varicosporium and Phialocephala in Kilpisjärvi and by Alatospora and Cadophora in Ny-Ålesund. In addition, Symphoventuriaceae (Family) of the Dothideomycetes were enriched in the Ny-Ålesund endophytic communities. In contrast, the genera Acicuseptoria and Herpotrichia (Dothideomycetes) were abundantly present in the Mayrhofen endospheres, followed by the genera Tetracladium and Cryptosporosis (class Leotiomycetes) (Figure 7b). In addition, Mycenella trachyspora (OTU 6) of the Agaricomycetes was predominantly observed in the Mayrhofen and Kilpisjärvi communities.

Figure 7 Ternary plots of all OTUs from (a) rhizosphere samples and (b) endosphere samples plotted based on distribution across three climatic regions. Each circle represents one OTU. The size, color and position of OTUs represents the relative abundance, fungal class and contribution of these to the nearby region respectively.

Irrespective of the influence of geographic region on the root-

associated (rhizosphere and endosphere) fungal community structures, we detected minimal overlap between the rhizosphere and endosphere fungal communities from the same geographic region. In short, the plants harbored different rhizosphere and endosphere communities in the same region. Using SIMPER, we detected 20 key OTUs, each from rhizosphere and endosphere (separately) that determined the difference between geographic region in the rhizosphere and endosphere fungal communities. Only the Mayrhofen communities had three OTUs (OTU 23, OTU 18 and OTU 36) that were shared between the rhizosphere and endosphere compartments. In contrast, the Kilpisjärvi and Ny-Ålesund rhizosphere and endosphere communities had no shared OTUs that determined regional influence. Finally, one OTU (OTU 15) from Mayrhofen rhizosphere was enriched in the Ny-Ålesund endosphere community (Table S2). No major plant species effect, but plant species specific fungi? Most of the plant-associated OTUs (852 OTUs) were only detected in the rhizosphere (490 OTUs) as compared to the endosphere-‘only’ (22) or OTUs

!!

93

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

present in both of plant associated compartments (340). However, the shared OTUs comprised about 80% of the total sequences, with dominance in the endosphere compartment (66% of the sequences) (Table 3). Furthermore, both the rhizosphere and endosphere OTUs shared about 90% of their sequences between the two plant species, with host-specific OTUs only accounting for 11% (endosphere) and 10% (rhizosphere) of the sequences (Table 3). Even though the fungal community structures of O. digyna or S. oppositifolia in the rhizospheres and endospheres were highly similar between the plant species, we identified a few OTUs and genera as potentially host plant specific. Two OTUs (Varicosporium, OTU 1 and Cryptosporiopsis, OTU 26) from the order Helotiales , Ascomycota, were enriched in the O. digyna endosphere, while two other OTUs (unidentified Sympoventuriaceae , order Venturiales [OTU 5] and genus Phialocephala, order Helotiales, [OTU 9]) belonging to the Ascomycota, were predominantly present in the S. oppositifolia endosphere. The rhizosphere communities of O. digyna were enriched with the genera Acicuseptoria and Herpotrichia (Pleosporales) and Tetracladium (Helotiales) from the phylum Ascomycota, while the S. oppositifolia rhizosphere communities were dominated by the genera Inocybe and Cortinarius (Agaricales) of phylum Basidiomycota.

Table 3 Overview of the distribution of compartment (rhizosphere and endosphere) and host-plant-specific (O. digyna and S. oppositifolia) OTUs from rhizosphere and endosphere samples.

Compartment/Plant species No. of OTUs No. of Sequences % of sequences Plant-associated 852 267,141 Rhizosphere specific 490 47,443 17.8% Endosphere specific 22 3,711 1.40% Shared 340 215,987 80.8% Rhizosphere 830 120,725 O. digyna specific 168 7761 6.43% S. oppositifolia specific 114 6448 5.34% Shared 548 106,516 88.3% Endosphere 352 146,416 O. digyna specific 81 7,145 4.88% S. oppositifolia specific 120 8,597 5.87% Shared 161 130,674 89.3%

Fungal OTUs highly conserved in the endospheres of O. digyna and S. oppositifolia in all three regions We considered OTUs from the endospheres that were consistently present in at least two of the three replicates per site from all three geographic regions as representing the fungal “core” endosphere communities. In total, three OTUs, all belonging to the phylum Ascomycota, were identified as the fungal core

!94

Cha

pter

4

community - one (OTU 26) being consistently present in O. digyna and two (OTU 8 and 18) in S. oppositifolia. OTU 26 (genus Cryptosporiopsis) was observed in all O. digyna samples, while OTU 8 (genus Cadophora) and OTU 18 (Leptodontidium) formed the core of the S. oppositifolia endophyte community. In addition, these three OTUs formed a major part of the endosphere sequences by covering almost 15% (sequences from both plants) of total endosphere sequences. Discussion Plant species has little effect on the endosphere or rhizosphere fungal community structure In a recent study on fungal endophytes from the high-Arctic, Zhang et al. (2015) reported host plant specificity across four different arctic plants, including S. oppositifolia from Ny-Ålesund, Svalbard. In contrast, we observed no clear host plant specificity in either O. digyna or S. oppositifolia endophytes on the basis of our GLLVM plots. This finding is also different from those with respect to the bacterial endophytic communities from the same samples, where we did observe host plant specificity (Kumar et al., 2016a). The lack of an overall effect of plant species was surprising, but may be explained by the following. O. digyna and S. oppositifolia, addressed in this study, prefer similar habitats, and are often found in close proximity to each other, while the plant species studied by Zhang et al. (2015) clearly represent different arctic habitats. In particular, the endosphere fungal communities of the two Saxifragaceae species S. oppositifolia and S. cespitosa showed the highest similarities between them when compared with two other plant species, i.e. Cassiopea tetragona from the Ericaceae and Silene acaulis of the Caryophyllaceae. This might also indicate that soil edaphic factors are prime determinants of the soil fungal communities, with the endophytic fungi being selected from the available soil communities (Compant et al., 2010). Furthermore, O. digyna and S. oppositifolia might secrete similar compounds, such as organic acids and amino acids, from the roots, which alter the pH of the surrounding soils alike and thus attract almost similar fungal communities from the latter. Unfortunately, we currently lack information as to the compounds released by the two plant species that presumably are perceived as attractants by the soil fungi. Alternatively, the fungi found may have developed a strong and tight, and possibly symbiotic, relationship with the two plant species, and such relationships may not be strongly host plant specific. Geographic region strongly determines plant-associated fungal community structures In our previous study on the structure of bacterial communities from the same samples studied here, we found a strong relationship between the bulk soil

!!

95

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

communities and geographic region. The influence of region diminished with the progressive tightening of the association with plants (Kumar et al., 2016a). In contrast, in the current study we - to our surprise - observed a very strong influence of geographic region on the plant-associated fungal communities, as opposed to a much looser effect of region on the bulk soil communities. Observations on fungal communities associated with Agave sp. from low nutrient soils by Coleman-Derr et al. (2016) corroborate our findings. They also reported fungal communities from Agave sp. to be more impacted by geographic distance than by plant species, in contrast to plant associated bacterial communities. They further hypothesized that dispersal limitation of fungi (as compared to bacteria) might be at the basis of fungal endemism in certain habitats. In contrast, Cox et al. (2016) relates strong dispersal capability and adaptation to cold nutrient-poor habitats as a reason for the ubiquitous dispersal of similar soil fungal communities in the Arctic and Antarctica. It should be noted that both studies were conducted in different habitats and may be the reason for the contention that fungal endemism is not straightforward. A detailed study on background environmental and soil edaphic factors may be needed for further conclusion. According to Taylor et al. (2006), fungi, being multicellular and eukaryotic, may behave more like plants than bacteria in this regard. Another hypothesis is that this effect is due to the effect of plant root exudates on the conditions (e.g. pH) of the rhizosphere soils. In our previous study at the same sites, we observed no significant difference between the pH values of the O. digyna and S. oppositifolia rhizosphere soils (Kumar et al., 2016b). Instead, the pH of the rhizosphere soils of the same plant species differed between the geographic regions, i.e: Kilpisjärvi and Longyearbyen, Svalbard (Kumar et al., 2016b).

The EM fungi Tomentella, Inocybe and Cortinarius, along with the saprotrophic Clavaria, have been reported to be the most common root-associated fungi in the Arctic (Blaalid et al., 2012b, 2014; Geml et al., 2012; Mundra et al., 2015). We also observed these EM fungi to be associated with the rhizosphere of S. oppositifolia (known as an EM plant) in our arctic sites, versus the saprotrophic Clavaria in the rhizosphere samples of O. digyna (known as a non-mycorrhizal plant) from the same sites. Tetracladium, which was selectively enriched in our Mayrhofen rhizosphere soils, has been reported to be an endophytic or plant-associated mycobiont in alpine environments (Lv et al., 2010; Mühlmann and Peintner, 2008). Fungal diversity is highest in the rhizosphere as compared to bulk soil and endosphere The finding that the rhizosphere soil fungal communities were more diverse and richer than those in the corresponding bulk soil (or endosphere) was

!96

Cha

pter

4

consistent with other findings on fungal (as well as bacterial) diversities in low-organic-matter soils (Coleman-Derr et al., 2016; Kumar et al., 2016b; Miniaci et al., 2007; Yergeau et al., 2007)(Coleman-Derr et al., 2016; Kumar et al., 2016b; Miniaci et al., 2007; Yergeau et al., 2007). In contrast, this is not always the case with agricultural soils (!nceo"lu et al., 2011; Kowalchuk et al., 2002; Smalla et al., 2001). These findings indicate that natural plants may offer propicious broad nutritional niches in the low-nutrient soils, allowing the persistence of a broad diversity of fungi. Furthermore, the observed decrease in the richness and diversity values of the endosphere fungal communities from the south (Alpine) to the north (Arctic) was consistent with data by Blaalid et al. (2014). They found the richness of fungal communities associated with the roots of B. vivipara to decrease with increasing latitude in the northern hemisphere, attributing this trend to environmental and soil edaphic factors such as temperature, nutrient availability, precipitation and soil pH. Furthermore, we can also relate the decrease in root-associated fungal diversity to the decrease in plant diversity with increasing latitude, since plants may need particular root-associated fungi for efficient nutrient collection in the low nutrient soils studied. Influence of soil pH on the bulk soil fungal communities Corroborating our results, Zhang et al. (2016) recently found soil pH to be a key determining factor that drives soil fungal communities in the high Arctic. The genera Polyblastia and Verrucaria, of the order Verrucariaceae, were often detected in our high alkaline Ny-Ålesund bulk soil samples. Verrucariaceae are common lichen-forming fungi which are reported to prefer recently deglaciated soils with high pH (8-9) (Zhang et al., 2016). Leotiomycetes, especially Helotiales (family) were present in bulk soils with pH 5-6 and were also enriched in our plant-associated fungal communities. Helotiales have been reported to prefer moderately acidic soils (pH 5-6), with their relative abundance decreasing with increasing pH (Rousk et al., 2010). Most members of the Leotiomycetes and Helotiales may act as EM fungi and be dominant in root-associated communities in arcto-alpine plants (Mühlmann and Peintner, 2008; Walker et al., 2011). They have been found to be specifically enriched in plants of the Saxifragaceae family (Zhang et al., 2015). Yamanaka (2003) reported that many EM fungal species grow well at pH 5 to 6, which was the pH of the O. digyna and S. oppositifolia rhizosphere soils, as described earlier (Kumar et al., 2016b).

Irrespective of soil pH, the classes Leotiomycetes in the Ascomycota and Agaricomycetes in the Basidiomycota were found to be more dominant in bulk soils than other phyla (i.e. Chytridiomycota, Glomeromycota and Zygomycota), as also found by Gittel et al. (2014) and Zhang et al. (2016). Here, the low abundance

!!

97

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

of Glomeromycota (one OTU) in our analyses is related to our primers that tend to exclude most members of the Glomeromycota. Moreover, our soil fungal communities were dominated by the psychrotropic fungus G. pannorum (OTU 58), especially in our Ny-Ålesund sites. The frequent isolation of G. pannorum from high arctic and antarctic soils reveals a circumpolar distribution (Cox et al., 2016; Gesheva and Negoita, 2011; Gonçalves et al., 2012; Timling and Taylor, 2012; Tosi et al., 2002). Robinson (2001), in his review, hypothesizes that G. pannorum is well adapted to cold habitats and might even be transported by migrating birds between the poles. Plant-associated fungal communities - Mycorrhizal association as a factor? Even though we observed low host plant specificity across the board, we did find some genera or OTUs to be specific to O. digyna and S. oppositifolia in their respective rhizosphere/endosphere samples. These may be attributed to potential mycorrhizal association of these plants: the S. oppositifolia root-associated EM fungal communities were dominated by Inocybe and Cortinarius (rhizosphere) and dark septate endophytes Leptodontidium and Phialocephala (endosphere). However, the root-associated communities of the non-mycorrhizal O. digyna were selectively enriched with the reportedly saprotrophic fungi Clavaria (rhizosphere) (Timling et al., 2014) and Varicosporium (endosphere). Leptodontidium and Phialocephala have been also reported as being part of the root-associated fungal community of the EM plant B. vivipara from Ny-Ålesund (Mundra et al., 2015). Cadophora and Leptodontidium which made part of the core community of S. oppositifolia were also reported to be abundant in D. octopetala, another EM forming plant from a similar habitat (Bjorbækmo et al., 2010). Cryptosporosis from the O. digyna core is known as an endophyte of ericoid plants in the Arctic (Walker et al., 2011). Interestingly, Cryptosporopsis quercina, isolated from Triptergyium wilfordii roots synthesizes the antimycotic tetramic acid cryptocin, with antifungal activity against a wide variety of plant pathogens (Li et al., 2000). Apart from cryptocin, C. quercina has been reported to produce several different anti-mycotics (Strobel et al., 1999) and it might act as a symbiotic endophyte that provides protection against various plant pathogens (Li et al., 2000). Thus, any host plant specificity we observe in relation with O. digyna and S. oppositifolia should be seen in relation to their differential mycorrhizal states.

The genus Mortierella (phylum Zygomycota), which was highly enriched in our rhizosphere samples, has also been reported to be abundant in the rhizosphere soils of several temperate plants (Curlevski et al., 2010; Gomes et al., 2003; Smit et al., 1999). In addition to rhizosphere soils, Mortierella has been

!98

Cha

pter

4

isolated from arctic, alpine and antarctic soils and is one of the most dominant fungal taxa isolated from Antarctic lakes (Gonçalves et al., 2012). Thus, it seems to have a very wide distribution range. They were also isolated as endophytes from Antarctic mosses (Tosi et al., 2002) and a GC-MS analysis of crude extracts revealed that they harbor several antimicrobial compounds (Melo et al., 2014). Dothideomycetes which were enriched in our root-associated communities have been reported to have genes for several small secreted proteins (SSPs), which play an important role in fungal-plant interactions. Fifteen of the 18 Dothideomycetes included in the study were pathogenic (Ohm et al, 2012). However, often such potential pathogens do not cause disease and the difference between so-called pathogens and endophytes is not well-defined and so local conditions may define the outcome of such interactions (Ohm et al., 2012).

In summary, compartment (bulk soil, rhizosphere soil, endosphere) was determined to be the primary driver of the fungal communities in three geographic regions representing an arcto-alpine climate. Further, the endosphere fungal community structures and diversities in two selected plant species were strikingly different from those of the soil communities. Furthermore, the endosphere community diversity and richness values decreased progressively with increasing latitude. Although the structures of the plant-associated fungal communities were primarily influenced by geographic region, we observed differential affiliation with several fungal genera or OTUs with one of the plant species in this study, O. digyna and S. oppositifolia. Funding This research was funded by NWO project 821.01.005 (for JE), Finnish cultural foundation Lapland regional fund (for RN) and by Finnish Academy, project 259180 (for RN) Acknowledgements Authors acknowledge Maarten Loonen and other members of Netherlands Arctic Station, Ny-Ålesund, Svalbard and Kilpisjärvi Biological Station of the University of Helsinki for assistance in sampling. We also thank Jolanda Brons for her assistance in soil physico-chemical analysis. References

Allan, T., Lim, S. R., Berch, S., and Berbee, M. (2005). Two new Cryptosporiopsis species from roots of ericaceous hosts in western North America. Studies in Mycology 53, 53–62. doi:10.3114/sim.53.1.53.

Anderson, I. C., and Cairney, J. W. G. (2004). Diversity and ecology of soil fungal

!!

99

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

communities: increased understanding through the application of molecular techniques. Environmental microbiology 6, 769–79. doi:10.1111/j.1462-2920.2004.00675.x.

Becklin, K. M., Hertweck, K. L., and Jumpponen, A. (2012). Host Identity Impacts Rhizosphere Fungal Communities Associated with Three Alpine Plant Species. Microbial Ecology 63, 682–693. doi:10.1007/s00248-011-9968-7.

Bengtsson-Palme, J., Ryberg, M., Hartmann, M., Branco, S., Wang, Z., Godhe, A., et al. (2013). Improved software detection and extraction of ITS1 and ITS2 from ribosomal ITS sequences of fungi and other eukaryotes for analysis of environmental sequencing data. Methods in Ecology and Evolution 4, 914–919. doi:10.1111/2041-210X.12073.

Bjorbækmo, M. F. M., Carlsen, T., Brysting, A., Vrålstad, T., Høiland, K., Ugland, K. I., et al. (2010). High diversity of root associated fungi in both alpine and arctic Dryas octopetala. BMC plant biology 10, 244. doi:10.1186/1471-2229-10-244.

Blaalid, R., Carlsen, T., Kumar, S., Halvorsen, R., Ugland, K. I., Fontana, G., et al. (2012a). Changes in the root-associated fungal communities along a primary succession gradient analysed by 454 pyrosequencing. Molecular Ecology 21, 1897–1908. doi:10.1111/j.1365-294X.2011.05214.x.

Blaalid, R., Carlsen, T., Kumar, S., Halvorsen, R., Ugland, K. I., Fontana, G., et al. (2012b). Changes in the root-associated fungal communities along a primary succession gradient analysed by 454 pyrosequencing. Molecular Ecology 21, 1897–1908. doi:10.1111/j.1365-294X.2011.05214.x.

Blaalid, R., Davey, M. L., Kauserud, H., Carlsen, T., Halvorsen, R., Høiland, K., et al. (2014). Arctic root-associated fungal community composition reflects environmental filtering. Molecular Ecology 23, 649–659. doi:10.1111/mec.12622.

Botnen, S., Vik, U., Carlsen, T., Eidesen, P. B., Davey, M. L., and Kauserud, H. (2014). Low host specificity of root-associated fungi at an Arctic site. Molecular Ecology 23, 975–985. doi:10.1111/mec.12646.

Caporaso, J. G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F. D., Costello, E. K., et al. (2010). QIIME allows analysis of high-throughput community sequencing data. Nature methods 7, 335–6. doi:10.1038/nmeth.f.303.

Coleman-Derr, D., Desgarennes, D., Fonseca-Garcia, C., Gross, S., Clingenpeel, S., Woyke, T., et al. (2016). Plant compartment and biogeography affect microbiome composition in cultivated and native Agave species. New Phytologist 209, 798–811. doi:10.1111/nph.13697.

Compant, S., Clément, C., and Sessitsch, A. (2010). Plant growth-promoting bacteria in the rhizo- and endosphere of plants: Their role, colonization, mechanisms involved and prospects for utilization. Soil Biology and Biochemistry 42, 669–678. doi:10.1016/j.soilbio.2009.11.024.

Cox, F., Newsham, K. K., Bol, R., Dungait, J. A. J., and Robinson, C. H. (2016). Not poles apart: Antarctic soil fungal communities show similarities to those of the distant Arctic. Ecology Letters 19, 528–536. doi:10.1111/ele.12587.

Curlevski, N. J. A., Xu, Z. H., Anderson, I. C., and Cairney, J. W. G. (2010). Diversity of soil and rhizosphere fungi under Araucaria bidwillii (Bunya pine) at an Australian tropical montane rainforest site. Fungal Diversity 40, 12–22. doi:10.1007/s13225-009-0001-0.

deCárcer, D. A., Denman, S. E., McSweeney, C., and Morrison, M. (2011). Evaluation of subsampling-based normalization strategies for tagged high-throughput sequencing data sets from gut microbiomes. Applied and Environmental Microbiology 77, 8795–8798. doi:10.1128/AEM.05491-11.

Edgar, R. C. (2010). Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461. doi:10.1093/bioinformatics/btq461.

!100

Cha

pter

4

Edgar, R. C. (2013). UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nature Methods 10, 996–998. doi:10.1038/nmeth.2604.

Fisher, P. J., Graj, F., Petrini, L. E., Sutton, B. C., and Wookey, P. A. (1995). Fungal endophytes of Dryas octopetala from a high arctic polar semidesert and from the Swiss Alps. Mycologia 87, 319–323. doi:10.2307/3760828.

Gardes, M., and Dahlberg, A. (1996). Mycorrhizal diversity in arctic and alpine tundra: an open question. New Phytologist 133, 147–157. doi:10.1111/j.1469-8137.1996.tb04350.x.

Geml, J., Timling, I., Robinson, C. H., Lennon, N., Nusbaum, H. C., Brochmann, C., et al. (2012). An arctic community of symbiotic fungi assembled by long-distance dispersers: phylogenetic diversity of ectomycorrhizal basidiomycetes in Svalbard based on soil and sporocarp DNA. Journal of Biogeography 39, 74–88. doi:10.1111/j.1365-2699.2011.02588.x.

Gesheva, V., and Negoita, T. (2011). Psychrotrophic microorganism communities in soils of Haswell Island, Antarctica, and their biosynthetic potential. Polar Biology 35, 291–297. doi:10.1007/s00300-011-1052-8.

Gittel, A., Bárta, J., Kohoutová, I., Mikutta, R., Owens, S., Gilbert, J., et al. (2014). Distinct microbial communities associated with buried soils in the Siberian tundra. The ISME journal 8, 841–53. doi:10.1038/ismej.2013.219.

Gomes, N. C. M., Fagbola, O., Costa, R., Rumjanek, N. G., Buchner, A., Mendona-Hagler, L., et al. (2003). Dynamics of Fungal Communities in Bulk and Maize Rhizosphere Soil in the Tropics. Applied and Environmental Microbiology 69, 3758–3766. doi:10.1128/AEM.69.7.3758-3766.2003.

Gonçalves, V. N., Vaz, A. B. M., Rosa, C. a, and Rosa, L. H. (2012). Diversity and distribution of fungal communities in lakes of Antarctica. FEMS microbiology ecology 82, 459–71. doi:10.1111/j.1574-6941.2012.01424.x.

Higgins, K. L., Arnold, A. E., Miadlikowska, J., Sarvate, S. D., and Lutzoni, F. (2007). Phylogenetic relationships, host affinity, and geographic structure of boreal and arctic endophytes from three major plant lineages. Molecular Phylogenetics and Evolution 42, 543–555. doi:10.1016/j.ympev.2006.07.012.

Hobbie, J. E., and Hobbie, E. A. (2006). 15 N in Symbiotic Fungi and Plants Estimates Nitrogen and Carbon Flux Rates in Arctic Tundra. Ecology 87, 816–822. doi:10.1890/0012-9658(2006)87[816:NISFAP]2.0.CO;2.

Hui, F. K. C., Taskinen, S., Pledger, S., Foster, S. D., and Warton, D. I. (2015). Model-based approaches to unconstrained ordination. Methods in Ecology and Evolution 6, 399–411. doi:10.1111/2041-210X.12236.

Ihrmark, K., Bödeker, I. T. M., Cruz-Martinez, K., Friberg, H., Kubartova, A., Schenck, J., et al. (2012). New primers to amplify the fungal ITS2 region--evaluation by 454-sequencing of artificial and natural communities. FEMS microbiology ecology 82, 666–77. doi:10.1111/j.1574-6941.2012.01437.x.

!nceo"lu, Ö., Al-Soud, W. A., Salles, J. F., Semenov, A. V., and van Elsas, J. D. (2011). Comparative Analysis of Bacterial Communities in a Potato Field as Determined by Pyrosequencing. PLoS ONE 6, e23321. doi:10.1371/journal.pone.0023321.

Koske, R. (1987). Distribution of VA mycorrhizal fungi along a latitudinal temperature gradient. Mycologia 79, 55–68. doi:10.2307/3807744.

Kowalchuk, G. A., Buma, D. S., de Boer, W., Klinkhamer, P. G. L., and van Veen, J. A. (2002). Effects of above-ground plant species composition and diversity on the diversity of soil-borne microorganisms. International Journal of General and Molecular Microbiology 81, 509–520. doi:10.1023/A:1020565523615.

Kumar, M., Brader, G., Sessitsch, A., Mäki, A., van Elsas, J. D., and Nissinen, R. M. (2016a). Plants Assemble Specific Bacterial Communities from Common Core Taxa in Three Arcto-Alpine Climate Zones. Frontiers in Microbiology, In review.

!!

101

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

Kumar, M., Männistö, M. K., van Elsas, J. D., and Nissinen, R. M. (2016b). Plants impact structure and function of bacterial communities in Arctic soils. Plant and Soil 399, 319–332. doi:10.1007/s11104-015-2702-3.

Kytöviita, M.-M. (2005). Asymmetric symbiont adaptation to Arctic conditions could explain why high Arctic plants are non-mycorrhizal. FEMS microbiology ecology 53, 27–32. doi:10.1016/j.femsec.2004.09.014.

Kytöviita, M.-M., and Ruotsalainen, A. L. (2007). Mycorrhizal benefit in two low arctic herbs increases with increasing temperature. American journal of botany 94, 1309–1315. doi:10.3732/ajb.94.8.1309.

Li, J. Y., Strobel, G., Harper, J., Lobkovsky, E., and Clardy, J. (2000). Cryptocin, a potent tetramic acid antimycotic from the endophytic fungus Cryptosporiopsis cf. quercina. Organic letters 2, 767–770. doi:10.1002/chin.200027221.

Lv, Y., Zhang, F., Chen, J., Cui, J., Xing, Y., Li, X., et al. (2010). Diversity and Antimicrobial Activity of Endophytic Fungi Associated with the Alpine Plant Saussurea involucrata. Biological & Pharmaceutical Bulletin 33, 1300–1306. doi:10.1248/bpb.33.1300.

Mäki, A., Rissanen, J. A., and Tiirola, M. (2016). A practical method for barcoding and size-trimming PCR templates for amplicon sequencing. Biotechniques 60, 88–90. doi:10.2144/000114380.

Marshall, W. A. (1998). Aerial Transport of Keratinaceous Substrate and Distribution of the Fungus Geomyces pannorum in Antarctic Soils. Microbial Ecology 36, 212–219. doi:10.1007/s002489900108.

Melo, I. S., Santos, S. N., Rosa, L. H., Parma, M. M., Silva, L. J., Queiroz, S. C. N., et al. (2014). Isolation and biological activities of an endophytic Mortierella alpina strain from the Antarctic moss Schistidium antarctici. Extremophiles 18, 15–23. doi:10.1007/s00792-013-0588-7.

Meyer, D., Zeileis, A., and Hornik, K. (2015). vcd: Visualizing Categorical Data. R package version. 1.4–1. Available at: https://cran.r-project.org/web/packages/vcd/citation.html.

Miniaci, C., Bunge, M., Duc, L., Edwards, I., Bürgmann, H., and Zeyer, J. (2007). Effects of pioneering plants on microbial structures and functions in a glacier forefield. Biology and Fertility of Soils 44, 289–297. doi:10.1007/s00374-007-0203-0.

Mühlmann, O., and Peintner, U. (2008). Mycobionts of Salix herbacea on a glacier forefront in the Austrian Alps. Mycorrhiza 18, 171–180. doi:10.1007/s00572-008-0169-2.

Mundra, S., Halvorsen, R., Kauserud, H., Müller, E., Vik, U., and Eidesen, P. B. (2015). Arctic fungal communities associated with roots of Bistorta vivipara do not respond to the same fine-scale edaphic gradients as the aboveground vegetation. New Phytologist 205, 1587–1597. doi:10.1111/nph.13216.

Ohm, R. A., Feau, N., Henrissat, B., Schoch, C. L., Horwitz, B. A., Barry, K. W., et al. (2012). Diverse Lifestyles and Strategies of Plant Pathogenesis Encoded in the Genomes of Eighteen Dothideomycetes Fungi. PLoS Pathogens 8, e1003037. doi:10.1371/journal.ppat.1003037.

Pietikäinen, A., Kytöviita, M.-M., Husband, R., and Young, J. P. W. (2007). Diversity and persistence of arbuscular mycorrhizas in a low-Arctic meadow habitat. The New phytologist 176, 691–8. doi:10.1111/j.1469-8137.2007.02209.x.

Robinson, C. H. (2001). Cold adaptation in Arctic and Antarctic fungi. New Phytologist 151, 341–353. doi:10.1046/j.1469-8137.2001.00177.x.

Rousk, J., Bååth, E., Brookes, P. C., Lauber, C. L., Lozupone, C., Caporaso, J. G., et al. (2010). Soil bacterial and fungal communities across a pH gradient in an arable soil. The ISME journal 4, 1340–51. doi:10.1038/ismej.2010.58.

Ryberg, M., Andreasen, M., and Björk, R. G. (2011). Weak habitat specificity in

!102

Cha

pter

4

ectomycorrhizal communities associated with Salix herbacea and Salix polaris in alpine tundra. Mycorrhiza 21, 289–296. doi:10.1007/s00572-010-0335-1.

Ryberg, M., Larsson, E., and Molau, U. (2009). Ectomycorrhizal Diversity on Dryas octopetala and Salix reticulata in an Alpine Cliff Ecosystem. Arctic, Antarctic, and Alpine Research 41, 506–514. doi:10.1657/1938-4246-41.4.506.

Smalla, K., Wieland, G., Buchner, A., Zock, A., Parzy, J., Roskot, N., et al. (2001). Bulk and Rhizosphere Soil Bacterial Communities Studied by Denaturing Gradient Gel Electrophoresis#: Plant-Dependent Enrichment and Seasonal Shifts Revealed. Applied and environmental microbiology 67, 4742–4751. doi:10.1128/AEM.67.10.4742.

Smit, E., Leeflang, P., Glandorf, B., van Elsas, J. D., and Wernars, K. (1999). Analysis of fungal diversity in the wheat rhizosphere by sequencing of cloned PCR-amplified genes encoding 18S rRNA and temperature gradient gel electrophoresis. Applied and environmental microbiology 65, 2614–21.

Strobel, G. A., Miller, R. V., Martinez-Miller, C., Condron, M. M., Teplow, D. B., and Hess, W. M. (1999). Cryptocandin, a potent antimycotic from the endophytic fungus Cryptosporiopsis cf. quercina. Microbiology 145, 1919–1926. doi:10.1099/13500872-145-8-1919.

Taylor, J. W., Turner, E., Townsend, J. P., Dettman, J. R., Jacobson, D., Adam, L., et al. (2006). Eukaryotic microbes, species recognition and the geographic limits of species: examples from the kingdom Fungi. Philosophical transactions of the Royal Society of London. Series B, Biological sciences 361, 1947–63. doi:10.1098/rstb.2006.1923.

Timling, I., Dahlberg, A., Walker, D. A., Gardes, M., Charcosset, J. Y., Welker, J. M., et al. (2012). Distribution and drivers of ectomycorrhizal fungal communities across the North American Arctic. Ecosphere 3, art111. doi:10.1890/ES12-00217.1.

Timling, I., and Taylor, D. L. (2012). Peeking through a frosty window: molecular insights into the ecology of Arctic soil fungi. Fungal Ecology 5, 419–429. doi:10.1016/j.funeco.2012.01.009.

Timling, I., Walker, D. a, Nusbaum, C., Lennon, N. J., and Taylor, D. L. (2014). Rich and cold: diversity, distribution and drivers of fungal communities in patterned-ground ecosystems of the North American Arctic. Molecular ecology 23, 3258–72. doi:10.1111/mec.12743.

Tosi, S., Begonã, A., Ae, C., Gerdol, R., and Caretta, G. (2002). Fungi isolated from Antarctic mosses. Polar Biology 25, 262–268. doi:10.1007/s00300-001-0337-8.

Väre, H., Vestberg, M., and Eurola, S. (1992). Mycorrhiza and root-associated fungi in Spitsbergen. Mycorrhiza 1, 93–104.

Walker, J. F., Aldrich-Wolfe, L., Riffel, A., Barbare, H., Simpson, N. B., Trowbridge, J., et al. (2011). Diverse Helotiales associated with the roots of three species of Arctic Ericaceae provide no evidence for host specificity. New Phytologist 191, 515–527. doi:10.1111/j.1469-8137.2011.03703.x.

Warton, D. I., Blanchet, F. G., O’Hara, R. B., Ovaskainen, O., Taskinen, S., Walker, S. C., et al. (2015). So Many Variables: Joint Modeling in Community Ecology. Trends in Ecology & Evolution 30, 766–779. doi:10.1016/j.tree.2015.09.007.

Yamanaka, T. (2003). The effect of pH on the growth of saprotrophic and ectomycorrhizal ammonia fungi in vitro. Mycologia 95, 584–589.

Yao, F., Vik, U., Brysting, A. K., Carlsen, T., Halvorsen, R., and Kauserud, H. (2013). Substantial compositional turnover of fungal communities in an alpine ridge-to-snowbed gradient. Molecular Ecology 22, 5040–5052. doi:10.1111/mec.12437.

Yergeau, E., Newsham, K. K., Pearce, D. A., and Kowalchuk, G. A. (2007). Patterns of bacterial diversity across a range of Antarctic terrestrial habitats. Environmental microbiology 9, 2670–82. doi:10.1111/j.1462-2920.2007.01379.x.

!!

103

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

Zhang, T., Wang, N.-F., Liu, H.-Y., Zhang, Y.-Q., and Yu, L.-Y. (2016). Soil pH is a Key Determinant of Soil Fungal Community Composition in the Ny-Ålesund Region, Svalbard (High Arctic). Frontiers in microbiology 7, 227. doi:10.3389/fmicb.2016.00227.

Zhang, T., Yao, Y.-F., Bergero, R., Girlanda, M., Varese, G., Intili, D., et al. (2015). Endophytic Fungal Communities Associated with Vascular Plants in the High Arctic Zone Are Highly Diverse and Host-Plant Specific. PLOS ONE 10, e0130051. doi:10.1371/journal.pone.0130051.

!104

Cha

pter

4

!!! !

Tab

le S

1 Li

st o

f 5 m

ost d

omin

ant O

TU

s fo

und

in a

ll ou

r sa

mpl

es li

sted

with

taxo

nom

ic in

form

atio

n.

OT

U #

Ph

yla

Cla

ss

Ord

er

Fam

ily

Gen

us

Spec

ies

Tot

al

sequ

ence

s E

ndos

pher

e se

quen

ces

Rhi

zosp

here

se

quen

ces

Bul

k so

il se

quen

ces

OT

U_9

A

scom

ycot

a Le

otio

myc

etes

H

elot

iale

s U

n_H

elot

iale

s Ph

ialoc

epha

la

uncu

lture

d_Ph

ialoc

epha

la

1034

8 97

81

325

242

OT

U_6

B

asid

iom

ycot

a A

garic

omyc

etes

Aga

rical

es

Tric

holo

mat

acea

e M

ycen

ella

Myc

enell

a tra

chys

pora

12

610

1229

3 30

7 10

OT

U_8

A

scom

ycot

a Le

otio

myc

etes

H

elot

iale

s H

elot

iale

s_IS

C

adop

hora

1541

4 13

351

1137

92

6

OT

U_1

A

scom

ycot

a Le

otio

myc

etes

H

elot

iale

s H

elot

iace

ae

Var

icosp

oriu

m

Var

icosp

oriu

m s

pp.

1380

2 13

082

632

88

OT

U_5

8 A

scom

ycot

a Le

otio

myc

etes

Le

otio

myc

etes

_IS

Leot

iom

ycet

es_I

S G

eom

yces

G

eom

yces

pan

noru

m

1297

6 21

39

24

9031

!!

105

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

!!! !T

able

S2

Ove

rvie

w o

f 20

mos

t dom

inan

t OT

Us

whi

ch d

eter

min

e th

e di

ffer

ence

bet

wee

n th

e ge

ogra

phic

reg

ion

in r

hizo

sphe

re/e

ndos

pher

e co

mpa

rtm

ents

list

ed w

ith ta

xono

mic

info

rmat

ion.

Onl

y th

e O

TU

s w

hich

det

erm

ines

the

diff

eren

ce o

f the

com

mun

ities

in a

t lea

st 2

geo

grap

hic

regi

on w

ere

only

list

ed.

OT

U

Reg

ion

Com

part

men

t Ph

yla

Cla

ss

Ord

er

Fam

ily

Gen

us

Spec

ies

OT

U_2

6 M

yr

End

o A

scom

ycot

a Le

otio

myc

etes

H

elot

iale

s D

erm

atea

ceae

C

rypt

ospo

riops

is un

cultu

red_

Cry

ptos

porio

psis

OT

U_1

5 M

yr,

NyA

E

ndo

(Myr

), R

hizo

(NyA

) A

scom

ycot

a D

othi

deom

ycet

es

Pleo

spor

ales

Pl

eosp

oral

es_I

S H

erpo

trich

ia

Her

potri

chia

_sp_

CR

G_2

011

OT

U_2

3 M

yr

End

o, R

hizo

A

scom

ycot

a Le

otio

myc

etes

U

n_Le

otio

myc

etes

U

n_Le

otio

myc

etes

U

n_L

eotio

myc

etes

uncu

lture

d_ect

omyc

orrh

iza_

(L

eotio

myc

etes)

OT

U_1

8 M

yr

End

o, R

hizo

A

scom

ycot

a U

n_A

scom

ycot

a U

n_A

scom

ycot

a U

n_A

scom

ycot

a L

epto

dont

idiu

m

uncu

lture

d_L

epto

dont

idiu

m

OT

U_2

95

Myr

E

ndo

Asc

omyc

ota

Sord

ario

myc

etes

X

ylar

iale

s A

mph

isph

aeria

ceae

U

n_A

mph

ispha

eria

ceae

Am

phisp

haer

iacea

e_sp

OT

U_2

9 M

yr

End

o A

scom

ycot

a D

othi

deom

ycet

es

Pleo

spor

ales

Le

ptos

phae

riace

ae

Acic

usep

toria

A

cicus

epto

ria_r

umici

s

OT

U_3

6 M

yr

End

o, R

hizo

A

scom

ycot

a Le

otio

myc

etes

H

elot

iale

s H

elot

iale

s_IS

T

etrac

ladi

um

Tetr

acla

dium

_sp

OT

U_2

2 M

yr

Rhi

zo

Bas

idio

myc

ota

Aga

ricom

ycet

es

Un_

Aga

ricom

ycet

es

Un_

Aga

ricom

ycet

es

Un_

Aga

ricom

ycete

s U

n_A

garic

omyc

etes

OT

U_4

8 M

yr

Rhi

zo

Bas

idio

myc

ota

Aga

ricom

ycet

es

Aga

rical

es

Hyd

nang

iace

ae

Un_

Hyd

nang

iacea

e U

n_H

ydna

ngia

ceae

OT

U_3

63

Myr

R

hizo

C

hytr

idio

myc

ota

Chy

trid

iom

ycet

es

Rhi

zoph

ydia

les

Rhi

zoph

ydia

ceae

U

n_R

hizo

phyd

iacea

e R

hizo

phyd

iacea

e_sp

OT

U_5

2 M

yr

Rhi

zo

Un_

Fung

i U

n_Fu

ngi

Un_

Fung

i U

n_Fu

ngi

Un_

Fung

i U

n_Fu

ngi

OT

U_1

152

Kil

End

o B

asid

iom

ycot

a A

garic

omyc

etes

A

garic

ales

T

richo

lom

atac

eae

Myc

enell

a M

ycen

ella_

trach

yspo

ra

OT

U_1

K

il E

ndo

Asc

omyc

ota

Leot

iom

ycet

es

Hel

otia

les

Hel

otia

ceae

V

arico

spor

ium

V

arico

spor

ium

_sp

OT

U_3

8 K

il E

ndo

Asc

omyc

ota

Leot

iom

ycet

es

Hel

otia

les

Hya

losc

ypha

ceae

L

achn

um

Lac

hnum

_sp

OT

U_1

4 K

il E

ndo

Asc

omyc

ota

Leot

iom

ycet

es

Un_

Leot

iom

ycet

es

Un_

Leot

iom

ycet

es

Un_

Leo

tiom

ycete

s L

eotio

myc

etes_

sp

OT

U_2

4 K

il E

ndo

Asc

omyc

ota

Leot

iom

ycet

es

Un_

Leot

iom

ycet

es

Un_

Leot

iom

ycet

es

Un_

Leo

tiom

ycete

s U

n_L

eotio

myc

etes

OT

U_4

K

il E

ndo

Asc

omyc

ota

Leot

iom

ycet

es

Hel

otia

les

Hel

otia

ceae

U

n_H

elotia

ceae

Un_

Helo

tiacea

e

OT

U_1

87

Kil

Rhi

zo

Zyg

omyc

ota

Zyg

omyc

ota_

IS

Mor

tiere

llale

s M

ortie

rella

ceae

M

ortie

rella

M

ortie

rella

_elon

gata

OT

U_4

61

Kil

Rhi

zo

Bas

idio

myc

ota

Aga

ricom

ycet

es

Aga

rical

es

Cla

varia

ceae

C

lava

ria

Cla

varia

_acu

ta

OT

U_1

18

Kil

Rhi

zo

Bas

idio

myc

ota

Tre

mel

lom

ycet

es

Filo

basi

dial

es

Filo

basi

diac

eae

Cry

ptoc

occu

s C

rypt

ococ

cus_

terric

ola

OT

U_4

4 K

il R

hizo

Z

ygom

ycot

a Z

ygom

ycot

a_IS

M

ortie

rella

les

Mor

tiere

llace

ae

Mor

tiere

lla

Mor

tiere

lla_h

umili

s

OT

U_3

0 K

il R

hizo

B

asid

iom

ycot

a A

garic

omyc

etes

T

hele

phor

ales

T

hele

phor

acea

e T

omen

tella

un

cultu

red_

Tom

entel

la

!106

Cha

pter

4

! !

OT

U

Reg

ion

Com

part

men

t Ph

yla

Cla

ss

Ord

er

Fam

ily

Gen

us

Spec

ies

OT

U_6

0 K

il R

hizo

B

asid

iom

ycot

a A

garic

omyc

etes

T

hele

phor

ales

T

hele

phor

acea

e T

omen

tella

un

cultu

red_

Tom

entel

la

OT

U_8

1 K

il R

hizo

B

asid

iom

ycot

a A

garic

omyc

etes

A

garic

ales

C

lava

riace

ae

Un_

Cla

varia

ceae

OT

U_7

N

yA

End

o A

scom

ycot

a U

n_A

scom

ycot

a U

n_A

scom

ycot

a U

n_A

scom

ycot

a U

n_A

scom

ycot

a U

n_A

scom

ycot

a

OT

U_8

N

yA

End

o A

scom

ycot

a Le

otio

myc

etes

H

elot

iale

s H

elot

iale

s_IS

C

adop

hora

OT

U_1

2 N

yA

End

o A

scom

ycot

a Le

otio

myc

etes

Le

otia

les

Leot

iace

ae

Ala

tosp

ora

Ala

tosp

ora_

sp

OT

U_5

N

yA

End

o A

scom

ycot

a D

othi

deom

ycet

es

Ven

turia

les

Sym

pove

ntur

iace

ae

Un_

Sym

pove

ntur

iacea

e Sy

mpo

vent

uria

ceae_

sp

OT

U_1

0 N

yA

End

o A

scom

ycot

a Le

otio

myc

etes

H

elot

iale

s H

elot

iale

s_IS

H

elotia

les_I

S H

elotia

les_I

S

OT

U_3

N

yA

End

o A

scom

ycot

a A

scom

ycot

a_IS

A

scom

ycot

a_IS

A

scom

ycot

a_IS

A

scom

ycot

a_IS

A

scom

ycot

a_IS

OT

U_6

2 N

yA

Rhi

zo

Asc

omyc

ota

Un_

Asc

omyc

ota

Un_

Asc

omyc

ota

Un_

Asc

omyc

ota

Un_

Asco

myc

ota

Un_

Asco

myc

ota

OT

U_2

N

yA

Rhi

zo

Asc

omyc

ota

Leot

iom

ycet

es

Hel

otia

les

Un_

Hel

otia

les

Un_

Helo

tiales

OT

U_3

1 N

yA

Rhi

zo

Zyg

omyc

ota

Zyg

omyc

ota_

IS

Mor

tiere

llale

s M

ortie

rella

ceae

M

ortie

rella

OT

U_4

0 N

yA

Rhi

zo

Zyg

omyc

ota

Zyg

omyc

ota_

IS

Mor

tiere

llale

s M

ortie

rella

ceae

M

ortie

rella

M

ortie

rella

_poly

goni

a

OT

U_1

08

NyA

R

hizo

A

scom

ycot

a Le

otio

myc

etes

U

n_Le

otio

myc

etes

U

n_Le

otio

myc

etes

L

eohu

mico

la

uncu

lture

d_L

eohu

mico

la

OT

U_2

93

NyA

R

hizo

Z

ygom

ycot

a Z

ygom

ycot

a_IS

M

ortie

rella

les

Mor

tiere

llace

ae

Mor

tiere

lla

Mor

tiere

lla_f

imbr

icysti

s

OT

U_3

3 N

yA

Rhi

zo

Asc

omyc

ota

Sord

ario

myc

etes

H

ypoc

real

es

Nec

tria

ceae

N

ectria

N

ectria

_ram

ular

iae

OT

U_5

8 N

yA

Rhi

zo

Asc

omyc

ota

Leot

iom

ycet

es

Leot

iom

ycet

es_I

S Le

otio

myc

etes

_IS

Geo

myc

es

Geo

myc

es_p

anno

rum

O

TU

_15

Shar

ed b

etw

een

two

regi

ons,

but

diff

eren

t com

part

men

ts

OT

U_2

3 Sh

ared

bet

wee

n di

ffer

ent c

ompa

rtm

ents

in th

e sa

me

regi

on

!

!!

107

Fung

al c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

! !

 108

 

!

!

Chapter 5

Manoj Kumar, Jan Dirk van Elsas, Riitta M. Nissinen

Strong regionality and dominance of anaerobic bacterial taxa

characterize diazotrophic bacterial communities of the arctic

plant species Oxyria digyna and Saxifraga oppositifolia

!110

Cha

pter

5

Abstract Arctic and alpine biomes developing in rocky areas are most often strongly nitrogen-limited, and hence biological nitrogen fixation is a strong driver of such biomes. On the basis of total community DNA, we here studied the nitrogen-fixing communities at two arcto-alpine pioneer plant species, Oxyria digyna and Saxifraga oppositifolia, using the nifH gene as a proxy. Replicate samples taken from two arctic regions, Kilpisjärvi and Ny-Ålesund, next to one alpine region, Mayrhofen, were included, to study a South-to-North cold-climate soil gradient. The data revealed strong regional effects on the dominating nitrogen fixers. As a major overall finding, anaerobic bacterial taxa dominated the diazotrophic bacterial communities of the two plant species. Specifically, Geobacter and Leptothrix spp. dominated the Mayrhofen and Kilpisjärvi regions, while members of the Clostridiales preferred the Kilpisjärvi and Ny-Ålesund regions. Endosphere communities were enriched with Goebacter and Clostridia, whereas the soil communities were dominated with Burkholderia and Leptothrix. Key words: nifH, arcto-alpine plants, Geobacter, Clostridia, Leptothrix

!

!!

111

Dia

zotr

ophi

c ba

cter

ial c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

Introduction Nitrogen (N) is considered to be the major limiting factor for both microbial and plant growth in arctic and alpine biomes. For example, Sistla et al. (2012) revealed - by N addition experiments - that microbial growth, enzyme synthesis and carbon allocation in tussock tundra soils during the arctic summer are N-limited. Furthermore, corroborating this finding, Wallenstein et al. (2009) observed overall decreases in microbial enzyme activities with the reduction of available soil nitrogen levels. In particular, the amount of available soil nitrogen is very low in low organic content mineral-rich tundra soils. In addition, a study of a chronosequence along the forefield of the Damma glacier in the Swiss Alps by Brankatschk et al. (2011) reported the presence of very low N nutrient contents in soils with deglaciation ages between 10 and 70 years. These soils also showed quite scarce vegetation. Furthermore, the authors observed no potential N-fixation activity and extremely low denitrification and nitrification rates. Finally, the nifH gene copy numbers were low, with the lowest numbers being found in the 10-year old soils. In contrast, a densely plant-covered 120-year old soil showed both higher enzyme activities and higher copy numbers of the nifH, nirK and nirS genes. Clearly, the older soils had regained the capacity to cycle nitrogen on the basis of a pioneer species based ecosystem build-up, in which plants found capacities to grow and, in turn, supported the evolved microbial communities and activity.

The N limitation is underlined by a higher demand for N by arctic plants compared to their temperate-climate relatives. Given the fact that arctic plants are adapted to low temperatures during the growing season and photosynthesis rates (at 68°N) are comparable to those at lower latitudes (Central Europe) (Chapin et al., 1987), higher amounts of enzymes (and so higher N) are required. For example, higher numbers of the key photosynthesis enzyme ribulose 1,5 bisphosphate (RuBiSco) are required to compensate for the slower kinetics at low temperatures. Thus, the tissues of arctic plants contain more N than those of plants in warmer climates (Weintraub and Schimel, 2005), underlining the demand for N availability. Clearly, in nascent ecosystems such as those forming in the low-nutrient arctic soils, provision of nitrogen is a key issue, as living systems critically depend on reduced nitrogen. Biological nitrogen fixation by free-living local bacteria (under which cyanobacteria), appears as a key strategy involved in the local primary production, yielding organic compounds containing reduced nitrogen. In the emerged systems, with plants in place as primary producers, the key nitrogen input may come from plant-symbiotic and/or associative nitrogen fixers, whereas nitrogen from the produced organic compounds is recycled/mineralized via deamination (yielding ammonia), followed by nitrification and denitrification reactions. Unfortunately, we currently lack

!112

Cha

pter

5

information as to the key organisms involved in these important steps of the nitrogen cycle in the Arctic.

In previous studies, we selected two plant species, Oxyria digyna and Saxifraga oppositifolia, as these are among the primary colonizers of the nutrient poor mineral soils that abound in both arctic and alpine regions. In these studies, we have found preliminary evidence for the contention that the soil bacterial communities in different arcto-alpine regions are region-specific, whereas the endophytic bacterial communities (using 16S rRNA gene as a proxy) are host plant specific (Kumar et al., 2016b). However, these host plant specific communities also include shared core taxa, which are associated with their host plants in all three climatic regions. Unfortunately, the role of these core taxa has been understudied, although, on the other hand, the data already provided glimpses of the presence of several potential nitrogen fixers in the core microbiome (Kumar et al., 2016b).

In this study, we place a focus on the nitrogen fixing bacteria in the two selected arcto-alpine plants, with the hypothesis that key potential nitrogen fixers identified as major members of the bacterial communities associated with these plants are indeed involved in local N fixation processes. In the light of the impossibility to measure local nitrogen fixation rates, we provide a thorough discussion of the likelihood of this tenet.

Figure 1 Sampling locations; Mayrhofen in Austrian Alps, Kilpisjärvi in low-arctic Finnish Lapland and Ny-Ålesund in high-arctic Svalbard archipelago

!

!!

113

Dia

zotr

ophi

c ba

cter

ial c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

Materials and methods Sampling and DNA isolation: Samples were collected in June 2012 from three different regions, i.e. two in the arctic (Kilpisjärvi (KJ) and Ny-Ålesund (NÅ)) and one in the European Alps (Mayrhofen (MA)) (Figure 1). NÅ is located in the high Arctic, where soil temperatures do not rise above 10oC, whereas alpine MA sites and KJ low Arctic sites experience higher annual temperature fluctuations. Three replicates of bulk soil samples and three plant samples of each O. digyna and S. oppositifolia (with adhering rhizosphere soils) were collected from all sites. All samples were processed as specified by Kumar et al. (2016). In brief, after removing rhizosphere soil parts, plant roots were thoroughly washed with water and then surface sterilized by immersion into 3% sodium hypochlorite (3 min), followed by sterile double-distilled water (3 x 90 s). 80-100 mg of processed root sample were snap-frozen with liquid nitrogen and stored at -80˚C for further DNA extraction and analysis. In addition, 250-300 mg of rhizosphere and bulk soil samples were stored at -80˚C. nifH library preparation and sequencing

The MoBio Power soil kit (MoBio, Carlsbad, CA, USA) and Invisorb Spin plant Midi kit (STRATEC, Biomedical AG, Germany) were used to extract DNA from soil and plant samples, respectively (Kumar et al., 2016b). The nifH gene was amplified using a nested approach. The first PCR was done with primers 19f/nifH3 (Yeager et al., 2004) and the second one with one forward primer nifH1f (as described by Zehr and Mcreynolds, 1989) and equimolar concentration of reverse primers nifH2r (Zehr and Mcreynolds, 1989) and nifH2 (Izquierdo and Nüsslein, 2006). The first PCR reaction had 2 !l of template DNA, 1x PCR buffer, 2.5mM of MgCl2, 1 mg/ml of BSA, 0.4 mM dNTP’s, 0.9 !M of each primer and 2500 U/ml of GoTaq DNA ploymerase (Promega, WI, USA) in a 30 !l reaction volume. For the second PCR, 1 !l of amplified product from the first PCR was added along with 1x PCR buffer, 1 mM of MgCl2, 1 mg/ml of BSA, 0.2 mM dNTP’s, 0.9 !M of each primer and 1250 U/ml of GoTaq DNA ploymerase (Promega) in a 30 !l reaction volume. Amplifications for both PCR reactions were performed as follows: 2 min denaturation at 95°C followed by 35 cycles of denaturing, annealing and extension at 95°C for 1 min, 51°C (first PCR)/ 53°C (second PCR) for 1 min and 72°C for 1 min, respectively. Final extension was carried out at 72°C for 2 min.

Sequence libraries were then prepared by running a third PCR to attach the M-13 barcode using the system developed by Mäki et al. (2016). An aliquot (1 !l) of the second PCR was re-amplified using the barcode-attached M13

!114

Cha

pter

5

system as forward primer and nifH2r/nifH2-P1 with adaptor A as reverse primer. PCR mix and conditions were similar to those described for the second PCR, except that only 8 cycles were used for amplification. The products were then purified using the Agencourt AMPure XP PCR purification system (Beckman Coulter, CA, USA), followed by quantification with a Qubit Fluorometer (Invitrogen). Equivalent DNA quantities of each sample were then pooled, size-fractionated (size selection range of 350-550 bp) using Pippin Prep (Sage Science, MA, USA) 2% agarose gel cassette (Marker E) according to the manufacturer’s protocol. The size-fractioned libraries were sequenced using Ion 314 chip kit V2 BC on an Ion Torrent PGM machine (Life Technologies, MA, USA) in the Biocenter, Oulu, Finland.

Bioinformatics and statistical analysis All reads from the Ion torrent sequencing were processed using QIIME (Caporaso et al., 2010) and Fungene pipelines (Fish et al., 2013), as described by Zhang et al. (2015). All sequences with quality scores below 22 and lengths below 240bps were removed. USEARCH (Edgar, 2010) algorithm in de novo was used for chimer removal. After chimera removal, all nucleotide sequences were translated into protein sequences and frameshift-corrected with Framebot (length cutoff = 80 AAs), aligned using HMMER3 Aligner (with nifH as representative gene) and then clustered by RDP mcClust (90% similarity) in Fungene. The resulting cluster file was then converted into an OTU table by RDP cluster file formatter in R (version 3.2.5; https://www.r-project.org/). Then, OTUs were then assigned to taxa by FunGene FrameBot, annotations. The OTUs were grouped into OPUs by clustering with the same reference sequence. Taxonomic affiliations of 12 OPUs were complemented by P-BLAST at NCBI.

The OTU table was normalized and OPU abundances were square-root transformed prior to community analysis with the PRIMER 6.1 (http://www.primer-e.com/) software package. In addition, PRIMER 6 was also used to determine the similarities between the samples (ANOSIM) and to assess the dissimilarity between dominating OPUs from different regions or compartments (SIMPER). PCoA graphs were also plotted with PRIMER 6 based on Bray-Curtis dissimilarity matrix. Venny (http://bioinfogp.cnb.csic.es/tools/venny/) was used for constructing a Venn diagram showing similarities and dissimilarities across the geographic region. Results We successfully produced pure total microbiome DNA from 165 samples taken from bulk soil, rhizosphere soil and the endosphere of O. digyna and S.

!

!!

115

Dia

zotr

ophi

c ba

cter

ial c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

oppositifolia from the Mayrhofen, Kilpisjärvi and Ny-Ålesund regions. Then, PCR amplifications were performed with the 19f/nifH3r and nifH1f/nifH2r and nifH2 primers (amplifying 350 bp of the nifH gene). This nested approach was optimized by us to have maximum coverage of all phylogenetic diazotrophic groups based on the review by Gaby and Buckley (2012). All quality-trimmed nifH sequences were translated to the respective amino acid sequences. These were used for clustering, yielding 296 OTUs (at 90% amino acid sequence identity). These 296 OTUs were further clustered based on alignments with nifH protein reference sequences in the RDP FunGene nifH database. This clustering finally resulted in 129 units, which were denoted ‘operational phylogenetic units’ (OPUs).

Figure 2 Nitrogen fixing OPUs (nifH gene sequences clustered with the closest reference sequences) in the three geographic regions. Figure legends: MA - Mayrhofen, the Alps, KJ - Kilpisjärvi, low Arctic, NÅ - Ny-Ålesund, high Arctic Distribution of nifH OPUs over geographic region Seventy-eight of the 129 OPUs (62%) were present across all three geographic regions used in this study. Thirty-nine OPUs (31%) were present in either one or both arctic regions, i.e. NÅ and KJ, whereas these were missing from MA in the alpine region (Figure 2). Ten OPUs (7%) were unique to MA or shared between MA and either one of the two arctic regions (Figure 2) Diazotrophic communities – an overview Overall, we detected high diversities of potential nitrogen fixing bacteria (PNFB) in all our samples. Remarkably, we detected nifH genes from the

!116

Cha

pter

5

bacterial taxa Proteobacteria, Firmicutes, Actinobacteria, Chlorobi, Bacteroidetes, Cyanobacteria and Verrucomicrobia. The most abundant nifH gene sequences in the dataset clustered with nifH gene sequences from the !-Proteobacteria (genera Geobacter and Desulfotignum), Clostridiales (Clostridium, Desulfosporosinus and Acetobacterium), "-Proteobacteria (Burkholderia [Burkholderiales] and Leptothrix [Comamonadaceae]), #-Proteobacteria (Bradyrhizobium [Rhizobia]).

Figure 3 Principal Coordinate Analysis (PCoA) plots based on Bray-Curtis similarity matrix of PNFB communities (nifH-based OPU sequences as proxy species) from bulk soil (B), rhizosphere soil (R) and endosphere (E) from three climatic regions Mayrhofen (MA), Kilpisjärvi (KJ) and Ny-Ålesund (NÅ) Community structures of PNFB Community structures of the PNFB were analyzed at the OPU level across all samples. We observed no influence of plant species on the community structures of either rhizosphere or endosphere PNFB (Supplemental figures S1, S2). ANOSIM further confirmed that Geographic region (rhizosphere - R = 0.57; endosphere – R = 0.36) was the primary determinant of the plant-associated PNFB, followed by plant species (rhizosphere - R = 0.15; endosphere – R = 0.20). Considering the negligible effect of plant species, we combined the O. digyna and S. oppositifolia communities and analysed the data set based on ’Compartment’ (bulk soil, rhizosphere soil, endosphere) across the three geographic regions. ANOSIM (Table 1) and PCoA ordination showed (Figure 3) that both geographic region and compartment strongly impacted the PNFB community structures. The endosphere PNFB

!

!!

117

Dia

zotr

ophi

c ba

cter

ial c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

communities were significantly different from those from both soil compartments, while we detected only small differences between bulk and rhizosphere soil PNFB communities. Curiously, the endosphere communities were less similar to the rhizosphere than to the bulk soil communities (Table 1). The NÅ communities (analyzed across the different compartments) were relatively more different from both the KJ and MA communities than the latter two from each other (Table 1).

Figure 4 Species richness (a) and shannon diversity (b) of PNFB communities in different regions and compartments. Diversity indices were calculated using nifH gene OPUs as a proxy for species. Figure legends: MA - Mayrhofen; KJ - Kilpisjärvi; NÅ - Ny-Ålesund; B - bulk soil; R - rhizosphere soil, E – endosphere Species richness as well as diversity of PNFB communities is lowest in the high Arctic The species richness (SR) and Shannon diversity (SD) values of the PNFB communities (based on the nifH gene amplicon OPUs as a proxy) were highest in the bulk soils, followed by the rhizosphere soils and the endosphere samples. The diversities in the endosphere samples were significantly lower than those in the two soil compartments (Figure 4a, 4b).

The differences in the diversity and richness measures between the three geographic regions were estimated separately for each compartment. Species richness and diversity values of bulk soil communities were both highest in the MA, followed by the KJ and NÅ regions. However, in the plant-associated compartments (rhizo- and endospheres), both species richness and

!118

Cha

pter

5

diversity values were highest in the KJ communities, followed by the MA and NÅ communities (Figure 4a, 4b).

Table 1 One-way ANOSIM statistics for comparisons of PNFB communities according to geographic region (MA = Mayrhofen, KJ = Kilpisjärvi and NÅ = Ny-Ålesund), sample compartment (E = endosphere, R = rhizosphere soil and B = bulk soil) or plant species (O = O. digyna, S = S. oppositifolia). using OPU similarity values calculated based on Bray-Curtis resemblance matrix with R-value and significance value in % (in brackets). Significance values are 0.1% unless it is mentioned in the table

Region

All Samples E R B All Samples 0.302 0.505 0.406 MA vs KJ 0.139 0.238 0.346 0.284 MA vs NÅ 0.249 0.48 0.626 0.291 KJ vs NÅ 0.179 0.186 0.535 0.530 Compartment All regions MA KJ NÅ B vs R 0.049 0.194 (2%) 0.127 (1.6%) 0.092 (3.1%) B vs E 0.258 0.26 (0.7%) 0.578 0.389 R vs E 0.359 0.436 0.459 0.54 Plant species (O vs S) Endosphere All regions MA KJ NÅ 0.105 (0.2%) 0.18 0.230 0.132 (2.1%)

Different bacteria dominate soil- and plant-associated PNFB communities in different regions and compartments SIMPER analysis confirmed the findings from our initial taxonomic analyses. Thus, different bacterial taxa dominated the PNFB communities in different geographic regions. The bulk and rhizosphere so i l PNFB diverged from the endosphere ones. Leptothrix and Burkholderia type nifH sequences dominated both the bulk and rhizosphere soil communities across all three regions (Figure 5a). Additionally, OPUs clustering with nifH sequences from the Opitutaceae (Verrucomicrobia) were highly abundant in the KJ bulk soils, whereas Acetobacterium bakii (Clostridiales) nifH sequences were predominantly present only in the rhizosphere soils from NÅ. Compared to MA and KJ, Geobacter-type nifH sequences were detected only at very low relative abundances in NÅ, while the soil communities were dominated by the genera Burkholderia, Leptothrix and species Nostoc punctiforme and Nodularia spumigena (Cyanobacteria).

The endosphere communities were often enriched with anaerobic nitrogen fixers from the genera Geobacter and Clostridium (Figure 5a). Thus the MA communities were strongly dominated by nifH-based OPUs related to several Geobacter species, including G. lovelyi, G. bemidjiensis and G. uraniireducens (Figure 5b). Specifically, six out of seven key OPUs (SIMPER analysis) of these communities represented the genus Geobacter (Table 2). In contrast, the main OPUs structuring the NÅ endophytic PNFB communities represented

!

!!

119

Dia

zotr

ophi

c ba

cter

ial c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

the genera Clostridium (three OPUs, including C. akagii and C. pasteurianum), Leptothrix (Comamonadaceae, two OPUs) and unidentified_bacterium (OPU 1364). While Clostridium-like OPUs were virtually absent in the MA endosphere communities and Geobacter-type OPUs likewise in NÅ, the major nifH OPUs in the KJ endosphere communities represented both the genera Clostridium and Geobacter, as well as Bradyrhizobium and Burkholderia (Table 2; Figure 5b). Specifically, OPU 3027 (Bradyrhizobium sp.) was predominantly found in the KJ endosphere samples. The endosphere and soil communities in NÅ were more dissimilar from each other (ANOSIM R=0.74, 0.1%) than the ones in KJ or in MA (ANOSIM R=0.68, and R=0.63 respectively).

Figure 5 Average relative abundance of PNFB communities plotted based on sample compartment (a) and geographic region separated based on compartment (b). Only selected major PNFB OPUs were shown, and OPUs with similar species/genus were makred with OPU numbers (in bracket) Geobacter and Bradyrhizobium diazotrophs are differentially enriched in S. oppositifolia and O. digyna endospheres As stated above, the compartment, in particular the endosphere-soil division, was the major determinant of the PNFB communities, with no strong host-plant specificity. However, we detected some trends in the affiliation of several OPUs with particular plant species. OPUs representing Bradyrhizobium, Burkholderia and OPU 1_783 (nifH from an unidentified bacterium) were consistently present in high abundances in the O. digyna endosphere samples (Figure 6), whereas S. oppositifolia endosphere communities were, instead, more enriched with Geobacter related OPUs, especially in the MA and KJ regions.

!120

Cha

pter

5

Furthermore, the S. oppositifolia endosphere communities harbored relatively more OPUs related to Desulfosopronius (Clostridiales) in KJ and Clostridium and Desulfuromonas related OPUs in NÅ, all in comparison to O. digyna (Figure 6).

Figure 6 Average relative abundance of PNFB communities plotted based on plant species O. digyna and S. oppositifolia across all three regions. Only selected major PNFB OPUs were shown, and OPUs with similar species/genus were marked with OPU numbers (in bracket)

Discussion Using the nifH gene as a proxy, we here analyzed the diversity and community composition of PNFB in two pioneer plant species growing in low-nutrient soils in three arcto-alpine regions. Overall, we detected high diversities across the samples, with nifH genes from seven different bacterial phyla being present. With respect to the breadth of detection, we argue that the primers selected theoretically allow to obtain maximum coverage from all the nifH phylogenetic groupings (Gaby and Buckley, 2012). Indeed, our libraries had good phylogenetic coverage, with all OPUs representing seven phyla, including representatives from nifH clusters I, II and III. In specific, the PNFB communities were largely dominated by nifH gene types from the $-, "-, #- and !-proteobacteria and the Clostridiales. The !-proteobacterial genus Geobacter, the "-proteobacterial genera Burkholderia and Leptothrix, as well as Clostridia (from the Firmicutes) were the most abundant. With the exception of Burkholderia and Leptothrix, the dominant diazotrophs in these communities have been mostly described from anoxic soils and sediments (Dang et al., 2013; Rösch et al.,

!

!!

121

Dia

zotr

ophi

c ba

cter

ial c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

2002; Wu et al., 2009; Yeager et al., 2004; You et al., 2005). This points at a key suggestive finding of this study, i.e. the likely prevalence of conditions enabling nitrogen fixation under anoxia in most of the soils. Both O. digyna and S. oppositifolia are known to be able to tolerate waterlogging in soil. The two are common snow patch species, as known from observations in the low-arctic sites in KJ. We sampled all sites relatively early in the growth season, when the plants were flowering, which was relatively shortly after snowmelt at these sites. Only the NÅ region had permafrost, with KJ having seasonal and patchy permafrost, with snow often melting away as late as mid-July. The MA site soils were, however, relatively less water-logged than the soils from the arctic sites.

Table 2 Data based on SIMPER analysis of normalized and square root transformed major OPU/species driving the dissimilarities between different regions in the endosphere samples. All OPUs which contribute to 90% dissimilarity were selected and presented in the table

OPU/Species Average Abundance

Average Similarity

Contribution to Dissimilarity %

Average dissimilarity = 30.23 (Mayrhofen) Geobacter lovleyi_SZ 3.92 9.14 30.25 Geobacter uraniireducens_Rf4 2.96 6.45 21.35 Geobacter uraniireducens sp 2.62 5.27 17.43 Geobacter CP001089 1.38 2.86 9.47 Geobacter bemidjiensis_Bem 1.57 1.61 5.32 Geobacter_2397398_2398267_CP001124 1.37 1.54 5.08 Bradyrhizobium elkanii 1.22 0.84 2.79 Average dissimilarity = 22.96 (Kilpisjärvi) Clostridium pasteurianum_4_1-like 2.23 3.26 14.19 Geobacter_2397398_2398267 2 2.9 12.62 Bradyrhizobium 2.27 2.76 12.03 Clostridium akagii 1.73 2.33 10.13 Geobacter_lovleyi_SZ 1.95 2.31 10.05 Geobacter_bemidjiensis_Bem 1.4 1.77 7.73 Clostridium_pasteurianum_1_ 1.14 1.69 7.37 Geobacter_CP001089 0.97 1.11 4.82 Geobacter uraniireducens_Rf4 0.98 0.78 3.38 Bradyrhizobium elkaniifor 0.58 0.63 2.74 Desulfosporosinus orientis_DSM_765 0.86 0.47 2.05 Burkholderiales_321041_321922 0.95 0.39 1.69 Burkholderia_sp_Ch1-1_ctg00023 0.68 0.38 1.65 Average dissimilarity = 18.71 (Ny-Ålesund) Clostridium pasteurianum_4_1-like 2.48 4.56 25.1 Clostridium akagii 2.31 3.47 19.09 Bradyrhizobium elkaniifor 2.07 3.25 17.88 1_783_AB094963 1.2 1.73 9.55 Opitutaceae_bacterium_TAV2_ctg796 0.88 1.31 7.23 Clostridium pasteurianum 1.21 0.88 4.87 Nostoc punctiforme_PCC_73102 0.44 0.5 2.77 Polaromonas naphthalenivorans_CJ2 0.49 0.25 1.39 Leptothrix_161893_162774_CP002016 0.59 0.24 1.33 Leptothrix cholodnii_SP-6 0.44 0.24 1.3

!122

Cha

pter

5

Compared to reports from high-organic-matter tundra soils from Arctic or alpine sites, where aerobic proteobacterial diazotrophs reportedly dominate (Tai et al., 2013), our soils were mostly enriched in !-proteobacteria and Clostridiales (depending on region and compartment). The current data, where we found no influence of plant species (O. digyna and S. oppositifolia) on the PNFB communities in the endosphere were different from those on the 16S bacterial community structures (Kumar et al., 2016b). However, a previous study on fungal community structures was in line with the contention that plant species was a minor driver (Kumar et al., 2016a). In addition, both fungal and PNFB community structure were influenced by geographic region. We here hypothesize that the two plant species most likely ‘behave’ in a similar manner in the soils, for instance secreting similar compounds that attract the diazotrophs as well as fungi. Possibly, although unproven, the diazotrophs are tightly associated with the selected fungi, or each of the three components of the system are similarly selected to make part of it.

As mentioned, clear regionality was also evident in the distribution of the diazotrophic communities. Thus Geobacter was abundantly present in the sites from MA and KJ, at altitudes of about 2400m and 900m above sea level. This was in line with the observation of this taxon in high mountain ranges where it could metabolize under both oxic and anoxic conditions (Ciccazzo et al., 2016). Geobacter has also been detected in nifH defined communities from switchgrass roots in oxic temperate-climate prairie ecosystems (Bahulikar et al., 2014), indicating adjustment to temperate and environmental plasticity to utilize both aerobic and anaerobic soil conditions.

Moreover, Clostridiales, an anaerobic taxon (Clostridium and Acetobacterium), were virtually absent from the MA samples, whereas they increased in abundance towards the north. In addition, this may reflect the different bioclimatic conditions between MA and NÅ, as discussed.

It should be noted, though, that - even in NÅ - Clostridium was mainly restricted to the endosphere, which is actually in agreement with our analysis of total bacterial communities of the same plants (based on the16S ribosomal RNA [rRNA] gene) (Kumar et al., 2016b). Specifically, we observed higher relative abundances of Clostridium with increasing latitude and restricted to the endospheres of both plant species. Such Clostridium PNFB (predominantly observed in the endosphere) also showed endosphere specificity in the previous 16S rRNA gene based analysis (Kumar et al., 2016b). In addition, Given et al. (personal communication) observed 16S rRNA sequences of Clostridium spp. in seeds of O. digyna, indicating possible vertical transmission. Both Geobacter spp. and Clostridium spp. have been detected as diazotrophs in rice roots in flooded rice paddy fields (Hardoim et al., 2012; Sessitsch et al., 2012).

!

!!

123

Dia

zotr

ophi

c ba

cter

ial c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

The diversity of PNFB communities was lowest in the high-arctic NÅ sites, with a similar trend being detectable in all (i.e. the bulk soil and plant associated (rhizo- and endosphere)) compartments. While the bulk soil communities were most diverse in the alpine sites in MA, the plant associated communities were most diverse in KJ, which is in contrast to the bacterial and fungal communities from the same sites, where we observed the highest diversity in rhizosphere samples (Kumar et al., 2016a; 2016b). Corroborating the previous 16S rRNA based analyses (Kumar et al., 2016b), O. digyna and S. oppositifolia shared some plant specific PNFB OPUs. OTUs previously identified as ‘core’ or as part of the endosphere community, e.g. Bradyrhizobium, Burkholderiales, Comamonadaceae and Clostridia (Kumar et al., 2016b) were also found to make part of the PNFB endosphere communities. Thus, such N fixers are often key dominant factors in the plant-associated microbiomes, indicating the importance of (plant-associated) nitrogen fixation for these plants. It may well be that plants recruit these bacteria from the soil, allowing N fixation to occur in the ‘protected’ environment offered by the plant. The outcome of this selective process might be improved plant growth in the highly N limited soils. Furthermore, this signifies importance of N fixating bacteria for plant fitness and growth in these conditions.

Concluding, the nature of the PNFB communities found at two plant species in three different cold-climate geographic regions points to selective processes that operate at different scales, i.e. that of compartment as well as of region. Plant type appeared to have a lesser influence, with the exception of some OPUs from the order Burkholderiales (Burkholderia sp.) in O. digyna and the order Clostridiales (Desulfosporsinus sp.) in S. oppositifolia. These organisms were also found to be enriched in the (16S rRNA defined) endosphere communities from the respective plants. Finally, on the basis of the predominance of anaerobic nitrogen fixers, we surmised that local conditions allowing nitrogen fixation might have been prevalently anoxic in the two northern regions, whereas those in the alpine region may have been more variable (oxic or anoxic).

Funding This research was funded by NWO project 821.01.005 (for JE), Finnish cultural foundation Lapland regional fund (for RN) and by Finnish Academy, project 259180 (for RN) Acknowledgements

!124

Cha

pter

5

Authors acknowledge Maarten Loonen and other members of Netherlands Arctic Station, Ny-Ålesund, Svalbard and Kilpisjärvi Biological Station of the University of Helsinki for assistance in sampling. References

Bahulikar, R. A., Torres-Jerez, I., Worley, E., Craven, K., and Udvardi, M. K. (2014). Diversity of Nitrogen-Fixing Bacteria Associated with Switchgrass in the Native Tallgrass Prairie of Northern Oklahoma. Applied and Environmental Microbiology 80, 5636–5643. doi:10.1128/AEM.02091-14.

Brankatschk, R., Töwe, S., Kleineidam, K., Schloter, M., and Zeyer, J. (2011). Abundances and potential activities of nitrogen cycling microbial communities along a chronosequence of a glacier forefield. The ISME journal 5, 1025–37. doi:10.1038/ismej.2010.184.

Caporaso, J. G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F. D., Costello, E. K., et al. (2010). QIIME allows analysis of high-throughput community sequencing data. Nature Methods 7, 335–336. doi:10.1038/nmeth.f.303.

Chapin, F., Bloom, A., Field, C., and Waring, R. (1987). Plant responses to multiple environmental factors. Bioscience 37, 49–57.

Ciccazzo, S., Esposito, A., Borruso, L., and Brusetti, L. (2016). Microbial communities and primary succession in high altitude mountain environments. Annals of Microbiology 66, 43–60. doi:10.1007/s13213-015-1130-1.

Dang, H., Yang, J., Li, J., Luan, X., Zhang, Y., Gu, G., et al. (2013). Environment-dependent distribution of the sediment nifH-harboring microbiota in the Northern South China Sea. Applied and environmental microbiology 79, 121–32. doi:10.1128/AEM.01889-12.

Edgar, R. C. (2010). Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461. doi:10.1093/bioinformatics/btq461.

Fish, J. A., Chai, B., Wang, Q., Sun, Y., Brown, C. T., Tiedje, J. M., et al. (2013). FunGene: the functional gene pipeline and repository. Frontiers in Microbiology 4, 291. doi:10.3389/fmicb.2013.00291.

Gaby, J. C., and Buckley, D. H. (2012). A comprehensive evaluation of PCR primers to amplify the nifH gene of nitrogenase. PloS one 7, e42149. doi:10.1371/journal.pone. 0042149.

Hardoim, P. R., Hardoim, C. C. P., van Overbeek, L. S., and van Elsas, J. D. (2012). Dynamics of seed-borne rice endophytes on early plant growth stages. PloS one 7, e30438. doi:10.1371/journal.pone.0030438.

Izquierdo, J. a, and Nüsslein, K. (2006). Distribution of extensive nifH gene diversity across physical soil microenvironments. Microbial ecology 51, 441–52. doi:10.1007/s00248-006-9044-x.

Kumar, M., Antonielli, L., Brader, G., Sessitsch, A., Niku, J., Taskinen, S., et al. (2016a). Fungal Communities Associated with Arcto-Alpine Plants are Strongly Shaped by Regional Effects. In Prep.

Kumar, M., Brader, G., Sessitsch, A., Mäki, A., van Elsas, J. D., and Nissinen, R. M. (2016b). Plants Assemble Specific Bacterial Communities from Common Core Taxa in Three Arcto-Alpine Climate Zones. Frontiers in Microbiology, In review.

Mäki, A., Rissanen, J. A., and Tiirola, M. (2016). A practical method for barcoding and size-trimming PCR templates for amplicon sequencing. Biotechniques 60, 88–90. doi:10.2144/000114380.

Rösch, C., Mergel, A., and Bothe, H. (2002). Biodiversity of denitrifying and dinitrogen-fixing bacteria in an acid forest soil. Applied and Environmental … 68, 3818–3829.

!

!!

125

Dia

zotr

ophi

c ba

cter

ial c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

doi:10.1128/AEM.68.8.3818. Sessitsch, A., Hardoim, P., Döring, J., Weilharter, A., Krause, A., Woyke, T., et al. (2012).

Functional Characteristics of an Endophyte Community Colonizing Rice Roots as Revealed by Metagenomic Analysis. Molecular Plant-Microbe Interactions 25, 28–36. doi:10.1094/MPMI-08-11-0204.

Sistla, S. a., Asao, S., and Schimel, J. P. (2012). Detecting microbial N-limitation in tussock tundra soil: Implications for Arctic soil organic carbon cycling. Soil Biology and Biochemistry 55, 78–84. doi:10.1016/j.soilbio.2012.06.010.

Tai, X. S., Mao, W. L., Liu, G. X., Chen, T., Zhang, W., Wu, X. K., et al. (2013). High diversity of nitrogen-fixing bacteria in the upper reaches of the Heihe River, northwestern China. Biogeosciences 10, 5589–5600. doi:10.5194/bg-10-5589-2013.

Wallenstein, M. D., Mcmahon, S. K., and Schimel, J. P. (2009). Seasonal variation in enzyme activities and temperature sensitivities in Arctic tundra soils. Global Change Biology 15, 1631–1639. doi:10.1111/j.1365-2486.2008.01819.x.

Weintraub, M. N., and Schimel, J. P. (2005). Nitrogen Cycling and the Spread of Shrubs Control Changes in the Carbon Balance of Arctic Tundra Ecosystems. BioScience 55, 408. doi:10.1641/0006-3568(2005)055[0408:NCATSO]2.0.CO;2.

Wu, L., Ma, K., and Lu, Y. (2009). Prevalence of Betaproteobacterial Sequences in nifH Gene Pools Associated with Roots of Modern Rice Cultivars. Microbial Ecology 57, 58–68. doi:10.1007/s00248-008-9403-x.

Yeager, C. M., Kornosky, J. L., Housman, D. C., Grote, E. E., Belnap, J., and Kuske, C. R. (2004). Diazotrophic Community Structure and Function in Two Successional Stages of Biological Soil Crusts from the Colorado Plateau and Chihuahuan Desert. Applied and Environmental Microbiology 70, 973–983. doi:10.1128/AEM.70.2.973-983.2004.

You, M., Nishiguchi, T., Saito, A., Isawa, T., Mitsui, H., and Minamisawa, K. (2005). Expression of the nifH Gene of a Herbaspirillum Endophyte in Wild Rice Species!: Daily Rhythm during the Light-Dark Cycle. 71, 8183–8190. doi:10.1128/AEM.71.12.8183.

Zehr, J. P., and Mcreynolds, L. A. (1989). Use of Degenerate Oligonucleotides for Amplification of the nifH Gene from the Marine Cyanobacterium Trichodesmium thiebautii. Applied and Environmental Microbioogy 55, 2522–2526.

Zhang, B., Ryan Penton, C., Xue, C., Wang, Q., Zheng, T., and Tiedje, J. M. (2015). Evaluation of the ion torrent personal genome machine for gene-targeted studies using amplicons of the nitrogenase gene nifH. Applied and Environmental Microbiology 81, 4536–4545. doi:10.1128/AEM.00111-15.

!126

Cha

pter

5

Supplemental figure S1 Principal Coordinate Analysis (PCoA) plots rhizosphere soil PNFB communities (nifH-based OPU sequences as proxy species) from O. digyna (O) and S. oppositifolia (S) plant species from three climatic regions Mayrhofen (MA), Kilpisjärvi (KJ) and Ny-Ålesund (NÅ)

Supplemental figure S2 Principal Coordinate Analysis (PCoA) plots endophytic PNFB communities (nifH-based OPU sequences as proxy species) from O. digyna (O) and S. oppositifolia (S) plant species from three climatic regions Mayrhofen (MA), Kilpisjärvi (KJ) and Ny-Ålesund (NÅ)

!

!!

127

Dia

zotr

ophi

c ba

cter

ial c

omm

uniti

es in

arc

to-a

lpin

e pl

ants

 128

!

!

Chapter 6

Manoj Kumar

Microbial diversity in arcto-alpine plants: a synthesis

!130

Cha

pter

6

!

!!

131

Mic

robi

al D

iver

sity

in A

rcto

-Alp

ine

Plan

ts: a

Syn

thes

is

Background of the study Microbes (bacteria and fungi) and terrestrial plants have co-evolved since the time higher plants appeared on the Earth (Zhang et al., 2006). Indeed, the association of mycorrhizal fungi with plants may have assisted the colonization of land by plants about 450 Myr ago (Redecker et al., 2000; Simon et al., 1993). In addition, we now know that particular bacteria can also be tightly associated with plants, often occurring inside plant tissue (Kloepper et al., 1992). In addition to the classically known rhizobia, a wealth of other bacteria have been isolated from inside root, stem, leaf, fruit and even seeds (Chanway, 1996). However, the diversity, ecology and functional properties of plant-associated microbes are still not well understood. All the more so, only few studies on arcto-alpine plants have been reported so far (Nissinen et al., 2012; Poosakkannu et al., 2015; Zhang et al., 2015). In this thesis, I set out to analyze the diversity of microbial (bacterial as well as fungal) communities that are associated with two selected pioneer plants, Oxyria digyna and Saxifraga oppositifolia, in arcto-alpine soils. Implicit in this study was the assumption that these plants, for their healthy growth in the cold and nutrient-poor soils, were to some extent dependent on their local microbiomes. In particular, I hypothesized that some of their functions, e.g. nitrogen provision, depended on plant-associated microorganisms. In the light of the overriding effects of edaphic and climatic factors on soil status, I also speculated that bulk soil microbial communities are mainly influenced by geographic region and that, consequently, the plant-associated microbial communities are selected from these, forming sub-communities. I selected the aforementioned two plant species, O. digyna and S. oppositifolia, on the premise that, even though they prefer similar growth habitats, they secrete different compounds in the exudates from their roots and thus influence their associated communities in different manners.

The sparse information that was available at the onset of this study was as follows:

Bacterial and fungal diversities in cold soil habitats do not follow similar pattern Chu et al. (2010), using next generation sequencing, found that soil bacterial diversity in the Arctic is similar to that in other biomes. In contrast, fungal communities from polar regions (Antarctic and Arctic) were indicated to exhibit endemism in these regions, with cold climate effectively acting as a habitat-specific filter (Cox et al., 2016) . Meanwhile, both bacterial and soil fungal communities in the Arctic region were shown to be influenced by soil pH, and so soil pH regulated these communities, albeit to different extents (Männistö et al., 2007; Zhang et al., 2016).

!132

Cha

pter

6

Contradicting statements about host specificity of plant-associated microbes in arcto-alpine soils Nissinen et al. (2012) observed that endophytes isolated from three arctic plants, i.e. O. digyna, Diapensia lapponica and Juncus trifidus, exhibit host plant specificity. They further identified members of these endosphere communities that closely resembled isolates from other cold habitats. A similar host plant specificity was recently reported for endophytic fungal communities from S. oppositifolia, Saxifraga cespitosa, Silene acaulis and Cassiope tetragona sampled from Ny-Ålesund, Svalbard (Zhang et al., 2015). However, other studies indicate fluctuating host specificity for both bacterial and fungal root-associated communities in cold-climate soils (Botnen et al., 2014; Fujimura and Egger, 2012; Teixeira et al., 2010), and so the level of host plant selectivity is controversial and may depend on differing factors, such as plant species and edaphic factors and the niche studied (endopshere vs rhizosphere). Diazotrophs from cold habitats are similar

Nitrogen is considered to be a major limiting factor for plant (and microbial) growth in arcto-alpine biomes. Deslippe and Egger (2006) analyzed the molecular diversity of nifH genes from bacteria associated with high-arctic plants and found a striking similarity between the composition of bacterial nifH clone libraries of bulk soils and roots of high Arctic shrubs from Canadian arctic and those from other cold habitats. They postulated that the distribution of nitrogen fixing bacteria can be predicted based on the habitat type (referring to the cold habitats here). Some other studies - not addressed here - have yielded somewhat erratic results; these are subject of the discussions in the research sections of this thesis.

On the basis of the sparse information available, as depicted above, I took as a strategy to determine the microbiomes of three cold climate regions, including one in the European alps (Mayrhofen, Austria) and two across the northern sphere (Kilpisjärvi, Finnish Lapland [low-arctic] and Ny-Ålesund, Svalbard archipelago; [high-arctic]).

In this study, I have thus been able to pinpoint the unknown/undiscovered plant-associated microbes that are associated with the aforementioned two pioneer plants in this pristine environment, as discussed in the following sections. In the light of the importance of such microbiomes for plant growth and health in soils of other climate zones, I placed a focus on both the bacterial and fungal communities in the samples. Moreover, I also

!

!!

133

Mic

robi

al D

iver

sity

in A

rcto

-Alp

ine

Plan

ts: a

Syn

thes

is

addressed the (diversity and composition of) diazotrophs given their postulated importance in the study systems. My hypotheses in this study were:

Bacterial community structure in arcto-alpine soils is highly influenced by geographic region or soil physico-chemical properties.

Plant-associated microbial communities are primarily dependent on plant species, indicating a strong selection by the plants.

Nitrogen acquisition by plants in arctic soils is driven strongly by plant-associated bacteria, and the diazotrophic communities are shaped by plant species.

Below, I come to some overriding conclusions on the basis of these hypotheses and the data generated in the research chapters of this thesis. Endosphere microbial communities have reduced diversities and are different from (bulk and rhizosphere) soil microbial communities There is compelling evidence, from all data generated in chapters 3, 4 and 5, that the diversities and community structures (of bacteria, fungi and potential nitrogen fixing bacteria (PNFB)) in the endospheres of the two selected plant species – in the three geographic regions studied, are different from those in the (bulk and rhizosphere) soil. Expectedly, the endosphere total bacteria (chapter 3), PNFB (chapter 5) and fungal (chapter 4) communities had very low richness and diversity values when compared to those of the (bulk and rhizosphere) soil samples. This generic key finding indicates that the two plant species that were selected act as effective ‘filters’, selecting a restricted set of microbes as endophytes from the highly diverse soil communities. The putative function of such endophytes in these cold-adapted plants has to await further study, but activities with respect to ‘protection’ from frost, low nutrient status and drying/osmotic tension are all possible (Subramanian et al., 2015; White and Torres, 2010).

We thus also identified the key bacterial and fungal taxa that contribute to the apparent ‘endophyte specificity’. The bacterial genera Clostridium (Firmicutes) and Leptothrix and Burkholderia (Burkholderiales), were selectively enriched in the endospheres, as became apparent from both the bacterial (chapter 3) and PNFB communities (chapter 5) communities. Coupled to the fact that such genera are presumably involved in (non-rhizobial) nitrogen fixation (and that they were also found by using the nifH gene proxy), it is logical to posit that they play an important role in the nitrogen acquisition by plants in these N limited soils. The endospheres of the reportedly EM plant S. oppositifolia were enriched with dark septate endophytic fungi from the genera Leptodontidium and Phialocephala. Both Leptodontidium and Phialocephala have been

!134

Cha

pter

6

reported as fungal endophytes in the high-arctic EM plants Dryas octopetala and Bistorta vivipara (Bjorbækmo et al., 2010; Mundra et al., 2015). Newsham et al. (2009) reported that the dark septate fungi such as the ones found by us can be present as endophytes in arcto-alpine plants and are predicted to enhance plant performance. The largely non-mycorrhizal O. digyna was predominated by the presumably saprotrophic Varicosporium. However the functional mechanisms of these fungi-plant symbiosis are unexplored and a potential future prospect.

!Figure 1 The interactions of bulk, rhizosphere and endospere communities of bacteria (red), fungi (green) and potential nitrogen fixing bacterial (orange) communities

Endosphere communities - Host plant specificity The bacterial community structures in the endosphere (chapter 3) were influenced by plant species, whereas the fungal (chapter 4) and PNFB (chapter 5) community structures had low host plant specificity (Figure 1). Thus, the two arctic plants may interact differently with respect to the ‘selection’ of bacterial and fungal types into their endospheres. In particular the fungi and the PNFB, as potential nutrient providers, may have been more ‘generically’ attracted, reflecting their potential key roles in plant support. Moreover, it would be intriguing to assess to what extent these groups interact between themselves, in addition to their interactions with the plant. Thus, a detailed study on fungal-bacterial interactions with focus on bacterial nitrogen fixers in the Arctic would constitute a sensible future research scenario.

!

!!

135

Mic

robi

al D

iver

sity

in A

rcto

-Alp

ine

Plan

ts: a

Syn

thes

is

Rhizosphere communities – Host plant specificity Overall, the influence of plant species on the rhizosphere communities was rather low when compared with that on the endosphere ones (Figure 1). The bacterial communities (chapters 2 and 3) revealed a moderate degree of host plant species specificity, whereas the fungal ones had low host plant specificity (chapter 4). Also the ‘functional profiles’ of the rhizosphere communities of O. digyna and S. oppositifolia were different. In particular, we observed a raised abundance of antibiotic resistance bacteria in the O. digyna rhizosphere, which can be also related to the non-mycorrhizal status of this plant host. On the basis of these collective profiles, we surmised that the provision of different conditions, possibly related to the secretion of different compounds in the root exudates by O. digyna and S. oppositifolia are at the basis of the establishment of different bacteriomes in the rhizosphere (and endosphere, as discussed in the foregoing). Rhizosphere communities of bacteria and fungi assemble via different rules Analysis of the bacterial communities (chapters 2 and 3) versus the fungal (chapter 4) and PNFB (chapter 5) ones revealed that the former responded differently from the two latter ones. Thus, specifically, the bacterial rhizosphere communities were more strongly influenced by plants than the fungal or PNFB rhizosphere communities. In several prior studies in temperate climate soils, strong selection of bacterial communities by the rhizosphere of different plants was also found, and most authors have attributed this rather strong effect to the action of plant root exudates on soil bacteria (Duineveld et al., 1998; !nceo"lu et al., 2011; Smalla et al., 2001).We here extend this rhizosphere selective effect fact to two plants in cold-climate soils. We also know, from previous work, that bacteria may show stronger responses to plant exudates than fungi, and so the selection of fungal types by community assessments is often more difficult to discern. Rather, our observations indicate that fungal rhizosphere communities are more strongly influenced by local soil properties. So, generalizing, one may posit that, within the limits of our observations and tools, bacteria appeared as the more avid responders to the two plants, as compared to the fungi.

In addition, both the bacterial (chapters 2 and 3) and fungal (chapter 4) rhizosphere communities had higher richness and diversity values than the corresponding bulk soil communities. This is in contradiction to multifold studies from agricultural soils where bulk soil communities tended to show highest diversity (!nceo"lu et al., 2011; Kowalchuk et al., 2002; Smalla et al., 2001). However, our observation is in support of other data from low-nutrient habitats (Coleman-Derr et al., 2016; Miniaci et al., 2007). It may be attributed to the sparse availability of nutrients in the bulk soil, which I speculate limits

!136

Cha

pter

6

the overall niche-defined diversity. Thus, the plant root exudates as well as other condition in the plant rhizospheres may have been acting as providers of diverse nutrients/conditions and so amplified the number of colonizable niches.

In contrast, the diversity of PNFB (chapter 5) was not raised in the rhizospheres as compared to that in the corresponding bulk soils. Hence, it is possible that the potentially strong selection by the two plants – in this functional group – indeed reduced diversity, as related to the selective role of these organisms. This facet of bacterial life in the rhizosphere of the arctic plants, much like for the endosphere, warrants dedicated future research endeavors. The influence of geographic region on microbiomes is dependent on compartment In the light of the differential effect of compartment (chapters 2, 3, 4 and 5), I will discuss the effect of region as per compartment (bulk soil, rhizosphere soil, endosphere), as below:

Bulk soils The assessments of bacterial (chapters 2 and 3), fungal (chapter 4) and PNFB (chapter 5) community structures across the bulk soil samples revealed a strong regional specificity, in particular for the bacterial communities (total and PNFB). In other words, the bacterial and PNFB community structures of the bulk soils were different per region. Moreover, the fact that the diversities of the bulk soil bacterial communities were positively correlated with soil pH (range 4.5 to 8.2) and soil organic matter indicated, in support of work by Fierer and Jackson (2006), that soil pH was a key ‘shaper’ of these communities. The effect of region on the bulk soil fungal communities was much weaker, however their community structures were, again, primarily influenced by soil pH. Overall, I thus hypothesize that the conditions in the local soils, under which climatic and soil physicochemical (e.g. pH) properties, were the key drivers of all bulk soil communities.

Rhizosphere soils Above, I addressed the selective effect of plants and plant species on the microbiomes, as observed in their rhizospheres. Here, I examine the influence of geographic region. First, the fungal (chapter 4) and PNFB (chapter 5) rhizosphere communities displayed a strong regional selection when compared to the bacterial communities (chapter 3). Clearly, whereas the bacterial communities had a more ‘ubiquitous’ nature, both the fungal and PNFB showed a more endemic one. Concerning the latter, I here speculate that the rhizosphere PNFB communities are strongly affected by the local climatic conditions or soil edaphic factors, and may shape the environment for the

!

!!

137

Mic

robi

al D

iver

sity

in A

rcto

-Alp

ine

Plan

ts: a

Syn

thes

is

local plants. For example, the cyanobacteria Nostoc punctiforme and Nodularia spumigena were predominantly observed in the Ny-Ålesund regions. They were reported as key primary producers mediating carbon and nitrogen fixation in early succession soils and also mediating plant establishment in low-nutrient soils (Borin et al., 2010).

Endosphere The endosphere communities of both fungi and PNFB were clearly influenced by region, whereas the total bacterial communities responded to a lesser extent (Figure 1). The strong regional influence on the PNFB communities was remarkable. I relate these to effects of the local climatic conditions (as mentioned above). Clearly, the anaerobic nitrogen-fixing genus Clostridia dominated the Arctic region samples, while the genus Geobacter (that is capable of fixing nitrogen under both aerobic and anaerobic conditions) dominated the endosphere communities of the alpine region. Indeed, predominantly anaerobic conditions were observed frequently in the soil pockets that held the plants in the Arctic, due to late melting soil and underlying permafrost, whereas such conditions were less frequently found in the alpine region. Is there a (beneficial) core microbiome that establishes in cold-soil-adapted plant species? The question whether the two pioneer plant species O. digyna and S. oppositifolia contain a core (trans-species) set of microorganisms (that may yield growth/fitness benefits) was addressed by comparing the key shared members of the endosphere communities (Table 1). Most of the taxa identified in the endosphere bacteriomes (Clostridium, Bradyrhizobiaceae, Comomonadaceae [Leptothrix]) were also found in the endosphere ‘PNFB-omes’, and I therefore strongly argue that they play a potentially vital role in nitrogen acquisition. Moreover, in a side project, we also detected Clostridium, Comamonadaceae and Bradyrhizobiae from the seeds of O. digyna (Given et al., in preparation), indicating seed transmission. However, the question whether such core microbes are transported via seeds, thus connecting plant generations, remained unanswered. In other work, such organisms have been found in seeds, which indicates they are vertically transmitted and may be selected by the host plants to be carried off to the next generations (Hardoim et al., 2012; Truyens et al., 2014). A case in point is offered by the seeds of cardon cactus (Pachycereus pringlei) which carry endophytes that allow them to establish in barren rocks by weathering these and dissolving the rock minerals (Puente et al., 2009). We suspect similar mechanisms might be involved in the colonization of the non-mycorrhizal O. digyna in the nutrient-limited rocky soils that it usually colonizes. Furthermore, several studies, referred to in Table 1, have revealed that, in addition to nitrogen fixation, these key OTUs also have diverse potential functional aspects like enhancement of stress tolerance,

!138

Cha

pter

6

phosphate acquisition, resistance against plant pathogens by secretion of antimycotic compounds, and, finally, promotion of plant growth by synthesis of growth promoting hormones like gibberellins. With reference to this work, I argue here that we are now at a stage where the functional implications of our arctic plant endophytes need to be addressed. Outlook and recommendations for further work The study of the interactions of arcto-alpine plants with local microbiomes is relatively new, and the current study has added a large body of information to this area. In the work, I have been able to pinpoint the bacterial, fungal and PNFB communities enriched in the endosphere but I did not yet obtain any information on the details of the functional properties of those communities. It is likely that, in addition to an involvement in nitrogen fixation, the community members have roles in, for instance, frost protection and acquisition of compounds like phosphorus (in addition to nitrogen). Moreover, a possible steering role of plant physiology might be assumed. Future studies on the functional aspects of (members of) the plant-associated microbiomes and why the arcto-alpine plants are in need of these microbial associations for their fitness is vital. I have thus posited that key endosphere specialists assist in the fitness/survival of plants in these cold and nutrient (specially N and P) limited soils (Table 1). For example, some of these endosphere specialists (Table 1) were predominantly present in the endospheres of two plant species across all the regions, and so this indicates they have acquired a niche that potentially also supported their host plant. As I mentioned in my conclusion, there is mounting evidence for the tenet that such plant-supportive endosphere dwellers can be vertically transmitted as they make part of the seed microbiome.

Future work should thus attempt to isolate key plant helper species (both bacteria and fungi), as this will allow studies on behavior and function at the plant. This will also give us vital information on the strategies employed by these pioneer plants in colonizing the low-nutrient mineral-rich deglaciated moraines.

As argued before, a key lack of knowledge concerns the cycles of N and P in the studied habitats. Clearly, the level of N and presumably also of P is the limiting factor for plant growth in the studied arcto-alpine habitats. Studies on the functional involvement of selected microbes in these processes under local conditions is needed. In addition, an investigation into the potential interaction of fungal-bacterial (nifH-endowed) interactions in the rhizosphere, especially in the context of the mycorrhizal associations (considering differential association of O. digyna (non-mycorrhizal) and S. oppositifolia (ecto-mycorrhizal)) will be necessary.

!

!!

139

Mic

robi

al D

iver

sity

in A

rcto

-Alp

ine

Plan

ts: a

Syn

thes

is

The study of plant-associated communities in arcto-alpine plants is new, and important in an era of global warming; so far only few papers have been published in this area. Thus the mechanistic underpinning of the presumed microbial support of plant growth and health in the challenges posed by the cold (although warming) and nutrient limited soils is in its infancy. Exploring the functional consequences of the microbial diversity in the Arctic, in both bulk soil and associated with plants, will be the “arctic expedition” of this century and I am eager to reach the pole References

Bahulikar, R. A., Torres-Jerez, I., Worley, E., Craven, K., and Udvardi, M. K. (2014). Diversity of Nitrogen-Fixing Bacteria Associated with Switchgrass in the Native Tallgrass Prairie of Northern Oklahoma. Applied and Environmental Microbiology 80, 5636–5643. doi:10.1128/AEM.02091-14.

Bjorbækmo, M. F. M., Carlsen, T., Brysting, A., Vrålstad, T., Høiland, K., Ugland, K. I., et al. (2010). High diversity of root associated fungi in both alpine and arctic Dryas octopetala. BMC plant biology 10, 244. doi:10.1186/1471-2229-10-244.

Borin, S., Ventura, S., Tambone, F., Mapelli, F., Schubotz, F., Brusetti, L., et al. (2010). Rock weathering creates oases of life in a high Arctic desert. Environmental Microbiology 12, 293–303. doi:10.1111/j.1462-2920.2009.02059.x.

Botnen, S., Vik, U., Carlsen, T., Eidesen, P. B., Davey, M. L., and Kauserud, H. (2014). Low host specificity of root-associated fungi at an Arctic site. Molecular Ecology 23, 975–985. doi:10.1111/mec.12646.

Chanway, C. P. (1996). Endophytes: they’re not just fungi! Canadian Journal of Botany 74, 321–322. doi:10.1139/b96-040.

Chu, H., Fierer, N., Lauber, C. L., Caporaso, J. G., Knight, R., and Grogan, P. (2010). Soil bacterial diversity in the Arctic is not fundamentally different from that found in other biomes. Environmental microbiology 12, 2998–3006. doi:10.1111/j.1462-2920.2010.02277.x.

Coleman-Derr, D., Desgarennes, D., Fonseca-Garcia, C., Gross, S., Clingenpeel, S., Woyke, T., et al. (2016). Plant compartment and biogeography affect microbiome composition in cultivated and native Agave species. New Phytologist 209, 798–811. doi:10.1111/nph.13697.

Cox, F., Newsham, K. K., Bol, R., Dungait, J. A. J., and Robinson, C. H. (2016). Not poles apart: Antarctic soil fungal communities show similarities to those of the distant Arctic. Ecology Letters 19, 528–536. doi:10.1111/ele.12587.

Cruz, A. F., Horii, S., Ochiai, S., Yasuda, A., and Ishii, T. (2008). Isolation and analysis of bacteria associated with spores of Gigaspora margarita. Journal of Applied Microbiology 104, 1711–1717. doi:10.1111/j.1365-2672.2007.03695.x.

Deslippe, J. R., and Egger, K. N. (2006). Molecular diversity of nifH genes from bacteria associated with high arctic dwarf shrubs. Microbial Ecology 51, 516–525. doi:10.1007/s00248-006-9070-8.

Duineveld, B., Rosado, A., van Elsas, J. D., and van Veen, J. A. (1998). Analysis of the dynamics of bacterial communities in the rhizosphere of the chrysanthemum via denaturing gradient gel electrophoresis and substrate utilization. Applied and Environmental Microbiology 64, 4950–4957.

Elbeltagy, A., Nishioka, K., Sato, T., Suzuki, H., Ye, B., Hamada, T., et al. (2001).

!140

Cha

pter

6

Endophytic Colonization and In Planta Nitrogen Fixation by a Herbaspirillum sp. Isolated from Wild Rice Species. Applied and Environmental Microbiology 67, 5285–5293. doi:10.1128/AEM.67.11.5285-5293.2001.

Fierer, N., and Jackson, R. B. (2006). The diversity and biogeography of soil bacterial communities. Proceedings of the National Academy of Sciences of the United States of America 103, 626–31. doi:10.1073/pnas.0507535103.

Francis, I., Holsters, M., and Vereecke, D. (2010). The Gram-positive side of plant-microbe interactions. Environmental Microbiology 12, 1–12. doi:10.1111/j.1462-2920.2009.01989.x.

Fujimura, K. E., and Egger, K. N. (2012). Host plant and environment influence community assembly of High Arctic root-associated fungal communities. Fungal Ecology 5, 409–418. doi:10.1016/j.funeco.2011.12.010.

Hallmann, J. (2001). “Plant interactions with endophytic bacteria,” in Biotic Interactions in Plant-Pathogen Associations, eds. M. J. Jeger and N. J. Spence (CABI), 87–118.

Han, J.-I., Choi, H.-K., Lee, S.-W., Orwin, P. M., Kim, J., LaRoe, S. L., et al. (2011). Complete Genome Sequence of the Metabolically Versatile Plant Growth-Promoting Endophyte Variovorax paradoxus S110. Journal of Bacteriology 193, 1183–1190. doi:10.1128/JB.00925-10.

Hardoim, P. R., Hardoim, C. C. P., van Overbeek, L. S., and van Elsas, J. D. (2012). Dynamics of seed-borne rice endophytes on early plant growth stages. PloS one 7, e30438. doi:10.1371/journal.pone.0030438.

!nceo"lu, Ö., Al-Soud, W. A., Salles, J. F., Semenov, A. V., and van Elsas, J. D. (2011). Comparative Analysis of Bacterial Communities in a Potato Field as Determined by Pyrosequencing. PLoS ONE 6, e23321. doi:10.1371/journal.pone.0023321.

Itakura, M., Saeki, K., Omori, H., Yokoyama, T., Kaneko, T., Tabata, S., et al. (2009). Genomic comparison of Bradyrhizobium japonicum strains with different symbiotic nitrogen-fixing capabilities and other Bradyrhizobiaceae members. The ISME Journal 3, 326–339. doi:10.1038/ismej.2008.88.

Izumi, H., Anderson, I. C., Alexander, I. J., Killham, K., and Moore, E. R. B. (2006). Endobacteria in some ectomycorrhiza of Scots pine (Pinus sylvestris). FEMS Microbiology Ecology 56, 34–43. doi:10.1111/j.1574-6941.2005.00048.x.

Kloepper, J. W., Schippers, B., and Bakker, P. A. H. M. (1992). Proposed Elimination of the Term Endorhizosphere. Phytopathology 82, 726–727.

Kohout, P., S!korová, Z., Ctvrtlíková, M., Rydlová, J., Suda, J., Vohník, M., et al. (2012). Surprising spectra of root-associated fungi in submerged aquatic plants. FEMS microbiology ecology 80, 216–35. doi:10.1111/j.1574-6941.2011.01291.x.

Kowalchuk, G. A., Buma, D. S., de Boer, W., Klinkhamer, P. G. L., and van Veen, J. A. (2002). Effects of above-ground plant species composition and diversity on the diversity of soil-borne microorganisms. International Journal of General and Molecular Microbiology 81, 509–520. doi:10.1023/A:1020565523615.

Li, J. Y., Strobel, G., Harper, J., Lobkovsky, E., and Clardy, J. (2000). Cryptocin, a potent tetramic acid antimycotic from the endophytic fungus Cryptosporiopsis cf. quercina. Organic letters 2, 767–770. doi:10.1002/chin.200027221.

Madhaiyan, M., Poonguzhali, S., Senthilkumar, M., Pragatheswari, D., Lee, J.-S., and Lee, K.-C. (2015). Arachidicoccus rhizosphaerae gen. nov., sp. nov., a plant-growth-promoting bacterium in the family Chitinophagaceae isolated from rhizosphere soil. INTERNATIONAL JOURNAL OF SYSTEMATIC AND EVOLUTIONARY MICROBIOLOGY 65, 578–586. doi:10.1099/ijs.0.069377-0.

Magalhaes Cruz, L., Maltempi de Souza, E., Weber, O. B., Baldani, J. I., Dobereiner, J., and de Oliveira Pedrosa, F. (2001). 16S Ribosomal DNA Characterization of Nitrogen-Fixing

!

!!

141

Mic

robi

al D

iver

sity

in A

rcto

-Alp

ine

Plan

ts: a

Syn

thes

is

Bacteria Isolated from Banana (Musa spp.) and Pineapple (Ananas comosus (L.) Merril). Applied and Environmental Microbiology 67, 2375–2379. doi:10.1128/AEM.67.5.2375-2379.2001.

Männistö, M. K., Tiirola, M., and Häggblom, M. M. (2007). Bacterial communities in Arctic fjelds of Finnish Lapland are stable but highly pH-dependent. FEMS Microbiology Ecology 59, 452–465. doi:10.1111/j.1574-6941.2006.00232.x.

Miniaci, C., Bunge, M., Duc, L., Edwards, I., Bürgmann, H., and Zeyer, J. (2007). Effects of pioneering plants on microbial structures and functions in a glacier forefield. Biology and Fertility of Soils 44, 289–297. doi:10.1007/s00374-007-0203-0.

Mundra, S., Halvorsen, R., Kauserud, H., Müller, E., Vik, U., and Eidesen, P. B. (2015). Arctic fungal communities associated with roots of Bistorta vivipara do not respond to the same fine-scale edaphic gradients as the aboveground vegetation. New Phytologist 205, 1587–1597. doi:10.1111/nph.13216.

Newsham, K. K., Upson, R., and Read, D. J. (2009). Mycorrhizas and dark septate root endophytes in polar regions. Fungal Ecology 2, 10–20. doi:10.1016/j.funeco.2008.10.005.

Nissinen, R. M., Männistö, M. K., and van Elsas, J. D. (2012). Endophytic bacterial communities in three arctic plants from low arctic fell tundra are cold-adapted and host-plant specific. FEMS microbiology ecology 82, 510–22. doi:10.1111/j.1574-6941.2012.01464.x.

Poosakkannu, A., Nissinen, R., and Kytöviita, M. M. (2015). Culturable endophytic microbial communities in the circumpolar grass, Deschampsia flexuosa in a sub-Arctic inland primary succession are habitat and growth stage specific. Environmental Microbiology Reports 7, 111–122. doi:10.1111/1758-2229.12195.

Puente, M. E., Li, C. Y., and Bashan, Y. (2009). Rock-degrading endophytic bacteria in cacti. Environmental and Experimental Botany 66, 389–401. doi:10.1016/j.envexpbot.2009.04.010.

Redecker, D., Kodner, R., Graham, L. E., Stahl, M., Pirozynski, K. A., Malloch, D. W., et al. (2000). Glomalean fungi from the Ordovician. Science (New York, N.Y.) 289, 1920–1. doi:10.1126/science.289.5486.1920.

Saito, A., and Minamisawa, K. (2006). Evaluation of the Nitrogen-fixing Ability of Endophytic Clostridia based on Acetylene Reduction and Reverse Transcription-PCR Targeting the nifH Transcript and Ribosomal RNA. Microbes and Environments 21, 23–35. doi:10.1264/jsme2.21.23.

Simon, L., Bousquet, J., Lévesque, R. C., and Lalonde, M. (1993). Origin and diversification of endomycorrhizal fungi and coincidence with vascular land plants. Nature 363, 67–69. doi:10.1038/363067a0.

Smalla, K., Wieland, G., Buchner, A., Zock, A., Parzy, J., Roskot, N., et al. (2001). Bulk and Rhizosphere Soil Bacterial Communities Studied by Denaturing Gradient Gel Electrophoresis#: Plant-Dependent Enrichment and Seasonal Shifts Revealed. Applied and environmental microbiology 67, 4742–4751. doi:10.1128/AEM.67.10.4742.

Subramanian, P., Mageswari, A., Kim, K., Lee, Y., and Sa, T. (2015). Psychrotolerant Endophytic Pseudomonas sp. Strains OB155 and OS261 Induced Chilling Resistance in Tomato Plants (Solanum lycopersicum Mill.) by Activation of Their Antioxidant Capacity. http://dx.doi.org/10.1094/MPMI-01-15-0021-R.

Teixeira, L. C. R. S., Peixoto, R. S., Cury, J. C., Sul, W. J., Pellizari, V. H., Tiedje, J., et al. (2010). Bacterial diversity in rhizosphere soil from Antarctic vascular plants of Admiralty Bay, maritime Antarctica. The ISME journal 4, 989–1001. doi:10.1038/ismej.2010.35.

Truyens, S., Weyens, N., Cuypers, A., and Vangronsveld, J. (2014). Bacterial seed endophytes: genera, vertical transmission and interaction with plants. Environmental Microbiology Reports, n/a–n/a. doi:10.1111/1758-2229.12181.

!142

Cha

pter

6

Wang, K., Yan, P., Ding, Q., Wu, Q., Wang, Z., and Peng, J. (2013). Diversity of culturable root-associated/endophytic bacteria and their chitinolytic and aflatoxin inhibition activity of peanut plant in China. World Journal of Microbiology and Biotechnology 29, 1–10. doi:10.1007/s11274-012-1135-x.

White, J. F., and Torres, M. S. (2010). Is plant endophyte-mediated defensive mutualism the result of oxidative stress protection? Physiologia Plantarum 138, 440–446. doi:10.1111/j.1399-3054.2009.01332.x.

Wu, L., and Guo, S. (2008). Interaction between an isolate of dark-septate fungi and its host plant Saussurea involucrata. Mycorrhiza 18, 79–85. doi:10.1007/s00572-007-0159-9.

You, Y.-H., Yoon, H., Kang, S.-M., Woo, J.-R., Choo, Y.-S., Lee, I.-J., et al. (2013). Cadophora malorum Cs-8-1 as a new fungal strain producing gibberellins isolated from Calystegia soldanella. Journal of Basic Microbiology 53, 630–634. doi:10.1002/jobm.201200002.

Zhang, H. W., Song, Y. C., and Tan, R. X. (2006). Biology and chemistry of endophytes. Natural product reports 23, 753–71. doi:10.1039/b609472b.

Zhang, T., Wang, N.-F., Liu, H.-Y., Zhang, Y.-Q., and Yu, L.-Y. (2016). Soil pH is a Key Determinant of Soil Fungal Community Composition in the Ny-Ålesund Region, Svalbard (High Arctic). Frontiers in microbiology 7, 227. doi:10.3389/fmicb.2016.00227.

Zhang, T., Yao, Y.-F., Bergero, R., Girlanda, M., Varese, G., Intili, D., et al. (2015). Endophytic Fungal Communities Associated with Vascular Plants in the High Arctic Zone Are Highly Diverse and Host-Plant Specific. PLOS ONE 10, e0130051. doi:10.1371/journal.pone.0130051.

!

!!

143

Mic

robi

al D

iver

sity

in A

rcto

-Alp

ine

Plan

ts: a

Syn

thes

is

! !

 144

 

!

!

Manoj Kumar

Summary

Uittreksel

!"#$!

Sum

mar

y

! "#%!

Eng

lish

Sum

mar

y

English Summary Terrestrial plants and microbes have co-evolved since the emergence of the former on Earth. Associations with microorganisms can be either beneficial or detrimental for plants. Microbes can be found in the soil surrounding the plant roots, but also in all plant tissues, including seeds. In arcto-alpine regions, plants face extraordinary challenges, and, although we know a lot about their growth and survival strategies, there is a paucity of information about their associated microbiomes. Here, I investigated the diversity of microbes associated with two arcto-alpine plants, Oxyria digyna and Saxifraga oppositifolia, from three distinct geographical regions, i.e. Mayrhofen, Austria (alpine climate), Kilpisjärvi, Finland (low-arctic climate) and Ny-Ålesund (high-arctic climate). Oxyria digyna and Saxifraga oppositifolia are key pioneer plant species in arcto-alpine soils. In chapter 2, I investigated the influence of these plants on the rhizosphere bacterial communities in two arctic regions. Geographic location and soil type, as well as plant species, significantly influenced the bacterial community structures. Furthermore, the ability to tolerate oxidative stress as well as specific antibiotics was highly pronounced in the rhizosphere-associated communities in both plant species. In chapters 3 and 4, I investigated the biogeographical diversity of the plant-associated microbial communities and the corresponding bulk soils in three arcto-alpine regions. The factor compartment (bulk soil, rhizosphere soil, endosphere) influenced the diversities and community structures of both fungi and bacteria. In particular, the endophytic communities had the lowest diversity and their community structures were different from those of the bulk and rhizosphere soils. In contrast to studies on plants from agricultural soils, I observed a higher diversity of bacterial and fungal communities in our rhizosphere soils when compared to bulk soils. In chapters 3 and 5, I show that geographic region significantly influences the bacterial and PNFB communities in the bulk soils. However, the effect of region on the bulk soil fungal communities was much weaker (chapter 4). These communities were primarily shaped by soil pH. The fungal (chapter 4) and PNFB (chapter 5) communities in the rhizospheres displayed a strong regional selection, while the bacterial (chapters 2 and 3) communities were ubiquitous. Similarly, the fungal and PNFB endosphere communities were clearly influenced by region, whereas the total endosphere bacterial communities responded to a lesser extent.

!"#&!

Sum

mar

y

With respect to the endophytic bacterial and PNFB communities, plant species and geographic region were the main determinants of the community structures. The bacterial community structures in the endosphere (chapter 3) were influenced by plant species, whereas the fungal (chapter 4) and PNFB (chapter 5) community structures had low host-plant specificity. Plants in the alpine region had higher relative abundances of Proteobacteria, as compared to plants from the low- and high-arctic regions, which were dominated by Firmicutes. The PNFB communities in the endosphere were enriched with Geobacter and Clostridium types. Moreover, I observed a strong regional influence (related to climatic conditions) on the endosphere PNFB communities. Anaerobic nitrogen fixing Clostridia dominated the Arctic regions, while Geobacter (capable of fixing nitrogen in both aerobic and anaerobic conditions) dominated the endosphere PNFB communities in the Alpine region. The fungal communities in the endosphere were highly influenced by geographic region, with low host plant specificity. The fungal genera Mycenella, Cadophora and Varicosporium dominated the endospheres. In conclusion, the two pioneer plant species act as effective ‘filters’, selecting a restricted set of microbes as endophytes from the highly diverse soil communities across three geographic regions. Most of the key taxa identified in the endosphere bacteriomes, i.e. Clostridium, Bradyrhizobiaceae, Comomonadaceae [Leptothrix], were also found in the endosphere ‘PNFB-omes’, and I therefore strongly argue that members of these taxa may play a key role in nitrogen acquisition in these low-nitrogen environments.!

!! "#$!

Ned

erla

ndse

Uitt

reks

el!

Nederlandse Uittreksel (Dutch Summary) Terrestrische planten en micro-organismen zijn samen geëvolueerd, sinds de opkomst van eerstgenoemden op aarde. De associatie met micro-organismen kan zowel nuttig als schadelijk zijn voor planten. Microben komen voor in de grond rond de wortels, alsook in alle plantweefsels, waaronder zaden. In Arctisch-alpine gebieden worden de aanwezige planten geconfronteerd met buitengewone uitdagingen. Ofschoon er veel bekend is over de groei en overlevingsstrategieën van deze planten, is er een gebrek aan informatie over de met hen geassocieerde microbiomen. In dit proefschrift beschrijf ik mijn onderzoek naar de diversiteit van micro-organismen die geassocieerd zijn met twee Arctisch-alpine planten, Oxyria digyna en Saxifraga oppositifolia, in drie verschillende geografische regio's, i.e. Mayrhofen, Oostenrijk (alpine klimaat), Kilpisjärvi, Finland (laag-arctisch klimaat) en Ny-Ålesund, Spitsbergen (hoog-arctisch klimaat). O. digyna en S. oppositifolia zijn belangrijke pionierplantsoorten in Arctisch-alpine bodems. In hoofdstuk 2 beschrijf ik het onderzoek naar de invloed van deze planten op de bacteriële gemeenschappen in de rhizosfeer in de twee arctische gebieden. Geografische locatie en bodemtype, alsmede plantsoort, hadden een sterke invloed op de bacteriële gemeenschapsstructuren. Bovendien werd het vermogen om oxidatieve stress en specifieke antibiotica te tolereren beïnvloed in de rhizosfeer-geassocieerde gemeenschappen bij beide plantsoorten. In hoofdstuk 3 en 4 heb ik onderzoek gedaan naar de biogeografische diversiteit van de plant-geassocieerde microbiële gemeenschappen en de bijbehorende bulkgronden in de drie Arctisch-alpine gebieden. De factor ‘compartment’ (bulkgrond, rhizosfeergrond, endosfeer) had invloed op de diversiteit en structuur van zowel de schimmel- als de bacterie-gemeenschappen. Met name de endofytische gemeenschappen hadden de laagste diversiteit en hun structuren verschilden van die van de bulk- en rhizosfeergronden. In tegenstelling tot studies over planten uit landbouwgronden, nam ik een hogere diversiteit aan bacteriën en schimmels waar in de bestudeerde rhizosfeergronden in vergelijking met de respectievelijke bulkgronden. In hoofdstukken 3 en 5 laat ik zien dat de factor geografische regio een sterke invloed had op de bacteriële en PNFB gemeenschappen in de bulkgronden. Het effect van de omgeving op de schimmelgemeenschappen in de bulkgrond was veel zwakker (hoofdstuk 4). Deze gemeenschappen werden voornamelijk gestuurd door de bodem-pH. De schimmel- (hoofdstuk 4) en PNFB- (hoofdstuk 5) gemeenschappen in de rhizosfeer vertoonden een sterke invloed van regio, terwijl de bacteriële (hoofdstuk 2 en 3) gemeenschappen dit niet vertoonden. Ook de schimmel- en PNFB-gemeenschappen in de endosfeer werden duidelijk beïnvloed door de

!"%&!

Uitt

reks

el!

factor regio, terwijl de bacteriële gemeenschappen daar in mindere mate door gestuurd werden.

De endofytisch bacteriële en PNFB gemeenschappen werden in belangrijke mate gestuurd door plantsoort en geografische regio. Waar de bacteriële gemeenschapsstructuren (hoofdstuk 3) werden beïnvloed door plantsoort, hadden de schimmel- (hoofdstuk 4) en PNFB- (hoofdstuk 5) gemeenschapsstructuren een lage gastheerspecificiteit. De planten in het Alpengebied hadden een relatief hoge prevalentie van Proteobacteria, terwijl de planten uit de lage en hoge arctische regio's werden gedomineerd door Firmicutes. De PNFB gemeenschappen in de endosfeer waren verrijkt met Geobacter en Clostridium typen. Hiernaast zag ik een sterke regionale invloed (gerelateerd aan klimatologische omstandigheden) op de PNFB gemeenschappen in de endosfeer. Anaëroob stikstofbindende Clostridia domineerden de Arctische gebieden, terwijl Geobacter (stikstofbinding onder zowel aërobe als anaërobe omstandigheden) de endosfeer-PNFB gemeenschappen in het Alpengebied domineerde. De schimmelgemeenschappen in de endosfeer werden sterk beïnvloed door de factor geografische regio, met lage waardplantspecificiteit. De schimmelgeslachten Mycenella, Cadophora en Varicosporium domineerden de endosferen.

De twee geselecteerde pionierplantsoorten fungeren als effectieve 'filters', die een beperkte set van micro-organismen als endofyten uit de zeer diverse bodemgemeenschappen selecteren in drie geografische regio's. Het merendeel van de sleutel taxa in de endosfeerbacteriomen, i.e. Clostridium, Bradyrhizobiaceae, Comamonadaceae [Leptothrix], werd ook gevonden in de endosfeer 'PNFB-omen', en daarom postuleer ik hier dat de leden van deze taxa een sleutelrol in de binding van atmosferische stikstof spelen in deze laag-stikstofomgevingen.

!!

151

Ack

now

ledg

emen

ts

Acknowledgements “A journey is best measured in friends, rather than miles.” – Tim Cahill My journey here has helped me to make a lot of friends from several countries and I can proudly say that this trip is a most wonderful and memorable one. From the time I left home to pursue my Masters, I have met many people who has had some tremendous impact on me, both personally and professionally and thereby, they have my special appreciation. But foremost, I would like to thank my family, my parents for all the love and support they provided to pursue my dreams ardently and always braced my decision, no matter what. I also extend my special gratitude to my brother Balaji for his love and all the good times we shared among the years, especially in the dark Finnish winters. Being away from home, it was your company that kept me working during some tough times. I could not have done this without him.

Next, I would like to thank my supervisors Riitta Nissinen and Jan Dirk Van Elsas for providing me an opportunity to work in this awesome project. I have learned a quite deal and also got to travel to so many unique places as a part of this project. Dick, you have been always supportive and encouraging, and I can always talk to you about any idea (even the crazy ones!). I am very grateful for all the discussions we had and your perpetual enthusiasm for the soil microbes. It’s very infectious!!! To Riitta, my sincere and heartfelt thanks for the continuous support you provided from start of my Master’s thesis to this day. Special thanks for educating me about dressing up appropriately and warmly for the sampling trips. I learned a great deal from you. Thanks for being tolerant of my mistakes and patiently correcting and teaching me write scientifically. Also, I can’t remember the innumerable times you went out of your way to help me even personally. I could not have a better advisor and mentor for my PhD study. Thank you for always inviting me to your Christmas dinner and also introducing me to Nathan, Osmo and Maisa. It was a pleasure knowing them. I also extend my thanks to the members of the reading committee, Prof. J.T.M. Elzenga, Prof. M.G.A. van der Heijden, Prof. G. Berg for taking the time to read and approve this thesis.

Besides my supervisors I also had an opportunity to work with many talented people in the course of these 4 years and would like to extend my sincere gratitude to each and every one of them. Minna Männistö you along with Riitta is a major influence on my early research career and I genuinely

 152

thank you for being patient and guiding me writing my first research article. Günter, thanks for helping me with the samples in the picturesque European Alps and I would have never done that sampling on the Cliff site without you. Another challenging sampling trip was in the Ny-Ålesund and I sincerely acknowledge Maarten Loonen for protecting us from Polar bears and arctic Skuas. Also thank you for that unofficial trip you took us to the Glacier, that was once in a lifetime experience. Livio, thanks for being patient with me and teaching the basics of sequence analysis in a quick time and making this thesis possible. You were always there to help me. Jenni, your R stat skills are amazing and thanks for teaching some basic R and all the awesome graphs you have produced for this thesis. Anita, thanks for helping me with the Ion-torrent sequencing and always extending your support even in your busiest hours. I also extend my sincerest gratitude to Joana, Sara and Angela for their continuous guidance. Finally, I acknowledge my support group members Miina-Maarit and Lutz from Jyväskylä. Thanks for encouraging me and providing guidance through the toughest times of my PhD. Lutz, thank you for teaching me to play a bit of foosball as well.

I also had an opportunity to work in two different universities as a part of my program and would like to thank everyone who helped me. My first introduction to Groningen happened in the best way possible, and for that, I should honestly thank Koen for letting me couch surf in his unique Dutch styled boathouse and also sharing my first ever-arctic experience with midnight sun. It was quite unique! Vijay, you need a special mention here because I am delighted to know you and our “potato-fry-with-rasam” dinners topped with “philosophical” discussions were just exclusive. I thank Pilar, Miaozhi, Muhammad Shahid and Rashid for helping me during the initial days in the Groningen and all the fun times we shared inside and outside the lab. Stephanie, your enthusiasm in the group is quite contagious and thanks for the friendly hangouts we shared outside the lab. Irshad, Fransico and Pu Yang, thanks for being my dear friends in Groningen and always helping me whenever I needed. Joukje, I always remember your warm greetings with gentle smile and thanks for assisting me in dealing with all the difficult stuff. I am also grateful to the other members of the microbial ecology group who helped me in progressing my career or shared a casual chat or laugh. Miao Wang, Adjie, Larissa, Maryam, Patricia, Michele, Jonas, Elisa, Sasha, Silvia, Maju, Sandy, Diego it was nice to know you all and hope our paths will cross again sometime. Finally, I thank Jolanda and Nelly for helping me with my never-ending soil analysis.

!!

153

Ack

now

ledg

emen

ts

Jyvaskyla holds a special place in my heart and I have met so many wonderful people here in the past three years. I would like to express my heartfelt gratitude to my co-worker Cindy for your help with DNA extractions, sampling and always being there for casual chit chats. I also like to thank my fellow office mates Jenni and Marina for keeping the environment as cheerful and never complaining about my messy workbench. I specially thank Andreas and Sara for the brainstorming statistical discussions we had and tolerating me whenever I asked for your help. I hope for more of such discussions in the near future. Claire and Maria, it was nice knowing you both and as an avid foodie, our “gastronomy club” gave me a chance to taste some of the best cuisine from Spain and Caribbean. Andrecia, I am grateful for your incredible friendship during the long harsh winters and the short amazing Finnish summer. I also get to check off one of my bucket list with you, which was seeing the Northern lights. Apart from doing research and enjoying the everlasting Finnish winter and darkness, I was fortunate to get introduced to the world of board games. Special thanks to Seba and Emily for that, but also to my fellow gamers Otto, Oopi, Jaana, Jimi, Matthew, David and Eppu for all the scheming and plotting we shared. I owe thanks to all my co-workers and friends whom I met here: Joannes, Swanne, Piret, Aigi, Beerik, Maz, Zhara, Rogi, Bibi, Juho, Matthieu, Sami, Diana, Jaakko, Manas, Hema, Rakesh, Jiby, Anbu, Tahti, and Sukithar. You all are an incredible part of this journey and I am glad to know you all. I also offer my sincere gratitude to all the faculty and administrative staff who helped me here. Finally, I should thank Steve for just being the amazing friend he is. I owe my special thanks to you for inviting me to the Poker nights and broadening my social circle in Jyväskylä.

A true friend really gets you and knows what to say, and more

importantly, when to say nothing at all. When you flip through the collage of people in your life, you may have some people in your corner that you can count on and I am fortunate to say I have you guys Sundar and Sathish. Thanks for being a part of this journey and always bringing joy to my life whenever I needed.

I would like to acknowledge both of my paranymphs Mukil and

Stéphanie. Thanks for accepting my request to be my paranymph. Mukil and Kavitha just mere thanks will not be enough to acknowledge all the help and support I received from you both. Whenever I visited Groningen you guys were happy to host me and I could even call your place as home away from home. Thanks again and I hope to maintain this friendship of ours forever.

 154

Stéphanie, there are some people in life that deserve a special place and you are one among them. The years that I spent in Groningen would have been completely different if not for your presence. Your earnest passion and enthusiasm for science is contagious and I thank you for contaminating me. I learned a great deal from you. I have never met someone who is so passionate as you and I remember even our casual walks from lab to home will turn into you chasing seagulls around and trying to read their tags or teaching me about different birds or plants in the park. Many thanks for all the times we shared along all these years.

Finally, I would like to acknowledge the person without whom this

book won’t be a possibility. Utham, it was your relentless support and encouragement that made me to leave India to travel and pursue my Master’s in Sweden. You have been my mentor, well-wisher and a friend through all these years and I sincerely thank you for all the love and support you give me.

!!

155

Aut

hor A

ffili

atio

ns

Author Aff i l ia t ions Manoj Kumar University of Groningen, Department of Microbial Ecology, Nijenborgh 7, 9747 AG, Groningen, The Netherlands University of Jyväskylä, Department of Biological and Environmental Science, Survontie 9, 40520, University of Jyväskylä, Finland e-mail: m.k.gopala.krishnan@rug.nl/makugopa@jyu.fi Angela Sessitsch AIT Austrian Institute of Technology GmbH, Health and Environment Department Bioresources, Konrad-Lorenz-Strasse 24, 3430, Tulln, Austria e-mail: angela.sessitsch@ait.ac.at Anita Mäki University of Jyväskylä, Department of Biological and Environmental Science, Survontie 9, 40520, University of Jyväskylä, Finland e-mail: anita.maki@jyu.fi Günter Brader AIT Austrian Institute of Technology GmbH, Health and Environment Department Bioresources, Konrad-Lorenz-Strasse 24, 3430, Tulln, Austria e-mail: guenter.brader@ait.ac.at Jan Dirk van Elsas University of Groningen, Department of Microbial Ecology, Nijenborgh 7, 9747 AG, Groningen, The Netherlands e-mail: j.d.van.elsas@rug.nl Jenni Niku University of Jyväskylä, Department of Mathematics and Statistics, Ahlmaninkatu 2, 40100, University of Jyväskylä, Finland e-mail: jenni.me.niku@student.jyu.fi Livio Antonielli AIT Austrian Institute of Technology GmbH, Health and Environment Department Bioresources, Konrad-Lorenz-Strasse 24, 3430, Tulln, Austria e-mail: livio.antonielli@ait.ac.at

 156

Minna K Männistö Finnish Forest Research Institute, Eteläranta 55, 96301, Rovaniemi, Finland e-mail: minna.mannisto@luke.fi Riitta M. Nissinen University of Jyväskylä, Department of Biological and Environmental Science, Survontie 9, 40520, University of Jyväskylä, Finland e-mail: riitta.m.nissinen@jyu.fi Sara Taskinen University of Jyväskylä, Department of Mathematics and Statistics, Ahlmaninkatu 2, 40100, University of Jyväskylä, Finland e-mail: sara.taskinen@jyu.fi  

 

Biogeographical diversity of microbialcommunities in arcto-alpine plants

Manoj Kumar

Biogeographical diversity of m

icrobial comm

unities in arcto-alpine plants

M

anoj Kum

ar

Invitation

You are invited to the publicdefense of my PhD thesis

Biogeographical diversity ofmicrobial communities in

arcto-alpine plants

Manoj Kumar

University of GroningenAcademy Buliding

Broerstraat 5, Groningen

Monday, 12.12.2016at 12:45

Paranymphs

Mukil Maruthamuthum.marutha.muthu@rug.nl

Stephanie Grizardstephanie.grizard@univ-lyon.fr

14255_Manoj_Kumar_Cover.indd 1 08-11-16 19:22