introduction of a putative biocontrol agent into a range ...sri.haifa.ac.il/images/vered_2.pdf ·...
TRANSCRIPT
Acc
epte
d A
rticl
eIntroduction of a putative biocontrol agent into a range of Phytoplasma- and
Liberibacter-susceptible crop plants
A putative biocontrol agent can penetrate a wide range of crop plants
Ofir Lidor,1 Orit Dror,2 Dor Hamershlak,2,3 Nofar Shoshana,2 Eduard Belausov,4 Tirtza
Zahavi,5 Netta Mozes-Daube,1 Vered Naor,6,7 Einat Zchori-Fein,1 Lilach Iasur-Kruh8 &
Ofir Bahar2, †
1Department of Entomology, Agricultural Research Organization, Newe Ya’ar
Research Center, Ramat Yishai Israel
2Department of Plant Pathology and Weed Research, Agricultural Research
Organization, Volcani Center, Rishon LeZion, Israel
3The Robert H. Smith Faculty of Agriculture, Food and Environment, the Hebrew
University of Jerusalem, Rehovot, Israel
4Microscopy Unit, Institute of Plant Sciences, Agricultural Research Organization,
Volcani Center, Rishon LeZion, Israel
5Extension Service, Ministry of Agriculture, Israel,
6Shamir Research Institute, Katzrin, Israel,
7Ohallo College, Katzrin, Israel,
8Department of Biotechnology Engineering, ORT Braude College of Engineering,
Karmiel, Israel.
† corresponding author: [email protected] Dr. Ofir Bahar Department of Plant Pathology and Weed Research A.R.O. Volcani Center HaMakkabbim Road 68, Rishon LeZion P.O.B 15159 7528809 ISRAEL Tel: 972-3-9683561
This article is protected by copyright. All rights reserved.
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/ps.4775
Acc
epte
d A
rticl
eAbstract
BACKGROUND
Phytoplasma, the causative agent of bois-noir disease of grapevines, are vectored by
the planthopper Hyalesthes obsoletus (Hemiptera: Cixiidae). A Dyella-like bacterium
(DLB) isolated from H. obsoletus inhibits the growth of Spiroplasma melliferum, a
cultivable relative of phytoplasma. Additional evidence suggests that DLB can reduce
the symptoms of yellows disease in grapevine plantlets. The present aim was to test
whether DLB could colonize a range of Phytoplasma- and Liberibacter-sensitive crop
plants, and thus assess its potential agricultural use.
RESULTS
Vitex agnus-castus – the preferred host plant of H. obsoletus was found to be a natural
host of DLB, which was successfully introduced into a range of crop plants belonging
to seven families. The most effective DLB application method was foliar spraying.
Microscopy observation revealed that DLB aggregated on the leaf surface and around
the stomata, suggesting this is its route of entry. DLB was also present in the vascular
tissues of plants, indicating that it moved systemically through the plant.
CONCLUSIONS
DLB is a potential biocontrol agent and its broad spectrum of host plants indicates the
possibility of its future use against a range of diseases caused by phloem-limited
bacteria.
Keywords: biocontrol, endophytes, Phytoplasma, Liberibacter
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
e1 INTRODUCTION
Candidatus Phytoplasma (Class: Mollicutes, hereafter ‘phytoplasma’) are Gram-
positive, wall-less, non-culturable phytopathogenic bacteria. Phytoplasma
transmission between plants strictly depends on phloem-feeding hemipteran insects,
mostly leafhoppers (Cicadellidae), planthoppers (Cixiidae, Delphacidae, Derbidae, and
Flatidae) and psyllids (Psyllidae).1 These insect vectors introduce phytoplasma cells
directly into the plant’s phloem sieve elements, where the pathogen multiplies and
spreads throughout the plant.2
Phytoplasma infects several hundred plant species, including perennial and annual
crops.3 Common phytoplasma-induced symptoms include leaf yellowing and curling,
shoot proliferation (witch’s broom), flower discoloration (virescence), and malformation
(phyllody).3,4 These abnormalities inevitably impair both quality and quantity of yield.
The most common means of controlling phytoplasma diseases is application of
insecticides to manage vector populations.1,5 The effectiveness of this approach is
limited, especially in perennial crops, where one transmission event by a single vector
is sufficient to cause infection.2 The absence of an epiphytic stage, in which the
bacterium is vulnerable to surface applications of chemicals or biological substances,
and the lack of known crop varieties that are genetically resistant to phytoplasma
diseases, leave farmers with very few options to control phytoplasma disease.1,5 Thus,
there is an urgent need to develop novel management tools against phytoplasma and
other phloem-inhabiting phytopathogenic bacteria. One such tool could be biocontrol
based on the use of beneficial bacteria.
Nonpathogenic endophytic bacteria inhabit plant inner tissues for at least part of their
life cycle without causing any apparent disease.6-9 The association of nonpathogenic
endophytes with plants benefits the latter by inducing plant growth through production
of plant phytohormones,7,10 improved nutrient availability,11,12 accelerated seedling
emergence,13 etc. In addition, some bacterial endophytes may protect plants from biotic
stresses via direct antibiotic activity, competition, and immune-response priming.14,15
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
eOne example of the last is Burkholderia phytofirmans, an endophyte of grapevine that
inhibits Botrytis cinerea mold, directly by antifungal activity and indirectly by priming
H2O2 production and salicylic acid (SA) and jasmonic acid (JA) related gene
expression.16
Nonpathogenic endophytic bacteria also reduce phytoplasma-induced symptoms. For
example, phytoplasma-infected Catharanthus roseus (periwinkle) plants showed fewer
symptoms when inoculated with the biocontrol agent Pseudomonas migulae, strain
8R6.17 The authors speculated that the reduction in symptoms was due to direct
damage to phytoplasma cells and to the 1-aminocyclopropane-1-carboxylate (ACC)
deaminase activity of 8R6, which led to reduced ethylene levels, thereby delaying
disease symptoms.17 Another example is the co-inoculation of daisy plants with
‘phytoplasma asteris’, the causal agent of chrysanthemum yellows phytoplasma (CYP)
and Pseudomonas putida as a protecting agent. Although the presence of P. putida
did not appear to reduce phytoplasma titer, it improved plant growth compared with
that of plants inoculated only with ‘phytoplasma asteris’.18 The mechanism in this case
was not clear, but it was tentatively attributed to bacterial production of indole acetic
acid (IAA), which would compensate for the phytoplasma-induced deficiency of the
phytohormone in infected plants.18
We have recently isolated a Gram-negative, Dyella-like bacterium (DLB) from
Hyalesthes obsoletus (Hemiptera: Cixiidae), the vector of bois-noir phytoplasma in
grapevines.19 This indicates that H. obsoletus can carry both DLB and bois-noir
phytoplasma; however, it is not known whether the two bacteria simultaneously co-
inhabit H. obsoletus. In Israel, H. obsoletus thrives and completes its life cycle on Vitex
agnus-castus (Abraham’s balm) shrubs, but it cannot complete its life cycle on
grapevines.20 Conversely, although this planthopper can transmit phytoplasma to
grapevine, phytoplasma were never detected in Abraham’s balm shrubs in Israel,20 but
they were detected in Abraham’s balm shrubs in Montenegro.21 Whether DLB is
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
etransmitted to and/or acquired from Abraham’s balm plants by H. obsoletus is not
known.
In a recent study,22 we demonstrated that DLB can penetrate grapevine plantlets when
applied by either foliar spraying, soil drenching, root dipping, or injection. Microscopic
and PCR analyses of DLB-inoculated grapevines indicated that DLB cells were located
inside the phloem tissue and could persist for up to a month following application.22
Importantly, DLB inhibited the growth of the Mollicute Spiroplasma melliferum19 in vitro,
and markedly reduced phytoplasma-induced symptoms in grapevine plantlets.22
If DLB prove to be an efficient biocontrol agent, it would be essential to determine its
rate of establishment in relevant susceptible crop plants. In the present study, we
examined the colonizing ability of DLB in a range of crop plants and one environmental
host (Abraham’s balm). We also examined DLB host penetration pathways,
localization patterns and in planta persistence following various inoculation
procedures. Our results indicate that this newly-discovered bacterium – and possible
biocontrol agent – is a natural plant colonizer with a potentially very broad host range.
Furthermore, the complete genome of DLB was recently sequenced and deposited in
the Genbank (NCBI) under accession no. LFQR00000000.23
2 MATERIALS AND METHODS
2.1 Bacterial strains and inoculum preparation
DLB stocks were kept at -80C in a 30% glycerol stock and were freshly streaked on a
modified CCT-agar,19 composed of sucrose (67 g L-1), sorbitol (10 g L-1), Lysogeny
Broth (LB, Difco) (2 g L-1) and Bacto-agar (13 g L-1), and cultivated at 28˚C. Liquid DLB
cultures were prepared by inoculating a 3-5 mL starter containing Nutrient Broth (NB,
Difco) or LB media with a single DLB colony from agar plates. The starters were
cultivated for ~24 h at 28ºC with shaking (220 rpm) until the culture was turbid; they
were then used to inoculate the necessary volume of broth culture at a ratio of 1:100
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
e(V:V) in an Erlenmeyer flask. The DLB cultures were cultivated until they reached a
concentration of ~108 colony forming units (CFU) mL-1 as determined by counts
following serial dilution and plating of cultures on CCT-modified plates incubated for
48 h at 28ºC.
2.2 Generating a gfp-expressing DLB strain
To monitor DLB localization in planta, a gfp-expressing strain (DLBGFP) was generated.
DLB cells were transformed according to Enderle et al (1998).24 In brief, DLB was
streaked on a CTT-modified agar plate and grown at 28ºC for 48 h. A 3-mm colony
was lifted with a sterile loop, and transferred into a 1.5-mL Eppendorf tube containing
500 µL of double-distilled water (DDW). Tube contents were centrifuged at 16,000 x g
(~12,300 rpm – Eppendorf centrifuge model 5810R, rotor F45-30-11) for 1 min and the
supernatant was discarded. The cell pellet was resuspended with 500 µL of DDW,
further washed with a repeated centrifugation step, and finally resuspended in 50 µL
of DDW. The cells were kept on ice and a plasmid harboring a green fluorescent protein
(gfp) and a kanamycin-resistance gene (‘pKT-Kn’)25 was added to the tube. The
mixture of cells and plasmid was then transferred to an electroporation cuvette and
placed in an electroporator (ECM 399 model, BTX, Harvard Apparatus, Holliston, MS,
USA). Electroporation was set at 150 Ω, 36 μFD, and field strength of 1.8-2.5 kV was
obtained, depending upon the gap size of the cuvette (1 mM gave 1.8 kV; 2 mM, 2.4
kV). A 1-mL aliquot of S.O.C medium was immediately added to the cuvette and then
collected and transferred into a 15-mL tube for incubation (1 h at 28˚C, 220 rpm).
Electroporated cells were then plated at two volumes (20 and 200 µL) on CCT-modified
agar plates containing kanamycin (50 µg mL-1). DLB colonies growing on the
kanamycin plates were confirmed to carry the ‘pKT-Kn’ plasmid by confocal
microscopy.
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
e2.3 Abraham’s balm shrubs and H. obsoletus insect sampling
Abraham’s balm shrubs grown outdoors at the Newe-Ya’ar Research Center (originally
collected in 1996 near the Kishon River) were used to determine whether DLB naturally
colonizes Abraham’s balm shrubs. The plants were drip-irrigated continuously
throughout the year (10 m3 per 3 weeks) apart from the winter months (December-
April). Four Abraham’s balm shrubs were sampled monthly from September to
December (2016) and again in April through June (2017); they were not sampled
during January through March because Abraham’s balm is a deciduous shrub and was
leafless during those months. from each sampling date, two leaves were randomly
collected from each plant and DNA was extracted from each leaf separately (n = 8).
DLB presence was determined by PCR with specific primers (as described in 2.6).
To assess the prevalence of DLB in H. obsoletus feeding on DLB-infected Abraham’s
balm shrubs, insects were collected at the same site in Newe Ya’ar in October 2016,
with a vacuum unit comprising a modified Echo model PB 1000 leaf blower, in which
the air intake and exhaust ports were interchanged as described elsewhere.26
Captured insects were anesthetized (-20ºC, 30 min) and H. obsoletus individuals were
separated from other insects. Ten H. obsoletus adults were recovered and placed
directly in 96% ethanol pending assay for the presence of DLB (see 2.6).
2.4 Plant material and growth conditions
Seeds of tested plant species (Tables 1 to 3) were sown in three types of substrate:
commercial potting soil (light-medium clay containing 63% sand, 12% silt, and 22%
clay, Gal-Marketing Ltd, Israel); perlite supplemented with NPK (20:20:20) fertilizer and
0.5% Hoagland agar,30 as indicated in Tables 1 to 3. Cucumis melo (melon), Cucumis
sativus (cucumber), Gossypium (cotton), Capsicum annuum (pepper), Nicotiana
benthamiana, Catharanthus roseus (periwinkle), and Sesamum indicum (sesame)
plants were grown under controlled conditions at 25-28ºC under cool white neon
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
efluorescent illumination (14/10 h light/dark cycle) at approximately 200 μmol m-2 s-1 light
intensity. Daucus carota (carrot), Solanum lycopersicum (tomato) and Solanum
tuberosum (potato) plants were grown under natural light (approximately 11/13 h
light/dark cycle) with controlled temperature (25-28 ºC). The plants were treated with
DLB (as described in 2.5) at the 4-5th true-leaf stage, and kept under the same growth
conditions. Citrus grandis (pomelo) and Citrus sinensis (orange) plants were grafted
on bitter orange and grown outdoors in 10-L buckets containing commercial potting
soil under drip-irrigation, and DLB was applied when they were 6 years old.
2.5 DLB application methods
Two DLB application methods, namely soil drenching and foliar spraying, were tested
to determine the preferred method to introduce DLB effectively into plant tissue. In both
methods, DLB cultures were prepared as described in section 2.1, and applied to
plants at a concentration of ~108 CFU mL-1. For foliar spraying, a surfactant (either
Tween-20 [Acros Organics, NJ, USA] at 0.1% or Silwet L-77 (Adama, Israel) at 0.07%)
was added before application; the foliage was sprayed with a 1-L hand sprayer until
runoff. For soil drenching 50 mL of DLB culture was poured directly into the potting
medium of each plant. Untreated plants were used as controls.
2.6 Total DNA extraction and DLB detection in plants and insects
To determine the presence of DLB in plant tissue, samples were collected at several
time points, as indicated in Tables 1 to 3. Leaves were surface-sterilized by immersion
in 70% ethanol for 30 s, followed by a 2-min wash in 0.6% NaOCl and two additional
10-s washes with DDW.27 Then, 300-400 mg of leaf tissue were used for total DNA
extraction by means of a cetyl trimethyl ammonium bromide (CTAB) protocol described
elsewhere.28 DNA extraction from pepper samples was conducted with a plant/fungi
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
eDNA isolation kit (Norgen Biotek, Thorold, Canada), according to the manufacturer’s
guidelines.
DNA from H. obsoletus insects was extracted by grinding each insect in 120 µL of lysis
buffer29 (5 mM of Tris-HCl, pH 8.0, containing 0.5 mM of EDTA, 0.5% of Nonidet P-40,
and proteinase K at 1 mg mL-1). The lysate was then incubated at 65 ºC for 15 min,
followed by 95 ºC for 10 min and then briefly centrifuged at 10,000 x g to pellet the
debris. A 3-µL aliquot of the aqueous supernatant were used as the template for PCR
analysis in a final volume of 25 µL. All DNA-lysate samples were kept at -20 ºC pending
screening for DLB.
The presence of DLB was determined by PCR with species-specific DLB primers
(DLBF, 5’-CTCTGTGGGTGGCGAGTGGC-3’ and DLBR, 5’-
ACCGTCAGTTCCGCCGGG-3’), as described elsewhere.19 PCR products of
environmental samples (i.e. Abraham’s balm shrubs and H. obsoletus insects) were
further tested by DNA sequencing by using BigDye Terminator v3.1 Ready Reaction
Mix (Thermo Fisher Scientific) according to the manufacturer’s instructions, in a 3130xl
Genetic Analyzer (Applied Biosystems).
2.7 Visualization of GFP-labelled DLB by confocal microscopy
In planta localization of DLB was examined in carrot and sesame plants by using a gfp-
expressing strain (DLBGFP) and confocal microscopy. For carrot, seeds were sown in
perlite medium and grown as described in section 2.4 until the 3-4 true-leaf stage and
then spray- or drench-inoculated with the DLBGFP strain. Transverse and longitudinal
sections of petioles and leaves were taken 48 h post inoculation.
For sesame, seeds were sown in Hoagland agar and grown under fluorescent
illumination as mentioned in section 2.4, until the emergence of two true leaves.
DLBGFP was applied to 15 plants by gently brushing a single leaf of each plant with a
cotton swab soaked in the bacterial suspension. Five plants were sampled at each
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
etime point, as indicated in Table 3. All samples were analyzed with an IX 81 confocal
laser-scanning microscope (Olympus, Tokyo, Japan).
2.8 Visualization of DLB with a scanning electron microscope
A scanning electron microscope (SEM) was used to visualize DLB on the leaf surface.
Two DLB-spray-inoculated sesame plants and one untreated plant were used.
Specimens were collected 3, 7 and 10 days post inoculation and prepared by chemical
fixation of plant leaves in a fixative solution (formaldehyde:glacial acetic acid:70%
ethanol, 5:5:90) for 24 h followed by dehydration with increasing amounts of ethanol
according to published procedures.31 The specimens were then critical-point dried with
a K850 Critical Point Drier (Quorum Technologies Ltd, Laughton, England) and sputter-
coated with gold by using a SC7620 mini-sputter coater (Quorum). They were
visualized with a TM3000 table top scanning electron microscope (Hitachi-High
Technologies, Toronto, Canada).
2.9 Statistics
The Pearson chi-square test was applied to determine statistical significance (α = 0.05)
of differences between treatments (inoculation type and growth media), using JMP
software (SAS, Cary, NC, USA) (Tables 1 to 3).
3 RESULTS AND DISCUSSION
3.1 DLB was present in environmental samples of Abraham’s balm shrubs
DLB was first isolated from the insect vector H. obsoletus. To determine whether DLB
also colonizes plants, we tested environmental samples of Abraham’s balm, the
preferred host for H. obsoletus in Israel; we detected DLB in all four shrubs tested from
September through December 2016 (Supporting information Fig. S1A), in one shrub
in April 2017 and in all shrubs in May and June (Data not shown). Nine out of ten H.
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
eobsoletus individuals collected from the same Abraham’s balm shrubs in October 2016
were positive for DLB (Supporting information Fig. S1B). DLB-positive PCR bands from
each of the sampling points on all four shrubs were sequenced and confirmed to be
DLB.
The fact that DLB could be found in environmental samples of Abraham’s balm shrubs
indicates that DLB is a natural endophyte. This result, in light of our previous finding of
DLB in H. obsoletus gut, might suggest that DLB can be acquired and transmitted by
feeding insects, but this was not empirically tested. An example of a phloem-feeding
insect vector transmitting endophyte populations from source to recipient plants has
recently been reported.32 Intriguingly, although both DLB and phytoplasma were found
in H. obsoletus, unlike DLB, phytoplasma could never be detected in Abraham’s balm
shrubs in Israel.20
The presence of DLB in Abraham’s balm plants from September through December
and from April through June may suggest that it is persistent in this host. The low
incidence of DLB in Abraham’s balm in April can perhaps be attributed to the fact that
Abraham’s balm is a deciduous plant, which remains leafless from January through
March. One possible explanation is that during this leafless period DLB cells hibernate
in Abraham’s balm stems and/or roots and then become active again when new leaves
emerge in spring. Therefore, in April, although leaves were already present, DLB may
not have propagated to high levels and their distribution in the plant would be limited;
hence, the low detection rate. Another possibility is that DLB was lost from Abraham’s
balm shrubs during winter and was re-inoculated onto the shrubs with the reoccurrence
of H. obsoletus in spring. According to this scenario, there would be relatively little time
from March to April for the DLB population to build up, hence the low detection rate.
3.2 DLB penetration to various host plants
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
eThe fact that DLB is a natural plant colonizer, together with its putative biocontrol
applications, has encouraged us to examine additional potential plant hosts to which
DLB could be applied. Twelve different plant species were inoculated with DLB by foliar
spraying or soil drenching, and then examined by PCR for the presence of DLB, which
was detected in the majority (9/12) of the tested plant species, as elaborated in Tables
1 to 3. DLB could not be detected in tomato, potato, or pepper, all belonging to the
Solanaceae family; but it was detected in N. benthamiana, which belongs to the same
family. These results suggest that when applied exogenously, DLB can colonize
multiple plant species, including annual and perennial crops. Our detection of DLB in
plant tissues following a fairly aggressive leaf-washing procedure suggests that DLB
was present, not merely on the leaf surface, but that it penetrated and established itself
inside the plant tissue. Furthermore, the fact that DLB could be detected in the foliage
of some plants following soil drenching supports previous observations19 that it also
penetrated the root system and systemically moved throughout the plant.
Whereas some endophytes, such as Azoarcus sp.,33 have a relatively restricted host
range and can colonize only grasses, DLB appears capable of adapting to many host
environments, at least for short periods of time. Although this trait is not unique to DLB
and is shared with other broad-host-range endophytes, such as B. phytofirmans,15 it is
an important trait to consider if DLB were to be used as a biocontrol agent.
3.3 DLB application and persistence in plants
To better understand the dynamics of DLB colonization in plants and to determine the
optimal application method, we conducted further experiments with two plants species:
carrot (Daucus carota) and sesame (Sesamum indicum) (Tables 2 and 3, respectively).
These plants were selected for further analysis specifically because both are known to
suffer from phloem-restricted pathogens such as phytoplasma4 and Ca. Liberibacter
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
esolanacearum.34 The experiments described below wills facilitate future examination
the ability of DLB to suppress disease in these two crops.
The inoculation method used significantly affected in planta detection of DLB in both
plant species. In carrot experiment I, the DLB detection rate was similar for both foliar
spraying and drenching inoculations; in experiment II, however, foliar spraying resulted
in a significantly higher DLB detection rate than drenching, at all four time points (Table
2). Soil-grown sesame plants also had higher DLB detection rates following foliar
spraying at three of four sampling time points, but for those in Hoagland medium the
results of the two respective inoculation methods did not differ (Table 3). These results
suggest that under non-sterile conditions, DLB establishes more effectively in the plant
when applied by spraying than by drenching. The soil drenching findings revealed that
DLB can penetrate through plant roots and move systemically through the vascular
tissue of the plant.
The decline of DLB populations observed in both carrot and sesame was also
observed in grapevines, where DLB was not detected 37 days post inoculation.22 It is
possible that despite the ability of DLB to colonize various plant species, many of these
species cannot support its reproduction and survival for long periods. Additionally, non-
natural hosts of DLB may activate their defense systems against DLB, and thereby
lead to its decline in the plant, as was observed in other transient endophytes.35 DLB
appears to adapt more successfully to Abraham’s balm, where it was detected during
several months; this supports the notion that Abraham’s balm is a natural host of the
bacterium. In light of possible use of DLB as a biocontrol agent these findings imply
that DLB should most likely be reapplied at certain time intervals to maintain an
effective titer in the plant.
3.4 Microscope analyses of DLB localization in planta
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
eThe results presented above indicate that DLB can penetrate the leaf surface and
move systemically within a plant. To better understand its penetration routes and
localization, a gfp-expressing DLB strain was applied to plants by either foliar spraying
or soil drenching, and was followed by both confocal microscopy and scanning electron
microscopy (SEM).
Time-scale monitoring of DLB following its application to sesame leaves showed that
DLB cells were randomly spread on the surface of the leaf by 3 days post inoculation
(Fig. 1); by seven and ten days post application, they were located mainly in the leaf
stomatal opening and partly in the intercellular spaces of the leaf cortex. SEM
observations agreed with confocal microscopy results in showing that DLB was located
around the stomatal openings of sesame leaves 3 days post exposure (Fig. 2A). Seven
and ten days post exposure DLB was clearly seen inside the stomatal openings (Fig.
2B, 2C). Bacterial cells were not observed in the control plants (Fig. 2D).
DLB was also detected in sesame roots, mainly in the outer periderm and in the inner
cortex layers, 10 days post exposure (Supporting information Fig. S2). Visual
examination of the root external surface (Supporting information Fig. S2A, washed
once in DDW), revealed that DLB was located around the root epidermal surface,
including the root hairs. Sections of the root reveal DLB also in the inner-cortex
(Supporting information Fig. S2B).
In carrots, DLBGFP applied to foliage by spraying could be easily detected covering the
leaf surface of unwashed leaves 48 h post inoculation (Fig. 3A, B). Leaf cross-sections
revealed that DLBGFP penetrated the leaf tissue within 48 h of inoculation and was
located in the xylem vessels and phloem cells (Fig. 3C, D). DLBGFP was also seen in
the xylem vessels of washed petioles (Fig. 3F). This further supports the notion of
systemic movement of DLB. The bacterium could not be observed in three attempts
(on three different plants) following soil drenching. In general, DLBGFP was unevenly
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
espread in carrot tissue and could not be detected by microscopic analysis of numerous
PCR-positive samples.
Both nonpathogenic endophytes and phytopathogenic bacteria gain entrance into
plant tissue via natural openings such as stomata and through wounds.36 Our
microscopic observations support a similar penetration route for DLB; it could be seen
aggregating in stomatal openings. Furthermore, localization of DLB inside the vascular
tissue accounts for the systemic movement of DLB within plants, and is consistent with
the behavior of other endophytes, which migrate within the plant tissue either
apoplastically or through the vascular tubes.37 In light of the fact that H. obsoletus feeds
on phloem tissue, one could expect that this would be the niche colonized by DLB in
environmental samples of its Abraham’s balm host; but this has yet to be
demonstrated. Conversely, the localization of DLB in the phloem tissue following foliar
spraying is very intriguing. Very few bacterial genera are known to inhabit intracellular
phloem; they include obligate bacterial parasites such as phytoplasma and
Liberibacter, both non-culturable organisms that possess reduced genomes and that
are directly introduced into the phloem tissue by phloem-feeding insects. However,
DLB appeared to make its own way into the phloem tissue, even without the aid of a
phloem-feeding vector. If indeed DLB could be targeted to the phloem tissue, it would
represent a very elegant method to combat phloem-limited bacterial pathogens
specifically, with minimal side effects.
4 CONCLUSIONS
The present study has added an important element to the prospective use of DLB as
a biocontrol agent: DLB has a broad range of host plants, and it localizes to the same
plant tissue as its target pathogen. Thus far, the effectiveness of DLB as a biocontrol
agent was shown in vitro against S. melliferum and recently with phytoplasma-infected
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
egrape plantlets. The present study widens the prospect for further testing the potential
use of DLB as a biocontrol agent in to protect numerous plant species.
ACKNOWLEDGEMENTS
The authors would like to thank Shira Gal for technical assistance with SEM operation,
Dr. Eric Palevsky and Sharon Hecht for supplying citrus plants, Dr. Phyllis Weintraub
for insect descriptions and scientific advice, Tamar Zakai for linguistic editing, and Dr.
Zvi Peleg for supplying sesame seeds. This work was supported by the Chief Scientist
of Israel’s Ministry of Agriculture and Rural Development and by Israel’s Ministry of
Economy and Industry.
This paper is contribution number 574/17 from the Agricultural Research Organization,
Volcani Center, Rishon LeZion, Israel.
REFERENCES
1. Weintraub PG and Beanland L, Insect vectors of phytoplasmas. Annu Rev
Entomol 51:91–111 (2006).
2. Christensen NM, Nicolaisen M, Hansen M and Schulz A, Distribution of
phytoplasmas in infected plants as revealed by real-time PCR and
bioimaging. Mol Plant-Microbe Interact 11:1175–1184 (2004).
3. Lee IM, Davis RE and Gundersen-Rindal DE, Phytoplasma: phytopathogenic
Mollicutes. Annu Rev Microbiol 54:221–255 (2000).
4. Salehi M, Esmailzadeh Hosseini SA, Salehi E and Bertaccini A, Genetic
diversity and vector transmission of phytoplasmas associated with sesame
phyllody in Iran. Folia Microbiol (Praha) 62:99–109 (2017).
5. Morone C, Boveri M, Giosuè S, Gotta P, Rossi V, Scapin I and Marzachì C,
Epidemiology of Flavescence dorée in vineyards in northwestern Italy.
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
ePhytopathology 97:1422–1427 (2007).
6. Stone J, Bacon C and White J, An Overview of Endophytic Microbes :
Endophytism Defined. Marcel Dekker Inc., New York, pp. 3-30 (2000).
7. Hoffman MT, Gunatilaka MK, Wijeratne K, Gunatilaka L, Arnold AE and
Sarniguet A, Endohyphal bacterium enhances production of indole-3-acetic
acid by a foliar fungal endophyte. PLoS One 24:e73132 (2013).
8. Hardoim PR, van Overbeek LS, Berg G, Pirttilä AM, Compant S, Campisano
A, Döring M and Sessitsch A, The hidden world within plants: ecological and
evolutionary considerations for defining functioning of microbial endophytes.
Microbiol Mol Biol Rev 79:293-320 (2015).
9. Brader G, Vescio K, Mitter B, Trognitz F, Ma LJ and Sessitsch A, Ecology and
genomic insights on plant-pathogenic and -nonpathogenic endophytes. Annu
Rev Phytopathol 10: 123-145 (2017).
10. Taghavi S, Wu X, Ouyang L, Zhang YB, Stadler A, McCorkle S, Zhu W,
Maslov S and van der Lelie D, Transcriptional responses to sucrose mimic
the plant-associated life style of the plant growth promoting endophyte
Enterobacter. PLoS One 10:e0115455 (2015).
11. Maheshwari DK, Dubey RC, Aeron A, Kumar B, Kumar S, Tewari S and Arora
NK, Integrated approach for disease management and growth enhancement
of Sesamum indicum L. utilizing Azotobacter chroococcum TRA2 and
chemical fertilizer. World J Microbiol Biotechnol 28:3015–3024 (2012).
12. Hallmann J, Miller WG, Sikora RA and Lindow SE, Endophytic colonization of
plants by the biocontrol agent Rhizobium etli G12 in relation to Meloidogyne
incognita infection. Phytopathology 91:415–422 (2001).
13. Bent E and Chanway CP, The growth-promoting effects of a bacterial
endophyte on lodgepole pine are partially inhibited by the presence of other
rhizobacteria. Canad J Microbiol 44:980–988 (1998).
14. Sessitsch A, Hardoim P, Döring J, Weilharter A, Krause A, Woyke T, Mitter B,
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
eHauberg-Lotte L, Friedrich F, Rahalkar M, Hurek T, Sarkar A, Bodrossy L,
van Overbeek L, Brar D, van Elsas JD and Reinhold-Hurek B, Functional
characteristics of an endophyte community colonizing rice roots as revealed
by metagenomic analysis. Mol Plant-Microbe Interact 25:28–36 (2012).
15. Mitter B, Petric A, Shin MW, Chain PSG, Hauberg-Lotte L, Reinhold-Hurek B,
Nowak J and Sessitsch A, Comparative genome analysis of Burkholderia
phytofirmans PsJN reveals a wide spectrum of endophytic lifestyles based on
interaction strategies with host plants. Front Plant Sci 4:120 (2013).
16. Miotto-Vilanova L, Jacquard C, Courteaux B, Wortham L, Michel J, Clément
C, Barka EA and Sanchez L, Burkholderia phytofirmans PsJN confers
grapevine resistance against Botrytis cinerea via a direct antimicrobial effect
combined with a better resource mobilization. Front Plant Sci 7:1236 (2016).
17. Gamalero E, Marzachì C, Galetto L, Veratti F, Massa N, Bona E, Novello G,
Glick BR, Ali S, Cantamessa S, D'Agostino G and Berta G, An 1-
Aminocyclopropane-1-carboxylate (ACC) deaminase-expressing endophyte
increases plant resistance to Flavescence dorée phytoplasma infection. Plant
Biosyst 151:331–340 (2017).
18. Gamalero E, D’Amelio R, Musso C, Cantamessa S, Pivato B, D’Agostino G,
Duan J, Bosco D, Marzachì C and Berta G, Effects of Pseudomonas putida
S1Pf1Rif against chrysanthemum yellows phytoplasma infection.
Phytopathology 100:805–813 (2010).
19. Iasur-Kruh L, Naor V, Zahavi T, Ballinger MJ, Sharon R, Robinson WE,
Perlman SJ and Zchori-Fein E, Bacterial associates of Hyalesthes obsoletus
(Hemiptera: Cixiidae), the insect vector of bois noir disease, with a focus on
cultivable bacteria. Res Microbiol 168:94–101 (2016).
20. Sharon R, Soroker V, Wesley SD, Zahavi T, Harari A and Weintraub PG,
Vitex agnus-castus is a preferred host plant for Hyalesthes obsoletus. J Chem
Ecol 31:1051–1063 (2005).
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
e21. Kosovac A, Radonjić S, Hrnčić S, Krstić O, Toševski I and Jović J, Molecular
tracing of the transmission routes of bois noir in Mediterranean vineyards of
Montenegro and experimental evidence for the epidemiological role of Vitex
agnus-castus (Lamiaceae) and associated Hyalesthes obsoletus (Cixiidae).
Plant Pathol 65:285–298 (2015).
22. Iasur-Kruh L, Zahavi T, Barkai R, Freilich S, Zchori Fein E and Naor V, Dyella-
like bacterium isolated from an insect as a potential biocontrol agent against
grapevine yellows. Phytopathology (under revision).
23. Lahav T, Zchori-Fein E, Naor V, Freilich S and Iasur-Kruh L, Draft genome
sequence of a Dyella-like bacterium from the planthopper Hyalesthes
obsoletus. Genome Announc 4:4 (2016).
24. Enderle PJ and Farwell MA, Electroporation of freshly plated Escherichia coli
and Pseudomonas aeruginosa cells. Biotechniques 25:954–956, 958 (1998).
25. Barel V, Chalupowicz L, Barash I, Sharabani G, Reuven M, Dror O, Burdman
S and Manulis-Sasson S, Virulence and in planta movement of Xanthomonas
hortorum pv. pelargonii are affected by the diffusible signal factor (DSF)-
dependent quorum sensing system. Mol Plant Pathol 16:710–723 (2015).
26. Weintraub PG, Horowitz AR. Spatial and diel activity of the pea leafminer
(Diptera: Agromyzidae) in potatoes, Solanum tuberosum. Environ Entomol
25:722–726 (1996).
27. Zhang S and Fu Y, First report of gray mold of German thyme (Thymus
vulgaris ) caused by Botrytis cinerea in Florida. Plant Dis 100:1012 (2016).
28. Zahavi T, Sharon R, Sapir G, Mawassi M, Dafny-Yelin M and Naor V, The long-
term effect of stolbur phytoplasma on grapevines in the Golan Heights. Aust J
Grape Wine Res 19:277–284 (2013).
29. Frohlich DR, Torres-Jerez I, Bedford ID, Markham PG and Brown JK, A
phylogeographical analysis of the Bemisia tabaci species complex based on
mitochondrial DNA markers. Mol Ecol 8:1683–1691 (1999).
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
e30. Klerks MM, Franz E, van Gent-Pelzer M, Zijlstra C and van Bruggen AHC,
Differential interaction of Salmonella enterica serovars with lettuce cultivars
and plant-microbe factors influencing the colonization efficiency. ISME J
1:620–631 (2007).
31. Bahar O, Goffer T and Burdman S, Type IV Pili are required for virulence,
twitching motility, and biofilm formation of Acidovorax avenae subsp. citrulli.
Mol Plant Microbe Interact 22:909–920 (2009).
32. Lòpez-Fernàndez S, Mazzoni V, Pedrazzoli F, Pertot I and Campisano A, A
phloem-feeding insect transfers bacterial endophytic communities between
grapevine plants. Front Microbiol 8:834 (2017).
33. Krause A, Ramakumar A, Bartels D, Battistoni F, Bekel T, Boch J, Böhm M,
Friedrich F, Hurek T, Krause L, Linke B, McHardy AC, Sarkar A, Schneiker S,
Syed AA, Thauer R, Vorhölter FJ, Weidner S, Pühler A, Reinhold-Hurek B,
Kaiser O and Goesmann A, Complete genome of the mutualistic, N2-fixing
grass endophyte Azoarcus sp. strain BH72. Nat Biotechnol 24:1385–1391
(2006).
34. Munyaneza JE, Fisher TW, Sengoda VG, Garczynski SF, Nissinen A and
Lemmetty A, Association of “Candidatus Liberibacter solanacearum” with the
psyllid, Trioza apicalis (Hemiptera: Triozidae) in Europe. J Econ Entomol
103:1060–1070 (2010).
35. Iniguez AL, Dong Y, Carter HD, Ahmer BMM, Stone JM and Triplett EW,
Regulation of enteric endophytic bacterial colonization by plant defenses. Mol
Plant-Microbe Interact 18:169–178 (2005).
36. Beattie GA and Lindow SE, Bacterial colonization of leaves: A spectrum of
strategies. Phytopathology 89:353–359 (1999).
37. Rosenblueth M and Martínez-Romero E, Bacterial endophytes and their
interactions with hosts. Mol Plant-Microbe Interact 19:827–837 (2006).
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
eFigure legends
Figure 1: Localization of DLBGFP in sesame leaves over time. Hoagland-grown
sesame seedlings were inoculated by gentle brushing of the first two true leaves with
a DLBGFP soaked cotton swab. Upper panel (A) represents the upper epidermis layer
and lower panel (B) represents the mesophyll cells imaged at 3, 7, and 10 days post
inoculation. Neg indicates untreated control leaves (St indicates stomata). Images
are representative of five plant repetitions. Red fluorescence results from auto
fluorescence of chloroplasts and green fluorescence from DLBGFP cells. All images
are merges of red auto-fluorescence, GFP fluorescence and bright-field. All images
were taken via confocal microscope.
Figure 2: SEM pictures of DLB on sesame leaf surface. Leaves were sampled at 3
(A), 7 (B), and 10 (C) days post DLB inoculation. DLB cells can be seen in and
around leaf stomata (St) opening. (D) untreated plants 10 days post treatment.
Bacterial elongated rod shapes are indicated by white arrows.
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
e
Figure 3: DLB cells located in the vascular tissues of carrot leaves and petioles.
Carrot plants were spray-inoculated with DLBGFP (A-E) or untreated (F), and
visualized under confocal microscopy 48 h post inoculation. (A) DLBGFP cells spread
across the surface of an unwashed carrot leaf. (B) DLBGFP cells located on the outer
surface of a carrot leaf cross-section. (C-D) Cross-section of two different carrot
petioles. DLBGFP cells are located in the xylem vessels and phloem cells. (E)
Longitudinal section of a washed petiole shows DLBGFP in a xylem vessel (indicated
by a white arrow). (St, stomata; Pc, palisade cells; Sc, spongy cells; Ep, epidermis;
Xy, xylem vessels; Ph, phloem tissue). Red fluorescence results from auto-
fluorescence of chloroplasts; green fluorescence from DLBGFP cells. All images are
merges of red auto-fluorescence, GFP fluorescence and bright-field microscopy.
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
e
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
eSupporting information
Figure S1: DLB is present in Abraham’s balm and H. obsoletus environmental
samples. (A) DLB detected in environmental Abraham’s balm samples. Four samples
(1–8) from Abraham’s balm shrubs, two leaves per shrub, sampled in October. (B) H.
obsoletus samples (1 to 10) collected from Abraham’s balm shrubs in October. In
both gels (+) and (-) represent positive and negative controls, respectively; M, size
marker (500-bp band is indicated by an arrow) Expected size of the specific DLB
PCR product was ~400 bp.
Figure S2: DLBGFP colonization of a sesame plant following leaf inoculation.
Hoagland-grown 5-week-old sesame seedlings were inoculated by gentle brushing of
the first two true leaves with a DLBGFP soaked cotton swab. (A) DLB colonizes the
outer layers of sesame root (10 days post exposure). (B) Root cross-section showing
DLB colonization of inner-root cortex. Periderm (Pr), cortex (Co), Phloem (Ph), Xylem
(Xy). Green fluorescence was emitted by the DLBGFP strain. All images were
recorded with a confocal microscope.
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
eTable 1: DLB detection in various plant species 5-7 days post application
†leaves were tested 14 days post inoculation without washing
Family Species (cV) Common name
Plant growth media
Inoculation method
Detection ratio
positive/total§
Malvaceae Gossypium Cotton Potting soil
Spray 5/5(a)
Potting soil
Drench 3/5(a)
Potting soil
Untreated 0/5(b)
Rutaceae Citrus sinensis
(Shamouti)
Orange Potting soil
Spray 4/5(a)
- Drench ND‡ Potting
soil Untreated 0/5(b)
Citrus grandis
Pomelo Potting soil
Spray 5/5(a)
- Drench ND‡ Potting
soil Untreated 0/5(b)
Cucurbitaceae Cucumis melo
(Cezanne)
Melon Potting soil
Spray 10/10(a)
Potting soil
Drench 6/10(b)
Potting soil
Untreated 0/10(c)
Perlite Spray 10/10(a)
Perlite Drench 2/10(b)
Perlite Untreated 0/10(b)
Cucumis sativus
(Beit-Alpha)
Cucumber Potting soil
Spray 10/10(a)
Potting soil
Drench 2/10(b)
Potting soil
Untreated 0/10(b)
Perlite Spray 9/10(a)
Perlite Drench 3/10(b)
Perlite Untreated 0/10(c)
Solanaceae Nicotiana benthamiana
Benthamiana Potting soil
Spray 7/8(a)
- Drench ND‡ Potting
soil Untreated 0/8(b)
Capsicum annuum
Pepper Perlite Spray 0/9(a)
Perlite Drench 1/9(a)
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
e
‡ND- not-determined. §Different letters indicate statistical significance (α = 0.05) among application types within each plant species in the same medium.
Perlite Untreated 0/9(a)
Solanum tuberosum (Désireé)
Potato Perlite Spray 0/12(a)
Perlite Drench 0/12(a)
Perlite Untreated 0/12(a)
Solanum lycopersicum (HA-29430)
Tomato Perlite Spray 0/12(a)
Perlite Drench 0/12(a)
Perlite Untreated 0/12(a)
Apocynaceae Catharanthus roseus
Periwinkle Potting soil
Spray 9/9†(a)
Potting soil
Untreated 0/9(b)
Perlite Spray 0/9(a) Perlite Drench 5/9(b) Perlite Untreated 0/9(a)
This article is protected by copyright. All rights reserved.
Acc
epte
d A
rticl
e Table 2: DLB penetration efficiency and persistence in carrot plants following foliar spraying and soil drenching inoculations Experiment I† Days after application 5 10 18 - Ratio of DLB positive plants following foliar spraying inoculation
12/12(a)
1/8(a)
2/12(a) -
Ratio of DLB positive plants following soil drenching inoculation
12/12(a)
1/8(a)
0/12(a)
-
Ratio of DLB positive plants in untreated plants 0/12(b)
0/8(a)
0/12(a)
-
Experiment II† Days after application 3 7 10 16 Ratio of DLB positive plants following foliar spraying inoculation
7/7(a)
7/8(a)
7/8(a)
3/8(a)
Ratio of DLB positive plants following soil drenching inoculation
3/5(b)
2/6(b)
0/7(b)
0/6(b)
Ratio of DLB positive plants in untreated plants 0/7(c) 0/8(c) 0/7(b) 0/6(b)†Different letters indicate statistically significant difference (α = 0.05) among application types within each sampling time point. Table 3: DLB penetration efficiency and persistence in sesame plants following foliar spraying and soil drenching inoculations. Experiment I – potting soil† Days after application 7 14 21 28 Ratio of DLB-positive plants following foliar spraying inoculation
10/10(a) 10/10(a) 6/10(a) 2/10(a)
Ratio of DLB-positive plants following soil drenching inoculation
0/10(b) 0/10(b) 0/10(b) 0/10(a)
Ratio of DLB-positive untreated plants
0/10(b) 0/10(b) 0/10(b) 0/10(a)
Experiment II – Hoagland agar† Days after application 7 14 21 28 Ratio of DLB-positive plants following foliar spraying inoculation
10/10(a) 10/10(a)
8/10(a) 2/10(b)
Ratio of DLB-positive plants following soil drenching inoculation
10/10(a) 9/10(a)
7/10(a) 3/10(ab)
Ratio of DLB-positive untreated plants
0/10(b) 0/10(b) 0/10(b) 0/10(b)
†Different letters indicate statistically significant differences (α = 0.05) among application types within each medium and sampling time point.
This article is protected by copyright. All rights reserved.