the role of erf transcription factors in defenses …
TRANSCRIPT
THE ROLE OF ERF TRANSCRIPTION FACTORS IN DEFENSES
AGAINST SPECIALIST AND GENERALIST HERBIVORES
IN ARABIDOPSIS THALIANA
_______________________________________
A Thesis
presented to
the Faculty of the Graduate School
at the University of Missouri-Columbia
_______________________________________________________
In Partial Fulfillment
of the Requirements for the Degree
Doctor of Philosophy
_____________________________________________________
by
ERIN MACNEAL REHRIG
MAY 2010
ii
The undersigned, appointed by the Dean of the Graduate School, have examined the thesis entitled
THE ROLE OF ERF TRANSCRIPTION FACTORS IN DEFENSES AGAINST SPECIALIST AND GENERALIST HERBIVORES IN ARABIDOPSIS THALIANA
presented by Erin MacNeal Rehrig a candidate for the degree of Doctor of Philosophy, and hereby certify that, in their opinion, it is worthy of acceptance. Dr. Jack Schultz Dr. Heidi Appel Dr. Shuqun Zhang Dr. Walter Gassmann
iii
ACKNOWLEDGEMENTS
I would like to thank Dr. Jack Schultz and Dr. Heidi Appel for their support, input and
advice during my thesis research. I thank Dr. Shuqun Zhang and Dr. Walter Gassmann
for serving on my committee and for their help. This thesis would not be possible
without all of the assistance from the Schultz-Appel lab, including Dr. Imgard Seidl-
Adams, Lucy Rubino, Ouassim Auhal, Dean Bergstrom, Abbie Ferierri, Clayton
Coffman, Jordan Richards, Erin Feinfield, Sree Depthi, and Caitlin Vore. I thank the
collaborators from the University of Missouri’s Computer Science and Electrical
Engineering Dept. including Dr. Chi-Ren Shyu, Jia-Fu Chang, Jason Green, Jaturon
Harnsomburana, and Dayong Fang for their help with the digital phenotyping algorithm.
Special thanks to Dr. Dong Xu and Dr. Trupti Joshi for their help with our bioinformatics
analysis. I appreciate the collaborations with Tim Havens, Jeff Anderson, and Guohong
Mao who helped with TF results, protein analysis, and ethylene measurements,
respectively. Finally, I would like to thank Dr. Dan Jones and Xioli Gao from Michigan
State University for their work on our JA and stress hormone analysis.
iv
TABLE OF CONTENTS
ACKNOWLEDGEMENTS............................................................................................... iii LIST OF FIGURES .......................................................................................................... vii LIST OF TABLES............................................................................................................. ix
Chapter 1: Transcription Factor Profiles and Cis-Element Distributions in Arabidopsis Genes Regulated by Herbivory and Wounding .................................................................. 1
Abstract ........................................................................................................................... 2 Introduction..................................................................................................................... 3 Material and Methods ..................................................................................................... 5
Plant and Insect Care .................................................................................................. 5 Insect and Wounding Treatments ............................................................................... 6 RNA Isolation and Tissue Preparation ....................................................................... 8 Microarray and Data Analysis .................................................................................. 10 Reverse Transcription-Real Time PCR (RT-PCR)................................................... 11 Transcription Factor Analysis................................................................................... 14 Bioinformatics Analysis of Promoter Regions ......................................................... 14
Results........................................................................................................................... 16 General Stress Responses ......................................................................................... 19 Differential Responses to Insect Treatments ............................................................ 26 Over-represented Cis-Elements ................................................................................ 30
Discussion..................................................................................................................... 34 References..................................................................................................................... 46 Supplemental Materials ................................................................................................ 57
Chapter 2: Measuring “Normalcy” in Plant Gene Expression After Herbivore Attack ... 62
Abstract ......................................................................................................................... 63 Introduction................................................................................................................... 64 Material and Methods ................................................................................................... 68
Plant Care and Insect Treatments ............................................................................. 68 RNA Isolation and Tissue Preparation ..................................................................... 69 Reverse Transcription-Real Time PCR (RT-PCR)................................................... 69 Luciferase Spike Analysis ........................................................................................ 71 Data Analysis and Assessment of Gene Stability ..................................................... 72
Results........................................................................................................................... 73 Discussion..................................................................................................................... 79 Conclusion .................................................................................................................... 84 References..................................................................................................................... 86
v
Chapter 3: Differences in ERF transcription factor expression, defense-related gene transcription, and stress hormone release reveal diverse signaling pathways elicited after attack by two different herbivores in Arabidopsis............................................................ 92
Abstract ......................................................................................................................... 93 Introduction................................................................................................................... 94 Materials and Methods.................................................................................................. 98
Plant Growth Conditions .......................................................................................... 98 Insect Treatments ...................................................................................................... 98 Plant Tissue Harvesting ............................................................................................ 99 Ethylene Measurements .......................................................................................... 100 JA and JA Conjugate Measurements ...................................................................... 100 Gene expression via Real Time RT-PCR ............................................................... 101 Data Analysis .......................................................................................................... 105
Results......................................................................................................................... 106 Insect Elicitation of Ethylene Release .................................................................... 106 ERF and Defense Gene Expression ........................................................................ 110
Discussion................................................................................................................... 114 Conclusions................................................................................................................. 120 References................................................................................................................... 122 Supplementary data..................................................................................................... 130
Thigmotrophic Responses....................................................................................... 130
Chapter 4: Insect performance on erf mutant plants suggests a major role for ERF transcription factors in Arabidopsis susceptibility to herbivory ..................................... 132
Abstract ....................................................................................................................... 133 Introduction................................................................................................................. 134 Materials and Methods................................................................................................ 137
Plant Growth Conditions ........................................................................................ 137 Insect Growth Conditions ....................................................................................... 138 Insect Feeding Assays............................................................................................. 138 Quantification of Amount Eaten Using Digital Photography................................. 140 Measurement of Leaf Glucosinolates ..................................................................... 141 Statistical Analysis.................................................................................................. 143
Results......................................................................................................................... 144 Insect Growth Rates and Feeding Assays............................................................... 144 Glucosinolate Analysis ........................................................................................... 147
Discussion................................................................................................................... 149 References................................................................................................................... 157 Supplementary Materials ............................................................................................ 165
Protein Analysis ...................................................................................................... 165
vi
Chapter 5: The role of ERF Transcription Factors in defense against the specialist and generalist caterpillars in Arabidopsis thaliana ................................................................ 167
Research Summary ..................................................................................................... 168 Overall Conclusions.................................................................................................... 184 References................................................................................................................... 188
VITA................................................................................................................................. 204
vii
LIST OF FIGURES
Figure PAGE Figure 1.1: Differentially Expressed Transcription Factor Families………………..…...18 Figure 1.2: Venn Diagrams of Transcription Factors up- (A) and down- (B) regulated within each treatment………………………………………….19 Figure 1.3: Heat Map of Cis-Element Distributions in Genes Up- and Down- Regulated by Insect Herbivory and Wounding……………………………...31 Supplemental Figure 1.1: Distribution of parametric p-values from ANOVA…………57 Supplemental Figure 1.2: Genes and Transcription Factors Affected by Insect and Wounding Treatments……………………………………………….……….59 Figure 2.1: Reference Gene Stability after Insect Herbivory…………………………...77 Figure 2.2: Luciferase Exogenous mRNA Spike……………………………………….78 Figure 2.3: Plant Cellular Trauma Caused by Herbivory……………………………….85 Figure 3.1: Ethylene production in WT Arabidopsis plants after short-term Pieris rapae and S. exigua feeding over a 72hr time course…………………………...107 Figure 3.2: JA levels in WT Arabidopsis plants after S. exigua (A) or P. rapae (B) feeding over a 24hr time course…………………………………….……109 Figure 3.3: JA-isoleucine (JA-IL) levels in WT Arabidopsis plants after S. exigua (A) or P. rapae (B) feeding over a 24hr time course …………………109 Figure 3.4: RT-PCR of ERF Transcription Factors and defense-related genes………...111 Figure 3.5: Heat Map of ERF Transcription Factor and Defense-Related Gene Expression Patterns over a 24-hour Time Course……………………………..…113 Supplemental Figure 3.1: Gene expression of ERFs and defense genes in response to caging (touch)…………………………………………..………………….131 Supplementary Figure 3.2: SA production in WT Arabidopsis plants after S. exigua (A) or P. rapae (B) feeding over a 24hr time course……………….………131
viii
Figure 4.1: Schematic showing quantification of amount eaten using digital photography…………………………………………………………………..…141 Figure 4.2: S. exigua and P. rapae relative growth rates (RGR) (A,B), relative consumption rates (RCR) (C,D) and efficiency of conversion of ingested food index (ECI) (E,F) on WT and erf mutant plants………………….…..146 Figure 4.3: Constitutive and induced Indolyl and Aliphatic Glucosinolate levels in WT and erf plants in S. exigua and P. rapae bioassays…………………….…149 Supplementary Figure 4.1: Total Protein Levels in WT and erf plants………………...165 Supplementary Figure 4.2: Initial Plant Size for each bioassay………………..………166 Supplementary Figure 4.3: Initial Insect Size for each bioassay………………..……...166 Figure 5.1: Hypothetical model for differential expression patterns by generalist and specialist insects ………………..………………………..……….…175
ix
LIST OF TABLES
Table Page Table 1.1: Insect treatments used in this study………………..…………………….…..16 Table 1.2: Transcription Factors statistically significantly up- and down-regulated by insect and wounding treatments (p<0.05). ………………..….…..…21 Table 1.3: RT-PCR Confirmation of AP2-ERF Transcription Factor Genes ……………28 Table 1.4: Characterized Enriched Cis-Elements in Genes Regulated by Insect Herbivory and Wounding………………..………………………..………………..……33 Supplemental Table 1.1: Estimation of differentially expressed genes in all treatments using different false discovery rates (FDR), p values for t-tests, and fold-change cutoffs………………..…………………………..………..……58 Supplemental Table 1.2: Primers used in this experiment………………..…………...…59 Supplemental Table 1.3: Putative Motifs in Insect-Regulated Genes Using Motif Sampler Algorithm………………..………..………..………………..……60 Supplemental Table 1.4: Transcription factors with known repressor activity that are down regulated by Myzus after 6 hours………………..………………61 Table 2.1: Reference Gene Primers used in this experiment ………………..……….…70 Table 2.3: Results of post hoc Tukey tests after ANOVA of Expression Levels of Reference and Stress-Related Genes………………..………………………...75 Table 2.2: Results of post hoc Tukey tests after ANOVA of Reference Gene Cts for 6hr Herbivory Treatments by 2 caterpillars in local and systemic tissue……………………………………..………………………..………………..……76 Table 3.1: ERF Family TFs and Defense-Related Gene Primers used in this Experiment………………..………………………..……………..…..……103 Table 4.1: Pearson product-moment correlations and regression a nalyses of relative consumption rates and glucosinolate levels after insect feeding. ………………..………………………..………………….………148
1
Chapter 1:
Transcription Factor Profiles and Cis-Element Distributions in Arabidopsis Genes Regulated by
Herbivory and Wounding
2
Abstract
Plant responses to insects and other environmental stresses are complex, involving
differential perception, multiple signaling pathways, and the transcription of appropriate
defense-responsive genes. Using a whole-genome array, we identified 193 unique
transcription factors from 34 different families that were up- or down-regulated in local
or systemic tissues at different time points after herbivore feeding or mechanical
wounding. The various treatments elicited the accumulation of different transcription
factor mRNAs. Our results indicate that differences in gene expression patterns by
mechanical wounding and differently feeding insects are largely controlled at the
transcription factor level. A large percentage of these transcription factors were members
of the AP2/ERF, MYB, Homeobox, C2H2, bHLH, and WRKY gene families. A
bioinformatics analysis revealed that enriched motifs in the promoters of insect-affected
genes are involved in water stress, circadian rhythms, and JA and SA signaling. Cluster
analysis of cis-element distributions in co-expressed genes did not reveal distinct “motif
signatures” between treatment groups, as most genes shared many of the same elements.
Although transcription factor profiles differed both quantitatively and qualitatively
among insect treatments, we also observed generalized stress responses where many of
the same TFs occurred in most, if not all, treatments. In many cases, the expression
patterns of specific transcription factors correspond to enrichments of their compatible
binding sites in down-stream affected genes. These results suggest that transcription
3
factor profiles and unique motif occurrences are an integral part of differential signaling
in response to insect herbivory.
Introduction
Plant responses to insects and wounding are complex, involving differential perception,
multiple signaling pathways, and extensive transcriptional reprogramming (deVos et al.
2005; Dellasert et al. 2004). Perception of insect attack by plants is thought to occur at
the site of herbivory via elicitors in insect oral secretions (OS) (Alborn et al. 1997; Pare
and Tumlinson 1999; Schmelz et al. 2003) as well as cell wall fragments. Insect traits
beyond OS composition such as feeding behavior may also affect perception and
response (Mattiacci et al. 1995; Wittstock et al. 2004) and evidence is accumulating to
suggest that plants can identify their attacker and activate defenses depending on the
insect attacker (Mewis et al. 2005; DeVos et al. 2005; Moran and Thompson 2004).
Early cellular signaling events following perception of insect attack include rapid bursts
of phytohormone release, including salicylic acid (SA), ethylene (ET), and jasmonic acid
(JA) (Kahl et l. 2000; Winz and Baldwin 2001; Kessler and Baldwin 2002, Thaler et al.
2002; Reymond and Farmer 1998). However, the subsequent activation and transcription
of signaling-related proteins, such as trans-acting elements or transcription factors
following insect attack, is less well characterized. The activation of transcription factors
in the WRKY and APETALA2/ETHYLENE RESPONSE FACTOR (AP2/ERF) families
has been shown to be a critical molecular event in Arabidopsis plants after JA treatment,
4
pathogensis, and wounding (see review by Eulgem et al. 2005; Delassert et al. 2004;
Reymond et al. 2004; Chen et al. 2002; Schenk et al. 2000). Several transcription factors
serve as points of cross talk within the pathways of the stress hormones including JA-
insensitive1 (JIN1/AtMYC2), a key point of regulation between JA and ABA pathways
(Abe et al. 2003; Yadav et al. 2005). WRKY70 has been shown to be a point of interaction
between the JA and SA signaling pathways (Li et al. 1999), and Ethylene Response
Factor1 (ERF1) arbitrates JA and ET signaling (Lorenzo et al. 2003). Because
transcription factors are both components of early signaling events and nodes of cross-
talk, they are likely to be critical in plant responses to specific biotic stresses, including
insect herbivory.
Upon activation, transcription factors induce or repress mRNA transcription by binding
to specific DNA sequences or cis-regulatory elements in gene promoters. The elucidation
of potential regulatory networks based on microarray and gene expression data in concert
with bioinformatics analysis of over-represented cis-elements in co-expressed genes in
plants has been the subject of several studies and reviews (Sreenivasulu et al. 2007;
Vandepoele et al. 2006; Mahalingam et al. 2003; Cheong et al. 2002). Several on-line
databases and computational tools for predicting Arabidopsis cis-elements and
transcription relationships are readily available (Obayashi et al. 2007; Palaniswamy et al.
2006; O’Connor et al. 2005; Steffens et al. 2004; Shah et al. 2003) but have not yet been
applied towards understanding plant-insect interactions.
Using a whole genome cDNA microarray, we identified transcription factor genes in
5
Arabidopsis thaliana (Columbia) in attacked (local) and unattacked (systemic) tissues at
two time points (6hr and 24 hr) after treatment by mechanical wounding or by specialist
or generalist insects from different feeding guilds (caterpillars and aphids). We
hypothesized that responses to these different stimuli (insects or mechanical wounding)
involve differential regulation of transcription via different pathways. The present
analysis specifically examined the expression of genes encoding known transcription
factors and the frequencies of putative and characterized transcription factor binding
motifs among all differentially-expressed genes. A bioinformatics analysis of co-
regulated gene promoters was also done to identify over-represented cis-elements in each
treatment. Our results suggest that differences in gene expression patterns elicited by
wounding and different insect species are largely controlled at the transcription factor
level and include unique enriched motifs. Distinct cis-element signatures in down-stream
affected genes were not apparent, as most genes contained most known transcription
factor binding sites.
Material and Methods
Plant and Insect Care
The aphids Brevicoryne brassicae (L.) and Myzus persicae (Sulzer) were maintained as
plant virus free clones on pak-choi plants (Brassica campestris L. ssp. chinensis cv.
Black Behi). Eggs of the caterpillar Spodoptera exigua Hübner (Noctuidae) were
obtained from Benzon Research (Carlisle, PA) and larvae were reared on artificial diet
(Bioserv, Frenchtown, NJ, USA). The caterpillar Pieris rapae L. (Pieridae) was
6
maintained as a culture in our lab on pak-choi and originated from the Carolina
Biological Supply Company (North Carolina). Both caterpillar species were transferred
to Col WT plants one day before the experiments to acclimate to the new host.
Arabidopsis thaliana (L.) ecotype Columbia (Col-0) seeds were vernalized in 2% agar
and sown into 6 x 5 cm pots containing sterile Metromix 200 soil (Sun Gro Horticulture).
Plants were chamber grown at 22 ± 1 °C, 65 ± 5 % relative humidity, and 200 µmol m-2 s-
1 light intensity on a 8:16 (L:D) photoperiod. Plants were watered as needed and fertilized
every other watering with 21-7-7 Miracle Gro (Scotts Company).
Insect and Wounding Treatments
Plants were treated with caterpillars and aphids in separate experiments summarized in
Table 1.1 The first sampling of plants occurred several hours after removal of insects so
that plant gene expression was not confounded by insect RNA or by plant gene
expression elicited by the physical movement of insect removal.
The caterpillar treatment was designed to capture early gene expression events and
minimize variation due to leaf age and amount of insect damage. All leaves selected for
treatment and harvest were fully-expanded mature leaves. Six to 10 second and third
instar S. exigua and P. rapae caterpillars were allowed to feed for 2-4 hours to generate 6
leaves of similar age per plant with 10-30% leaf area removed. Caterpillars were
wrangled as needed with camel hair brushes (size 0) to concentrate their feeding on 6
leaves, and leaves of control plants were jiggled with a camel-hair brush to simulate the
7
leaf movement caused by wrangling. Once sufficient damage was achieved, caterpillars
were removed and the plants were returned to the growth chamber. The mechanical
wounding treatment was designed to approximate insect damage to tissues by running a
sterile pattern wheel across both sides of the midrib of 6 leaves of similar age on each
plant, once at the beginning of the caterpillar treatment and again half way through the
caterpillar wrangling period. Control plants were jiggled with a camel hair brush to
simulate the leaf movement caused by mechanical wounding. Leaves were harvested for
gene expression at 6 and 24 hr after the start of caterpillar damage or wounding.
Unwounded leaves were harvested separately from size-matched damaged or wounded
leaves. Leaves from 3-4 plants were pooled for each of the four bioreplicates.
Because aphids have effects on plants that are much weaker and slower to develop than
those of caterpillars (Mewis et al 2005, 2006) and aphids cannot be readily contained on
individual leaves, the design of their treatment was different from that for caterpillars.
Twenty sub-adult (final instar) and adult aphids were placed on plants whose rosettes
were caged at the soil line by transparent mylar cylinders (5 cm diameter, 9 cm high)
with tops of fine mesh gauze (< 0.01 mm mesh wide) to maintain air exchange. Controls
were caged plants without aphids, and all plants were returned to the growth chamber.
After 1 week of feeding, all cages and aphids were removed and control plants were
jiggled with a camel hair brush to simulate the leaf movement caused by aphid removal.
Plants were returned to the growth chamber and whole plants were harvested for gene
expression at 6 and 24 hr after aphid removal. All insect treatments and sample
collections were scheduled to avoid perturbing the plants’ circadian cycles.
8
RNA Isolation and Tissue Preparation
Total RNA was isolated from individual plants using a modified TRIZOL extraction
method as follows. Plant material was ground in liquid nitrogen using a mortar and
pestle, resuspended in 6 ml TRIZOL reagent (Invitrogen, Carlsbad CA, USA), vortexed
and incubated at 65°C for 5 min with regular mixing. Cell debris was pelleted by
centrifugation for 30 min at 12,000 g and 4°C and the supernatant was extracted with 3
ml chloroform twice. After centrifugation for 20 min at 12,000 g, the aqueous phase was
recovered and RNA was precipitated at room temperature for 5 min with 0.5 volumes of
0.8 M sodium citrate and 0.5 volumes isopropanol. After centrifugation for 30 min at
12,000 g, the pellet was washed with 70% ethanol and re-centrifuged. The RNA pellet
was air dried for 5 min and resuspended in 200 µl RNAse free water. Following a
spectrophotometric determination of RNA concentration, the RNA was precipitated with
2.5 volumes of ethanol and a 1/10 volume of 3 M sodium acetate at –20°C overnight, and
subsequently pelleted at 20,000 g for 30 min at 4°C. The precipitate was washed with
70% ethanol, re-centrifuged, air dried and resuspended in RNAse free water to an
approximate concentration of 5 µg/µl. Actual concentration was determined
spectrophotometrically, and RNA quality of randomly selected samples was determined
using a 2100 Bioanalyzer (Agilent Technologies, Mississauga ON, Canada).
Total RNA was used for a direct labeling procedure. 80 µg total RNA was incubated with
0.27 µM T17VN primer, 0.15 mM dATP, dCTP, and dGTP, 0.05 mM dTTP (Invitrogen),
9
0.025 mM Cyanidin3- or Cyanidin5-conjugated dUTP (Amersham, Piscataway, NJ,
USA), 40 U RNAseInh (Promega, San Luis Obispo CA, USA), and 400 U SuperscriptII
(Invitrogen) in 10 mM DTT and 1 x first strand buffer in a total volume of 40 µl. In
addition, 0.3 fmole human cRNAs complementary to the human negative control
oligonucleotides were used in labelling reactions (HsD17B1, KRT1, and MB). Prior to
addition of enzymes the solution was heated to 65°C for 5 min and for primer annealing
cooled to 42°C. Following an incubation at 42°C for 2.5 h, the RNA was degraded with 8
µl 1 M sodium hydroxide for 15 min at 65°C, neutralized with 8 µl 1 M hydrochloric acid
and buffered with 4 µl 1M Tris-pH 7.5. Subsequently, the labelled cDNA was purified
using a PCR purification kit according to the manufacturer’s protocol (Qiagen,
Mississauga, ON, Canada). DNA was eluted in 100 µl 10 mM Tris, pH 8.5, the two
labeling reactions were combined, and 1µl Cyanidin5-labelled GFP was added.
Following an ethanol/sodium acetate precipitation (Sambrook and Russel, 2001) the air-
dried cDNA pellet was resuspended in 3 µl water, denatured at 95°C for 3 min, added to
50 µl pre-warmed array hybridization buffer #1 (Ambion, Austin, TX, USA), and kept at
65°C until use. We pre-hybridized microarray slides for 45 min at 48°C in 5 x SSC, 0.1
% SDS, 0.2 % BSA. Slides were washed twice with water for 1 min, dipped 5 times in
isopropanol, and spun dry in Falcon tubes at 100 g for 3 min. The hybridization solution
was applied to the microarray slides and covered with untreated glass cover slips (Fisher
Scientific, Nepean, ON, Canada). Arrays were incubated over night in CMT
hybridization chambers (Corning, Corning, NY, USA) submerged in a water bath at 42°C
with moderate vertical shaking. Hybridization chambers were disassembled and slides
were washed for 15 min at 42°C in 2 x SSC, 0.5 % SDS, and for 2 times 15 min in 0.5 x
10
SSC, 0.5 % SDS. Subsequently, arrays were dipped five times in 0.1 x SSC and spun dry
as described above. Microarrays were scanned with a ScanArray Express (Perkin Elmer,
Woodbridge, ON, Canada) scanner with laser power set to 95% and photo-multiplier-
tube set to 54 to 64.
Microarray and Data Analysis
The microarray used in this study comprised a set of 26,090 Arabidopsis gene specific
70-mer oligonucleotides (Operon v1) and analysis was as described in Ehlting et al.
(2005, 2008). Briefly, after removal of manually flagged spots, background correction,
and flooring, an average of 12.5 % of all spots were non-detectable and excluded from
further analyses. Signal intensities were used for loess normalization, generating log2-
ratios comparing each treatment with the corresponding control. For each time point, we
first used the data from the 3-4 replicate arrays to perform a Student’s t-test and to
calculate mean expression ratios for each treatment sample relative to the corresponding
control. To assess the type I error rate, we calculated q-values estimating the false
discovery rate based on the parametric p-values obtained from the t-statistic (Storey and
Tibshirani 2003). We then used the four normalized expression ratios from each of the
two time points (for a total of 8 data points) to perform an analysis of variance (ANOVA)
and again estimated the false discovery rate based on the distribution of parametric p-
values. Normalized mean expression ratios for all probes on the array and results for all
statistical analysis are provided in the Supplementary Materials.
11
To identify differentially expressed genes, we used a low p-value cut-off (0.01).
Although this reduces the number of falsely discovered genes, it misses a substantial
number of truly differentially expressed genes. Therefore, assuming that high fold change
difference is associated with a lower likelihood of being a false positive (Pylatuik &
Fobert 2005), we initially defined genes as ‘differentially expressed’ (i.e. genes with
treatment-induced change in transcript abundance) as those genes for each time point that
were associated with a t-test p-value of less than 0.05 (accepting a false discovery rate of
up to 0.3) and also displayed a more than two-fold change between treatment and control.
A comparison of the number of expressed genes in all treatments using different false
discovery rates (FDR), p values for t-tests, and fold-change cutoffs can be seen in the
Supplementary Materials.
Reverse Transcription-Real Time PCR (RT-PCR)
Primers for AP2/ERF transcription factors, 18S, and G6PD5 were designed using Primer
3 Software (Rozen and Skaletsky, 2000) and further analyzed for primer dimers using
Invitrogen’s Vector NTI Software (Carlsbad, CA). All primers were BLASTed in NCBI
to ensure specificity of amplification. We performed gel electrophoresis of PCR products
and detected single bands of expected size. Additionally, melting curve analysis of all
PCR products was done via real-time PCR. Only primer pairs that produced one clear
peak were used for experiments. All PCR products were sequenced to ensure that only
gene products of interest were being amplified. A list of primer sequences used in this
analysis can be found in the Supplementary Materials.
12
The same RNA used for the microarray was used for RT-PCR. Three biological
replicates were available for the 24 hour treatments and four biological replicates were
available for the 6 hour treatments. RNA concentrations were measured using a
Nanodrop Spectrophotometer (ThermoScientific, Wilmington, DE) and diluted to 1
µg/uL. RNA quality was confirmed via gel electrophoresis. To eliminate genomic DNA
contamination, samples were treated with Turbo DNAse according to the manufacturer’s
specifications (Ambion, Austin, TX). We used Omniscript Reverse Transcriptase kit and
protocol for RT reactions (Qiagen, Valencia, CA). DNAse-treated RNA was synthesized
into first strand cDNA using a mix of random and oligo-dT primers. To produce enough
cDNA for the subsequent qPCR reactions, 8 RT reactions per bioreplicate were done in
20 µL reactions then pooled.
For real time qPCR standard curves, a pool of cDNA from each biorep was serially
diluted to match fold changes encompassed within the microarray data. All PCR
reactions were run in 96-well plates. Each biorep was run in triplicate on the array and
analyzed for technical variation. For PCR reactions, we used 5 µL of cDNA template, 5
µM primers, water, and Platinum SYBR Green qPCR Super-Mix UDG (Invitrogen,
Calsbad, CA) for a total of 20 µL. Amplification was then conducted under the following
conditions on a MJ Research Opticon 2 DNA Engine (Hercules, CA): 50°C UDG
treatment for 2 minutes, 95°C denaturation for 2 minutes, followed by 40 cycles of 95°C
denaturation for 15 seconds, 56°C annealing for 30 seconds and 72°C extension for 30
seconds. After extension, but prior to fluorescence measurement reads, the temperature
13
was ramped to approximately 1.5-2.0°C below the gene product melting curve start (Tm,
–dl/dT min) to melt off primer dimers and non-specific random amplicons. This ensured
that only the specific gene product was contributing towards SYBR Green activity. A
final 5 minute extension at 72°C followed by a complete melting curve analysis from
72°C to 95°C were then conducted.
RT-qPCR data were acquired using the standard curve method (Larionov et al., 2005).
All data were initially analyzed using Opticon 3 Monitor Software (BioRad Industries,
Hercules, CA) and imported into a customized Excel spreadsheet (Microsoft Corp.
Redmond, WA). We used an in-house algorithm to identify cycle threshold values (Cts)
of fluorescence within the exponential phase of the PCR curve. Unit-less expression
values were then calculated automatically from the Ct values based on the regression
equation of the standard curve. Expression values for 24 hour data were normalized
against the geometric mean of 18S and G6PD5. Because a suitable housekeeping gene
could not be found for the 6 hour data, all 6 hour expression levels were normalized to
the total amount of cDNA in the PCR reaction using a correction factor. Statistically
significant differences in expression levels between treatments and respective controls
were identified using GLM and Dunnett’s T statistic (SAS Institute, Cary, NC)
We compared RT-qPCR patterns of gene expression with those from the array. As
expected, the more sensitive qPCR detected more statistically significant changes in
expression than did the array. A majority of those identified by the array as significant
were confirmed by qPCR as significant (20/26). Four of the 6 array false positives had
14
qPCR values in the same direction as those of the array even though they failed the test of
significance. In order for RT-PCR data to “confirm” differentially expressed genes in the
array, gene expression levels measured by RT-qPCR also had to be statistically
significantly different than controls. Often authors use directionality or less quantitative
methods including Northern blotting to confirm array data. Our confirmation rates are
within the range of expectation for microarray data.
Transcription Factor Analysis
We conducted a literature search and used the online databases, Gene Annotation tool
(GO) from the TAIR website (www.Arabidopsis.org) and DATF: Database of
Arabidopsis Transcription Factors (Guo et al. 2005) to identify transcription factors in
our differentially expressed gene set. Four-way Venn Diagrams showing shared TFs
among treatments were constructed using the online tool
http://www.pangloss.com/seidel/Protocols/venn4.cgi. Five-way diagrams then were
generated manually using GNU Image Manipulation Program (GIMP) (www.gimp.com).
Bioinformatics Analysis of Promoter Regions
We analyzed promoter regions for the presence of enriched cis-elements or transcription
factor binding sites for all up- and down-regulated genes in each treatment. Using AGI
gene annotations, we downloaded gene promoter sequences up to1000 bp upstream of the
transcription start site, ATG, from the TAIR Sequence Database (http://www.
15
Arabidopsis.org). First, we used the MotifSampler tool (Thijs et al. 2001, 2002) to
identify any putative motifs in co-expressed genes up-regulated by each insect. To search
for known transcription factor binding sites and cis-elements in gene promoters, we used
the ATHENA search tool which uses a library of 105 known Arabidopsis transcription
factor binding sites and 30,067 predicted promoters (O’Connor et al. 2005). The
ATHENA algorithm conducts a student’s T-test to determine whether motifs in a given
sample set are significantly different from a random distribution in the genome. Motif
occurrences with a p-value less than 1.0 E-5 were designated as “enriched”.
To determine if co-expressed genes in each treatment set contained a unique profile or
signature of cis-element distributions, we collated cis-element totals for each treatment
then conducted a Principle Component Analysis using SAS. A cluster analysis of cis-
element distributions across all treatments was done using Cluster 3.0 (Eisen et al. 1998).
16
Table 1.1: Insect treatments used in this study- A total of 16 different measurements/treatments were conducted. Four insects, including Generalist and Specialist Caterpillars and Aphids were used. Tissue samples from either local (L) or systemic leaves (S) (caterpillars) or whole plant (aphids) were taken 6 or 24 hrs after insects fed on the plants. Pi= P. rapae, Sp=S. exigua, Br= B. brassicae, My= M. persicae, Wo= Wounding
Results
We used a large-scale cDNA 70-mer oligonucleotide microarray consisting of 26,090
genes representing the Arabidopsis genome to examine the expression of transcription
factor genes after herbivory and wounding and performed quantitative Reverse
Transcriptase- Real Time Polymerase Chain Reaction (RT-PCR) to confirm the array
Table 1.1 Treatments analyzed in this study
Abbreviation Insect tissue time # of
biological replicates
Pi-L-6h Pieris brassicae local 6 h 3 Pi-L-24h Pieris brassicae local 24 h 4 Pi-S-6h Pieris brassicae systemic 6 h 4 Pi-S-24h Pieris brassicae systemic 24 h 4 Sp-L-6h Spodoptera exigua local 6 h 3 Sp-L-24h Spodoptera exigua local 24 h 4 Sp-S-6h Spodoptera exigua systemic 6 h 4 Sp-S-24h Spodoptera exigua systemic 24 h 4 Br-6h Brevicoryne brassicae - 6 h 4 Br-24h Brevicoryne brassicae - 24 h 4 My-6h Myzus persicae - 6 h 4 My-24h Myzus persicae - 24 h 4 Wo-L-6h Wounding local 6 h 3 Wo-L-24h Wounding local 24 h 4 Wo-S-6h Wounding systemic 6 h 3 Wo-S-24h Wounding systemic 24 h 4
17
results. The robust statistical analysis and stringent guidelines for determining and
defining differentially expressed genes we used (p (t-test)<0.05, FCb>2) make us
confident that these data reflect biologically significant patterns of gene expression in
local and systemic tissues after insect attack and wounding at each treatment time.
Of the 1500 putative transcription factors in the Arabidopsis genome (Riechmann et al.
2000), we identified 193 genes encoding TFs that were affected by insect herbivory or
wounding. As seen in Figure 1.1, the transcription factors differentially regulated on this
array represented 34 of the 50 families outlined on the AgrisTF Database (Davuluri et al.
2003). Families with the most members represented were the MYB and MYB-Related
families (13%), AP2/ERF (12%), Homeodomain, NAC, and bHLH (each with 7%).
In many cases, the same gene was affected by multiple treatments, either insect or
mechanical wounding, at different time points. Figure 1.2 shows a 5-way Venn Diagram
of genes unique to and shared among treatments. The generalist S. exigua up-regulated
the greatest number of TF genes (65), whereas the generalist aphid, M. persicae (41)
down-regulated the most TF genes. S. exigua transcriptionally elicited 8 of the same up-
regulated TF genes as those elicited by the other caterpillar, P. rapae, but shared 10 TF
genes also transcribed in response to the aphid B. brassicae. As indicated by a large
number zeros in overlapping boxes in Figure 1.2,TF genes were unique to treatments,
suggesting that each insect (and wounding) triggered different signaling pathways.
18
Figure 1.1: Differentially Expressed Transcription Factor Families – The expression of 193 unique transcription factors from 34 different families were altered by insect and wounding treatments. Labeled pieces represent families with the highest percentage (>3%) of members affected in the array.
19
General Stress Responses
We identified several TF genes that were up-regulated in multiple treatments and are
characteristic of general stress response functions. These included ZIM (JAZ) genes, the
AP2/ERF family members ERF4, ERF11, and AtERF1 as well as a C2H2 gene and
several WRKYs. Of the 193 TF genes elicited by insects, 8 of them were members of the
JAZ/ ZIM family including JAZ1, which was up-regulated by all insects except M.
persicae and wounding treatments in our array. In addition to AP2/ERFs, WKRYs, and
JAZ TFs, NAC family transcripts were also up-regulated by several insects and
wounding. Wounding treatments and all insects except P. rapae activated the
transcription of the C2H2-family member protein ZAT10. The expression of the stress-
responsive TFs (Zheng et al. 2006, 2007) WRKY 33, WRKY25 and WKRY40 were
Figure 1.2: Venn Diagrams of Transcription Factors up- (A) and down- (B) regulated within each treatment. Values within boxes represent the number of genes shared between and among treatments. Values in parenthesis next to the treatment represent the total number of transcription factors affected. Rectangles are not drawn to scale based on gene set.
A B
20
affected by 8 treatments, including both caterpillars, wounding, and B. brassicae. Table
1.2 lists all up-and down- regulated TF genes whose expression was significantly affected
by insect treatments and wounding in this study.
21
Table 1.2: Transcription Factors statistically significantly up- and down-regulated by insect and wounding treatments (p<0.05). Genes are organized by TF family.
TF Family AGI Gene Name Insect Direction Tissue Time Fold X
Alfin-like At1g14510 Alfin-Like 7 M. persicae Down Whole Plant 6hrs 0.47 At5g26210 Alfin-Like 4 M. persicae Up Whole Plant 6hrs 2.07
AP2/ERF At1g22190 Similar to RAP2.4 S. exigua Up Local 24hrs 3.02 At1g25470 B-6 SubFamily Protein B. brassicae Down Whole Plant 6hrs 0.45 At1g28370 ERF11 B. brassicae Up Whole Plant 24hrs 8.50 S. exigua Up Systemic 24hrs 6.27 At1g43160 RAP2.6 B. brassicae Down Whole Plant 24hrs 0.37 M. persicae Down Whole Plant 24hrs 0.23 P. rapae Up Systemic 6hrs 3.15 At1g46768 RAP2.1 M. persicae Down Whole Plant 6hrs 0.50 At1g53170 ERF8 B. brassicae Up Whole Plant 24hrs 2.38 S. exigua Up Systemic 24hrs 2.67 S. exigua Up Local 24hrs 3.92 At1g63030 DDF2 S. exigua Up Local 6hrs 2.47 At1g64380 RAP2.4 M. persicae Down Whole Plant 6hrs 0.50 At1g71450 DREB SubFamily Protein M. persicae Down Whole Plant 6hrs 0.43 At1g74930 DREB SubFamily Protein B. brassicae Up Whole Plant 24hrs 3.18 S. exigua Up Systemic 6hrs 6.92 S. exigua Up Systemic 24hrs 6.14 S. exigua Up Local 24hrs 8.11 Wounding Up Local 6hrs 2.05 At2g35700 DREB SubFamily Protein S. exigua Up Local 6hrs 2.06 S. exigua Up Local 24hrs 2.55 At2g39250 SMZ (SCHLAFMUTZE) M. persicae Up Whole Plant 6hrs 2.34 At2g41710 ODP (putative) P. rapae Up Local 24hrs 2.13 At3g15210 ERF4 B. brassicae Up Whole Plant 24hrs 4.00 S. exigua Up Systemic 24hrs 3.33 S. exigua Up Local 6hrs 4.00 S. exigua Up Local 24hrs 4.70 Wounding Up Local 6hrs 2.47 At4g17500 AtERF1 S. exigua Up Systemic 6hrs 2.62 At4g32800 DREB SubFamily Protein M. persicae Down Whole Plant 24hrs 0.46 S. exigua Up Systemic 24hrs 2.39 At5g11590 TINY2 P. rapae Up Systemic 24hrs 3.70 At5g25810 TINY Wounding Up Local 6hrs 2.59 At5g47230 ERF5 S. exigua Up Systemic 24hrs 4.15 S. exigua Up Local 24hrs 7.63 At5g61590 B-3 SubFamily Protein S. exigua Down Systemic 6hrs 0.48 At5g61600 B-3 SubFamily Protein S. exigua Up Systemic 24hrs 2.28 At5g64750 ABR1 P. rapae Up Local 6hrs 7.16 At5g67180 AP2/ERF TF (putative) B. brassicae Down Whole Plant 6hrs 0.32 M. persicae Down Whole Plant 6hrs 0.49 B. brassicae Up Whole Plant 24hrs 3.41 M. persicae Down Whole Plant 6hrs 0.27
ARF At1g34390 ARF22 P. rapae Up Systemic 6hrs 2.22 At5g37020 ARF8 P. rapae Up Local 24hrs 2.16 Wounding Down Systemic 6hrs 0.48
ARR-B At2g25180 ARR-12 M. persicae Up Whole Plant 6hrs 2.31 AUX/IAA At1g04240 IAA3 B. brassicae Down Whole Plant 6hrs 0.30
M. persicae Down Whole Plant 6hrs 0.17 Wounding Down Local 6hrs 0.48 At1g52830 IAA6 S. exigua Up Systemic 6hrs 2.42 At3g04730 IAA16 M. persicae Down Whole Plant 6hrs 0.47 At3g23050 IAA7 M. persicae Down Whole Plant 6hrs 0.39 At3g62100 IAA30 Wounding Up Local 6hrs 2.18
bHLH At1g09530 PIF3 P. rapae Up Systemic 6hrs 2.20 At1g22490 bHLH94 M. persicae Down Whole Plant 6hrs 0.31 At1g26260 bHLH76 Wounding Up Local 6hrs 2.01 At2g18300 bHLH064 P. rapae Down Systemic 6hrs 0.34
22
S. exigua Down Local 6hrs 0.40 At2g22750 bHLH18 P. rapae Down Systemic 24hrs 0.46 At2g27230 AtbHLH156 M. persicae Down Whole Plant 24hrs 0.43 P. rapae Up Local 6hrs 2.16 At2g42280 bHLH130 P. rapae Down Systemic 24hrs 0.44 At2g43010 PIF4 M. persicae Down Whole Plant 6hrs 0.33 At2g47270 bHLH151 P. rapae Up Systemic 24hrs 2.02 P. rapae Up Local 6hrs 2.22 S. exigua Up Local 6hrs 4.72 At3g56980 bHLH39/ORG3 S. exigua Up Systemic 24hrs 2.66 At3g61950 bHLH67 P. rapae Down Local 6hrs 0.50 At4g01460 bHLH57 M. persicae Down Whole Plant 6hrs 0.36 M. persicae Down Whole Plant 24hrs 0.33 Wounding Down Local 6hrs 0.38 At4g16430 bHLH3 S. exigua Up Local 6hrs 2.19
bZIP At1g42990 AtbZip160 S. exigua Up Systemic 6hrs 2.08 At2g40620 AtbZip18 P. rapae Down Systemic 24hrs 0.22 Wounding Down Local 24hrs 0.43 At2g42380 AtbZip34 P. rapae Down Systemic 24hrs 0.39 P. rapae Down Local 6hrs 0.32 P. rapae Down Local 24hrs 0.46 S. exigua Down Local 24hrs 0.41 At4g34000 ABF3, AtbZip37 Wounding Up Local 24hrs 2.06 At4g34590 AtbZip181/ATB2 S. exigua Up Local 6hrs 2.60 At4g36730 GBF1 S. exigua Up Local 6hrs 3.23 At4g37730 AtbZip7 P. rapae Up Systemic 6hrs 2.15 P. rapae Up Local 24hrs 3.49 S. exigua Up Systemic 6hrs 2.63 S. exigua Up Systemic 24hrs 2.27 S. exigua Up Local 6hrs 5.48 At5g11260 AtbZip56, HY5 P. rapae Down Systemic 6hrs 0.45 At5g60830 AtbZip70 M. persicae Down Whole Plant 6hrs 0.48
C2C2-CO-like At1g68190 Zinc Finger Family Protein M. persicae Down Whole Plant 6hrs 0.40
At1g68520 COL6 Wounding Down Local 6hrs 0.46 At2g47890 COL13 M. persicae Down Whole Plant 6hrs 0.43 At3g02380 COL P. rapae Up Systemic 24hrs 2.82 At3g07650 COL9 M. persicae Down Whole Plant 6hrs 0.47 At3g21890 Zinc Finger Family Protein S. exigua Up Systemic 24hrs 2.10 At4g39070 Zinc Finger Family Protein M. persicae Down Whole Plant 6hrs 0.22 At5g15850 COL1 S. exigua Up Systemic 24hrs 2.48
C2C2-Dof At1g28310 DOF1.4 S. exigua Down Local 24hrs 0.48 At3g50410 OBP1 B. brassicae Up Whole Plant 24hrs 2.33 At5g60200 DOF5.3 B. brassicae Down Whole Plant 24hrs 0.24
C2C2-Gata At3g24050 GATA1 P. rapae Down Local 6hrs 0.40 At5g56860 GATA21 P. rapae Down Systemic 6hrs 0.36 P. rapae Down Local 6hrs 0.46
C2C2-YABBY At4g00180 YABBY3 M. persicae Down Whole Plant 6hrs 0.31
C2H2 At1g27730 ZAT10 B. brassicae Up Whole Plant 24hrs 4.04 M. persicae Down Whole Plant 6hrs 0.40 S. exigua Up Systemic 24hrs 5.57 S. exigua Up Local 24hrs 7.84 Wounding Up Systemic 6hrs 3.55 Wounding Up Local 6hrs 2.90 At2g28200 ZAT5 B. brassicae Up Whole Plant 24hrs 2.87 S. exigua Up Systemic 24hrs 4.51 S. exigua Up Local 24hrs 5.87 Wounding Up Systemic 6hrs 2.29 Wounding Up Local 6hrs 2.33
At3g10470 C2H2-Type Zinc Finger Protein S. exigua Up Systemic 6hrs 2.15
At3g20880 WIP4 B. brassicae Down Whole Plant 6hrs 0.47 Wounding Down Local 24hrs 0.46
At3g44750 Histone Deacetylase (putative) M. persicae Up Whole Plant 6hrs 2.46
At5g04340 C2H2-Type Zinc Finger B. brassicae Up Whole Plant 24hrs 2.56
23
Protein S. exigua Up Systemic 6hrs 2.89 S. exigua Up Systemic 24hrs 2.59 At5g43170 AZF3 S. exigua Up Systemic 24hrs 2.20 At5g59820 ZAT12 S. exigua Up Local 24hrs 2.04
C3H At2g19810 CCCH-Type Zinc Finger Protein M. persicae Down Whole Plant 6hrs 0.16
At2g25900 AtCTH B. brassicae Up Whole Plant 24hrs 2.25 M. persicae Down Whole Plant 6hrs 0.24 P. rapae Down Systemic 6hrs 0.25 At3g55980 SZF1 B. brassicae Up Whole Plant 24hrs 4.75 M. persicae Up/down Whole Plant 24hrs 2.49 S. exigua Up Systemic 24hrs 5.07 S. exigua Up Local 6hrs 3.26 S. exigua Up Local 24hrs 4.78
At4g29190 CCCH-Type Zinc Finger Protein M. persicae Down Whole Plant 6hrs 0.33
S. exigua Up Local 24hrs 3.45 S. exigua Down Systemic 6hrs 0.49
At5g15820 C3HC4-Type RING Finger Protein M. persicae Down Whole Plant 6hrs 0.33
At5g37200 C3HC4-Type RING Finger Protein S. exigua Up Systemic 24hrs 2.98
At5g37250 C3HC4-Type RING Finger Protein P. rapae Down Systemic 6hrs 0.48
At5g58620 CCCH-Type Zinc Finger Protein P. rapae Up Local 6hrs 2.72
CCAAT-HAP2 At2g34720
CCAAT-Binding TF, NFY-A4 P. rapae Up Local 24hrs 2.69
At5g06510 CCAAT-Binding TF, NFY-A10 S. exigua Up Local 6hrs 3.22
CCAAT-HAP3 At1g21970 LEC1 S. exigua Up Local 24hrs 2.14
At2g38880 HAP3 P. rapae Up Systemic 24hrs 2.33 S. exigua Up Local 24hrs 2.34
CCAAT-HAP5 At5g50470 HAP5a, NFY-C7 Wounding Up Systemic 6hrs 3.41
At5g50480 HAP5A, NFY-C6 P. rapae Up Local 6hrs 2.38 GRAS At1g66350 RGAL1 P. rapae Down Local 24hrs 0.49
At2g29060 AtGRAS12 P. rapae Down Local 6hrs 0.29 At4g17230 AtGRAS24, SCL13 P. rapae Up Local 6hrs 2.15 At5g66770 AtGRAS32, SCL4 M. persicae Up Whole Plant 6hrs 2.28
Homeobox At1g17920 HDG12 P. rapae Down Local 24hrs 0.43 At1g52150 ATHB15 P. rapae Up Local 6hrs 2.06 S. exigua Up Local 6hrs 2.30 At1g62360 SHOOT MERISTEMLESS B. brassicae Down Whole Plant 6hrs 0.40 At2g27990 POUND-FOOLISH P. rapae Up Systemic 24hrs 2.08 P. rapae Up Local 6hrs 2.19 S. exigua Up Systemic 6hrs 2.01 S. exigua Up Local 6hrs 2.76 S. exigua Up Local 24hrs 2.10 At2g35940 BLH1 M. persicae Up Whole Plant 6hrs 2.06 P. rapae Up Systemic 24hrs 2.48 At2g46680 ATHB7 P. rapae Up Local 6hrs 2.13 S. exigua Up Systemic 24hrs 2.45 P. rapae Down Local 24hrs 0.47 At4g17460 HAT1 P. rapae Down Local 6hrs 0.28 P. rapae Down Local 24hrs 0.49 At4g17710 HomeoDomain Protein B. brassicae Up Whole Plant 6hrs 2.49 At4g29940 PRHA P. rapae Up Systemic 6hrs 2.21 At4g32980 ATH1 M. persicae Up Whole Plant 6hrs 2.14 At4g37790 HAT22 S. exigua Down Systemic 6hrs 0.45 At5g06710 HAT14 S. exigua Down Local 6hrs 0.49 Wounding Up Systemic 24hrs 2.08 At5g59340 WOX2 P. rapae Up Systemic 6hrs 2.16
HSF At3g22830 HSFA6B B. brassicae Up Whole Plant 24hrs 2.39 S. exigua Up Systemic 6hrs 2.03
24
At3g24520 HSFC1 M. persicae Down Whole Plant 6hrs 0.43 At5g45710 HSFA4C S. exigua Up Systemic 24hrs 2.06
JAZ/ZIM At1g17380 JAZ5 P. rapae Up Systemic 24hrs 3.89 S. exigua Up Systemic 6hrs 5.66 S. exigua Up Local 24hrs 4.84 Wounding Up Systemic 6hrs 3.33 At1g19180 JAZ1 B. brassicae Up Whole Plant 24hrs 2.53 P. rapae Up Systemic 6hrs 2.16 P. rapae Up Systemic 24hrs 2.58 P. rapae Up Local 6hrs 2.48 P. rapae Up Local 24hrs 5.47 S. exigua Up Systemic 6hrs 10.61 S. exigua Up Systemic 24hrs 7.25 S. exigua Up Local 6hrs 8.68 S. exigua Up Local 24hrs 5.07 Wounding Up Systemic 6hrs 3.08 Wounding Up Local 6hrs 3.91 At1g70700 JAZ9 M. persicae Down Whole Plant 6hrs 0.49 P. rapae Up Systemic 6hrs 2.25 S. exigua Up Systemic 6hrs 2.22 S. exigua Up Local 6hrs 3.23 S. exigua Up Local 24hrs 2.13 Wounding Up Systemic 6hrs 2.70 At1g72450 JAZ6 B. brassicae Up Whole Plant 24hrs 2.02 P. rapae Up Systemic 6hrs 2.47 S. exigua Up Systemic 6hrs 2.87 S. exigua Up Systemic 24hrs 3.38 S. exigua Up Local 6hrs 5.60 At1g74950 JAZ2 P. rapae Up Systemic 24hrs 3.78 P. rapae Up Local 6hrs 6.02 S. exigua Up Systemic 6hrs 3.55 S. exigua Up Systemic 24hrs 3.69 S. exigua Up Local 6hrs 7.43 S. exigua Up Local 24hrs 4.79 Wounding Up Systemic 6hrs 3.88 At2g34600 JAZ7 B. brassicae Up Whole Plant 24hrs 2.80 At3g17860 JAZ3/JAI3 M. persicae Down Whole Plant 6hrs 0.30 S. exigua Up Local 6hrs 3.00 At5g13220 JAZ10 P. rapae Up Systemic 6hrs 3.00
JUMONJI At4g21430 Jumonji DomainTF Wounding Up Local 6hrs 2.15 LIM At1g10200 WLIM1 S. exigua Down Local 24hrs 0.48 LOB At2g45410 LBD19 Wounding Up Local 6hrs 2.43
At3g47870 LBD27 Wounding Down Local 6hrs 0.48 At3g58190 LBD29 Wounding Up Systemic 24hrs 2.62 At4g37540 LBD39 B. brassicae Up Whole Plant 6hrs 2.14 At5g67420 LBD37 Wounding Up Systemic 6hrs 2.01
MADS At4g09960 AGL11, SEEDSTICK S. exigua Down Local 24hrs 0.41 At5g37420 AGL105 Wounding Up Local 24hrs 2.37 At5g60440 AGL62 S. exigua Down Systemic 6hrs 0.45
MYB At1g18330 EPR1 P. rapae Down Systemic 6hrs 0.48 S. exigua Down Local 6hrs 0.50 At1g25340 AtMYB116 M. persicae Down Whole Plant 6hrs 0.49 Wounding Down Local 6hrs 0.28 At1g25550 MYB Family TF B. brassicae Up Whole Plant 24hrs 2.50 At1g35515 AtMYB8, HOS10 S. exigua Up Systemic 24hrs 2.34 At1g68670 MYB Family TF B. brassicae Up Whole Plant 24hrs 2.22 P. rapae Up Systemic 24hrs 2.07 At1g70000 MYB-like TF M. persicae Up Whole Plant 6hrs 2.10 At1g74430 AtMYB95 P. rapae Up Local 6hrs 3.13 S. exigua Up Systemic 24hrs 3.63 S. exigua Up Local 6hrs 5.30 At1g74840 MYB Family TF, M. persicae Down Whole Plant 6hrs 0.40 At1g75250 MYB Family TF Wounding Down Local 6hrs 0.48
25
At2g03470 MYB Family TF S. exigua Down Local 6hrs 0.40 At2g21650 MYB Family TF B. brassicae Down Whole Plant 6hrs 0.48 M. persicae Down Whole Plant 24hrs 0.33 At2g23290 AtMYB70 P. rapae Down Systemic 6hrs 0.41 At2g46410 CAPRICE M. persicae Down Whole Plant 6hrs 0.41 At2g46830 CCA1 P. rapae Up Systemic 24hrs 4.12 At3g23250 AtMYB15 Wounding Up Local 6hrs 2.83 At3g47600 AtMYB94 P. rapae Down Local 24hrs 0.43 At3g50060 AtMYB77 M. persicae Down Whole Plant 6hrs 0.43 At4g05100 AtMYB74 Wounding Up Local 6hrs 3.43 At4g28110 AtMYB41 S. exigua Down Local 6hrs 0.38 Wounding Down Local 6hrs 0.39 At5g01200 MYB Family TF M. persicae Down Whole Plant 6hrs 0.47 At5g08520 MYB Family TF M. persicae Down Whole Plant 6hrs 0.49 S. exigua Down Local 24hrs 0.42 At5g15310 AtMYB16, AtMIXTA P. rapae Down Local 6hrs 0.35 At5g26660 AtMYB4, AtMYB86 S. exigua Up Systemic 6hrs 2.44 At5g59430 TRFL1 Wounding Up Local 24hrs 2.10 At5g60890 ATR1, MYB34 B. brassicae Down Whole Plant 24hrs 0.35 M. persicae Down Whole Plant 24hrs 0.33 P. rapae Up Local 6hrs 3.67 S. exigua Up Local 6hrs 2.31
NAC At1g01010 ANAC001 M. persicae Up Whole Plant 6hrs 2.06 At1g03490 ANAC006 S. exigua Down Local 6hrs 0.43 At1g52890 ANAC019 B. brassicae Up Whole Plant 24hrs 3.84 M. persicae Down Whole Plant 6hrs 0.43 P. rapae Up Systemic 24hrs 3.51 S. exigua Up Local 6hrs 4.08 At1g69490 ANAC029 Wounding Up Systemic 24hrs 2.34 At2g24430 ANAC038 B. brassicae Up Whole Plant 24hrs 2.06 Wounding Up Systemic 6hrs 3.51 At3g15500 ANAC055 M. persicae Down Whole Plant 6hrs 0.35 At3g29035 ANAC059 Wounding Up Systemic 24hrs 2.18 At4g01550 ANAC069 S. exigua Up Local 6hrs 2.87 At4g27410 RD26 B. brassicae Up Whole Plant 24hrs 4.92 P. rapae Up Systemic 24hrs 7.32 At4g35580 NTL9 S. exigua Down Local 24hrs 0.50 At5g13180 ANAC083 M. persicae Down Whole Plant 6hrs 0.44 At5g41090 ANAC095 B. brassicae Down Whole Plant 6hrs 0.40 At5g63790 ANAC102 B. brassicae Up Whole Plant 24hrs 3.15 P. rapae Up Systemic 24hrs 2.00 S. exigua Up Local 24hrs 2.80
NIN At1g20640 AtNLP4 S. exigua Up Systemic 6hrs 2.69 OTHER At2g47590 PHR2 M. persicae Up Whole Plant 6hrs 2.22
RAV At1g13260 RAV1 B. brassicae Up Whole Plant 24hrs 3.17 S. exigua Up Systemic 24hrs 2.52 S. exigua Up Local 24hrs 2.77 At1g68840 RAV2 B. brassicae Up Whole Plant 24hrs 2.78 P. rapae Down Systemic 6hrs 0.42 S. exigua Up Systemic 24hrs 2.13 At2g46870 TF Similar to RAV1 B. brassicae Down Whole Plant 24hrs 0.44
REM At2g24680 REM12 S. exigua Up Local 6hrs 5.84 At2g24700 TF B3 Family Protein P. rapae Down Systemic 24hrs 0.44 At4g34400 TF B3 Family Protein Wounding Up Local 24hrs 2.00 At5g18090 TF B3 Family Protein M. persicae Down Whole Plant 6hrs 0.40
SBP At1g53160 SPL4 S. exigua Up Local 6hrs 2.06 TAZ At5g63160 BTB/TAZ Like Protein 1 B. brassicae Up Whole Plant 24hrs 2.74
M. persicae Down Whole Plant 6hrs 0.30 P. rapae Up Systemic 24hrs 4.30 S. exigua Up Local 24hrs 2.06 At5g67480 BTB/TAZ Like Protein 4 M. persicae Down Whole Plant 6hrs 0.39 P. rapae Up Local 6hrs 2.13 S. exigua Up Local 6hrs 2.32
TCP At1g35560 TCP23 M. persicae Up Whole Plant 6hrs 2.29 At2g45680 TCP9 S. exigua Up Systemic 6hrs 2.37
TUBBY At1g47270 TLP6 P. rapae Down Local 6hrs 0.43
26
WRKY At1g29280 AtWRKY65 S. exigua Up Local 24hrs 3.85 At1g29860 AtWRKY71 M. persicae Down Whole Plant 6hrs 0.49 At1g66560 AtWRKY64 B. brassicae Down Whole Plant 6hrs 0.43 At1g80840 AtWRKY40 B. brassicae Up Whole Plant 24hrs 5.36 M. persicae Down Whole Plant 6hrs 0.49 P. rapae Up Systemic 6hrs 2.06 P. rapae Up Local 6hrs 2.50 S. exigua Up Systemic 6hrs 13.09 S. exigua Up Systemic 24hrs 5.32 S. exigua Up Local 6hrs 14.48 S. exigua Up Local 24hrs 7.53 Wounding Up Local 6hrs 3.20 At2g04880 AtWRKY1, ZAP1 S. exigua Up Local 6hrs 3.19 At2g30250 AtWRKY25 B. brassicae Up Whole Plant 24hrs 2.79 S. exigua Up Systemic 24hrs 2.07 At2g38470 AtWRKY33 B. brassicae Up Whole Plant 24hrs 2.59 S. exigua Up Systemic 24hrs 3.21
Zinc Finger At1g76590
Zinc-Binding Family Protein B. brassicae Up Whole Plant 24hrs 3.10
M. persicae Down Whole Plant 6hrs 0.48
At4g17900 Zinc-Binding Family Protein M. persicae Up Whole Plant 6hrs 2.32
At5g46710 Zinc-Binding Family Protein B. brassicae Up Whole Plant 24hrs 2.04
S. exigua Up Systemic 6hrs 3.02 Gene Names: JA- Jasmonic Acid, SA- Salicylic Acid, ABA- Abscisic Acid, GA- Gibberellic Acid, TF- Transcription Factor, AP2- APETELA, ERF- Ethylene Response Factor, RAP2-Related to AP2, DREB- Drought Responsive Element Binding, DDF- Dwarf and Delayed Flowering, ODP- Ovule Development Protein, ABR- Abscisic Acid Repressor, ARF- Auxin Response Factor, IAA- Indole Acetic Acid-Induced Protein, ARR- Two-Component Response Regulator, bHLH- Basic Helix-Loop-Helix, PIF- Phytochrome Interacting Factor 3, ORG- OBP-3 Responsive Gene , OBP-OBF Binding Protein, OBF- OCS Element Binding Factor, OCS- Octopine Synthase, ABF- ABA-Responsive Transcription Factor (ABRE Binding), bZIP-Basic Leucine Zipper Interacting Protein, ATB- Arabidopsis thaliana bZip , GBF- G-box Binding Factor, HY- Elongated Hypotcotyl, CO- CONSTANS, COL- CONSTANS-Like Protein, Dof- DNA Binding with One Finger, ZAT- Zinc Finger Arabidopsis thaliana, HD2A- Histone Deacetylase, WIP- WPP-Domain-Interacting Protein, AZF- Arabidopsis Zinc Finger , ATCTH- Cys3His zinc finger protein, NFY- Nuclear Factor Y Subunit, SZF- Salt-Inducible Zinc Finger Protein, LEC- Leafy Cotyledon, HAP- Heme Activated Protein, GAI- Gibberellic Acid Insensitive, RGA- Regulator of Gibberellic Acid Responses, SCL- Scare Crow-Like, GRAS- GAI, RGA, SCR, HDG- Homeodomain Glabrous, ATHB- Arabidopsis thaliana Homeobox Protein, BLH- BELL1-like Homeodomain, HAT- Homeodomain Arabidopsis thaliana, HD- Homeodomain Protein (homeobox-leucine zipper), PRHA- Pathogenesis-Related HomeoDomain Protein, WOX- WUSCHEL- Related Homeobox Protein, HSF- Heat Shock Factor, JAZ-Jasmonic Acid Inducible, ZIM- Zinc-finger Protein Expressed in Inflorescence Meristem., LOB- Lateral Organ Boundary , LBD- Lateral Organ Boundaries (LOB) Domain Protein, AGL- AGAMOUS-like, MADS- MCM1, Agamous, Deficiens, SRF, MCM1- Myocyte enhancer factor Mef2a core, SRF- Serum Response Factor, EPR- Early Phytochrome Responsive, HOS- High Response to Osmotic Stress, CCA- Circadian Clock Associated, TRFL- Telomere repeat-Binding Protein, ATR- Altered Tryptophan Regulation, NAM- No Apical Meristem, NAC- NAM, ATAF1, CUC2 , CUC-Cup-Shaped Cotyledon, ANAC-Arabidopsis NAC domain containing, NTL-NAC TRANSCRIPTION FACTOR-LIKE , NIN- Nodule Inception Protein, NLP- Nin-Like Protein, PHR- Photolyase/blue light Photoreceptor, RAV-Regulator of ATPase of the Vacuolar Membrane, REM- Reproductive Meristem Gene, SBP- Squamosa Promoter Binding Protein, SPL- SBP-Like , TAZ- Transcriptional Adaptor Zinc Bundle, BTB- broad-complex, tramtrack and bric a` brac Region. , POZ- Poxvirus zinc fingers, TCP- TB1, CYC and PCFs Proteins, TB- Teosinte Branched Protein, CYC- Cyclodea, PCF- From Rice Proliferating Cell Nuclear Antigen (PCNA) Gene Factor, TLP- TUBBY-Like Protein, ZAP- Zinc Dependant Activator Protein
Differential Responses to Insect Treatments
The contrast in TF gene expression was particularly great between S. exigua and P. rapae
treatments. We confirmed this stark contrast by conducting RT-PCR on 17 of the 23
27
affected AP2/ERF TFs (including 2 RAV genes). We were able to statistically validate
20 out of 26 microarray expression values for the ERFs; however, if we include those
instances where gene expression RT-PCR levels were in the same direction as
microarray, we achieved over 92% confirmation (Table 1.3). RT-PCR data in
conjunction with our array data clearly show that TINY2 is only P. rapae responsive,
while four TF genes, including SIMRAP2.4, DREBb, ERF11, and ERF104 were solely
responsive to S. exigua feeding.
28
Table 1.3: RT-PCR Confirmation of AP2-ERF Transcription Factor Genes- ERF Transcription Factor genes affected by P. rapae and S. exigua in the array were amplified and quantified using RT-PCR. The number of cases of the listed genes that were significantly up- or down-regulated in the array is 26 and noted by red coloring. We confirmed 20 of these expression values using RT-qPCR as indicated by double asterisks and encountered 2 false positives (FP). Single asterisks represent expression level changes that were found to be significant through RT-PCR but not by the array. (+)* indicates an unconfirmed positive where fold change measured by RT-qPCR was directionally similar to the array, but not significantly different from controls.
29
Another clear difference between the specialist and generalist caterpillars was in the
expression of members of the C2H2 transcription factor families. Genes encoding
ZAT10, ZAT5, ZAT12, and AZF3, were up-regulated by S. exigua, but not P. rapae in any
tissue or treatment.
We observed differences in MYB TF genes in response to caterpillar vs. aphid
treatments. MYBs have a broad spectrum of functions in Arabidopsis, ranging from cell
cycle regulation to plant secondary metabolism and abiotic stress responses (Abe et al.
2003; Celenza et al. 2005; Cominelli et al. 2005; Baudry et al. 2006). Among the
differences, MYB95 expression was increased by caterpillars, while MYB34/ATR1 was
not only up by caterpillars, but down-regulated by both aphids. Also, caterpillars alone
significantly increased the expression of 12 homeobox transcription factor genes,
including Pound-Foolish, ATHB15, and ATHB7.
Both specialists, P. rapae and B. brassicae induced the expression of RD26, a NAC
family gene that is highly responsive to water stress and ABA treatment (Fujita et al.
2004). Furthermore, B. brassicae and wounding induced the expression of several LOB
genes (Lateral Organ Boundary TF). LOBs encode a diverse, plant-specific class of
proteins with distinct spacial expression in adaxial bases of lateral organs from the shoot
apical meristem, including lateral roots (Shuai et al. 2002) and might be important in re-
growth after wounding.
30
Over-represented Cis-Elements
Using available bioinformatics tools, we also identified several cis-elements that were
over-represented (p<0.00001) in genes co-expressed by each insect (Table 1.4). Because
transcription factors serve as both activators and repressors (McGrath et al. 2005),
identifying enriched binding sites in the upstream regions of the genes that are up- or
down-regulated can provide insight into the regulatory networks involved in insect
defense signaling. Several cis-elements, including ABREs, G-Boxes, TATA boxes,
WRKY boxes, and I-boxes were significantly enriched, indicating their importance in
plant responses to insects. We also conducted a hierarchical cluster analysis of all the
known TF binding sites regulated by each treatment against a random gene set. Using the
clade feature in Cluster 3.0 (Eisen et al. 1998), we observed similar cis-element profiles
in most of the treatments. However, genes down-regulated by M. persicae had a unique
cis-element signature, whereas cis-regulatory sequences of genes down-regulated by the
two caterpillars, S. exigua and P. rapae, were very similar and share the same sub-clade
(Figure 3). Although we did not find any insect species-specific promoter motifs from our
Principal Component Analysis (data not shown), it is evident that the combination of
transcription factor profiles as well as the binding sites in the down stream genes
comprise a complex regulatory network that is unique to each insect treatment.
31
Figure 1.3: Heat Map of Cis-Element Distributions in Genes Up- and Down- Regulated by Insect Herbivory and Wounding- Transcription factor binding sites were found in all affected genes using known databases and customized Perl Scripts. We used Cluster 3.0 for Cluster Analysis and SAS for Principal Component Analysis. Most gene promoters contained similar sets of binding sites. Data do not reveal any clear patterns or signatures of cis-elements within specific treatment groups. Myzus persicae down-regulated genes show the least similarity in cis-element distributions with other treatments.
32
TATA boxes are widely over-represented in many of the genes up- regulated by
caterpillars, while I-Boxes, found in many light-regulated genes (Giuliano et al. 1988),
were enriched in down-regulated genes. Interestingly, most treatments repressed genes
that were enriched in “Evening Element” motifs. This cis-element was found to be over-
represented in many genes whose expression peaks at dusk (Harmer et al. 2000, Table
1.4). We also conducted a cis-element analysis using the MotifSampler tool (Thujs et al.
2002). Several unknown motifs were enriched in each sample set (Supplementary
Materials), while others matched the sequences for ABRE elements in several of the
treatments (Table 1.4). Unknown motifs will need to be experimentally characterized to
confirm their role in plant-insect responses. Taken together, these data suggest that
several cis-elements are important players in the gene regulatory networks in Arabidopsis
following insect attack.
33
Table 1.4: Characterized Enriched Cis-Elements in Genes Regulated by Insect Herbivory and Wounding. “Enriched” signifies that motif numbers found in each treatment set are higher than would randomly be found in the genome (p<1.e-5).
34
Discussion
The goal of this study was to highlight the importance and diversity of transcription
factor genes transcriptionally elicited after attack by four different herbivores and
wounding in Arabidopsis. We hypothesized that differential responses to insects are
largely under the control of transcription factors. In many instances, transcription factors
are points of cross-talk between signaling pathways and thus could be key players in how
plants recognize and respond to various stresses (Abe et al. 2003; Li et al. 1998; Lorenzo
et al. 2003; Chini et al. 2007; Thines et al. 2007). Indeed, we identified 193 TF genes that
were differentially expressed in at least one treatment, and discovered several TF genes
affected by multiple treatments as part of a generalized stress response. Most TF gene
profiles differed greatly among treatments, suggesting that differential gene responses to
insects are shaped by transcriptional activation of stimulus-specific transcription factors.
Furthermore, TF genes detected in our array are also components in light, drought, and
oxidative stress pathways, demonstrating clear intersections between insect-responses
and these physiological pathways. Finally, a bioinformatics analysis of putative
transcription factor binding sites (cis-elements) in the promoters of differentially
regulated genes revealed several cis-elements that are playing a significant role in
responses to herbivorous insects.
Our results indicate that different insect species differentially regulate a multitude of
genes in many pathways. Previous microarray studies have also shown that the genes up-
regulated during herbivory function in drought responses, secondary metabolite
35
production, and defense signaling (DeVos et al. 2005, Thompson and Goggin 2006).
These studies also found marked differences between insect treatments. However, a
microarray analysis of Arabidopsis gene expression changes induced by a specialist (P.
rapae) and generalist (Spodoptera littoralis) herbivore reported little or no difference in
the expression profiles between the two insects (Reymond et al. 2004), the opposite of
what we saw in our study. Due to known elicitors in insect saliva, the authors anticipated
significant differences, but observed a very conserved transcriptional profile from both
insects. One potential problem in interpreting these data was the size of the array used.
Only about 25% of the entire Arabidopsis genome was represented on the authors’
customized array so results do not reflect true global transcriptional changes caused by
different herbivores. Using a whole-genome array, De Vos et al. (2005) found that gene
expression changes elicited by P. syringae, M. persicae, A. brassicicola, F. occidentalis,
and P. rapae were distinct. Taken together with our data, this suggests that plants have a
way of perceiving and responding differently to their attacker and that this information is
transmitted via response-specific transcription factors.
Surprisingly, the largest number of up- or down-regulated genes seen in the study by
DeVos et al. (2005) and in our study was caused by M. persicae, although little to no
visible damage was seen on the plants (H. Appel, personal communication). For example
JAZ3 and JAZ9, which were highly activated by insect and wounding, were both down-
regulated by M. persicae. Expression of Speckle-type POZ/ TAZ family TF genes were
up-regulated by all insects, but are repressed by M. persicae. These are a specific group
of calmodulin-binding proteins that are responsive to H2O2 and SA treatment (Du and
36
Poovalah 2004) and may provide insight to signaling pathways triggered by insects but
eluded by M. persicae. All TF genes up-regulated by M. persicae are unique to this
insect, including the histone deacytelase gene, HD2A, which represses gene transcription
by adding acetyl groups to histone proteins affecting chromatin remodeling (Lusser et al.
1997; Wu et al. 2000). A quick search for functional annotations found that 10 of 48 TFs
down-regulated by M. persicae are negative regulators of gene activity (Supplemental
Materials, Table 1.4). Although some research suggests that phloem-feeding insects are
“stealthier” feeders because their stylets cause significantly less tissue damage than
chewing caterpillars (Kempema et al. 2007; Zarate et al. 2007; Hao et al. 2008), our
results suggest the opposite of M. persicae. “Stealth” entails minimal reaction via
blocked perception or suppressed recognition to an insect despite a large amount of
damage (Karban & Agrawal 2002; Schultz 2002). A recent study by DeVos and Jander
et al. (2009) found M. persicae saliva to be a strong elicitor of defense induction in
Arabidopsis, which would explain why we observed little damage but drastic
transcriptional changes after treatment by this insect.
Due to the large number of TF genes that were affected my multiple treatments, we
propose their function in a generalized role in plant defense. Many TF genes including
members of the JAZ/ZIM, WRKY, NAC, and POZ/TAZ families were universally up- or
down-regulated across all treatments. Insect-induced genes with the strongest
transcriptional response belonged to the JAZ (ZIM) family. Others (DeVos et al. 2005;
Reymond et al. 2004) have reported significant JAZ transcription after P. rapae, S.
exigua, and JA treatments. JAZ proteins are key regulators of the JA-signaling pathway
37
(Chini et al. 2007; Thines et al. 2007) and have been found to be activated in response to
Malacosoma disstria feeding in poplar (Major & Constabel 2006), and S. exigua feeding
in Arabidopsis (Chung et al. 2008). We provide additional evidence that JAZ genes are
physiologically significant players after herbivory by diverse insect attackers.
In addition to JAZ TF genes, genes encoding WRKY TFs are responsive to numerous
biotic stresses including pathogen infection and are critical in defense-related hormone
cross-talk (Eulgem et al. 1999; Yu et al. 2001; for review see Ulker and Somssich 2004).
In our study WKRY40 was activated by every treatment except feeding by M. persicae.
WRKY40 is an active component in resistance against necrotropic fungi and its expression
may confer susceptibility to bacterial pathogens, depending on its interaction with other
WKRY genes (Xu et al. 2006). Chen et al. (2002) found that WRKY40 and WRKY33 were
quickly up-regulated in wounded plants after 30 minutes. In our array, S. exigua and B.
brassicae induced the expression of WRKY33 and WRKY25. Both of these TF genes are
needed for resistance to necrotrophic fungi, ROS signaling and repression of SA-
mediated responses (Zheng et al. 2006, 2007), and are substrates for MPK4 (Andreasson
et al. 2005), a negative regulator of the SA pathway. We also found WRKY binding
sites, or W-Boxes, to be highly enriched in genes up-regulated by S. exigua 24 hrs after
treatment in both local and systemic tissue. This suggests that WRKY transcription
factors mediate generalized responses to S. exigua by regulating the expression of down-
stream defense genes possibly involved in SA signaling. This implicates S. exigua
feeding may be manipulating JA-inducible defense responses by triggering SA pathways,
as SA is a negative regulator or JA responses (Spoel et al. 2003; See review by Beckers
38
& Spoel 2005).
Insect feeding differentially regulated seven C2H2 TF gene family members. C2H2
proteins have been implicated in abiotic stress signaling (Englbrecht et al. 2004). In
particular ZAT12 functions in oxidative and light-related stress management in
Arabidopsis (Davletova et al. 2005. Other C2H2 TF genes such as ZAT10 (STZ) and
AZF3 are responsive to salt, cold, (Sakamoto et al. 2000) and ethephon (an ethylene
mimic) treatment (Sakamoto et al. 2004). S. exigua up-regulated 5 of the 7 C2H2 genes.
We recently found that S. exigua elicits greater amounts of ET than P. rapae immediately
after herbivore feeding (Rehrig, unpublished data). In combination with the AP2/ERF
data, this suggests that the expression of C2H2 genes by S. exigua might also be driven
by secondary biotic stresses caused by insect feeding or increased ethylene production.
Transcriptional responses to wounding were minimal outside of generalized stress
signaling, but did include the expression of LOB TF genes. LOBs encode a diverse,
plant-specific class of proteins with distinct spatial expression in adaxial bases of lateral
organs arising from the shoot apical meristem (Shuai et al. 2002) and might be important
for re-growth after wounding damage.
Although the expression of TFs are shared across many treatments as part of a general
stress response, subsets of transcription factors specific to responses to caterpillars vs.
aphids or generalists vs. specialists were observed. For example, both MYB95 and
MYB34 transcripts were up-regulated by caterpillars, but not by aphids. In fact, MYB34
39
was down-regulated by both B. brassicae and M. persicae. MYB34/ATR1 trans-activates
tryptophan synthesis genes and affects auxin and indolyl glucosinolate biosynthesis
(Celenza et al. 2005). This would suggest that caterpillars up-regulate the production of
indolyl glucosinolates, while aphids decrease their levels. However, Mewis et al. 2006
found that both M. persicae and S. exigua induced the production of indolyl
glucosinolates in Arabidopsis, while B. brassicae did not and P. rapae did only slightly.
Additionally, M. persicae feeding induces the production of the toxic indolyl
glucosinolate 4MI3M (Kim & Jander 2007) and increases glucosinolate-dependent O-
methyltransferases gene transcription (DeVos and Jander 2009). Thus, further
experimentation on the involvement of ATR1 in the regulation of indolyl glucosinolates
in response to caterpillars and aphids is needed.
Dramatic differences between TF expression profiles elicited by the specialist caterpillar,
P. rapae, and the generalist caterpillar S. exigua, particularly involving AP2/ERF
transcription factors were observed. Subsequent RT-PCR found many of the genes to be
affected by both caterpillars. Nonetheless, an ERF TF, TINY2, was specifically
upregulated by P.rapae whereas DREBb, SIMRAP2.4, ERF104, and ERF11 were only
affected by S. exigua (Table 1.3). TINY2 binds to the DRE element and increases in
response to ABA, drought, salt, cold, wounding, and SA treatment (slightly), but not
ethylene (Wei et al. 2005).
According to the TAIR website, SIMRAP2.4 responds to H2O2 treatment and Tobacco
Mosaic Virus infection, but is down-regulated by SA. Wang et al. (2007) found that the
40
DREBb Family TF gene is COI1- dependent and JA-inducible with a peak expression 30
minutes after JA treatment or wounding. Furthermore, this gene is highly wound-
inducible and peaks in expression 15 minutes after mechanical wounding (Walley et al.
2007). Adding insect regurgitant from T. ni larvae or oligouronides to the wound site
significantly increases DREBb (ERF18) expression in both local and systemic tissue
(Walley et al. 2007). Over-expression of DREBb increased the expression of VSP2 but
not LOX3, which is different from the activity of ERF1 (Wang et al. 2007). ERF11 is an
ethylene-inducible transcriptional repressor with an EAR motif (Yang 2005) and is
highly induced by chitin treatment (Libault et al. 2007), MeJA application and Alternaria
brassicicola infection (McGrath et al. 2005). ERF104 serves as a substrate for MAPK6
and is required for FLG22-induced ET signaling (Bethke et al. 2009). Plants over-
expressing ERF104 had increased transcripts of pathogensis-related genes that are not
induced by ERF1 activation or JA and ET treatment; ERF104 signaling may represent a
novel function for ERF TFs, specifically after insect attack. All four ERF transcription
factors uniquely up-regulated by S. exigua were found to be chitin-responsive by Libault
et al. 2007. From our ERF data, it is tempting to speculate that Arabidopsis responses to
P. rapae are organized through ABA-dependent but ethylene-independent pathways,
while S. exigua feeding produces a chitin-like signal that is ET-dependent. Efforts are
currently underway in our lab to elucidate the differential signaling mechanisms induced
by each insect after herbivory in Arabidopsis.
Both TF expression profiles as well as cis-element analysis demonstrate clear overlaps
between insect-induced responses and drought and light signaling. During caterpillar
41
feeding large portions of tissue are damaged or removed by wounding due to chewing,
increasing transpiration rates (Aldea et al. 2005) and affecting the water status of the
plant. Insect-induced expression of TFs involved in drought, ABA responses, or
meristem regulation including DREBs, RAVs, NACs, and Homeobox TFs, (Sukuma et
al. 2002, Ooka et al. 2003; Olssen et al. 2004; Sohn et al. 2006) was observed.
Transcriptional regulation of these genes may enable the plant to recover more quickly
from damage by mediating ABA signaling or redirecting venation to induce the growth
of new tissue. Our cis-element analysis provides additional evidence for insect-induced
drought stress. One core element appearing in genes responding to multiple treatments
was (A/C)AC(A/G)TG. This motif is similar to ABRE sites for ABA-responsive TFs and
is directly involved in the activation of genes needed to maintain optimal water status in
the plant. (Choi et al. 2000; Fujita et al. 2005).
Our results also suggest that insect feeding may interact with light-related pathways and
circadian rhythms in Arabidopsis. For example, several genes encoding Phytochrome
Interacting Factors (PIFs) (See review by Castillon et al. 2007), light-responsive GATA
TFs (Chan et al. 2001) and EPR1, a circadian oscillator and phytochrome responsive
gene (Kuno et al. 2003) were altered after insect treatments. Additionally, most
treatments repressed expression of genes that were enriched in “Evening Element” motifs.
This cis-element is found to be over-represented in many genes whose expression peak at
dusk (Harmer et al. 2000, Supplementary Materials, Table 1.3). If insects down-regulate
genes whose expression peaks at dusk, they may be manipulating the plant’s central
circadian oscillator. Alternatively, this simply may be a plant response to damage that
42
permits re-growth and recovery at a later time when insect feeding has ceased. Circadian
Clock Associated Protein (CCA1), a crucial component of the plant’s internal clock, is a
MYB-related protein that was significantly up-regulated by P. rapae after 24 hours in
systemic tissue. CCA1 binds to the TOC1 promoter to negatively regulate its expression
(Alabadi et al. 2001), which could explain why genes down regulated by P. rapae after
24 hours are enriched in this motif. Furthermore, insect-repressed genes were enriched in
I-boxes, which regulate light responses and circadian rhythms (Giuliano et al. 1988; Rose
et al. 1990; Chan et al. 2001). In a recent review, Ballare (2009) provides evidence for
PIF interaction and far-red light signaling as anti-herbivore responses via reallocation of
resources throughout the plant. Because we took measures to ensure that no circadian
cycles affected the results and RNA levels were normalized to those from control plants
grown in the same environment, our data in combination with previous studies suggest
that plants may be tightly regulating the expression of light-regulated developmental
genes as well as defense-related pathways to optimize recovery after severe damage
caused by insect herbivory.
Why are insect species causing different transcription factor expression profiles? To
date, no known receptors have been identified for insect-specific elicitors. Although
salivary components are probably an important part of elicitation, (Hailitschke et al.
2001; 2003; Schmelz et al. 2009), many studies have shown that the subsequent interplay
and levels of plant hormones such as JA, SA, ABA, and ET after herbivory are critical
for appropriate defense responses (Stotz et al. 2000; Thaler et al. 2002; Mewis et al.
2005; vonDahl et al. 2007). For example, a study conducted by Stoltz et al. (2000) using
43
ethylene insensitive and signaling mutants found that resistance of Arabidopsis to a
generalist herbivore S. littoralis, but not a specialist insect, requires mediation through an
ethylene pathway. Similarly, Mewis et al. (2005) found that responses to generalist but
not specialist insects required functional NPR1 and ETR1 genes, which regulate SA- and
ET-pathways, respectively. The differences in TF profiles observed in this experiment
may be the result of down-stream changes caused by rapid phtyohormone signaling after
insect attack.
In addition to searching for motifs unique to each treatment, we conducted a cluster
analysis of all of the cis-elements in the 1000 bp- upstream promoter regions of all
affected genes by treatment (Eisen et al. 1998). No known insect-specific regulatory
elements have been identified in plants and little is known about gene regulatory
networks in plant-insect interactions. Segal et al. (2003) and Beer and Tavazoie (2004)
proposed that a substantial part of the patterns in any gene expression data could be
explained by “cis-element profiles”. Using computational predictions with yeast stress
data, Segal et al. (2005) argued that motif profiles in conjunction with microarray
analysis can help understand important regulatory networks including those involved in
the pathology of cancer development. Our results showed that the vast majority of genes
differentially expressed in any treatment had many of the known cis-elements in their
promoter regions and no single element appeared to be “insect specific”. Nonetheless, a
small group of elements were associated specifically with particular insect treatments and
allowed some separation from randomly-selected TF genes. Cis-elements of TF genes
down-regulated by M. persicae were very different from any other set. Several matches
44
to ABRE- and ABRE-like elements were identified from our cis-element analysis using
MotifSampler. These may be important in plant-herbivory transcriptional responses
because ABREs regulate ABA and drought-responsive genes (Fujita et al. 2005).
Likewise, insect herbivores increase transpiration rates and severely affect the water
status of plants (Aldea et al. 2005). However, others that were determined to be enriched
by ATHENA were not confirmed by this secondary Gibb’s (Thijs et al. 2001) analysis.
This could be due to discrepancies in sequence consensi and high nucleotide substitution
rates that often appear in novel motif searches. Furthermore, for our MotifSampler
search we used all genes up- or down- regulated by each insect, whereas ATHENA
searches were more refined, and further broken down by treatment and tissue type. It
appears likely that regulation of transcriptional responses to insects via TFs may be under
combinatorial control, requiring the action of several TFs acting in concert to initiate a
broad array of expression patterns (Lindlof et al. 2009; see review by Singh 1998).
Because TATA boxes were enriched in genes up-regulated by both caterpillars, these
elements may also be important regulatory components in biotic stress. According to
Basehoar et al. (2004), genes of Saccharomyces sp. with TATA boxes were found to play
a more stress-responsive role when osmolarity, pH-balance, or nutrient availability
became abnormal. However, genes without TATA boxes performed more housekeeping
functions and may not need as much transcriptional regulation due to constitutive
expression. The authors suggested that TATA-containing genes induced by
environmental cues may need “finer regulatory tuning” to be switched on or off at the
appropriate times and at the appropriate levels. Arabidopsis plants may respond to insects
45
as general stress stimuli and activate genes enriched in TATA-binding motifs. Whether
the same biological phenomenon of large transcriptional regulation of potentially stress-
related genes with TATA boxes is also occurring in Arabidopsis in response to insect
herbivory requires further investigation.
Our results show that insect attack elicits expression of many Arabidopsis transcription
factor genes normally involved in generalizes stress responses, as well as insect-specific
sets of defense-related TFs. This is evident in the number and identities of transcription
factors differentially regulated by our 4 insect treatments and mechanical wounding.
Furthermore, we have identified key cis-elements over-represented in co-expressed genes
after insect attack providing insight into the complex networks and regulatory pathways
insects elicit in plants during herbivory.
46
References
1. Abe H, Urao T, Ito T, Seki M, Shinozaki K, Yamaguchi-Shinozaki K (2003)
Arabidopsis AtMYC2 (bHLH) and AtMYB2 (MYB) function as transcriptional activators in abscisic acid signaling. The Plant Cell 15: 63-78
2. Alabadi D, Oyama T, Yanovsky MJ, Harmon FG, Mas P, Kay SA (2001) Reciprocal
regulation between TOC1 and LHY/CCA1 within the Arabidopsis circadian clock. Science 293: 880-883
3. Alborn HT, Hansen TV, Jones TH, Bennett DC, Tumlinson JH, Schmelz EA, Teal
PEA (2007) Disulfooxy fatty acids from the American bird grasshopper Schistocerca americana, elicitors of plant volatiles. Proceedings of the National Academy of Sciences 104: 12976-12981
4. Andreasson E, Jenkins T, Brodersen P, Thorgrimsen S, Petersen NHT, Zhu S, Qiu J-
L, Micheelsen P, Rocher A, Petersen M, Newman M-A, Bjorn Nielsen H, Hirt H, Somssich I, Mattsson O, Mundy J (2005) The MAP kinase substrate MKS1 is a regulator of plant defense responses. The EMBO journal 24: 2579-2589
5. Ballare CL (2009) Illuminated behaviour: phytochrome as a key regulator of light
foraging and plant anti-herbivore defence. Plant, Cell & Environment 32: 713-725 6. Basehoar AD, Zanton SJ, Pugh BF (2004) Identification and distinct regulation of
yeast TATA box-containing genes. Cell 116: 699-709 7. Baudry A, Caboche M, Lepiniec L (2006) TT8 controls its own expression in a
feedback regulation involving TTG1 and homologous MYB and bHLH factors, allowing a strong and cell-specific accumulation of flavonoids in Arabidopsis thaliana. The Plant Journal : for Cell and Molecular Biology 46: 768-779
8. Beckers GJ, Spoel SH (2006) Fine-tuning plant defence signalling: salicylate versus
jasmonate. Plant Biology 8: 1-10 9. Beer M, Tavazoie S (2004) Predicting gene expression from sequence. Cell 117:
185-198 10. Bethke G, Unthan T, Uhrig JF, Poschl Y, Gust AA, Scheel D, Lee J (2009) Flg22
regulates the release of an ethylene response factor substrate from MAP kinase 6 in Arabidopsis thaliana via ethylene signaling. Proceedings of the National Academy of Sciences 106: 8067-8072
47
11. Castillon A, Shen H, Huq E (2007) Phytochrome Interacting Factors: central players in phytochrome-mediated light signaling networks. Trends in Plant Science 12: 514-521
12. Celenza JL, Quiel JA, Smolen GA, Merrikh H, Silvestro AR, Normanly J, Bender J
(2005) The Arabidopsis ATR1 Myb transcription factor controls indolic glucosinolate homeostasis. Plant Physiology 137: 253-262
13. Chan CS, Guo L, Shih MC (2001) Promoter analysis of the nuclear gene encoding
the chloroplast glyceraldehyde-3-phosphate dehydrogenase B subunit of Arabidopsis thaliana. Plant Molecular Biology 46: 131-141
14. Chen W, Provart NJ, Glazebrook J, Katagiri F, Chang H-S, Eulgem T, Mauch F,
Luan S, Zou G, Whitham SA, Budworth PR, Tao Y, Xie Z, Chen X, Lam S, Kreps JA, Harper JF, Si-Ammour A, Mauch-Mani B, Heinlein M, Kobayashi K, Hohn T, Dangl JL, Wang X, Zhu T (2002) Expression profile matrix of Arabidopsis transcription factor genes suggests their putative functions in response to environmental stresses. The Plant Cell 14: 559-574
15. Cheong YH, Chang H-S, Gupta R, Wang X, Zhu T, Luan S (2002) Transcriptional
profiling reveals novel interactions between wounding, pathogen, abiotic stress, and hormonal responses in Arabidopsis. Plant Physiology 129: 661-677
16. Chini A, Fonseca S, Fernandez G, Adie B, Chico JM, Lorenzo O, Garcia-Casado G,
Lopez-Vidriero I, Lozano FM, Ponce MR, Micol JL, Solano R (2007) The JAZ family of repressors is the missing link in jasmonate signalling. Nature 448: 666-671
17. Choi H, Hong J, Ha J, Kang J, Kim SY (2000) ABFs, a family of ABA-responsive
element binding factors. Journal of Biological Chemistry 275: 1723-1730 18. Chung HS, Koo AJK, Gao X, Jayanty S, Thines B, Jones AD, Howe GA (2008)
Regulation and function of Arabidopsis JASMONATE ZIM-domain genes in response to wounding and herbivory. Plant Physiology 146: 952-964
19. Cominelli E, Galbiati M, Vavasseur A, Conti L, Sala T, Vuylsteke M, Leonhardt N,
Dellaporta SL, Tonelli C (2005) A guard-cell-specific MYB transcription factor regulates stomatal movements and plant drought tolerance. Current Biology 15: 1196-1200
20. Davletova S, Schlauch K, Coutu J, Mittler R (2005) The zinc-finger protein Zat12
plays a central role in reactive oxygen and abiotic stress signaling in Arabidopsis. Plant Physiology 139: 847-856
21. Davuluri RV, Sun H, Palaniswamy SK, Matthews N, Molina C, Kurtz M, Grotewold
E (2003) AGRIS: Arabidopsis gene regulatory information server, an information
48
resource of Arabidopsis cis-regulatory elements and transcription factors. BMC Bioinformatics 4: 25
22. Delessert C, Wilson IW, Van Der Straeten D, Dennis ES, Dolferus R (2004) Spatial
and temporal analysis of the local response to wounding in Arabidopsis leaves. Plant Molecular Biology 55: 165-181
23. DeVos M, Jander G (2009) Myzus persicae (green peach aphid) salivary components
induce defence responses in Arabidopsis thaliana. Plant, Cell & Environment 32: 1548-1560
24. DeVos M, Van Oosten VR, Van Poecke RMP, Van Pelt JA, Pozo MJ, Mueller MJ,
Buchala AJ, Metraux J-P, Van Loon LC, Dicke M, Pieterse C, M. J. (2005) Signal dignature and transcriptome changes of Arabidopsis during pathogen and insect attack. Molecular Plant-Microbe Interactions 18: 923-937
25. Du L, Poovaiah BW (2004) A novel family of Ca2+/calmodulin-binding proteins
involved in transcriptional regulation: interaction with fsh/Ring3 class transcription activators. Plant Molecular Biology 54: 549-569
26. Ehlting J, Chowrira S, Mattheus N, Aeschliman D, Arimura G-I, Bohlmann J (2008)
Comparative transcriptome analysis of Arabidopsis thaliana infested by diamond back moth (Plutella xylostella) larvae reveals signatures of stress response, secondary metabolism, and signalling. BMC genomics 9: 154
27. Ehlting J, Mattheus N, Aeschliman DS, Li E, Hamberger B, Cullis IF, Zhuang J,
Kaneda M, Mansfield SD, Samuels L, Ritland K, Ellis BE, Bohlmann J, Douglas CJ (2005) Global transcript profiling of primary stems from Arabidopsis thaliana identifies candidate genes for missing links in lignin biosynthesis and transcriptional regulators of fiber differentiation. The Plant Journal 42: 618-640
28. Eisen MB, Spellman PT, Brown PO, Botstein D (1998) Cluster analysis and display
of genome-wide expression patterns. Proceedings of the National Academy of Sciences 95: 14863-14868
29. Englbrecht C, Schoof H, Bohm S (2004) Conservation, diversification and
expansion of C2H2 zinc finger proteins in the Arabidopsis thaliana genome. BMC genomics 5: 39
30. Eulgem T (2005) Regulation of the Arabidopsis defense transcriptome. Trends in
Plant Science 10: 71-78 31. Fujita M, Fujita Y, Maruyama K, Seki M, Hiratsu K, Ohme-Takagi M, Tran LS,
Yamaguchi-Shinozaki K, Shinozaki K (2004) A dehydration-induced NAC protein,
49
RD26, is involved in a novel ABA-dependent stress-signaling pathway. The Plant Journal 39: 863-876
32. Fujita Y, Fujita M, Satoh R, Maruyama K, Parvez MM, Seki M, Hiratsu K, Ohme-
Takagi M, Shinozaki K, Yamaguchi-Shinozaki K (2005) AREB1 Is a transcription activator of novel ABRE-dependent ABA signaling that enhances drought stress tolerance in Arabidopsis. The Plant Cell 17: 3470-3488
33. Giuliano G, Pichersky E, Malik PS, M P Timko MP, Scolnik PA, Cashmore AR
(1988) An evolutionarily conserved protein binding sequence upstream of a plant light-regulated gene Proceedings of the National Academy of Sciences 85: 7089-7093
34. Guo A, He K, Liu D, Bai S, Gu X, Wei L, Luo J (2005) DATF: a database of
Arabidopsis transcription factors. Bioinformatics 21: 2568-2569 35. Halitschke R, Gase K, Hui D, Schmidt DD, Baldwin IT (2003) Molecular
interactions between the specialist herbivore Manduca sexta (Lepidoptera, sphingidae) and its natural host Nicotiana attenuata. VI. Microarray analysis reveals that most herbivore-specific transcriptional changes are mediated by fatty acid-amino acid conjugates. Plant Physiology 131: 1894-1902
36. Halitschke R, Schittko U, Pohnert G, Boland W, Baldwin IT (2001) Molecular
interactions between the specialist herbivore Manduca sexta (Lepidoptera, Sphingidae) and its natural host Nicotiana attenuata. III. Fatty acid-amino acid conjugates in herbivore oral secretions are necessary and sufficient for herbivore-specific plant responses. Plant Physiology 125: 711-717
37. Hao P, Liu C, Wang Y, Chen R, Tang M, Du B, Zhu L, He G (2008) Herbivore-
induced callose deposition on the sieve plates of rice: an important mechanism for host resistance. Plant Physiology 146: 1810-1820
38. Harmer SL, Hogenesch JB, Straume M, Chang H-S, Han B, Zhu T, Wang X, Kreps
JA, Kay SA (2000) Orchestrated transcription of key pathways in Arabidopsis by the circadian clock. Science 290: 2110-2113
39. Kahl J, Siemens DH, Aerts RJ, Gäbler R, Kühnemann F, Preston CA, Baldwin IT
(2000) Herbivore-induced ethylene suppresses a direct defense but not a putative indirect defense against an adapted herbivore. Planta 210: 336-342
40. Karban R, Agrawal AA (2002) Herbivore offense. Annual Review of Ecology and
Systematics 33: 641-664
50
41. Kempema LA, Cui X, Holzer FM, Walling LL (2007) Arabidopsis transcriptome changes in response to phloem-feeding silverleaf whitefly nymphs. Similarities and distinctions in responses to aphids. Plant Physiology 143: 849-865
42. Kessler A, Baldwin IT (2002) Plant responses to insect herbivory: the emerging
molecular analysis. Annual Review of Plant Biology 53: 299-328 43. Kim JH, Jander G (2007) Myzus persicae (green peach aphid) feeding on
Arabidopsis induces the formation of a deterrent indole glucosinolate. The Plant Journal: for Cell and Molecular Biology 49: 1008-1019
44. Kuno N, Moller SG, Shinomura T, Xu X, Chua N-H, Furuya M (2003) The Novel
MYB Protein EARLY-PHYTOCHROME-RESPONSIVE1 Is a Ccomponent of a slave circadian oscillator in Arabidopsis. The Plant Cell 15: 2476-2488
45. Li J, Brader G, Palva ET (1999) The WRKY70 transcription factor: a mode of
convergence for jasmonate-mediated and salicylate mediated signals in plant defense. The Plant Cell 16: 319-331
46. Libault M, Wan J, Czechowski T, Udvardi M, Stacey G (2007) Identification of 118
Arabidopsis transcription factor and 30 Ubiquitin-Ligase genes responding to chitin, a plant-defense elicitor. Molecular Plant-Microbe Interactions 20: 900-911
47. Lindlof A, Brautigam M, Chawade A, Olsson O, Olsson B (2009) In silico analysis
of promoter regions from cold-induced genes in rice (Oryza sativa L.) and Arabidopsis thaliana reveals the importance of combinatorial control. Bioinformatics 25: 1345-1348
48. Lorenzo O, Chico JM, Sanchez-Serrano JJ, Solano R (2004) JASMONATE-
INSENSITIVE1 encodes a MYC transcription factor essential to discriminate between different jasmonate-regulated defense responses in Arabidopsis. The Plant Cell 16: 1938-1950
49. Lorenzo O, Piqueras R, Sanchez-Serrano JJ, Solano R (2003) ETHYLENE
RESPONSE FACTOR1 integrates signals from ethylene and jasmonate pathways in plant defense. The Plant Cell 15: 165-178
50. Lusser A, Brosch G, Loidl A, Haas H, Loidl P (1997) Identification of maize histone
deacetylase HD2 as an acidic nucleolar phosphoprotein. Science 277: 88-91 51. Mahalingam R, Gomez-Buitrago A, Eckardt N, Shah N, Guevara-Garcia A, Day P,
Raina R, Fedoroff NV (2003) Characterizing the stress/defense transcriptome of Arabidopsis. Genome Biology 4: R20
51
52. Major IT, Constabel CP (2006) Molecular analysis of poplar defense against herbivory: comparison of wound- and insect elicitor-induced gene expression. The New Phytologist 172: 617-635
53. Mattiacci L, Dicke M (2004) β-Glucosidase: an elicitor of herbivore-induced plant
odor that attracts host-searching parasitic wasps. Proceedings of the National Academy of Sciences 92: 12837-12842
54. Mewis I, Appel HM, Hom A, Raina R, Schultz JC (2005) Major signaling pathways
modulate Arabidopsis glucosinolate accumulation and response to both phloem-feeding and chewing insects. Plant Physiology 138: 1149-1162
55. Mewis I, Tokuhisa JG, Schultz JC, Appel HM, Ulrichs C, Gershenzon J (2006) Gene
expression and glucosinolate accumulation in Arabidopsis thaliana in response to generalist and specialist herbivores of different feeding guilds and the role of defense signaling pathways. Phytochemistry 67: 2450-2462
56. Moran PJ, Thompson GA (2001) Molecular responses to aphid feeding in
Arabidopsis in relation to plant defense pathways. Plant Physiology 125: 1074-1085 57. O'Connor TR, Dyreson C, Wyrick JJ (2005) Athena: a resource for rapid
visualization and systematic analysis of Arabidopsis promoter sequences. Bioinformatics 21: 4411-4413
58. Obayashi T, Kinoshita K, Nakai K, Shibaoka M, Hayashi S, Saeki M, Shibata D,
Saito K, Ohta H (2007) ATTED-II: a database of co-expressed genes and cis-elements for identifying co-regulated gene groups in Arabidopsis. Nucleic Acids Research 35: D863-869
59. Olsson AS, Engstrom P, Soderman E (2004) The homeobox genes ATHB12 and
ATHB7 encode potential regulators of growth in response to water deficit in Arabidopsis. Plant Molecular Biology 55: 663-677
60. Ooka H, Satoh K, Doi K, Nagata T, Otomo Y, Murakami K, Matsubara K, Osato N,
Kawai J, Carninci P, Hayashizaki Y, Suzuki K, Kojima K, Takahara Y, Yamamoto K, Kikuchi S (2003) Comprehensive analysis of NAC family genes in Oryza sativa and Arabidopsis thaliana. DNA Res 10: 239-247
61. Palaniswamy SK, James S, Sun H, Lamb RS, Davuluri RV, Grotewold E (2006)
AGRIS and AtRegNet. a platform to link cis-regulatory elements and transcription factors into regulatory networks. Plant Physiology 140: 818-829
62. Pare PW, Tumlinson JH (1999) Plant volatiles as a defense against insect herbivores.
Plant Physiology 121: 325-332
52
63. Pylatuik J, Fobert P (2005) Comparison of transcript profiling on Arabidopsis microarray platform technologies. Plant Molecular Biology 58: 609-624
64. Reymond P, Bodenhausen N, Van Poecke RMP, Krishnamurthy V, Dicke M,
Farmer EE (2004) A conserved transcript pattern in response to a specialist and a generalist herbivore. The Plant Cell 16: 3132-3147
65. Reymond P, Farmer EE (1998) Jasmonate and salicylate as global signals for
defense gene expression. Current Opinion in Plant Biology 1: 404-411 66. Riechmann JL, Heard J, Martin G, Reuber L, -Z. C, Jiang, Keddie J, Adam L, Pineda
O, Ratcliffe OJ, Samaha RR, Creelman R, Pilgrim M, Broun P, Zhang JZ, Ghandehari D, Sherman BK, -L. Yu G (2000) Arabidopsis transcription factors: Genome-wide comparative analysis among eukaryotes. Science 290: 2105-2110
67. Rose A, Meier I, Wienand U (1999) The tomato I-box binding factor LeMYBI is a
member of a novel class of myb-like proteins. The Plant Journal: for Cell and Molecular Biology 20: 641-652
68. Rozen S, Skaletsky H (2000) Primer3 on the WWW for general users and for
biologist programmers. Methods in Molecular Biology 132: 365-386 69. Sakamoto H, Araki T, Meshi T, Iwabuchi M (2000) Expression of a subset of the
Arabidopsis Cys(2)/His(2)-type zinc-finger protein gene family under water stress. Gene 248: 23-32
70. Sakamoto H, Maruyama K, Sakuma Y, Meshi T, Iwabuchi M, Shinozaki K,
Yamaguchi-Shinozaki K (2004) Arabidopsis Cys2/His2-type zinc-finger proteins function as transcription repressors under drought, cold, and high-salinity stress conditions. Plant Physiology 136: 2734-2746
71. Sakuma Y, Maruyama K, Osakabe Y, Qin F, Seki M, Shinozaki K, Yamaguchi-
Shinozaki K (2006) Functional analysis of an Arabidopsis transcription factor, DREB2A, involved in drought-responsive gene expression. The Plant Cell 18: 1292-1309
72. Schenk PM, Kazan K, Wilson I, Anderson JP, Richmond T, Somerville SC, Manners
JM (2000) Coordinated plant defense responses in Arabidopsis revealed by microarray analysis. Proceedings of the National Academy of Sciences 97: 11655-11660
73. Schmelz EA, Alborn HT, Tumlinson JH (2003) Synergistic interactions between
volicitin, jasmonic acid and ethylene mediate insect-induced volatile emission in Zea mays. Physiologia Plantarum 117: 403-412
53
74. Schmelz EA, Engelberth J, Alborn HT, Tumlinson JH, Teal PEA (2009) Phytohormone-based activity mapping of insect herbivore-produced elicitors. Proceedings of the National Academy of Sciences 106: 653-657
75. Schultz JC (2002) Biochemical ecology: how plants fight dirty. Nature 416: 267 76. Segal E, Friedman N, Kaminski N, Regev A, Koller D (2005) From signatures to
models: understanding cancer using microarrays. Nature Genetics 37 Suppl: S38-45 77. Segal E, Yelensky R, Koller D (2003) Genome-wide discovery of transcriptional
modules from DNA sequence and gene expression. Bioinformatics 19 Suppl 1: i273-282
78. Shah NH, King DC, Shah PN, Fedoroff NV (2003) A tool-kit for cDNA microarray
and promoter analysis. Bioinformatics 19: 1846-1848 79. Shuai B, Reynaga-Pena CG, Springer PS (2002) The lateral organ boundaries gene
defines a novel, plant-specific gene family. Plant Physiology 129: 747-761 80. Singh KB (1998) Transcriptional regulation in plants: the importance of
combinatorial control. Plant Physiology 118: 1111-1120 81. Sohn KH, Lee SC, Jung HW, Hong JK, Hwang BK (2006) Expression and
functional roles of the pepper pathogen-induced transcription factor RAV1 in bacterial disease resistance, and drought and salt stress tolerance. Plant Molecular Biology 61: 897-915
82. Spoel SH, Koornneef A, Claessens SMC, Korzelius JP, Van Pelt JA, Mueller MJ,
Buchala AJ, Matraux J-P, Brown R, Kazan K, Van Loon LC, Dong X, Pieterse CMJ (2003) NPR1 modulates cross-talk between salicylate- and jasmonate-dependent defense pathways through a novel function in the cytosol. The Plant Cell 15: 760-770
83. Sreenivasulu N, Sopory SK, Kavi Kishor PB (2007) Deciphering the regulatory
mechanisms of abiotic stress tolerance in plants by genomic approaches. Gene 388: 1-13
84. Steffens NO, Galuschka C, Schindler M, Bülow L, Hehl R (2004) AthaMap: an
online resource for in silico transcription factor binding sites in the Arabidopsis thaliana genome. Nucleic Acids Research 32: D368-372
85. Storey JD, Tibshirani R (2003) Statistical significance for genome wide studies.
Proceedings of the National Academy of Sciences 100: 9440-9445
54
86. Stotz HU, Pittendrigh BR, Kroymann J, Weniger K, Fritsche J, Bauke A, Mitchell-Olds T (2000) Induced plant defense responses against chewing insects. Ethylene signaling reduces resistance of Arabidopsis against Egyptian cotton worm but not diamondback moth. Plant Physiology 124: 1007-1018
87. Thaler JS, Farag MA, Pare PW, Dicke M (2002) Jasmonate-deficient plants have
reduced direct and indirect defences against herbivores. Ecology Letters 5: 764-774 88. Thijs G, Lescot M, Marchal K, Rombauts S, De Moor B, Rouze P, Moreau Y (2001)
A higher-order background model improves the detection of promoter regulatory elements by Gibbs sampling. Bioinformatics 17: 1113-1122
89. Thijs G, Marchal K, Lescot M, Rombauts S, De Moor B, Rouza P, Moreau Y (2002)
A Gibbs sampling method to detect overrepresented motifs in the upstream regions of coexpressed genes. Journal of Computational Biology 9: 447-464
90. Thines B, Katsir L, Melotto M, Niu Y, Mandaokar A, Liu G, Nomura K, He SY,
Howe GA, Browse J (2007) JAZ repressor proteins are targets of the SCF(COI1) complex during jasmonate signalling. Nature 448: 661-665
91. Thompson GA, Goggin FL (2006) Transcriptomics and functional genomics of plant
defence induction by phloem-feeding insects. Journal of Experimental Botany 57: 755-766
92. Ulker B, Somssich IE (2004) WRKY transcription factors: from DNA binding
towards biological function. Current Opinion in Plant Biology 7: 491-498 93. Vandepoele K, Casneuf T, Van de Peer Y (2006) Identification of novel regulatory
modules in dicotyledonous plants using expression data and comparative genomics. Genome Biology 7: R103
94. von Dahl CC, Havecker M, Schlogl R, Baldwin IT (2006) Caterpillar-elicited
methanol emission: a new signal in plant-herbivore interactions? The Plant Journal : for Cell and Molecular Biology 46: 948-960
95. Walley JW, Coughlan S, Hudson ME, Covington MF, Kaspi R, Banu G, Harmer SL,
Dehesh K (2007) Mechanical stress induces biotic and abiotic stress responses via a novel cis-element. PLoS Genetics 3: 1800-1812
96. Wang Q, Guan Y, Wu Y, Chen H, Chen F, Chu C (2008) Overexpression of a rice
OsDREB1F gene increases salt, drought, and low temperature tolerance in both Arabidopsis and rice. Plant Molecular Biology 67: 589-602
55
97. Wang Z, Cao G, Wang X, Miao J, Liu X, Chen Z, Qu LJ, Gu H (2007) Identification and characterization of COI1-dependent transcription factor genes involved in JA-mediated response to wounding in Arabidopsis plants. Plant Cell Reports
98. Wei G, Pan Y, Lei J, Zhu Y-X (2005) Molecular cloning, phylogenetic analysis,
expressional profiling and in vitro studies of TINY2 from Arabidopsis thaliana. Journal of Biochemistry and Molecular Biology 38: 440-446
99. Winz RA, Baldwin IT (2001) Molecular interactions between the specialist
herbivore Manduca sexta (Lepidoptera, Sphingidae) and its natural host Nicotiana attenuata. IV. Insect-Induced ethylene reduces jasmonate-induced nicotine accumulation by regulating putrescine N-methyltransferase transcripts. Plant Physiology 125: 2189-2202
100. Wittstock U, Agerbirk N, Stauber EJ, Olsen CE, Hippler M, Mitchell-Olds T,
Gershenzon J, Vogel H (2004) Successful herbivore attack due to metabolic diversion of a plant chemical defense. Proceedings of the National Academy of Sciences 101: 4859-4864
101. Wu K, Tian L, Malik K, Brown D, Miki B (2000) Functional analysis of HD2
histone deacetylase homologues in Arabidopsis thaliana. The Plant Journal 22: 19-27
102. Xu X, Chen C, Fan B, Chen Z (2006) Physical and functional interactions between
pathogen-induced Arabidopsis WRKY18, WRKY40, and WRKY60 transcription factors. The Plant Cell 18: 1310-1326
103. Yadav V, Mallappa C, Gangappa SN, Bhatia S, Chattopadhyay S (2005) A basic
helix-loop-helix transcription factor in Arabidopsis, MYC2, acts as a repressor of blue light-mediated photomorphogenic growth. The Plant Cell 17: 1953-1966
104. Yu D, Chen C, Chen Z (2001) Evidence for an important role of WRKY DNA
binding proteins in the regulation of NPR1 gene expression. The Plant Cell 13: 1527-1540
105. Zarate SI, Kempema LA, Walling LL (2007) Silverleaf whitefly induces salicylic
acid defenses and suppresses effectual jasmonic acid defenses. Plant Physiology 143: 866-875
106. Zheng Z, Mosher SL, Fan B, Klessig DF, Chen Z (2007) Functional analysis of
Arabidopsis WRKY25 transcription factor in plant defense against Pseudomonas syringae. BMC Plant Biology 7: 2
56
107. Zheng Z, Qamar SA, Chen Z, Mengiste T (2006) Arabidopsis WRKY33 transcription factor is required for resistance to necrotrophic fungal pathogens. The Plant Journal 48: 592-605
57
Supplemental Materials
Supplemental Figure 1.1: Distribution of parametric p-values from ANOVA. For each of the 26,090 probes present on the microarray, normalized expression ratios from four replicate arrays for each of the four time points were used for an analysis of variance (ANOVA). Shown is the frequency distribution of the resulting p-values. A horizontal line indicates the estimated NULL distribution separating the number of true positive tests (above the line) from negative tests within a given p-value bin (falsely discovered genes).
58
Supplemental Table 1.1: Estimation of differentially expressed genes in all treatments using different false discovery rates (FDR), p values for t-tests, and fold-change cutoffs.
p(t-test)<0.01 p(t-test)<0.05 p(t-test)<0.05, FCb>2 Treatment
genes max. FDRa genes max.
FDRa genes
up genes down
Pr-A-6h 210 0.48 978 0.55 198 73 Pr-A-24h 375 0.38 1597 0.45 158 120 Pr-U-6h 270 0.53 1281 0.56 132 71 Pr-U-24h 307 0.46 1422 0.51 202 118 Se-A-6h 332 0.34 1349 0.42 354 137 Se-A-24h 251 0.50 1356 0.51 262 139 Se-U-6h 359 0.39 1447 0.48 214 73 Se-U-24h 376 0.38 1523 0.47 208 82 Bb-6h 232 0.61 1130 0.64 98 111 Bb-24h 378 0.41 1578 0.50 225 61 Mp-6h 606 0.20 2029 0.30 410 484 Mp-24h 231 0.67 1136 0.69 187 99 Wo-A-6h 234 0.45 1068 0.51 131 76 Wo-A-24h 170 0.76 896 0.76 62 22
Wo-U-6h 156 0.84 800 0.84 99 35 Wo-U-24h 191 0.92 936 0.94 29 18
p(ANOVAc)<0.01 p(ANOVAc)<0.05 p(ANOVA)<0.05, FCd>
genes max. FDRa genes max.
FDRa twofold threefold
all 1315 0.15 3165 0.30 3123 2514 a the highest expected false discovery rate (FDR); i.e. the maximal q-value observed for the given p(t-test) cut-point. b fold-change between treatment and control. c Note that expression ratios, log2(treatment/control), were used for analysis of variance (ANOVA) and therefore, d fold-change is between treatments and not between treatment/control; i.e. (treatmentx/controlx) / (treatmenty/controly).
59
Supplemental Figure 1.2: Genes and Transcription Factors Affected by Insect and Wounding Treatments- Gene profiles varied greatly across treatments with very little overlap in shared genes (seen in Figure 2). (A) Gene profiles are mostly congruent to those of transcription factors (B) in corresponding treatments, with the exception of Myzus persicae, which has very few up-regulated transcription factors despite many up-regulated genes. Negative numbers indicate down-regulated genes. Supplemental Table 1.2: Primers used in this experiment
A B
60
Supplemental Table 1.3: Putative Motifs in Insect-Regulated Genes Using Motif Sampler Algorithm Spod UP Pieris Up TTSCTT TTSCTT Brevi UP Myzus Up Wounding UP TTACTT TTACTT TKGRCC TTCAGK CGWSSS TTTATT TTTATT GCNRCS GAWYCG TARTTA TTSAGT TTSAGT TTGACT GTGRCK TTGACT TTCAGK TTCAGK GACTSA TGGNCC TTAATT TTCCTT SGCNGS SAGTSA
Spod Down Pieris Down
GATAAG TCWCTG Brevi Down Myzus Down
Wounding Down
CGRWTC SSTGRC GANYCG SCSRCG NCTGMC GCNGCK TTCCTT GAMTTG SGWCSG GGMCCA SCNGCS GAWTCG GAYWCG SGYGSS WCSGGT KCNAGC CGTTGA AGTGAS CGKTNC SSAGCT SRCCSG NCAGCC SCASCT Nucleotide Substitution Chart G G A A Guanine T T Adenine C C Thymine R R|G|A Cytosine Y Y|T|C Purine M M|A|C Pyrimidine K K|G|T Amino S S|G|C Ketone
W W|A|T Strong Interaction
H H|A|C|T Weak Interaction
B B|G|T|C Not-G, H V V|G|C|A Not-A, B D D|G|A|T Not-T, V N N|A|G|C|T Not-C, D
61
Supplemental Table 1.4: Transcription factors with known repressor activity that are down regulated by Myzus after 6 hours Family AGI Gene Name Reference
AUX/IAA At1g04240 IAA3 Tian et al. 2002 AUX/IAA At3g23050 IAA7 Nakamoto et al. 2005
AP2/ERF At5g67180 AINTEGUMENTA Krizek et al. 2000
bHLH At2g43010 PIF4 Khanna et al. 2005 C2C2-CO-like At3g07650 COL9 Cheng & Wang 2005 C2C2-YABBY At4g00180 YAB3 Kumaran et al. 2002
C2H2 At1g27730 ZAT10 Mittler et al. 2006 MYB At2g46410 CAPRICE Schellmann et al. 2002
JAZ/ZIM At1g70700 JAZ9 Chini et al.2007; Thines et al. 2007
JAZ/ZIM At3g17860 JAZ3 Chini et al.2007; Thines et al. 2007
62
Chapter 2:
Measuring “Normalcy” in Plant Gene Expression After Herbivore Attack
63
Abstract
Plants make drastic changes to their transcriptome to appropriately respond to
environmental change, and the regulation of genes that are specific to abiotic and biotic
stresses is key to plant survival. The coordination of defense gene transcription is often
coupled with significant adjustments in the levels of expression of primary metabolic and
structural genes to relocate resources, repair damage, and/or induce senescence. This
complicates the process of finding suitable “housekeeping” or reference genes to use in
measurements of gene expression by real time Reverse Transcription (RT)-PCR in
response to herbivore attack. Several software programs have been developed to identify
candidate reference genes, but measurement of their expression may still not yield an
appropriate gene or suite of genes for normalization. This is especially true in plant-
herbivore interactions where tissue damage is immediate and continuous. Here we show
that 12 traditional reference genes customarily used in RT-PCR analysis are not stably
expressed after insect attack. We describe the pitfalls for using traditional reference
genes and why insect attack may be affecting whole cell metabolism. We propose a
method using RNA quantification in combination with an external spike of commercially
available mRNA as normalization factors for Arabidopsis studies involving herbivory or
multiple stress treatments.
64
Introduction
Plants exhibit dramatic transcriptional reprogramming in response to environmental
stimuli. For example, response to and recovery from insect attack requires the co-
ordination of multiple gene pathways including those involved in secondary defense
chemistry, water status, photosynthesis, cellular signaling, and re-growth (Mewis et al.
2006, Aldea et al. 2005; Tang et al. 2006; Zangerl et al. 2002). Transcriptional analyses
using microarrays have shown that Arabidopsis plants have a broad spectrum of genes
that are significantly induced or repressed after herbivore attack (Ehlting et al. 2008;
DeVos et al. 2005; Reymond et al. 2004; Moran & Thompson 2001). The subsequent
confirmation of gene expression in microarrays is usually done with RT-PCR using one
or more reference genes as internal standards or “normalization” factors. These help
correct for differences in starting nucleic acid concentrations and variations in RT-PCR
efficiencies. However, methods for standardizing amounts of nucleic acids going into
RT-PCR reactions are often lacking and evidence for the stability of these reference
genes after insect attack are often absent.
When measuring gene expression using relative RT-PCR, there are three major factors
other than the effect of a treatment that can influence the magnitude of gene expression in
samples: the starting RNA quality, the starting RNA quantity, and the efficiency of the
reverse transcriptase reaction (Udvardi et al. 2008). Therefore it is critical to ensure that
the total amounts of RNA and cDNA in reactions are equal across treatments so that
65
differences in gene expression are not simply a result of more or less RNA or cDNA in
the reactions. Methods to accurately measure starting RNA quality and quantity before
conducting the reverse transcription reaction step are fundamental to quantifying gene
expression by RT-PCR.
The quality of RNA can be estimated by the absorbance ratio of the sample and its
migration as a single band in an agarose gel. The quantity of RNA can be determined
fairly accurately by RNA or cDNA Bioanalyzers and Spectrophotometers like a
NanoDrop (ThermoScientific, Wilmington, DE) (Thellin et al. 1999; Tricarico et al.
2002) which also provide absorbance ratios. It is assumed that all samples that have a
consistent amount of total RNA also contain consistent ratios of mRNA, rRNA, and
tRNA (Johnson et al. 2005; Bustin 2002).
RT-PCR also requires robust control for differences in cDNA that may arise from highly
variable reverse transcription reactions (Stahlberg et al 2004; Mannhalter et al. 2000),
which is usually provided by dividing by the expression of an internal reference gene.
Because this is sample-specific, the same cDNA from one sample will be amplified by
multiple primers, including reference gene primers, thus if there is low enzyme efficiency
for the reverse-transcription reaction, this will be reflected in the low expression of both
the reference and target genes. However, this method relies on the assumption that the
reference gene being amplified is unaffected by treatments.
66
An ideal reference gene should have consistent and unchanging expression levels across
all treatments, tissue types, and developmental stages (Udvardi et al. 2008; Wong &
Medrano 2005). Typical reference genes used in Arabidopsis research include ribosomal
RNA (18S), genes involved in cell structure (Actin, Tubulin, Clathrin, or Elongation
Factor) primary metabolism (Glyceraldehyde 3-Phosphate Dehydrogenase), and
ubiquination (Ubiquitin, Ubiquitin Binding Proteins). Because these serve maintenance
or infrastructural functions in cells, they have been assumed to be stably expressed during
stress (Czechowski et al. 2005). However, new evidence suggests that the transcription
of many commonly used reference genes can fluctuate even if no treatment is
administered (Gutierrez et al. 2008).
Software programs including “BestKeeper” (Pfaffl et al. 2004) and geNorm
(Vandesomple et al. 2002) are available to facilitate the selection of multiple reference
genes as internal standards. geNorm has been used to facilitate reference gene selection in
Arabidopsis (Czechowski et al. 2005) and poplar (Gutierrez et al. 2008). More recently,
suitable reference genes for various pathologies, such as in plant cells after
Pectobacterium atrosepticum infection (Takle et al. 2007), mammalian cells with breast
cancer (McNeill et al. 2007), and rat retinal cells after hypoxia (van Wijngaarden 2007)
have been adequately vetted and published to help researchers avoid using unstable
genes. However, these tools may be unavailable for some treatments and species
conditions. In those cases, a simple ANOVA of Cycle Threshold (Ct) or “crossing point”
values combined with post-hoc generalized linear models can be applied to determine
variation in treatments and controls (Brunner et al. 2004). Unfortunately, each of these
67
methods requires expression data from many different reference genes, which can be
problematic if RNA amounts are limited or the costs of reagents like SYBR green are
prohibitive. Furthermore analyses of expression values after ANOVA may still fail to
identify stably expressed reference genes.
If invariant reference genes are identified, their stability must be verified if any changes
are made in subsequent experimental designs (Guenin et al. 2009). For example, a
majority of the studies conducted on gene expression in plants after herbivory have been
done in largely controlled laboratory or greenhouse settings. Schmidt and Baldwin (2006)
found that plants grown in a greenhouse environment had more stable gene expression
than did plants grown in field conditions after methyl jasmonate (MeJA) treatment. The
increased background “noise” in gene expression may have been caused by exposure to
multiple stresses and potentially several treatments imposed by field conditions. The
impact of multiple elicitors or unexpected or unknown transcriptional changes caused by
elicitors further complicate the selection of suitable reference genes when studying the
molecular biology of plant-insect interactions. Given the complex networks and
metabolic pathways that are affected by insect attack, it is not clear that established
reference genes are suitable for responses to new or novel stimuli.
We tested the stability of 12 commonly-used reference genes in tissue samples taken
from Arabidopsis plants 6 hours after treatment with the lepidopteran herbivores, Pieris
rapae L. or Spodoptera exigua (Hubner), in a laboratory setting. We show that expression
of all putative reference genes in our study varied in expression across treatments and that
68
alternative methods are needed to normalize gene expression after insect herbivory. We
propose using RNA quantification in combination with an external spike of commercially
available mRNA as normalization factors for Arabidopsis studies involving herbivory or
multiple stress treatments.
Material and Methods
Plant Care and Insect Treatments
Arabidopsis Col-0 wild-type seeds were germinated on sterile Metromix 200 with
Osmocote (Scott’s, Maryville, OH) in 2.5” pots and kept in Growth Chambers under
short day conditions (8:16, L:D; 200 µEinsteins illumination) to delay bolting and
prolong rosette stage at 22°C and 62% RH until they were used for experimentation.
Feeding assays were conducted when plants were approximately 1.5” in diameter and 4-5
weeks old. A. thaliana Col plants were fed upon by the larvae of two leaf chewing
caterpillar species; Spodoptera exigua (Hubner) a dietary generalist, and Pieris rapae
(L.), a dietary specialist of the family Brassicaceae, of which Arabidopsis is a member.
Caterpillars were restricted on middle-rosette leaves until 30% of the leaf area was
consumed. Time of feeding initiation was noted and transcription was examined 6 and
24 hours after feeding started. Local tissue included leaves on which the insects fed, and
systemic tissue leaves unattacked. Control plants received no insect treatments. Both
insect and control insect treatments and sample collections were scheduled to avoid
69
upsetting the plants’ circadian cycles. Four bioreplications were used for both control
and treatment plants at each time point and tissue type.
RNA Isolation and Tissue Preparation
Total RNA was isolated from individual plants using a modified TRIZOL extraction
method as follows as described in Chapter 1. Briefly, tissue was flash frozen and ground
in TRIZOL, vortexed and incubated. Samples were then centrifuged and subjected to a
chloroform extraction. After another centrifugation, the aqueous phase was recovered
and RNA was precipitated with sodium citrate and isopropanol. We re-centrifuged the
samples, washed the RNA in ethanol, and re-suspended it in RNase free water.
Reverse Transcription-Real Time PCR (RT-PCR)
Final RNA concentrations were measured using a Nanodrop Spectrophotometer
(Wilmington, DE) and diluted to 1 µg/uL and quality was confirmed via gel
electrophoresis. To eliminate genomic DNA contamination, samples were treated with
Turbo DNAse according to the manufacturer’s specifications (Ambion, Austin, TX). We
re-measured RNA quantity after DNAse treatment to ensure that the exact same of
amount of RNA (1.2 µg) was added to each RT reaction. We used the Omniscript
Reverse Transcriptase kit and protocol for RT reactions (Qiagen, Valencia, CA).
DNAse-treated RNA was synthesized into first strand cDNA using a mix of random and
70
oligo-dT primers. To produce enough cDNA for the subsequent qPCR reactions, 8 RT
reactions per bioreplicate were done in 20 µL reactions then pooled.
Primers were designed using Primer 3 Software (Rozen and Skaletsky, 2000) and further
analyzed for primer dimers using Invitrogen’s Vector NTI Software (Carlsbad, CA). We
conducted BLAST searches of all primers using the NCBI GENBANK database to
ensure specificity of amplification. Gel electrophoresis of PCR products detected single
bands of expected size. Additionally, melting curve analysis of all PCR products was
done via real-time PCR. All PCR products were sequenced to ensure that only gene
products of interest were amplified. Table 2.1 lists all primers used in this experiment.
Table 2.1: Reference Gene Primers used in this experiment – Melting curve analysis, gel electrophoresis and sequencing was conducted on all primer products to ensure that only the gene of interest was being amplified. Primers were designed using Primer 3 and IDT’s Oligo Analyzer Software.
For real time qPCR standard curves, a pool of cDNA from each of 4 bioreplicates was
serially diluted to match fold changes encompassed within the microarray data. All PCR
reactions were run in 96-well plates. Each bioreplicate was run in triplicate on the plate
and analyzed for technical variation. For PCR reactions, we used 5 µL of cDNA
71
template, 5 µM primer pair mixes, water, and Platinum SYBR Green qPCR Super-Mix
UDG (Invitrogen, Calsbad, CA) for a total of 20 µL. Amplification was conducted on a
MJ Research Opticon 2 DNA Engine (Hercules, CA) under the following conditions
50°C UDG treatment for 2 minutes, 95°C denaturation for 2 minutes, followed by 40
cycles of 95°C denaturation for 15 seconds, 56°C annealing for 30 seconds and 72°C
extension for 30 seconds. After extension, but prior to fluorescence measurement reads,
the temperature was ramped to approximately 1.5-2.0°C below the gene product melting
curve start (Tm, –dl/dT min) to melt off primer dimers and non-specific random
amplicons. This ensured that only the specific gene product was contributing to SYBR
Green activity. A final 5 minute extension at 72°C followed by a complete melting curve
analysis from 72°C to 95°C were then conducted.
Luciferase Spike Analysis
We ordered Luciferase positive control RNA (Cat L4561) from Promega (Madison, WI).
The 1800 base-pair nucleotide RNA is derived from in vitro transcription from the
pSP64POLY(A)-luc plasmid luciferase construct (Robert Deyes, Promega, personal
communication). From the sequence of this construct, we designed and tested PCR
primers as previously described. The sequence of the Luciferase RNA and its
corresponding primers are not homologous to any gene sequence in Arabidopsis, making
this an excellent candidate as an exogenous Reverse Transcription control spike. To test
the dose-responsiveness of the foreign LUC spike, we added 200 pg, 100 pg, 50 pg, and
10 pg of LUC mRNA to RNA samples. Based on these results, we chose a 50 pg spike
72
per reaction. To assess the efficacy of LUC mRNA as an exogenouse spike, a 50 pg spike
per reaction was added to RNA extracted from insect-treated Arabidopsis leaves
immediately before reverse transcription. Reverse transcription was done using
Invitrogen’s Superscript III Reverse Transcription Kit which uses a mix of random and
oligodT primers. To minimize error due to repetitive pipetting, a larger volume aliquot of
LUC mRNA was added directly to the RT mastermix. Controls with no RT enzyme were
done to confirm that no PCR product could be amplified from a template prior to reverse
transcription. A melting curve analysis using real-time PCR and subsequent sequencing
verified that the LUC primers produced a single product homologous to the LUC
construct. Primers for PR3 (Basic chitinase, At3g12500, Forward 5’-
GGGGCTACTGTTTCAAGCAA-3’, Reverse 5’- GCAACAAGGTCAGGGTTGTT-3’)
and AtMYC2/JIN1 (Jasmonate Insensitive 1, Forward 5’-
TGTCGTCTTCGTGTTCTTCG-3’, Reverse 5’- ACCGTCGCTTGTTGAATCAT-3’)
were used to amplify defense-related genes in a cDNA pool also containing LUC cDNA.
To assess the efficacy of exogenous LUC mRNA as a normalization factor, we found that
the presence of LUC mRNA did not affect the expression of other genes and that LUC
levels could be used as the denominator in relative gene expression ratios.
Data Analysis and Assessment of Gene Stability
RT-qPCR data were acquired using the standard curve method (Larionov et al., 2005).
All data were initially analyzed using Opticon 3 Monitor Software (BioRad Industries,
Hercules, CA) and imported into a customized Excel spreadsheet (Microsoft Corp.
73
Redmond, WA) (Irmgard Siedl-Adams, data unpublished). A customized algorithm was
used to identify cycle threshold values (Cts) of fluorescence within the exponential phase
of the PCR curve similar to that described in Chapter 1. Unit-less expression values were
then calculated automatically from the Ct values based on the regression equation of the
standard curve. Statistically significant differences in expression levels (Ct values)
between treatments and controls were identified using a Generalized Liner Model (GLM)
and post-hoc Tukey tests (SAS Institute, Cary, NC). We used the program geNorm
(Vandesomple et al. 2002) to determine the most stable reference genes based on the Ct
values of both treatment and control treatments.
Results
To make sure that RNA quantity was not a factor contributing to differences in Ct values
in later PCR reactions, we obtained uniform amounts of RNA for all reverse transcription
reactions using a NanoDropTM spectrophotometer (ThermoScientific, Wilmington, DE).
We then measured the stability of expression of 12 Arabidopsis reference genes after
insect herbivory by two different species of caterpillars using RT-PCR. RT-PCR is robust
only if the researcher can be sure that the reference genes of interest do not change their
transcriptional pattern in response to treatment. An ANOVA and post-hoc Tukey test on
the Ct values of the 12 reference genes found expression of all genes to differ
significantly among treatments (see Table 2.2). This indicates that insect herbivory
affects the transcription of primary metabolic and structural genes commonly used by
74
investigators, and a different method must be used to account for variation in reverse
transcription reactions.
We used the program geNorm to calculate a stability factor (M) for our 12 genes.
Although no “ideal” M value is defined (Vandesomple et al. 2002), genes with the lowest
M value are considered the most stable across treatment. As shown in Figure 2.1, a
SAND family gene and a CLATHRIN-related gene had the lowest M values. This is
consistent with results from Czechowski et al. (2005), who found these to be the best
reference genes in Arabidopsis. However, these genes were not stable across our insect
treatments, especially in systemic tissue after S.exigua feeding (Table 2.2).
To test the functionality of a foreign RNA spike, we used Luciferase control mRNA with
primers specific to its sequence. As seen in Figure 2.2, RNA spiked with 10pg, 50pg,
100pg, and 200pg LUC mRNA shows a satisfactory dose response curve after reverse
transcription using real time PCR. Primers specific to Luciferase produced amplicons
with clean, sigmoidal amplification patterns and a melting curve with one, distinct
product peak at 81.2°C (Figure 2.2). The addition of the LUC spike did not affect the
amplification of other genes in the cDNA pool and was used as a normalization factor
instead of the expression level of an internal reference gene. As seen in Table 2.3,
correcting for variations in RNA quantity as well as reverse transcription efficiency
among samples can help elucidate differences in significance levels across treatments.
Significant differences between treatments and controls can change depending on the
normalization strategy used. For example, after 24 hours after P. rapae feeding, both
75
CLATH and PR3 levels are not different from controls if either no normalization
(CLATH Ct, PR3 Ct) or only normalization against RNA quantity (CLATH RNA, PR3
RNA) is performed. However, if we normalize the expression levels against both RNA
quantity and variations caused by the reverse transcription reaction (CLATH RNA, LUC;
PR3 RNA, LUC), the gene expression values between treatments and controls are
statistically significantly different.
Table 2.2: Results of post hoc Tukey tests after ANOVA of Reference Gene Cts for 6hr Herbivory Treatments by 2 caterpillars in local and systemic tissue- Statistics were conducted in SAS 9.1 on Ct values generated from the Standard Curve Method of Real Time PCR analysis using a customized algorithm to determine the fluorescence threshold in the exponential phase (Irmgard Seidl-Adams, personal communication). P value > 0.05. Labels: Optimum Reference Gene (Example)), Ubiquitin 10 (UBQ10), ribosomal RNA (18S), Elongation Factor alpha-1 (EFa-1), Beta Tubulin 2 (TUB2), Unknown Expressed Protein (UNXP), Glucose 6 Phosophate Dehydrogenase 5 (G6PD5), Actin 8 (ACT8), Homeodomain Glabrous 2 (HDG2), Ubiquitin Binding Protein 2 (UBP2), Actin 7 (ACT7), SAND Protein (SAND), Clathrin-Related Protein (CLATH)
76
Table 2.3: Results of post hoc Tukey tests after ANOVA of Expression Levels of Reference and Stress-Related Genes- Levels are generated from P. rapae and S. exigua feeding in Arabidopsis after 6 or 24 hrs in local tissue. Letters represent Tukey values after ANOVA (p<0.05). Letters indicate the statistical status of expression values before normalization for RNA quantity (Ct), expression values after normalization for RNA quantity, and expression values after normalization for both RNA quantity and efficiency of the RT reaction using LUC expression. Grey areas highlight expression values that were statistically different between controls and treatments.
77
Figure 2.1: Reference Gene Stability after Insect Herbivory- Ct values from RT-PCR data were entered into geNorm and M values were calculated for each reference gene. Genes to the right are those that are most stably expressed in Arabidopsis after insect attack. Abbreviations are listed in Table 2 caption.
78
Figure 2.2: Luciferase Exogenous mRNA Spike- A. Dose Response Curve-Arabidopsis RNA spiked with varying amounts of Luciferase mRNA was reverse transcribed and used as PCR template. Real-Time PCR was done using SYBR Green on a MJ Research DNA Engine 2. B. Melting Curve Analysis of reverse transcribed Luciferase RNA and primers shows a single peak with an optimal melting temperature of 81.2°C. Data was analyzed using Opticon Monitor 3.1 Software. Red line=10pg spike, green line=50pg spike, yellow line= 100pg spike, and blue line=200pg spike.
79
Discussion
The goal of this study was to identify a suitable normalization method for RT-PCR in
studies of plant gene expression in response to insect herbivores. We first tested the
stability of 12 commonly used reference genes in tissue samples taken from Arabidopsis
plants 6 hours after treatment with the lepidopteran herbivores, Pieris rapae or
Spodoptera exigua, in a laboratory setting. We showed that all of the reference genes in
our study vary in expression among samples and that alternative methods are needed to
normalize gene expression after insect herbivory.
Using reverse transcription real time PCR, we found that the Ct values of twelve
reference genes in Arabidopsis were statistically significantly different after insect
herbivory treatment (Table 2.2) and that correcting for differences in RNA quantity and
reverse transcription can reveal underlying differences in gene expression (Table 2.3).
Although a few studies have used reference genes for normalization after herbivory
(DeVos et al. 2006, Mewis et al. 2005, 2006; Chen et al. 2003), the severity of herbivory
damage, timing after feeding initiation, tissue analyzed, and plant growth conditions
could lead to differences in gene expression between studies. We performed over 1000
PCR reactions using numerous primer sets and technical replicates, yet did not find a
single, stable, reference gene. Because each PCR reaction requires 10 uL of SYBR
green, the prohibitive cost of this type of experiment warrants investigation of simpler,
less-expensive means of normalizing data for RT-PCR analysis.
80
If microarray data for a particular system and treatment are available, genes that are not
different from controls in the arrays serve as good initial candidates for RT-PCR controls
(Maccoux et al. 2007; Lee et al. 2007). We tested the expression of Glucose 6-Phosphate
Dehydrogenase 5 as a potential reference gene because of its ANOVA p value of 1.00 in
our microarray data (unpublished data). Nonetheless, upon analyzing the transcript levels
of G6PD5 using the more-sensitive RT-PCR, we found that its expression does change
after insect attack. We also used geNORM to assess the stability of our 12 reference
genes and found that two of the genes recommended by Czechowski et al. (2005) were
the most stable (a SAND family gene, and a CLATHRIN-related gene) (Figure 1.1).
However, when we conducted an ANOVA and post hoc Tukey test on the Ct values of
these genes across controls and insect treatments, they were significantly different (Table
2.2). These results suggest that the selection of reference genes should be treatment-
dependent and supports the suggestions of Udvardi et al. (2008), that the expression and
stability of reference genes must be statistically assessed with each experiment.
We proposed a method using RNA quantification in combination with an external spike
of commercially available mRNA as normalization factors for Arabidopsis studies
involving herbivory or multiple stress treatments. The results of our LUC spike analysis
show that a foreign LUC spike added to an RT master mix can serve as an excellent
method to control for the efficiency of reverse transcription reactions.
An interesting question that emerges from this research is “what could be causing the
changes in reference gene expression after herbivory?” There are several explanations
81
for this. First, these genes have additional stress-related functions. Seocond, they are part
of the regulatory intersection of stress- and primary metabolic signaling pathways.
Finally, severe stress caused by insect feeding drastically alters the overall physiology
and transcriptome of the plant.
It has been demonstrated that reference genes fluctuated greatly after treatments because
their secondary, stress-related function was uncharacterized. For example,
Glyceraldehyde 3-Phosphate, a gene that encodes a protein involved in glycolysis, was
shown to regulate age-related cell death in brain tissue (Ishitiana et al. 1996) and suppress
H2O2-induced apoptosis in both Arabidopsis protoplasts and yeast cells (Baek et al.
2008). Currently, it is unclear whether the reference gene stested here are also serving
alternative stress-related functions in Arabidopsis.
Differential expression across treatments may be related to the regulation of stress- and
primary metabolic signaling pathways. The carbon or nitrogen backbones for many
secondary compounds needed for defense responses come from primary metabolites and
proteins, shifting substrate availability for primary metabolic pathways. Schwachtje and
Baldwin (2008) make the argument that herbivory reconfigures primary metabolism
through four possible mechanisms: changes in resource allocation, shifts in physiology to
improve tolerance and fitness, utilization of primary metabolites as defense signals, and
alterations in metabolic machinery leading to inherent defense. For example,
glucosinolates, which are directly synthesized as anti-feeding and anti-microbial agents in
response to insect feeding and pathogen infection, are directly synthesized from the
82
amino acids methionine and tryptophan in Arabidopsis (reviewed by Grubb & Abel 2006;
Mikkelson et al. 2003).
Damage imposed by chewing insects is also a diverse and extreme physiological stressor
(Figure 3). For example, chewing insects remove a large portion of photosynthetic tissue,
triggering a reduction in photosynthesis activity (Hui et al. 2003; Zangerl et al. 2002;
Hermsmeier et al., 2001) and create altered sugar signaling throughout the plant (Oriens
et al. 2005; Rolland and Sheen 2005; Zhu-Salzman 2004; Arnold & Schultz 2002). After
Popilla japaonica and Helicoverpa zea feeding in soybean, Aldea et al. (2005) measured
up to a 90% increase in transpiration rates that can affect overall plant water status.
Feeding also releases vacuolar contents into the cytosol (Bown et al. 2006) potentially
inducing oxidative stress, changes in pH, calcium fluxes, electrical signals or senescence
(for reviews see Maffei et al. 2007ab). Both aphids and chewing caterpillars induce the
production of reactive oxygen species (ROS), which have been shown to accelerate leaf
senescence and programmed cell death in plants (Maffei et al. 2006; Bi & Felton 1995).
ROS function during the hypersensitive response by serving as second messengers during
cell signaling (for review see Apel and Hirt 2004 or Gechev & Hille 2005).
In addition to mechanical damage, elicitors in insect saliva may exacerbate the wound
response. Insect salivary components, such as fatty-acid-amino-acid conjugates (FACs),
like volicitin, have been shown to induce the synthesis of the plant stress hormones JA
and ET, SA (Schmelz et al. 2009, 2006, 2003; vonDahl et al. 2007; Alborn et al. 1997).
The role of ET and JA leaf and floral senescence is well established (Kao et al. 1983;
83
Parthier et al. 1990; Gan & Amisino 1997; He et al. 2002). Whether these hormones are
also triggering senescence signaling after herbivory is uncertain, but this model would
explain the down-regulation of primary metabolic genes in our study. Phloem-feeding
insects secrete elicitor directly into the wound site through their stylets (Kaloshian and
Walling 2005) and hasten plant defenses such as callose deposition near the feeding site
(Hao et al. 2008). Insect regurgitant also contains measurable amounts of RNases, which
are speculated to play a role in subsequent resistance against pathogens or viruses
(Gergerich et al., 1986; Musser et al. 2002). Little is known about their overall effect on
RNA quality in harvested plant tissue, which is critical if gene expression levels are to be
measured.
These explanations for the impacts on plants imposed by insect feeding are consistent
with large scale gene expression profiling. Microarray studies have shown that expression
of hundreds of genes, both metabolic and defense-related, are affected by insect feeding
in Arabidopsis (DeVos et al. 2005; Reymond et al. 2004; Moran et al. 2002). Vogel et al.
(2007) found that approximately 27% of genes affected by insect feeding in the
Arabidopsis relative Boechera divaricarpa could be functionally categorized as primary
metabolism, energy-related, or growth and development genes, while 23% were known to
be “stress-related”. Similarly, DeVos et al. (2005, Suppl. Materials) demonstrated that
several structural genes such as actin depolymerizing factors, an actin family member,
beta tubulins (TUB5, a-TUB6-like), and kinesins could also be significantly affected by
either caterpillar or aphid feeding.
84
Conclusion
Our results, in addition to the plethora of physiological effects of insect herbivory,
strongly support the hypothesis that insect herbivory affects primary metabolism and
endogenous reference gene expression, rendering them unsuitable as normalization
factors for RT-PCR in plant-herbivore interactions. We propose a method using RNA
quantification in combination with an external spike of commercially available mRNA as
normalization factors for Arabidopsis studies involving herbivory or multiple stress
treatments. This method requires that RNA quantity first be assessed using a highly
sensitive spectrophotometer, such as a Nanodrop which has an accuracy of 2%
(NanoDrop User Manual, Wilmington, DE). We suggest running RNA quantiation
samples in triplicate on the same day of the reverse transcription reaction to avoid any
RNA degradation caused by the freeze-thaw process. Then, an RNA correction factor
should be calculated for all samples to account for variations in the amount of RNA in a
treatment set. Once RNA has been quantified, a known concentration of commercially
available Luciferase mRNA is added to the RT-PCR master mix. Because this is
uniformly added to each sample, is single-stranded, and cannot be amplified by PCR until
it has been successfully converted to cDNA by reverse transcription, an exogenous
Luciferase spike adequately controls for variations in RT enzyme efficiencies across
samples. Using the gene-specific primers for Luciferase, real time PCR can be conducted
on Luciferase cDNA and serve as the normalization factor for subsequent gene
expression analysis. We found this method to be time- and cost-effective because it
eliminates the arduous, expensive, and potentially unsuccessful task of finding stable
85
reference genes.
Figure 2.3: Plant Cellular Trauma Caused by Herbivory
86
References
1. Alborn HT, Hansen TV, Jones TH, Bennett DC, Tumlinson JH, Schmelz EA, Teal
PEA (2007) Disulfooxy fatty acids from the American bird grasshopper Schistocerca americana, elicitors of plant volatiles. Proceedings of the National Academy of Sciences 104: 12976-12981
2. Aldea M, Hamilton JG, Resti JP, Zangerl AR, Berenbaum MR, DeLucia EH (2005)
Indirect effects of insect herbivory on leaf gas exchange in soybean. Plant, Cell & Environment 28: 402-411
3. Apel K, Hirt H (2004) Reactive oxygen species: metabolism, oxidative stress, and
signal transduction. Annual Review of Plant Biology 55: 373-399 4. Arnold T, Schultz J (2002) Induced sink strength as a prerequisite for induced tannin
biosynthesis in developing leaves of Populus. Oecologia 130: 585-593 5. Baek D, Jin Y, Jeong JC, Lee H-J, Moon H, Lee J, Shin D, Kang CH, Kim DH, Nam
J, Lee SY, Yun D-J (2008) Suppression of reactive oxygen species by glyceraldehyde-3-phosphate dehydrogenase. Phytochemistry 69: 333-338
6. Bi JL, Felton GW, J. (1995) Foliar oxidative stress and insect herbivory: primary
compounds, secondary metabolites, and reactive oxygen species as components of induced resistance. Ecology Letters 21: 1511-1530
7. Bown AW, Macgregor KB, Shelp BJ (2006) Gamma-aminobutyrate: defense against
invertebrate pests? Trends in Plant Science 11: 424-427 8. Brunner AM, Yakovlev IA, Strauss SH (2004) Validating internal controls for
quantitative plant gene expression studies. BMC Plant Biology 4: 14 9. Chen W, Provart NJ, Glazebrook J, Katagiri F, Chang H-S, Eulgem T, Mauch F,
Luan S, Zou G, Whitham SA, Budworth PR, Tao Y, Xie Z, Chen X, Lam S, Kreps JA, Harper JF, Si-Ammour A, Mauch-Mani B, Heinlein M, Kobayashi K, Hohn T, Dangl JL, Wang X, Zhu T (2002) Expression profile matrix of Arabidopsis transcription factor genes suggests their putative functions in response to environmental stresses. The Plant Cell 14: 559-574
10. Czechowski T, Stitt M, Altmann T, Udvardi MK, Scheible W-R (2005) Genome-
wide identification and testing of superior reference genes for transcript normalization in Arabidopsis. Plant Physiology 139: 5-17
87
11. DeVos M, Van Oosten VR, Van Poecke RMP, Van Pelt JA, Pozo MJ, Mueller MJ, Buchala AJ, Metraux J-P, Van Loon LC, Dicke M, Pieterse C, M. J. (2005) Signal dignature and transcriptome changes of Arabidopsis during pathogen and insect attack. Molecular Plant-Microbe Interactions 18: 923-937
12. Ehlting J, Chowrira S, Mattheus N, Aeschliman D, Arimura G-I, Bohlmann J (2008)
Comparative transcriptome analysis of Arabidopsis thaliana infested by diamond back moth (Plutella xylostella) larvae reveals signatures of stress response, secondary metabolism, and signalling. BMC genomics 9: 154
13. Gan S, Amasino RM (1997) Making sense of senescence: Molecular genetic
regulation and manipulation of leaf senescence. Plant Physiology 113: 313-319 14. Gechev TS, Hille J (2005) Hydrogen peroxide as a signal controlling plant
programmed cell death. Journal of Cell Biology 168: 17-20 15. Gergerich RC, Scott HA, Fulton J (1986) Evidence that ribonuclease in beetle
regurgitant determines the transmission of plant viruses. Journal of Gene Virology 67: 367-370
16. Grubb CD, Abel S (2006) Glucosinolate metabolism and its control. Trends in Plant
Science 11: 89-100 17. Guenin S, Mauriat M, Pelloux J, Van Wuytswinkel O, Bellini C, Gutierrez L (2009)
Normalization of qRT-PCR data: the necessity of adopting a systematic, experimental conditions-specific, validation of references. Journal of Experimental Botany 60: 487-493
18. Gutierrez L, Mauriat M, Guenin S, Pelloux J, Lefebvre J-F, Louvet R, Rusterucci C,
Moritz T, Guerineau F, Bellini C, Van Wuytswinkel O (2008) The lack of a systematic validation of reference genes: a serious pitfall undervalued in reverse transcription-polymerase chain reaction (RT-PCR) analysis in plants. Plant Biotechnology Journal 6: 609-618
19. Hao P, Liu C, Wang Y, Chen R, Tang M, Du B, Zhu L, He G (2008) Herbivore-
induced callose deposition on the sieve plates of rice: an important mechanism for host resistance. Plant Physiology 146: 1810-1820
20. He Y, Fukushige H, Hildebrand DF, Gan S (2002) Evidence supporting a role of
jasmonic acid in Arabidopsis leaf senescence. Plant Physiology 128: 876-884 21. Hermsmeier D, Schittko U, Baldwin IT (2001) Molecular interactions between the
specialist herbivore Manduca sexta (Lepidoptera, Sphingidae) and its natural host Nicotiana attenuata. I. Large-scale changes in the accumulation of growth- and defense-related plant mRNAs. Plant Physiology 125: 683-700
88
22. Hui D, Iqbal J, Lehmann K, Gase K, Saluz HP, Baldwin IT (2003) Molecular
interactions between the specialist herbivore Manduca sexta (lepidoptera, sphingidae) and its natural host Nicotiana attenuata: V. microarray analysis and further characterization of large-scale changes in herbivore-induced mRNAs. Plant Physiology 131: 1877-1893
23. Ishitani R, Sunaga K, Hirano A, Saunders P, Katsube N, Chuang DM (1996)
Evidence that glyceraldehyde-3-phosphate dehydrogenase is involved in age-induced apoptosis in mature cerebellar neurons in culture. Journal of Neurochemistry 66: 928-935
24. Kaloshian I, Walling LL (2005) Hemipterans as plant pathogens. Annual Review of
Phytopathology 43: 491-521 25. Kao CH, Yang SF (1983) Role of ethylene in the senescence of detached rice leaves.
Plant Physiology 73: 881-885 26. Larionov A, Krause A, Miller W (2005) A standard curve based method for relative
real time PCR data processing. BMC Bioinformatics 6: 62 27. Lee S, Jo M, Lee J, Koh SS, Kim S (2007) Identification of novel universal
housekeeping genes by statistical analysis of microarray data. Journal of Biochemistry and Molecular Biology 40: 226-231
28. Maccoux LJ, Clements DN, Salway F, Day PJR (2007) Identification of new
reference genes for the normalisation of canine osteoarthritic joint tissue transcripts from microarray data. BMC Molecular Biology 8: 62
29. Maffei ME, Mithofer A, Arimura G-I, Uchtenhagen H, Bossi S, Bertea CM,
Cucuzza LS, Novero M, Volpe V, Quadro S, Boland W (2006) Effects of feeding Spodoptera littoralis on lima bean leaves. III. Membrane depolarization and involvement of hydrogen peroxide. Plant Physiology 140: 1022-1035
30. Maffei ME, Mithofer A, Boland W (2007) Before gene expression: early events in
plant-insect interaction. Trends in Plant Science 12: 310-316 31. Maffei ME, Mithofer A, Boland W (2007) Insects feeding on plants: rapid signals
and responses preceding the induction of phytochemical release. Phytochemistry 68: 2946-2959
32. Mannhalter C, Koizar D, Mitterbauer G (2000) Evaluation of RNA isolation
methods and reference genes for RT-PCR analyses of rare target RNA. Clinical Chemistry 38: 171-177
89
33. McNeill RE, Miller N, Kerin MJ (2007) Evaluation and validation of candidate endogenous control genes for real-time quantitative PCR studies of breast cancer. BMC Molecular Biology 8: 107
34. Mewis I, Tokuhisa JG, Schultz JC, Appel HM, Ulrichs C, Gershenzon J (2006) Gene
expression and glucosinolate accumulation in Arabidopsis thaliana in response to generalist and specialist herbivores of different feeding guilds and the role of defense signaling pathways. Phytochemistry 67: 2450-2462
35. Mikkelsen MD, Naur P, Halkier BA (2004) Arabidopsis mutants in the C-S lyase of
glucosinolate biosynthesis establish a critical role for indole-3-acetaldoxime in auxin homeostasis. The Plant Journal : for Cell and Molecular Biology 37: 770-777
36. Moran PJ, Cheng Y, Cassell JL, Thompson GA (2002) Gene expression profiling of
Arabidopsis thaliana in compatible plant-aphid interactions. Archives of Insect Biochemistry and Physiology 51: 182-203
37. Moran PJ, Thompson GA (2001) Molecular responses to aphid feeding in
Arabidopsis in relation to plant defense pathways. Plant Physiology 125: 1074-1085 38. Musser RO, Hum-Musser SM, Slaten-Bickford SE, Felton GW, Gergerich RC
(2002) Evidence that ribonuclease activity present in beetle regurgitant is found to stimulate virus resistance in plants. Journal of Chemical Ecology 28: 1691-1696
39. Orians C (2005) Herbivores, vascular pathways, and systemic induction: facts and
artifacts. Journal of Chemical Ecology 31: 2231-2242 40. Parthier B (1990) Jasmonates: Hormonal regulators or stress factors in leaf
senescence? Journal of Plant Growth Regulation 9: 57-63 41. Pfaffl MW, Tichopad A, Prgomet C, Neuvians TP (2004) Determination of stable
housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper--Excel-based tool using pair-wise correlations. Biotechnology Letters 26: 509-515
42. Reymond P, Bodenhausen N, Van Poecke RMP, Krishnamurthy V, Dicke M,
Farmer EE (2004) A conserved transcript pattern in response to a specialist and a generalist herbivore. The Plant Cell 16: 3132-3147
43. Rolland F, Sheen J (2005) Sugar sensing and signalling networks in plants.
Biochemical Society transactions 33: 269-271 44. Rozen S, Skaletsky H (2000) Primer3 on the WWW for general users and for
biologist programmers. Methods in Molecular Biology 132: 365-386
90
45. Schmelz EA, Alborn HT, Banchio E, Tumlinson JH (2003) Quantitative relationships between induced jasmonic acid levels and volatile emission in Zea mays during Spodoptera exigua herbivory. Planta 216: 665-673
46. Schmelz EA, Carroll MJ, LeClere S, Phipps SM, Meredith J, Chourey PS, Alborn
HT, Teal PEA (2006) Fragments of ATP synthase mediate plant perception of insect attack. Proceedings of the National Academy of Sciences 103: 8894-8899
47. Schmelz EA, Engelberth J, Alborn HT, Tumlinson JH, Teal PEA (2009)
Phytohormone-based activity mapping of insect herbivore-produced elicitors. Proceedings of the National Academy of Sciences 106: 653-657
48. Schmidt DD, Baldwin IT (2006) Transcriptional responses of Solanum nigrum to
methyl jasmonate and competition: a glasshouse and field study. Functional Ecology 20: 500-508
49. Schwachtje J, Baldwin IT (2008) Why does herbivore attack reconfigure primary
metabolism? Plant Physiology 146: 845-851 50. Slansky F, Scriber JM (1985) Food consumption and utilization. In GA Kerkut, LI
Gilbert, eds, Comprehensive Insect Physiology, Biochemistry and Pharmacology, Vol 4: Regulation: Digestion Nutrition Excretion. Pergamon Press, N.Y., pp 87-163
51. Stahlberg A, Hakansson J, Xian X, Semb H, Kubista M (2004) Properties of the
reverse transcription reaction in mRNA quantification. Clinical Chemistry 50: 509-515
52. Takle G, Toth I, Brurberg M (2007) Evaluation of reference genes for real-time RT-
PCR expression studies in the plant pathogen Pectobacterium atrosepticum. BMC Plant Biology 7: 50
53. Tang B, Chen X, Liu Y, Tian H, Liu J, Hu J, Xu W, Zhang W (2008)
Characterization and expression patterns of a membrane-bound trehalase from Spodoptera exigua. BMC Molecular Biology 9: 51
54. Thellin O, Zorzi W, Lakaye B, De Borman B, Coumans B, Hennen G, Grisar T,
Igout A, Heinen E (1999) Housekeeping genes as internal standards: use and limits. Journal of Biotechnology 75: 291-295
55. Tricarico C, Pinzani P, Bianchi S, Paglierani M, Distante V, Pazzagli M, Bustin SA,
Orlando C (2002) Quantitative real-time reverse transcription polymerase chain reaction: normalization to rRNA or single housekeeping genes is inappropriate for human tissue biopsies. Analytical Biochemistry 309: 293-300
91
56. Udvardi MK, Czechowski T, Scheible W-R (2008) Eleven golden rules of quantitative RT-PCR. The Plant Cell 20: 1736-1737
57. van Wijngaarden P, Brereton HM, Coster DJ, Williams KA (2007) Stability of
housekeeping gene expression in the rat retina during exposure to cyclic hyperoxia. Molecular Visualization 13: 1508-1515
58. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A,
Speleman F (2002) Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biology 3: RESEARCH0034
59. Vogel H, Kroymann J, Mitchell-Olds T (2007) Different transcript patterns in
response to specialist and generalist herbivores in the wild Arabidopsis relative Boechera divaricarpa. PLoS ONE 2: e1081
60. von Dahl C, Baldwin I (2007) Deciphering the role of ethylene in plant–herbivore
interactions. Journal of Plant Growth Regulation 26: 201-209 61. Zangerl AR, Hamilton JG, Miller TJ, Crofts AR, Oxborough K, Berenbaum MR, de
Lucia EH (2002) Impact of folivory on photosynthesis is greater than the sum of its holes. Proceedings of the National Academy of Sciences 99: 1088-1091
62. Zhu-Salzman K, Salzman RA, Ahn J-E, Koiwa H (2004) Transcriptional regulation
of sorghum defense determinants against a phloem-feeding aphid. Plant Physiology 134: 420-431
92
Chapter 3:
Differences in ERF transcription factor expression, defense-related gene transcription,
and stress hormone release reveal diverse signaling pathways elicited after attack by two
different herbivores in Arabidopsis
93
Abstract
Plant defenses against insects require the coordination of molecular, biochemical, and
physiological events. Attacker-specific changes in gene expression and production of
secondary defense compounds have been observed, including some in which the plant
failed to identify or respond to particular pest species. These differential responses
involve the fine-tuning of, or crosstalk among, jasmonate (JA), salicylate (SA), and
ethylene (ET) hormone signaling pathways. Although response to insects requires JA,
SA, and ET interaction, the details of their interaction are still unclear, especially in
Arabidopsis. Some studies have suggested that insect-elicited ethylene signaling may
suppress plant defense responses, particularly against specialist attackers. We examined
the transcriptional changes in Arabidopsis thaliana after herbivory by dietary generalist
(Spodoptera exigua) and dietary specialist (Pieris rapae) herbivores after a 24-hour time
course. Defense responses to the two insects differed and were frequently weaker or
absent in response to the specialist P. rapae. Using RT-qPCR, we found members of the
AP2/ERF (Ethylene Response Factor) transcription factor family and AtMYC2 to be
differentially regulated in response to the two insect attackers. Measurements of
increased ethylene levels after herbivory indicated that ethylene was produced in
response to both insect species, although the amounts and timing of ethylene production
differed. Additionally, rapid and increased JA and JA-isoleucine production by
Arabidopsis plants after attack by both insects confirms a role for JA and its amino acid
conjugate in general herbivory responses. We present evidence in support of the
hypothesis that ethylene signaling is involved in the differential response to these two
94
insects and the resulting divergent signaling pathways entail the coordination of JA-
signaling events.
Introduction
Plants have to cope continuously with changing external environments, including insect
attack, in order to survive and reproduce. Plant responses to insects and other
environmental stresses are complex, involving differential perception, numerous signal
cascades, and the transcriptional regulation of many defense- and stress-related genes
(see review by van Poecke 2007). However, plant responses differ depending on the
insect attacker (DeVos et al. 2005; Mewis et al. 2005, 2006; Vogel et al. 2007). The
resulting differential defense responses to insects likely involve the carefully coordinated
and timed release of stress-related plant hormones, activation of transcription factors, and
defense gene expression.
The plant hormones jasmonic acid (JA), salicylic acid (SA), abscisic acid (ABA), and
ethylene (ET) are critical players in the events following abiotic and biotic stress,
including insect attack. Hormones modulate the fine-tuning of defenses in response to
different insects (Mewis et al. 2005; DeVos et al. 2005; Thompson and Goggin 2006;
Zhu-Salzman et al. 2004; Kessler and Baldwin 2002), yet the mechanisms of pathway
control and integration remain elusive. Some of the first signaling events following
perception of insect attack are the rapid accumulation and transport of JA and a quick
“burst” of ET production (Babst et al. 2005; Kessler and Baldwin 2002, Thaler et al.
2002; Winz and Baldwin 2001; Reymond and Farmer 1998). The release of ethylene after
95
wounding and herbivory depends on timing, the plant species being attacked, and the
herbivore species involved (for review see vonDahl & Baldwin 2007). vonDahl et al.
(2007) hypothesized that ethylene may not act alone, but may serve to mediate or fine-
tune responses based on its interaction or cross talk with other phytohormones like JA,
SA, and ABA. For example, treatment with exogenous ET stimulated the production of
polyphenols in fir trees (Hudgins & Franceschi 2004), cysteine proteinases (Harfouche et
al. 2006) and predator-attracting volatiles in corn (Schmelz et al. 2003), while ET
decreased JA-inducible nicotine production in tobacco (Kahl et al. 2000). Similarly,
larvae of the specialist Pieris rapae grew poorly on mutant plants with compromised
ethylene signaling, which was associated with an increase in JA-inducible indolyl
glucosinolates (Mewis et al. 2006). In contrast, Stotz et al. (2000) found that ethylene
signaling increased susceptibility to the generalist Spodoptera littoralis in Arabidopsis.
DeVos et al. (2006) found that ethylene production after P. rapae feeding primed plants
against future viral infection. So, both specialist and generalist insects have been shown
to induce ethylene production and/or respond to plants treated with it, but the nature of its
involvement and the implications for insect performance and plant response remain
unclear.
The plant hormone JA is a long-chain fatty acid derivative in the octadecanoid pathway
of the chloroplasts that is stimulated by wounding, necrotrophic fungi, and insect feeding
(Stotz et al. 2002; DeVos et al. 2005, 2006). Application of JA to Arabidopsis leaves
induces defense phenotypes such as the formation of trichomes (Traw and Bergelson,
2003) and secondary metabolites like glucosinolates (Cipollini et al. 2004) and terpenoids
96
(Faldt et al. 2003). Although most studies related to plant defense have been conducted
using exogenous JA or Methyl JA application, the JA-amino acid conjugate, JA-
Isoleucine, was found to be the biologically active form of this hormone (Chini et al.
2008; Thines et al. 2007) by interacting with the SCF-COI1 complex to trigger JA
responses, including the degradation of JAZ proteins and activation of AtMYC2. In
addition to exogenous JA treatments, genetic approaches using plants with mutations in
the JA signaling pathway, such as coi1, have shown that impaired JA signaling increases
susceptibility to insect attack (Thaler et al. 2002; Mewis et al. 2005). JA typically acts
synergistically with ET to induce defenses (Lorenzo et al. 2004) and antagonistically with
SA (for review see Beckers and Spoel 2006) and ABA (DeVos 2006).
After insect attack, there are additional downstream signaling events, including the
activation and transcription of signaling-related proteins, such as trans-acting elements or
transcription factors (Reymond et al. 2004; Major & Constabel 2006; DeVos et al. 2005).
Because transcription factors can serve as both components of early signaling events and
nodes of phytohormone cross-talk (Abe et al. 2003; Yadav et al. 2005; Li et al. 2004;
Lorenzo et al. 2003), they may mediate a plant’s specific responses to various external
stresses, including wounding, pathogens, and herbivorous insects. For example,
members of the WRKY and ETHYLENE RESPONSE FACTOR (ERF) families have
been shown to be critical regulators in Arabidopsis plants treated with JA, pathogens, or
in response to wounding (see review by Eulgem et al. 2005; Delassert et al. 2004;
Reymond et al. 2004; Chen et al. 2002; Schenk et al. 2000, 2003).
97
In a previous study, we found that Arabidopsis ERF transcription factors are important in
differential responses to two caterpillar species (Chapter 1). AP2/ERF Transcription
Factors comprise about 120 members (Nakano et al. 2006), are exclusive to plants, and
consist of ERF or B3 DNA binding domains and several sub-families including AP2 and
RAV (McGrath et al. 2005). Many ERFs are responsive to the hormones ET and JA
(Lorenzo and Solano 2005), although individual gene responses to either ET or JA can
differ. For example, expression of ERF1 is compromised in both coi1 and ein1 plants
(Berrocal-Lobo et al. 2004), and the expression of ERF1, as well as ER2F2, ERF3, and
ERF4 is rapidly induced by exogenous application of both hormones in WT plants
(Brown et al. 2003). Fujimoto et al. (2000) showed that ERFs directly activate the
transcription of defense-related genes such as PDF1.2, B-chitinase (PR3), and Hevein-
like protein (PR4) by binding to GCC-boxes in their promoters. Lorenzo et al. (2004)
found the JA-inducible gene AtMYC acts to repress the expression of these genes, while
activating other wound-responsive genes such as VSP2, Thi2.1 and LOX3. Several
members of the ERF family, including ERF11, ERF3, and ERF4 contain an EAR motif
that functions in negative regulation of ethylene-responsive genes via the GCC box
(McGrath et al. 2005; Brown et al. 2003; Fujimoto et al. 2000).
Transcriptional responses to wounding and JA treatment often include the up-regulation
of genes involved in plant defense such as PDF1.2, VSP2, LOX2, and chitinases (Boter et
al. 2004; Wasternack et al. 2005). Since ERFs operate at a signaling intersection between
ET- and JA-regulated defense genes, the measurement of hormone levels and the
expression of ERFs and defense genes over a controlled time course are needed to
98
determine their roles in plant responses to herbivory. In this study, we measured the
expression of several ERF transcription factors and down-stream defense related genes
over a 24-hour and 72-hour time course after feeding by the generalist herbivore S.
exigua and the specialist herbivore, P. rapae. We also quantified the release of ET and
levels of JA after insect attack and found that both insects elicit JA and ET, but the timing
and amount of their production differed.
Materials and Methods
Plant Growth Conditions
Col-0 WT were grown in 4” pots with Metro-Mix 200 soil and Osmocote (Scott’s,
Maryville, OH) in growth chambers at 22°C and 62% RH under short day conditions
(8:16, L:D; 130 µEinsteins illumination) to delay bolting and prolong the rosette stage
until they were used for experimentation at the 6-week stage. All plants were watered as
needed.
Insect Treatments
The two lepidopteron chewing insects used for this study were Pieris rapae L. and the
generalist herbivore, S. exigua Hubner. S. exigua eggs were obtained from Benzon
Research (Carlisle, PA). Larvae were reared on artificial diet (Carolina Biologicals,
Burlington, NC) and acclimated to Arabidopsis plants for 24 hours before the
99
experiments. P. rapae larvae were taken from a colony currently maintained in our insect
rearing facility (Bond Life Sciences Center, University of Missouri, Columbia, MO), fed
a mix of Pak-Choi and Arabidopsis plants throughout their larval stages, and then pre-fed
on Arabidopsis plants for 24 hours before the experiment. Insects then were removed
from pre-feeding plants for a maximum of 1-2 hours before the feeding assay. Post-
ecdysial third or fourth instar caterpillars were placed in individual leaf cages put on four
mid-sized leaves and allowed to feed until 20-30% of the leaf was removed, a treatment
usually lasting 10-30 minutes. Cages without insects were put on paired control plants
and removed at the same time as the treatment plants.
Plant Tissue Harvesting
Tissue collection during the insect protocol was coordinated to ensure that insect treated
plants and their time-matched cage controls were sampled at 15 minutes, 30 minutes, 1
hour, 2 hours, 6 hours, and 24 hours after insect treatment. Experiments for S. exigua and
P. rapae treated plants were conducted on Feb. 4 and March 10, 2008, respectively. Four
plants were used for each bioreplicate and four bioreplicates were collected per treatment.
We harvested the four treated leaves on each plant for different assays: two for RNA
tissue, one for JA measurement, and one for an initial ethylene analysis. RNA and JA
sample leaves were weighed, flash frozen in grinding tubes immersed in liquid nitrogen,
and then stored at -80°C. Ethylene samples were processed immediately.
100
Ethylene Measurements
For ethylene analysis, additional experiments were conducted using similar methods as
described above with some exceptions. First, these experiments were done with P. rapae
and S. exigua on the same day (November 10, 2008, repeated April 6, 2009). The time
course was also extended to include the original time points (15 min, 30min, 1hr, 2hr,
6hr, 24hr) as well as 12hr, 36hr, 48hr and 72hr treatments. Four leaves from 1 plant (1
bioreplicate) from either insect-attacked or control plants were placed in sealed 10cc
glass vials and allowed to incubate for 30-90 minutes. Four bioreplicates were taken for
each treatment. Air was then drawn off using a 5cc syringe and manually injected into an
HP Gas Chromatograph. ET levels were calculated using a regression equation of a
standard curve and corrected for fresh weight and incubation time.
JA and JA Conjugate Measurements
JA and JA conjugate (JA-amino acid) levels were quantified using an ethyl acetate
extraction method in conjunction with HPLC/MS similar to that described in Chung et al.
2008. Briefly, samples (approximately 150 mg tissue) were frozen in liquid nitrogen and
hormones were extracted using 1 mL of extraction solvent (80:20 methanol:water + 0.1%
formic acid) for 18 hours at -20 C. Samples were then centrifuged (10,000 x g for 10 min
at 4 degrees C) and the supernatant was transferred to autosampler vials. Five µL of
sample were injected into a UPLC BEH C18 column (1.7 mM, 2.1 3 50 mm) on a
Water’s (Milford, MA) Acquity ultraperformance liquid chromatography system at 50°C.
101
A 0.15% aqueous formic acid/ methanol gradient solvent flow in a 3-min program for a
mobile phase flow rate of 0.4 mL/min was used for separation. Samples were then
identified using a Water’s (Milford, MA) Quattro Premier XE tandem quadrupole mass
spectrometer using negative mode electrospray ionization.
Gene expression via Real Time RT-PCR
The expression of 12 ERFs (AtERF1, ERF4, ERF5, ERF5/6, ERF6, ERF8, ERF11,
ERF105, ORA59, DREBb, TINY2, SimRAP2.6), AtMYC2 (JIN1), and 6 defense-related
marker genes (PR3, PR4, LOX3, PDF1.2, Thi2.1, VSP2) was measured by semi-
quantitative Real-Time PCR. Total RNA from insect-attacked and control tissue samples
was extracted using Sigma Total Plant RNA kits (STRN50, St. Louis, MO). RNA quality
was then confirmed by the DNA Core facility at the University of Missouri using a Bio-
Rad Experion automated electrophoresis system (Hercules, CA) and a Bio-Rad RNA
standard sensitivity kit which adequately detects and quantifies nanogram levels of RNA.
Primers were designed and tested using methods described in Chapter 1. We used Primer
3 Software (Rozen and Skaletsky, 2000) and Invitrogen’s Vector NTI Software
(Carlsbad, CA) as well as IDT’s on-line tool, OligoAnalyzer for further prediction of
primer dimers. All primers were BLASTed in NCBI to ensure specificity of
amplification. We performed gel electrophoresis of PCR products and detected single
bands of expected size. Additionally, melting curve analysis of all PCR products was
done via real-time PCR. All PCR products were sequenced to ensure that only gene
102
products of interest were being amplified. Table 3.1 lists all primers used in this
experiment.
103
Table 3.1: ERF Family TFs and Defense-Related Gene Primers used in this Experiment
104
We treated samples with Turbo DNAse (Ambion, Austin, TX) according to the
manufacturer’s specifications. RNA quantity after DNAse treatment was measured using
a NanoDrop (ThermoScientific, Wilmington, DE) in triplicate for each sample
immediately before the reverse transcription reaction. This allowed us to obtain accurate
normalization factors for RNA quantity prior to the RT reaction. We followed the
protocol for Invitrogen’s Superscript III 2-step RT-PCR kit with Platinum SYBR Green
qPCR Super-Mix UDG (Calsbad, CA) with minor modifications. As described in
Chapter 2, a foreign Luciferase mRNA Spike (Promega, Madison Wisconsin) was added
to a reverse transcription master mix so that the final concentration of the spike was 50 pg
per sample. To acquire sufficient amounts of cDNA for all of the subsequent real time
PCR reactions, 4 reverse transcription reactions were performed for each RNA sample.
These were done in 96-well plates and the volumes of 4 technical replicates for each
sample were pooled, sub-sampled for a standard curve mix, and diluted 5X.
For real time qPCR standard curves, we followed the methods outlined in Larionov et al.,
(2005) and serially diluted a pool of cDNA aliquotted from each bioreplicate. All PCR
reactions were run in 96-well plates. Each bioreplicate was run in triplicate on the PCR
plates. For PCR reactions, we used 5 mL of cDNA template, 5 mM primer pair mixes,
molecular-grade water, and Platinum SYBR Green for a total of 20 mL. Amplification
was then conducted under the following conditions on a MJ Research Opticon 2 DNA
Engine: 50°C UDG treatment for 2 minutes, 95°C denaturation for 2 minutes, followed
by 40 cycles of 95°C denaturation for 15 seconds, 56°C annealing for 30 seconds and
105
72°C extension for 30 seconds. After extension, but prior to fluorescence measurement
reads, the temperature was ramped to approximately 1.5-2.0°C below the gene product
melting curve start (Tm, –dl/dT min). A final 5 minute extension at 72°C followed by a
complete melting curve analysis from 72°C to 95°C were then conducted.
Data Analysis
RT-qPCR data were acquired using the standard curve method (Larionov et al., 2005).
All data were initially analyzed using Opticon 3 Monitor Software. We used LinReg PCR
(Ramakers et al. 2003) to identify a value for the threshold of florescence. We entered
this value into the Opticon Software Program, which automatically calculated expression
values from the Ct values based on the regression equation of the standard curve. Final
expression values were then corrected for starting RNA amounts and normalized to the
expression levels of Luciferase in corresponding samples. Outliers for RT-PCR and
hormone measurements were identified using a one-pass Extreme Studentized Deviate
(ESD) test (Pillai & Tienzo 1956) and eliminated from the analyses. Statistically
significant differences in final gene expression ratios between treatments and controls for
both the P. rapae and the S.exigua experiment were identified using the PROC
NPAR1WAY command in SAS and Kruskal-Wallis analyses (SAS Institute, Cary, NC).
Gene expression data displayed in Figure 5 were transformed using the Log2 values of
fold changes. A Hierarchal cluster analysis was done using the Spearman Rank
Correlation feature in the software Cluster 3.0 (Eisen et al. 1998) and monitored with
Java TreeView 1.1.3 (Saldanha 2004). To identify differences in ethylene, JA, JA-
isoleucine, and SA levels among treatments, we conducted an ANOVA in SAS (Cary,
106
NC). Statistically significant differences between treatments at a p-value of 0.05 or lower
were determined using the PROC GLM command and post-hoc Tukey values.
Results
Insect Elicitation of Ethylene Release
To determine if insects induce ethylene production as a potential signaling mechanism in
Arabidopsis, we used gas chromatography to measure ethylene levels emitted by locally
attacked tissue at selected time points. Both P. rapae and S. exigua feeding induced the
production of ethylene, but the times when levels became significantly different from
controls differed. S. exigua induced significantly higher levels of ethylene than controls
in Arabidopsis tissue by 30 min, while ethylene production by P. rapae -attacked plants
did not differ from controls until after 2hrs (Figure 3.1). Moreover, P. rapae -induced
ethylene production remained higher than controls at later time points. Our results
indicate that ethylene production induced by S.exigua occurs as a rapid burst shortly after
the insects feed on the plants, while ethylene production after P.rapae feeding is delayed.
107
Insect Elicitation of Jasmonic Acid, Jasmonic Acid-‐Isoleucine, and Salicylic Acid JA-isoleucine (JA-IL), and SA levels were measured using HPLC-MS. We found no
significant increases in SA in response to insect treatments (Supplementary Materials,
Figure 3.2). However, levels of JA (Figure 3.2) and JA-isoleucine (Figure 3.3) were
rapidly increased in plants after insect feeding. The induction of JA was much greater by
S. exigua than by P. rapae. S. exigua elicited a statistically significant increase above
controls in JA at 0.5, 2, and 6 hrs after feeding. The higher mean at 1 hr was not
statistically significant, probably due to the larger standard deviation at this time point
(p=0.1104). In response to P. rapae, JA levels increased significantly immediately (after
Figure 3.1: Ethylene production in WT Arabidopsis plants after short-term Pieris rapae and S. exigua feeding over a 72hr time course- Ethylene was measured as nanoliters /gram fresh weight/ minutes of incubation time. Asterisks represent data points that are significantly different than controls.
108
15 minutes) and remained above controls until 24 hours after treatment, although its
levels never approached those induced by S. exigua. Patterns of JA-isoleucine production
after S. exigua and P. rapae feeding matched those of JA (Figure 3.3).
109
Figure 3.2: JA levels in WT Arabidopsis plants after S. exigua (A) or P. rapae (B) feeding over a 24hr time course- JA was measured as pmol/g fresh weight. Blue bars represent S. eixuga feeding; light gray bars represent controls in the S. exigua experiment. Red bars in (B) represent P. rapae treatment, dark gray bars represent controls in the P. rapae experiment. Asterisks represent data points that are significantly different than controls. Error bars are standard errors of the mean.
JA (p
mol
/g F
W) * * * * * * * *
A B
Figure 3.3: JA-isoleucine (JA-IL) levels in WT Arabidopsis plants after S. exigua (A) or P. rapae (B) feeding over a 24hr time course- JA-IL was measured as pmol/g fresh weight. Blue bars represent S. exigua feeding; light gray bars represent controls in the S. exigua experiment. Red bars in (B) represent P. rapae treatment, dark gray bars represent controls in the P. rapae experiment. Asterisks represent data points that are significantly different than controls. Error bars are standard errors of the mean.
JA-I
L (p
mol
/ g F
W)
A B
* * * * * * * *
110
ERF and Defense Gene Expression
Using RT-PCR, we measured differences in the expression of genes encoding ERF
transcription factors and defense-related genes in Arabidopsis in response to feeding by
caterpillars of the same two lepidopteran species. We monitored gene expression
patterns of ERFs and down-stream defense genes through time (Figure 3.4). In general,
the expression of both ERFs and defense-related genes was higher after feeding by S.
exigua, the dietary generalist, than by P. rapae, the dietary specialist. There were only
three instances, all occurring at 15 minutes, where transcriptional changes were higher in
P.rapae-attacked tissue, namely ORA59 15 ERF5, and AtERF1 (p value<0.08).
Plants exposed to S.exigua feeding showed dramatic transcriptional responses. Every
gene measured except AtMYC2, which is a JA-responsive gene, had a stronger response
to S.exigua than to P. rapae at a given time point. This was especially true for ERF104,
ERF8, PR4, PR3,and ERF11, whose expression rose in response to S.exigua, but declined
in response to P. rapae.
In our experiment with S. exigua, we observed several instances where gene expression
was increased in control plants, especially during the initial time points, suggesting that
thigmotropic stimuli while caging insects on the plant may have contributed to elevated
gene expression (Supplementary Materials, Figure 3.1). Similarly, we found that the
starting control levels for JA in the S. exigua bioassay were higher than in the P. rapae
assay (Figure 3.2). We conducted an experiment to determine whether cages put on
111
plants elicited similar patterns on the expression of ERF8, ERF11, PDF1.2 and Thi2.1 as
those found in our insect experiments. Our results suggest that touch may be a small
contributing factor as gene expression in untouched plants was less than in touched
plants, but the transcriptional response elicited by thigmotropic stimulation was not
enough to explain the large expression changes in the insect experiments (Supplemental
Materials, Figure 3.1).
Figure 3.4: RT-PCR of ERF Transcription Factors and defense-related genes- Arabidopsis leaf tissue was collected 15 min, 30min, 1 hr, 2 hr, 6 hr and 24 hr after herbivory by S. exigua or P. rapae caterpillars. Red bars indicate P. treatments and blue bars represent S. treatments. RT-PCR data were normalized to RNA quantity and the levels of an exogenous LUC spike. Y-axes represent fold changes of treatment/controls. Note differences in fold change scale between each graph. Control plants (cage only, no insects) were paired with treatment plants. Error bars represent the standard error of the means of the bioreplicates for each treatment and time point. Asterisks (p<0.05) and lower case t’s (p<0.08) indicate statistically significant differences between insect treatments as determined by Kruskal-Wallis Analyses.
ERF104
112
113
Figure 3.5: Heat Map of ERF Transcription Factor and Defense-Related Gene Expression Patterns over a 24-hour Time Course- Using RT-PCR, the expression of ERFs and defense genes was monitored after 15 min, 30min, 1 hr, 2 hrs, 6 hrs, and 24 hrs after feeding by Pieris rapae or Spodoptera exigua larva. Expression values were calculated using RNA quantity and LUC expression levels as normalization factors. Treatments were then referenced to the respective controls and converted to Log2 values. The heat map and cladogram was created using the software program Cluster3.0. Trees were visualized using Java TreeView. Green pixels indicate a decrease in expression relative to controls while red pixels are indicative of an increase in transcription. Black pixels represent no change.
114
Discussion
Our results indicate that plant responses to chewing insects can differ even when the
insects feed on the same part of the plant for the same amount of time. This may reflect
the dietary breadth of the insect. Differences in plant response to different insects are
likely to arise from elicitors in insect saliva that may either trigger different pathways to
elicit the rapid production of different patterns of hormones and defense genes or that
may suppress signaling to impede or stifle responses.
Signaling after herbivory is complex, involving the production and interplay of JA, ET,
and SA as well as the regulation of transcription factors and defense related genes (Zhu-
Salzman et al 2004; Reymond et al. 2004; Mewis et al. 2006; Vogel et al. 2007; see
review by Wu & Baldwin 2009). In this study we show that herbivory by two different
insects elicits increases in both ET and JA, but not SA. However, the timing of ET and
JA responses and the total concentrations induced by the insects were different. In most
cases, S. exigua elicited stronger, and often earlier, responses which may shape
downstream responses. This is highlighted by the differential expression of ERF
transcription factors and notable defense “marker” PR genes in response to feeding by
each insect.
Increased ethylene emissions after insect herbivory is well documented (for review see
von Dahl & Baldwin 2007). In our study, ethylene production in Arabidopsis plants after
S. exigua attack occurred as a rapid burst peaking at 1 hour and achieved levels that were
115
significantly greater than controls as early as 15 minutes after insect removal (Figure
3.1). ET levels continued to remain above control levels until 6 hours, after which they
attenuated. Conversely, P. rapae feeding did not induce ET levels that were different
from controls until after 2 hours, and they remained elevated throughout most of the time
course. Our results suggest that ET could serve as an important signal in defense
responses to the generalist insect, S. exigua as well as the specialist P. rapae, but that the
timing of peak ethylene production may be crucial to organizing down-stream events.
The variation in ET-related plant response to different herbivores has given rise to several
hypotheses about the role of ethylene in plant responses to insects that vary in dietary
specialization. In one hypothesis, it is argued that from an energy standpoint, plants
attacked by dietary specialists should minimize ET signaling to avoid induction of plant
defenses that are ineffective against insects adapted to that food source (Winz & Baldwin
2001; Voelckel et al. 2001; von Dahl et al. 2007). Several specialist insects, including P.
rapae, have evolved the ability to reduce the toxicity of defense chemicals made by their
host plants (Wittstock et al. 2004) and often use those same secondary metabolites as
feeding and oviposition stimulants (Smallegange et al. 2007; Clauss et al. 2006; Renwick
& Lopez 1999). In one example, Diezel et al. (2009) found that ET production in
Nicotiana attenuata after feeding by the specialist Manduca sexta significantly increased,
thereby suppressing defenses, whereas feeding by the generalist S. exigua did not.
In an alternative hypothesis, it is argued that insects that are dietary generalists are more
susceptible to secondary metabolite production (Hansen et al. 2008) and they may induce
116
ethylene as a general mechanism to suppress plant responses. In addition to our results,
there are others demonstrating an ability of dietary generalists to elicit ET. For example,
voliticin, a component of oral secretions (OS) isolated from a dietary generalist (Alborn
et al. 1997) induced both ET and JA production in soybean, eggplant, and maize
(Schmelz et al. 2009), demonstrating that elicitors from generalist insect oral secretions
can induce ET in a broad range of plant species. However, unlike our study, they failed to
find S. exigua induction of ET in Arabidopsis, finding ET and JA induction in
Arabidopsis only when treated with grasshopper caeliferins. Discrepancies among these
studies may be a result of using insect feeding vs. mechanical wounding to which OS or
individual elicitors were applied. This is an interpretation consistent with Schmelz et
al.’s (2003) earlier study, in which only feeding by S.exigua, but not the application of
volicitin, induced ET production in corn.
It is well known that JA is an important component in defense responses, especially
biotic or herbivore stress (Schmelz et al. 2003a; see review by Halitschke & Baldwin
2004). In fact, adding insect oral secretions to wounding sites in N. attenuata potentiates
the JA response (Halitschke et al. 2001). Levels of JA as well as the oxylipins OPDA and
dnOPDA gradually increased over a 24-hour time course after P. rapae feeding
(Reymond et al. 2004). DeVos et al. (2006) reported an increase in JA production after P.
rapae feeding that peaked at 48 hours after feeding. Herbivory by S. exigua also
increased JA levels in Zea mays (Schmelz et al. 2003b). However, there is little research
available on the induction of JA by generalist insects in Arabidopsis. In this study we
found that both insect species increased the production of JA after feeding. Both insects
117
induced levels that were significantly different from controls at early time points (S.
exigua, 30 minutes; P. rapae, 15 minutes) and which tapered off after 24 hours.
However, feeding by S. exigua elicited almost twice the concentration of JA as did P.
rapae feeding. Our results suggest that in response to S. exigua and P. rapae, the timing
of ET released by Arabidopsis plants may be a crucial regulatory mechanism while the
concentration of JA modulates downstream signaling.
Gene expression of ERF TFs and defense-related genes was very different in response to
the two insect treatments. ERF transcription factors have gotten significant attention
recently as their role in stress responses, cooperative regulation of JA and ET signaling,
and repressor and activator functions in multiple plant species has become of interest
(Fujimoto et al. 2000; Chakravarthy et al. 2003; Lorenzo et al. 2003; 2004; McGrath et
al. 2005; Yang et al. 2005; Nakano et al. 2006; Argarwal et al. 2006). In a previous study
(Rehrig et al.2010) we found 4 ERF transcription factors, DREBb, SIMRAP2.4, ERF104,
and ERF11, that were responsive to S. exigua at either 6 or 24 hours after feeding, while
TINY2 was primarily P. rapae-responsive. Much of our data support this preliminary
finding. Using Cluster 3.0, we created a heat map of gene expression values and were
able to identify three main gene expression clades. As shown in Figure 5, the first clade
consists of all P. rapae treatments, while the third is comprised of only S. exigua
treatments. In the second clade, only the expression patterns of AtMYC2 and PR4 were
similar between both P. rapae and S.exigua. This demonstrates that these two insects are
triggering markedly different signaling responses in Arabidopsis.
118
What may be causing the differential gene expression patterns by each insect? Mithofer
and Boland (2008) suggested that early events after insect herbivory, including hormone
production, may be analogous to the initial regulatory mechanisms involved in plant-
pathogen interactions. The authors hypothesized that plants respond to molecular signals
provided by insects (HAMPs, or Herbivory-Associated Molecular Patterns) and argued
that elicitors in insect oral secretions (OS) could be the primary triggers. Indeed, many
insect elicitors isolated from Spodoptera oral secretions have been identified, including
voliticin (Alborn et al. 1997), glucose oxidase (GOX) (Bede et al. 2006) and inceptins
(Schmelz et al. 2006). While little is reported about P. rapae salivary components, b-
glucosidase from Pieris brassicae OCs was found to induce indirect defenses and VOC
production in cabbage (Mattiaci et al. 1995). Therefore, the discrete differences in
ethylene signaling through ERFs may be originating at the feeding site.
Our results suggest that defense responses after S. exiuga feeding require the activation of
ERF transcription factors. Fold change increases in ERF4, ERF104, ERF, SIMRAP2.4,
ORA59, ERF5, ERF105, DREBb, AtERF1, and ERF11 were significantly higher in S.
exigua treatments. In fact, in only 2 cases, ORA59 and AtERF1 after 15 minutes, is gene
expression significantly higher in P. rapea-treated plants. This is especially highlighted
with the expression of ERF104 and ERF11, which are significantly increased by S.
exigua, and often repressed by P. rapae at these time points. Interestingly, ERF104 and
ERF11 were increased 9.5 and 3.9 fold in Arabidopsis plants after chitooctaose (chitin)
treatment (Libault et al. 2007). Furthermore, AtERF1, which was also chitin-responsive,
is increased by S. exigua, but not by P. rapae after 6 hours. Found in the cell wall of
119
necrotrophic fungi as well as arthropod exoskeletons, chitin is an N-acetylglucosamine
polymer whose perception is mediated through a LysM Receptor-Like kinase and
actively degraded by β-chitinases in plant cells (Boller 1985; Kaku et al. 2006; Wan et al.
2008). PR3 or Basic Chitinase gene expression was highly up-regulated by S. exigua with
little to no transcriptional changes after P. rapae treatment. Why chitin-sensing might be
involved in responses to S. exigua but not P. rapae is not clear.
We specifically selected the defense-related genes in this study because of their previous
characterizations as targets by ERF transcription factors via activation of the GCC box in
their promoters (Pre et al. 2008; McGrath et al. 2005, Brown et al. 2003; Fujimoto et al.
2000).
In general, transcriptional responses to S. exigua were consistently increased in most
genes, while responses to P. rapae were attenuated or in some cases, absent. In a similar
study, DeVos et al. 2005 also found that P. rapae did not significantly increase PDF1.2
or HEL (PR4) transcription, although an increase in PDF1.2::GUS activity at the
periphery of P.-damaged tissue was seen (DeVos et al 2005, Figure 4). Using
mechanically damaged plants plus regurgitant treatments, DeVos (2006) found that
simulated P. rapae ‘feeding’ suppressed PDF1.2 through the ABA-activation of AtMYC2
(Lorenzo et al. 2004).
Many of the defense-related genes analyzed in this study are known JA responsive genes
including PDF1.2, VSP2, LOX3, PR3, and PR4 (Lorenzo et al. 2003, 2004; Koornreef
120
and Pieterse 2008). We did not find any measurable increases in VSP2 transcripts present
in any treatment (data not shown). Although both insects elicited the production of JA
and ET, P. rapae did not increase the transcription of JA-inducible genes, providing
additional evidence that P. rapae is suppressing defense-related signaling. Schultz
(2002) spesculated that specialist insects may be “stealthy” and avoid triggering a barrage
of defense-related responses during herbivory. Patterns in expression of both ERFs and
down-stream defense genes in response to P. rapae treatment are consistent with this
hypothesis. Furthermore, JA and JA-IL levels elicited after P. rapae feeding were almost
half of what was observed for S. exigua. Reymond et al. (2000) found that wounding
induced far more genes than P. rapae, including water-stress related genes. Our results
suggest that the delayed timing of ET induced by P. rapae feeding as well as attenuated
JA production compared to S. exigua in Arabidopsis have broad implications in down-
stream signaling, specifically in the elicitation of ERF Transcription Factors and genes
coding for proteins involved in JA biosynthesis (LOX3), β-chitinase activity (PR3), and
feeding deterrents (PDF1.2, PR4).
Conclusions
Our results suggest that plant responses to different insects are not “one size fits all”
phenomena. Elicitors from insect saliva may initially trigger different pathways that
elicit the rapid production of different hormones and defense genes or may suppress
signaling to impede or stifle responses. In our study, plant reactions to the generalist S.
exigua entailed ERF activation through ET and JA signaling, while ET and ERF
121
transcriptional responses to the specialist, P. rapae, were delayed or attenuated.
Although JA concentrations in plants attacked by P. rapae were half those in plants
attacked by S. exigua, both insects induced the production of JA that was significantly
greater than controls, suggesting JA as a broad-spectrum response to herbivory. We also
confirmed that the expression of several genes, including ERF104 and ERF11, were
exclusively affected by S. exigua treatment. These genes may be key signaling
components in response to either insect; therefore, additional experimentation to assess
the insect resistance phenotypes in erf104 or erf11 mutants should be conducted.
122
References
1. Abe H, Urao T, Ito T, Seki M, Shinozaki K, Yamaguchi-Shinozaki K (2003)
Arabidopsis AtMYC2 (bHLH) and AtMYB2 (MYB) function as transcriptional activators in abscisic acid signaling. The Plant Cell 15: 63-78
2. Alborn HT, Hansen TV, Jones TH, Bennett DC, Tumlinson JH, Schmelz EA, Teal
PEA (2007) Disulfooxy fatty acids from the American bird grasshopper Schistocerca americana, elicitors of plant volatiles. Proceedings of the National Academy of Sciences 104: 12976-12981
3. Babst BA, Ferrieri RA, Gray DW, Lerdau M, Schlyer DJ, Schueller M, Thorpe MR,
Orians CM (2005) Jasmonic acid induces rapid changes in carbon transport and partitioning in Populus. The New Phytologist 167: 63-72
4. Beckers GJ, Spoel SH (2006) Fine-Tuning plant defence signalling: salicylate
versus jasmonate. Plant Biology 8: 1-10 5. Bede J, Musser R, Felton G, Korth K (2006) Caterpillar herbivory and salivary
enzymes decrease rranscript levels of Medicago truncatula genes encoding early enzymes in terpenoid biosynthesis. Plant Molecular Biology 60: 519-531
6. Berrocal-Lobo M, Molina A, Solano R, Plant J (2004) Constitutive expression of
ETHYLENE-RESPONSE-FACTOR1 in Arabidopsis confers resistance to several necrotrophic fungi. Molecular Plant-Microbe Interactions 29: 23-32
7. Boller T (1985) Induction of hydrolases as a defence reaction against 8. pathogens. . In J Key, T Kosuge, eds, Cellular and Molecular Biology of Plant
Stress. Alan R Liss, Alan R Liss, pp 247–262 9. Brown RL, Kazan K, McGrath KC, Maclean DJ, Manners JM (2003) A role for the
GCC-box in jasmonate-mediated activation of the PDF1.2 gene of Arabidopsis. Plant Physiology 132: 1020-1032
10. Chakravarthy S, Tuori RP, D'Ascenzo MD, Fobert PR, Despres C, Martin GB
(2003) The tomato transcription factor Pti4 regulates defense-related gene expression via GCC box and non-GCC box cis elements. The Plant Cell 15: 3033-3050
11. Chen W, Provart NJ, Glazebrook J, Katagiri F, Chang H-S, Eulgem T, Mauch F,
Luan S, Zou G, Whitham SA, Budworth PR, Tao Y, Xie Z, Chen X, Lam S, Kreps JA, Harper JF, Si-Ammour A, Mauch-Mani B, Heinlein M, Kobayashi K, Hohn T,
123
Dangl JL, Wang X, Zhu T (2002) Expression profile matrix of Arabidopsis transcription factor genes suggests their putative functions in response to environmental stresses. The Plant Cell 14: 559-574
12. Chini A, Fonseca S, Fernandez G, Adie B, Chico JM, Lorenzo O, Garcia-Casado G,
Lopez-Vidriero I, Lozano FM, Ponce MR, Micol JL, Solano R (2007) The JAZ family of repressors is the missing link in jasmonate signalling. Nature 448: 666-671
13. Chung HS, Koo AJK, Gao X, Jayanty S, Thines B, Jones AD, Howe GA (2008)
Regulation and function of Arabidopsis JASMONATE ZIM-domain genes in response to wounding and herbivory. Plant Physiology 146: 952-964
14. Cipollini D, Enright S, Traw MB, Bergelson J (2004) Salicylic acid inhibits
jasmonic acid-induced resistance of Arabidopsis thaliana to Spodoptera exigua. Molecular Ecology 13: 1643-1653
15. Clauss MJ, Dietel S, Schubert G, Mitchell-Olds T (2006) Glucosinolate and
trichome defenses in a natural Arabidopsis lyrata population. Journal of Chemical Ecology 32: 2351-2373
16. Delessert C, Wilson IW, Van Der Straeten D, Dennis ES, Dolferus R (2004) Spatial
and temporal analysis of the local response to wounding in Arabidopsis leaves. Plant Molecular Biology 55: 165-181
17. DeVos M (2006) Signal signature, transcriptomics, and effectiveness of induced
pathogen and insect resistance in Arabidopsis. Ph.D. University of Utrecht, Utrecht, Netherlands
18. DeVos M, Van Oosten VR, Van Poecke RMP, Van Pelt JA, Pozo MJ, Mueller MJ,
Buchala AJ, Metraux J-P, Van Loon LC, Dicke M, Pieterse C, M. J. (2005) Signal dignature and transcriptome changes of Arabidopsis during pathogen and insect attack. Molecular Plant-Microbe Interactions 18: 923-937
19. DeVos M, Van Zaanen W, Koornneef A, Korzelius JP, Dicke M, Van Loon LC,
Pieterse CMJ (2006) Herbivore-induced resistance against microbial pathogens in Arabidopsis. Plant Physiology 142: 352-363
20. Diezel C, von Dahl CC, Gaquerel E, Baldwin IT (2009) Different lepidopteran
elicitors account for cross-talk in herbivory-induced phytohormone signaling. Plant Physiology 150: 1576-1586
21. Eisen MB, Spellman PT, Brown PO, Botstein D (1998) Cluster analysis and display
of genome-wide expression patterns. Proceedings of the National Academy of Sciences 95: 14863-14868
124
22. Eulgem T (2005) Regulation of the Arabidopsis defense transcriptome. Trends in
Plant Science 10: 71-78 23. Faldt J, Arimura G, Gershenzon J, Takabayashi J, Bohlmann J (2003) Functional
identfication of AtTPS03 as (E)-beta-ocimene synthase: a monoterpene synthase catalyzing jasmonate-and wound-induced volatile formation in Arabidopsis thaliana. Planta 216: 745-751
24. Fujimoto SY, Ohta M, Usui A, Shinshi H, Ohme-Takagi M (2000) Arabidopsis
ethylene-responsive element binding factors act as transcriptional activators or repressors of GCC box-mediated gene expression. The Plant Cell 12: 393-404
25. Halitschke R, Schittko U, Pohnert G, Boland W, Baldwin IT (2001) Molecular
interactions between the specialist herbivore Manduca sexta (Lepidoptera, Sphingidae) and its natural host Nicotiana attenuata. III. Fatty acid-amino acid conjugates in herbivore oral secretions are necessary and sufficient for herbivore-specific plant responses. Plant Physiology 125: 711-717
26. Hansen BG, Kerwin RE, Ober JA, Lambrix VM, Mitchell-Olds T, Gershenzon J,
Halkier BA, Kliebenstein DJ (2008) A novel 2-oxoacid-dependent dioxygenase Involved in the formation of the goiterogenic 2-hydroxybut-3-enyl hlucosinolate and generalist insect resistance in Arabidopsis. Plant Physiology 148: 2096-2108
27. Harfouche AL, Shivaji R, Stocker R, Williams PW, Luthe DS (2006) Ethylene
signaling mediates a maize defense response to insect herbivory. Molecular Plant-Microbe Interactions 19: 189-199
28. Hudgins JW, Franceschi VR (2004) Methyl jasmonate-induced ethylene production
is responsible for conifer phloem defense responses and reprogramming of stem cambial zone for traumatic resin duct formation. Plant Physiology 135: 2134-2149
29. Kahl J, Siemens DH, Aerts RJ, Gäbler R, Kühnemann F, Preston CA, Baldwin IT
(2000) Herbivore-induced ethylene suppresses a direct defense but not a putative indirect defense against an adapted herbivore. Planta 210: 336-342
30. Kaku H, Nishizawa Y, Ishii-Minami N, Akimoto-Tomiyama C, Dohmae N, Takio
K, Minami E, Shibuya N (2006) Plant cells recognize chitin fragments for defense signaling through a plasma membrane receptor. Proceedings of the National Academy of Sciences 103: 11086-11091
31. Kessler A, Baldwin IT (2002) Plant responses to insect herbivory: the emerging
molecular analysis. Annual Review of Plant Biology 53: 299-328
125
32. Koornneef A, Pieterse CMJ (2008) Cross Talk in Defense Signaling. Plant Physiology 146: 839-844
33. Larionov A, Krause A, Miller W (2005) A standard curve based method for relative
real time PCR data processing. BMC Bioinformatics 6: 62 34. Li J, Brader G, Palva ET (1999) The WRKY70 transcription factor: a mode of
convergence for jasmonate-mediated and salicylate mediated signals in plant defense. The Plant Cell 16: 319-331
35. Libault M, Wan J, Czechowski T, Udvardi M, Stacey G (2007) Identification of 118
Arabidopsis transcription factor and 30 Ubiquitin-Ligase genes responding to chitin, a plant-defense elicitor. Molecular Plant-Microbe Interactions 20: 900-911
36. Lorenzo O, Chico JM, Sanchez-Serrano JJ, Solano R (2004) JASMONATE-
INSENSITIVE1 encodes a MYC transcription factor essential to discriminate between different jasmonate-regulated defense responses in Arabidopsis. The Plant Cell 16: 1938-1950
37. Lorenzo O, Piqueras R, Sanchez-Serrano JJ, Solano R (2003) ETHYLENE
RESPONSE FACTOR1 integrates signals from ethylene and jasmonate pathways in plant defense. The Plant Cell 15: 165-178
38. Lorenzo O, Solano R (2005) Molecular players regulating the jasmonate signalling
network. Current Opinion in Plant Biology 8: 532-540 39. Major IT, Constabel CP (2006) Molecular analysis of poplar defense against
herbivory: comparison of wound- and insect elicitor-induced gene expression. The New Phytologist 172: 617-635
40. Mattiacci L, Dicke M (2004) β-Glucosidase: an elicitor of herbivore-induced plant
odor that attracts host-searching parasitic wasps. Proceedings of the National Academy of Sciences 92: 12837-12842
41. McGrath KC, Dombrecht B, Manners JM, Schenk PM, Edgar CI, Maclean DJ,
Scheible W-Rd, Udvardi MK, Kazan K (2005) Repressor- and activator-type ethylene response factors functioning in jasmonate signaling and disease resistance identified via a genome-wide screen of Arabidopsis transcription factor gene expression. Plant Physiology 139: 949-959
42. Mewis I, Appel HM, Hom A, Raina R, Schultz JC (2005) Major signaling pathways
modulate Arabidopsis glucosinolate accumulation and response to both phloem-feeding and chewing insects. Plant Physiology 138: 1149-1162
126
43. Mewis I, Tokuhisa JG, Schultz JC, Appel HM, Ulrichs C, Gershenzon J (2006) Gene expression and glucosinolate accumulation in Arabidopsis thaliana in response to generalist and specialist herbivores of different feeding guilds and the role of defense signaling pathways. Phytochemistry 67: 2450-2462
44. Mithofer A, Boland W (2008) Recognition of Herbivory-Associated Molecular
Patterns. Plant Physiology 146: 825-831 45. Nakano T, Suzuki K, Ohtsuki N, Tsujimoto Y, Fujimura T, Shinshi H (2006)
Identification of genes of the plant-specific transcription-factor families cooperatively regulated by ethylene and jasmonate in Arabidopsis thaliana. Journal of Plant Research 119: 407-413
46. Pillai KCS, Tienzo P (1959) On the distribution of the extreme studentized deviate
from the sample mean. Biometrika 46: 467-472 47. Pre M, Atallah M, Champion A, De Vos M, Pieterse CMJ, Memelink J (2008) The
AP2/ERF domain transcription factor ORA59 integrates jasmonic acid and ethylene signals in plant defense. Plant Physiology 147: 1347-1357
48. Ramakers C, Ruijter JM, Deprez RHL, Moorman AFM (2003) Assumption-free
analysis of quantitative real-time polymerase chain reaction (PCR) data. Neuroscience Letters 339: 62-66
49. Renwick JAA, Lopez-Vidriero I (1999) Food consumption by larvae of Pieris
rapae: addiction to glucosinolates? . Entomologia Experimentalis et Applicata 91: 51-58
50. Reymond P, Bodenhausen N, Van Poecke RMP, Krishnamurthy V, Dicke M,
Farmer EE (2004) A conserved transcript pattern in response to a specialist and a generalist herbivore. The Plant Cell 16: 3132-3147
51. Reymond P, Farmer EE (1998) Jasmonate and salicylate as global signals for
defense gene expression. Current Opinion in Plant Biology 1: 404-411 52. Reymond P, Weber H, Damond M, Farmer EE (2000) Differential gene expression
in response to mechanical wounding and insect feeding in Arabidopsis. The Plant Cell 12: 707-720
53. Rozen S, Skaletsky H (2000) Primer3 on the WWW for general users and for
biologist programmers. Methods in Molecular Biology 132: 365-386 54. Schenk PM, Kazan K, Manners JM, Anderson JP, Simpson RS, Wilson IW,
Somerville SC, Maclean DJ (2003) Systemic gene expression in Arabidopsis during
127
an incompatible interaction with Alternaria brassicicola. Plant Physiology 132: 999-1010
55. Schenk PM, Kazan K, Wilson I, Anderson JP, Richmond T, Somerville SC,
Manners JM (2000) Coordinated plant defense responses in Arabidopsis revealed by microarray analysis. Proceedings of the National Academy of Sciences 97: 11655-11660
56. Schmelz EA, Alborn HT, Banchio E, Tumlinson JH (2003) Quantitative
relationships between induced jasmonic acid levels and volatile emission in Zea mays during Spodoptera exigua herbivory. Planta 216: 665-673
57. Schmelz EA, Alborn HT, Tumlinson JH (2003) Synergistic interactions between
volicitin, jasmonic acid and ethylene mediate insect-induced volatile emission in Zea mays. Physiologia Plantarum 117: 403-412
58. Schmelz EA, Carroll MJ, LeClere S, Phipps SM, Meredith J, Chourey PS, Alborn
HT, Teal PEA (2006) Fragments of ATP synthase mediate plant perception of insect attack. Proceedings of the National Academy of Sciences 103: 8894-8899
59. Schmelz EA, Engelberth J, Alborn HT, Tumlinson JH, Teal PEA (2009)
Phytohormone-based activity mapping of insect herbivore-produced elicitors. Proceedings of the National Academy of Sciences 106: 653-657
60. Schultz JC (2002) Biochemical ecology: how plants fight dirty. Nature 416: 267 61. Smallegange RC, van Loon JJA, Blatt SE, Harvey JA, Agerbirk N, Dicke M (2007)
Flower vs. leaf feeding by Pieris brassicae: glucosinolate-rich flower tissues are preferred and sustain higher growth rate. Journal of Chemical Ecology 33: 1831-1844
62. Stotz HU, Koch T, Biedermann A, Weniger K, Boland W, Mitchell-Olds T (2002)
Evidence for regulation of resistance in Arabidopsis to Egyptian cotton worm by salicylic and jasmonic acid signaling pathways. Planta 214: 648-652
63. Thaler JS, Farag MA, Pare PW, Dicke M (2002) Jasmonate-deficient plants have
reduced direct and indirect defences against herbivores. Ecology Letters 5: 764-774 64. Thines B, Katsir L, Melotto M, Niu Y, Mandaokar A, Liu G, Nomura K, He SY,
Howe GA, Browse J (2007) JAZ repressor proteins are targets of the SCF(COI1) complex during jasmonate signalling. Nature 448: 661-665
65. Thompson GA, Goggin FL (2006) Transcriptomics and functional genomics of
plant defence induction by phloem-feeding insects. Journal of Experimental Botany 57: 755-766
128
66. Tian H, Peng H, Yao Q, Chen H, Xie Q, Tang B, Zhang W (2009) Developmental
control of a lepidopteran pest Spodoptera exigua by ingestion of bacteria expressing dsRNA of a non-midgut gene. PLoS ONE 4: e6225
67. Traw MB, Bergelson J (2003) Interactive effects of jasmonic acid, salicylic acid,
and gibberellin on induction of trichomes in Arabidopsis. Plant Physiology 133: 1367-1375
68. Van Poecke Remco MP (2009) Arabidopsis-Insect Interactions. In The Arabidopsis
Book. The American Society of Plant Biologists, pp 1-34 69. Voelckel C, Schittko U, Baldwin I, T. (2001) Herbivore-induced ethylene burst
reduces fitness costs of jasmonate- and oral secretion-induced defenses in Nicotiana attenuata. Oecologia 127: 274-280
70. Vogel H, Kroymann J, Mitchell-Olds T (2007) Different transcript patterns in
response to specialist and generalist herbivores in the wild Arabidopsis relative Boechera divaricarpa. PLoS ONE 2: e1081
71. von Dahl C, Baldwin I (2007) Deciphering the role of ethylene in plant–herbivore
interactions. Journal of Plant Growth Regulation 26: 201-209 72. von Dahl CC, Winz RA, Halitschke R, Kuhnemann F, Gase K, Baldwin IT (2007)
Tuning the herbivore-induced ethylene burst: the role of transcript accumulation and ethylene perception in Nicotiana attenuata. The Plant Journal 51: 293-307
73. Wan J, Zhang X-C, Neece D, Ramonell KM, Clough S, Kim S-y, Stacey MG,
Stacey G (2008) A LysM Receptor-Like Kinase plays a critical role in chitin signaling and fungal resistance in Arabidopsis. The Plant Cell: tpc.107.056754
74. Winz RA, Baldwin IT (2001) Molecular interactions between the specialist
herbivore Manduca sexta (Lepidoptera, Sphingidae) and its natural host Nicotiana attenuata. IV. Insect-Induced ethylene reduces jasmonate-induced nicotine accumulation by regulating putrescine N-methyltransferase transcripts. Plant Physiology 125: 2189-2202
75. Wu J, Baldwin IT (2009) Herbivory-induced signaling in plants: perception and
action. Plant, Cell & Environment 32: 1161-1174 76. Yadav V, Mallappa C, Gangappa SN, Bhatia S, Chattopadhyay S (2005) A basic
helix-loop-helix transcription factor in Arabidopsis, MYC2, acts as a repressor of blue light-mediated photomorphogenic growth. The Plant Cell 17: 1953-1966
129
77. Yang Z, Tian L, Latoszek-Green M, Brown D, Wu K (2005) Arabidopsis ERF4 is a transcriptional repressor capable of modulating ethylene and abscisic acid responses. Plant Molecular Biology 58: 585-596
78. Zhu-Salzman K, Salzman RA, Ahn J-E, Koiwa H (2004) Transcriptional regulation
of sorghum defense determinants against a phloem-feeding aphid. Plant Physiology 134: 420-431
130
Supplementary data
Thigmotrophic Responses
Because we observed some unexpected peaks in gene expression (ERF4, ERF8, ERF11,
PDF1.2) in control samples after 15 min and 30 minutes, we conducted a simple
experiment to measure the expression of these genes in plants given only a cage/touch
treatment vs. no contact. Six-week old plants grown in the exact same conditions as
described above were used. To mimic the caged controls in the original experiments, we
placed cages on 4 middle-rosette leaves, occasionally manually adjusted them for 40
minutes, and removed them in 2, 20-minute increments. Control plants received no cages
and were minimally disturbed so not to induce a touch response. Plant tissue was
harvested for RNA/gene expression analysis 15 or 30 minutes later as described in
Materials and Methods.
131
Supplemental Figure 3.1: Gene expression of ERFs and defense genes in response to caging (touch). Black bars represent control plants (no touch) while gray bars represent samples that were caged and handled to mimic the control plants in the insect experiments. Error bars are +/- standard error of the means (n=3).
Supplementary Figure 3.2: SA production in WT Arabidopsis plants after S. exigua (A) or P. rapae (B) feeding over a 24hr time course- SA was measured as pmol/g fresh weight. Blue bars represent S. exigua feeding; light gray bars represent controls in the S. exigua experiment. Red bars in (B) represent P. rapae treatment, dark gray bars represent controls in the P. rapae experiment. Error bars are standard errors of the mean.
132
Chapter 4:
Insect performance on erf mutant plants suggests a major role for ERF transcription factors in
Arabidopsis susceptibility to herbivory
133
Abstract
Plant defenses against insects require the coordination of molecular, biochemical, and
physiological events. Previously, we examined the transcriptional and phytohormone
changes in Arabidopsis thaliana after herbivory by dietary generalist (Spodoptera exigua)
and dietary specialist (Pieris rapae) herbivores. Measurements of ethylene levels after
herbivory indicated that ethylene was produced in response to both species, although the
amounts and temporal pattern of ethylene, jasmonic Acid, and jasmonic acid-isoleucine
production differed. We found several members of the APETALA2/ERF transcription
factor gene family and downstream defense genes to be differentially regulated in
response to the two insect attackers. Four genes were only responsive to S. exigua,
including ERF104 (At5g61600) and two others, ERF105 and ERF5, had distinctly
different temporal patterns of expression in response to S. exigua and P. rapae. To further
elucidate the role of ERFs in defense responses to insects, we assessed the performance
and feeding behavior of S. exigua and P. rapae in wild-type (WT) Columbia ecotype and
the ERF mutants erf102 (erf5), erf103 (erf6), erf104, and erf105 using a novel digital
phenotyping technique. S. exigua maintained similar growth rates despite consuming less
mutant tissue. Although induced aliphatic and indolyl glucosinolate (GS) levels were
significantly higher in erf104 plants in response to S. exigua feeding, we found no
consistent relationships between GS and insect behavior or performance. Our results
challenge the effectiveness of GS “defense” compounds. ERF function does not appear
to be required for insect-induced GS increases, at least in response to P. rapae attack.
134
Differences in feeding behavior are critical to understanding such plant-insect
interactions, especially when they differ so greatly among insect species.
Introduction
During their development, plants encounter numerous abiotic and biotic stresses that
threaten their reproduction and fitness. Insect feeding removes and wounds plant tissue,
increases transpiration and water loss (Aldea et al. 2005), and upsets sugar balance
(Orians 2005). Because of this, plant response to herbivory requires the coordination of
many molecular events aimed at thwarting attackers through the production of defense-
related proteins and metabolites (Beekwilder, J. et al. 2008; Kessler et al. 2006; von Dahl
et al. 2007; Mewis et al. 2005, 2006). One of the first events to occur after plant stress is
the activation of defense-related transcription factors (Chen et al. 2002). Transcription
factors are then responsible for activating or repressing down-stream related defense
genes. Here, we focused on the role of ERF transcription factors in resistance to the
dietary specialist (Pieris rapae) and the dietary generalist (Spodoptera exigua) herbivores
in Arabidopsis.
ERF Transcription factors are a large family of plant-specific proteins with a DNA
binding domain specific to the GCC-box (Fugimoto et al. 2000; Brown et al. 2003) and
consist of several classes of subfamilies (Nakano et al. 2006). Originally identified in
tobacco (Ohme-Takagi & Shinski 1995), ERFs have since been found to be critical
players in jasmonic acid (JA)-inducible defenses as well as in resistance against fungal
135
and bacterial pathogens in tomato (Gu et al. 2000), Arabidopsis (Berrocal-Lobo, M. et al.
2004), rice (Cao et al. 2005), periwinkle (van der Fits & Mimelink 2000) and cotton
(Champion et al. 2009). Several studies have reported the role of ERFs as both
transcriptional activators and repressors in Arabidopsis (Ohta et al. 2001; Yang et al.
2005; McGrath et al. 2005). According to these studies, ERFs can up- or down-regulate
the expression of downstream genes including PDF1.2 and PR3, via the GCC-box.
Repression by ERFs is coordinated through the activity of an EAR motif (Ohta et al.
2001) which binds to DNA and prevents the expression of the downstream gene.
After insect attack, the plant hormones jasmonic acid (JA) and ethylene (ET) accumulate
and modulate the fine-tuning of defenses in response to different insects (Reymond et al.
2004, Mewis et al. 2004,2005; DeVos et al. 2005; Schmelz et al. 2003; Thompson and
Goggin 2006; Zhu-Salzman et al. 2009, 2004; Kessler and Baldwin 2002). ERFs have
been shown to be involved in responses to biotic stress and are rapidly induced after
treatment with JA and ET (Nakano et al. 2006; McGrath et al. 2005; Lorenzo et al. 2003;
Brown et al. 2003). In a seminal paper, Lorenzo et al. (2003) showed that intact JA and
ET signaling were required for the transcription of ERF1 as mutations in either of these
pathways failed to induce the expression of ERF1. Thus, ERFs may be able to regulate
specific abiotic and biotic stress responses in plants by turning on and off defense related
genes after ET and JA signaling. Despite the depth of research on JA and ET interactions
after insect attack and the known involvement of ERFs in JA-ET signaling, little is
known about the role of ERFs in mediating plant defense responses against insects in
Arabidopsis, especially in mediating glucosinolate-related defenses.
136
Glucosinolates (GSs) are the major chemical defenses in the family Brassicaeae (Hopkins
et al. 2009). Many of the transcription factors implicated in the regulation of GS-related
genes are in the MYB family, including ATR1, MYB28, and MYB29 (Celenza et al.
2005; Hirai et al. 2007). To date, no direct relationship between ERF transcription factors
and glucosinolate production has been identified. However, JA and ET signaling may
link ERF signaling and GS. Exogenous JA application positively regulated the
expression of genes involved in the production of indolyl glucosinolates (Brader et al.
2001; Mikkelsen et al. 2003, 2004). Similarly, application of exogenous JA significantly
increased total indolyl glucosinolates in Arabidopsis plants (Brader et al. 2001;
Mikkelsen et al. 2003; Mewis et al. 2005). Using a genetics approach, Mewis et al.
(2005) found that elicitation of GS after insect attack required intact ET signaling through
the ET receptor, ETR1. Bartlet et al. (1999) reported that both JA treatment and feeding
by specialist cabbage stem beetles increased indolyl GS levels in Brassica. These studies
highlight the importance of both JA and ET as well as insect feeding in the induction of
GSs, but whether this induction requires functional ERF activity is unclear.
In a previous study, we found that the generalist caterpillar S. exigua, but not the
specialist P. rapae, drastically induced the expression of several up-stream ERF
transcription factors as well as defense-related genes, although both insects induced
ethylene, JA, and JA-Isoleucine production (Chapter 3). These results suggest that ERFs
may be important signaling molecules in the differential responses to specialist and
generalist insects. We also found that ERF104 was transcriptionally affected by S. exigua
137
but not by P. rapae. In this study, we used a genetics approach combined with insect
bioassays to determine the resistance phenotypes of Arabidopsis erf105, erf5, erf6, and
erf104 mutant plants with impaired ERF signaling. Insect bioassays are effective ways to
measure the ecological consequences of gaining or losing a desired trait that may be
involved in nutrient status, mating, ovipositing or foraging (DeVos et al. 2008; Dorn et al.
2001; Foss & Rieske 2003). Furthermore, we utilized a novel, objective method for
measuring insect damage and thus assessing plant resistance phenotypes. Finally, we
tested the hypothesis that the induction of indolyl and aliphatic glucosinolates in
Arabidopsis by S. exigua and P. rapae depends on the activity of these ERF transcription
factors.
Materials and Methods
Plant Growth Conditions
Mutant seeds were provided by Dr. Shuquan Zhang (University of Missouri, Columbia,
MO) and came from either GABI or Salk insertion lines. (erf5:GABI_681E07, erf6:
Salk_087356, erf104: Salk_057720, erf105: GABI_680C11) Mutant plants were
homozygous with TDNA insertions in the open reading frame (S. Zhang, data
unpublished). Col-0 wild-type (WT), erf5, erf104,erf6 and erf105 seeds were sterilized,
imbibed in Millipore water for 5 days in the dark at 4°C then germinated on MS media
with 1X Sucrose. Plates and plants were kept in growth chambers at 22°C and 62% RH
on short -day conditions (8:16, L:D; 130 micro Einsteins illumination) to delay bolting
cycles and prolong rosette stage until they were used for experimentation. All plants were
watered as needed. After 7-10 days on MS/Sucrose media, plants were carefully
138
transplanted into Metromix 200 with Osmocote (Scott’s, Maryville, OH). Feeding assays
were conducted when plants were approximately 1.5” in diameter and 4-5 weeks old.
Insect Growth Conditions
The two lepidopteran chewing insects used for this study were P. rapae L. and the
generalist herbivore, S. exigua Hubner. P. rapae is a dietary specialist which feeds
exclusively on plant in the Brassicaceae family, while S. exigua is a dietary generalist that
feeds on a broad range of host plant families. S. exigua eggs were attained from Benzon
Research (Carlisle, PA) and reared on artificial diet (Carolina Biologicals, Burlington,
NC). Pieris were taken from a colony currently maintained in our insect rearing facility
(Bond Life Sciences Center, University of Missouri, Columbia, MO) and fed a mix of
Pak-Choi and Arabidopsis plants throughout their larval stages. Both insect species were
acclimated to eating Arabidopsis plants for 24 hours before the experiments. Larvae were
removed from pre-feeding plants and weighed 1-2 hours before the experiments started.
Insect Feeding Assays
Insect experiments with P. rapae and S. exigua were conducted separately. Weighed
insects were placed on plants of each genotype and enclosed on whole plants using
customized plastic cages with mesh lids. One insect per plant was used. Treatment
bioreplicates varied in number between experiments and ranged from 12 to 31 plants per
genotype. Plants were placed under growing lights under 12 hour days and insects were
139
allowed to feed for 24-48 hours. After the feeding assay, insects were reweighed for
growth analysis. To ensure that enough tissue remained for glucosinolate analysis, insects
were removed if too much of the tissue was being consumed. Growth rates were
calculated according to actual time spent feeding on the plants. Cages without insects
were placed on control plants. The performance of the insects was determined by a suite
of nutritional indices that describe the consumption, growth, and efficiency with which
food is converted to growth (Slansky & Scriber 1985). Growth of the insects during the
experiment was expressed as relative growth rate (RGR), which accounts for different
starting sizes of insects because differently sized insects grow at different rates even on
the same food. RGR is calculated using the equation: [(final weight (mg) – initial
weight (mg)] / initial weight (mg) * time of feeding (days). The amount of leaf eaten
during the experiment was expressed as relative consumption rate (RCR) which accounts
for different starting sizes of insects because differently sized insects can eat at different
rates on the same food. RCR is calculated using the equation: [specific mass of total
tissue eaten (mg) /[final weight (mg) – initial weight (mg)] * time of feeding (days).
Initial weights were used for RGR and RCR as recommended by Ferrar et al. 1989. The
ability of a given amount of consumed food to support insect growth was expressed as
the efficiency of conversion of ingested food (ECI). ECI is calculated using the equation:
[(final weight (mg) – initial weight (mg))]/ specific mass of total tissue eaten (mg) * time
of feeding (days). Because we conducted no-choice assays and caged insects on
individual plants, our experiments only ran for 2 days in the S. exigua assay and 1 day in
the P. rapae assay; otherwise, we risked having the whole plant consumed by the insect.
In addition to larval weight, development time to pupation has also been used for P.
140
rapae and S. exigua on Arabidopsis as a method of assessing insect performance (Van
Oosten et al. 2008), but this would have required more than one plant per insect to have
enough food to grow to pupation. Any insects that died, pupated or molted during the
experiment were eliminated from the analysis.
Quantification of Amount Eaten Using Digital Photography
Green et al. (2010, in prep) developed an algorithm for converting digital pictures of
insect infested plants into useful information about plant leaf removal by insects. A
schematic of the process of digital phenotyping can be seen in Figure 4.1. Using a 10
Megapixel Canon Rebel Digital camera in a customized stand, we took pictures of plants
before herbivory and then again after the feeding assay. The computer algorithm
automatically corrected for image size, setting the field of view to 10,000 pixels per cm2,
then masked the images to calculate pixels (leaf tissue) removed by the insect. Untreated
plants served as growth controls and increases in leaf area due to growth expansion were
factored into the analysis afterwards. To determine specific leaf mass (g/cm2), four-hole
punches were cut from 4-5 week-old control plants of each genotype (N=30) and
weighed. Samples were placed in an 80°C oven overnight and reweighed for dry weight.
Total mass eaten was calculated as a function of leaf removed (cm2) by fresh weight per
cm2 tissue. We then calculated the total mass of tissue (g) eaten by multiplying the
specific mass of leaf tissue disks (g/cm2) by the total area eaten by insects (cm2) (Figure
4.3). Control plants received no insect treatments and served as base lines for growth and
constitutive glucosinolates. Growth rates were determined for each genotype and factored
141
into the final amount of leaf tissue removed by insects after the feeding trials. This
method for assessing insect feeding is superior to those based on damage estimates after
feeding (Green et al 2010).
Figure 4.1: Schematic showing quantification of amount eaten using digital photography
Measurement of Leaf Glucosinolates
Since GS levels vary with leaf age and S exigua and P rapae preferentially consume
tissues of different ages (Appel and Schultz, unpublished), an analysis of GS in the
remaining tissue does not provide useful information on whole plant GS levels with
142
insect feeding. As a result, to obtain estimates of relative GS concentrations among
genotypes and between insect-free (constitutive GS) and insect-attacked (induced GS)
plants, we chose to focus on the youngest, innermost rosette leaves which we have also
found to be the most inducible by insect feeding (Appel and Schultz, unpublished).
Glucosinolates were extracted from flash frozen and freeze-dried tissues of the youngest
rosette leaves using a protocol similar to that used in Mewis et al. (2006), with
modifications. To extract GSs, 200 uL of 70% methanol and 10 uL of 3 mM sinalbin
standard as an internal control were added to 20-50 mg ground tissue, samples were
vortexed and extracted at 80°C for 5 minutes, and centrifuged at 4000 rpm for 4 minutes.
We removed the supernatant, added another 200 uL methanol to the pellet, and incubated
at 80°C for 5 min. The methanol wash and centrifugation step was repeated 3 times and
the supernatants pooled. Samples were placed in a centrivap until dry and pellets were
resuspended in 40 µL of 0.4 M barium acetate and 370 µL deionized water. Tubes were
sonicated for 5 minutes, centrifuged for 20 minutes at 4000 rpm and contents were
divided into two volumes for additional myrosinase (A) and desulfinase assays. For the
desulfation step, DEAE Sephadex A-25 solution in 2M acetic acid was packed into
Millipor MultiScreen 96-well filter plates and vacuumed. Two 200 µL washes of 6 M
imidazole formate followed by vacuum filtration and three water washes was done. The
entire crude glucosinolate extract from each sample was then added to the filter plates
and vacuum filtered. We washed samples twice with sodium acetate buffer solution (pH
= 4) and added 30 uL of sulfatase solution to each sample. Plates were placed in ZipLock
bags with a wet paper towel over the lid at room temperature overnight. To elute the final
desulfated glucosinolates from the samples, the plate was placed in a vacuum manifold
143
and eluted twice with 150 uL of distilled water. Detection and quantification of individual
desulfated indolyl and aliphatic glucosinolates was done using a Waters Alliance 2695
High-Performance Liquid Chromatograph in tandem with a Waters Acquity TQ Detector
Mass Spectrometer, on a C18 RP column using a water/acetonitrile linear gradient,
monitored at 229 nm, and quantified using an internal spike (sinalbin) added prior to
extraction. Calculations of molar concentration of individual GS relative response factors
(RF) (Brown et al., 2003; Buchner, 1987) were used to correct for absorbance difference
between the reference standard (4-hydroxybenzyl GS, RF 0.51) and other compounds.
Statistical Analysis
We conducted ANOVAs using the PROC GLM command in SAS (SAS Institute, Cary,
NC) followed by post-hoc Tukey tests to determine statistically significant differences
among genotypes in insect relative growth rates (RGR), relative consumption rates
(RCR), leaf tissue eaten, efficiency of conversion of ingested food indices (ECI), protein
levels, and glucosinolate (GS) levels. We recognized a statistical interaction between
growth and plant size and initial caterpillar weight during the S. exigua assay and
therefore used both plant size and insect size as covariates in the analyses. There was no
statistically significant interaction of these variables in the P. rapae assay. To determine
the relationship between insect growth (mg) and the amount of glucosinolates induced,
Pearson’s product-moment correlations (R-values) were calculated using the
CORREL(array1, array2) function in Microsoft (Redmond, WA) Excel. Using the PROC
REG command in SAS, we conducted a regression analysis on RCR against the amounts
144
of induced aliphatic and indolyl concentrations in the entire insect bioassays and by
treatment (genotype). Relationships with p-values less than 0.05 were designated as
significant.
Results
Insect Growth Rates and Feeding Assays
Overall, the two insects in this study responded differently to feeding on erf mutants
compared to WT plants. This is reflected in differences in the amount eaten,
consumption rates, and efficiency of conversion of ingested food (ECI) indices for each
insect on the different genotypes. As expected, P. rapae, which is adapted to feeding on
glucosinolate-containing plants, was insensitive to the mutants and maintained similar
relative consumption rates, relative growth rates, and efficiencies of conversion of
ingested food on all the genotypes (Figure 4. 2 b, d, f). In contrast, S. exigua reacted
differently to several of the erf mutants compared to WT plants (Figure 4.2 a, c, e). S.
exigua had significantly lower RCR on erf104 and erf6. Although higher specific leaf
masses can cause lower RCR because there is more nutrition per unit volume of leaf
consumed, this was not the case here because the specific leaf masses did not differ
statistically among the genotypes (data not shown). Despite eating significantly less of
erf104 and erf6, S. exigua growth rates did not differ among genotypes. This
compensatory feeding to maintain a constant growth rate is a common behavior of S.
145
exigua. The quality of the erf mutants as food for growth also differed for S. exigua. ECI
was significantly higher on erf104 than on the other mutants and WT.
Interestingly, no mortality was observed in the S. exigua assay, but we measured 38%,
13%, and 4% mortality rates for P. rapae larva on erf5, erf105, and erf6 plants,
respectively (data not shown). The high mortality rates in P. rapae on erf5 plants may
have resulted from using small caterpillars on this genotype to match plant size
(Supplementary Figure 4.3).
146
Figure 4.2: S. exigua and P. rapae relative growth rates (RGR) (A,B), relative consumption rates (RCR) (C,D) and efficiency of conversion of ingested food index (ECI) (E,F) on WT and erf mutant plants. RGR represents the total weight gained by an insect relative to its initial weight and total feeding time. RCR is calculated as the total material (mg) eaten divided by initial insect weight (mg) multiplied by the feeding period (days). ECI is calculated as the difference in insect mass before and after the feeding assay divided by the total material (g) eaten. Sample sizes differed between treatments and ranged between 11 and 31 insects. Error bars represent standard errors of the mean. Letters above columns represent post-hoc Tukey values. Different letters indicate significant differences between genotypes. p<0.05.
A B
C D
E F
147
Glucosinolate Analysis
The ability of plants to respond to insect feeding by increasing indolyl glucosinolates did
not depend on functional ERFs; both S. exigua and P. rapae feeding induced higher
concentrations of indolyl glucosinolates in the youngest leaves in all genotypes (Figure
4.3). In contrast, changes in aliphatic glucosinolates were heterogeneous. S. exigua did
not elicit an increase of aliphatic GS concentrations in WT, erf104 or erf105 plants,
whereas P. rapae elicited increases in all but erf6. Plants with the erf104 mutation were
more responsive to S. exigua than were the other genotypes. After S. exigua feeding,
erf104 plants had significantly higher induced levels of indolyl GSs than all the other
genotypes and induced aliphatic GSs that were higher than WT.
In the P. rapae experiments, erf104 plants had lower constitutive indolyl GS levels than
other genotypes, while erf104, erf105, and erf5 had lower constitutive aliphatic levels
(Figure 4.3). Induced levels of indolyl GSs after P. rapae feeding were not significantly
different among genotypes. However, P. rapae feeding resulted in lower induced
aliphatic GS in erf5 and erf6 plants. Although erf plants had, in many cases, lower
constitutive and induced GS levels, this did not have an effect on RGR, RCR or ECI in
the P. rapae bioassay. Despite eating significantly less erf5 tissue than all other
genotypes, P. rapae feeding metrics, including ECI, do not suggest that erf5 is a superior
food source for this insect.
148
Using a regression analysis, we calculated the relationships between total aliphatic or
indolyl levels and relative consumption rates (Table 4.1). When data from all genotypes
were combined, there was a significant relationship between total indolyl GS levels and
RCR that was negative for S. exigua and positive for P. rapae, consistent with their
tolerance for GS; however in neither case was the relationship strong. Within genotypes,
there were only 5 significant relationships between GS levels and RCR and these were all
positive correlations. These are highlighted in gray in Table 4.1.
Table 4.1: Pearson product-moment correlations and regression analyses of relative consumption rates and glucosinolate levels after insect feeding. p-values highlighted in gray represent significant correlations. Negative (-) correlation values indicate an inverse relationship between RCR and GSs induced, while positive values represent direct correlations.
149
Discussion
Ethylene Response Factors (ERFs) are a family of transcription factors that can serve as
both transcriptional activators and repressors by binding specifically to the GCC-cis-
C D
A B
Figure 4.3: Constitutive and induced Indolyl and Aliphatic Glucosinolate levels in WT and erf plants in S. exigua and P.rapae bioassays- Using HPLC-MS, constitutive levels of total indolyl (A, B) and aliphatic (C, D) GSs were determined in control plants and induced levels were measured after S. exigua (A, C) or P. rapae (B, D) treatments. Error bars represent standard errors of the mean. Letters above columns represent post-hoc Tukey values. Different letters indicate significant differences between genotypes. Lowercase letters represent Tukey values for constitutive GS levels, while capital letters indicate differences between induced GS levels. Asterisks represent significant differences between constitutive and induced GS levels within each genotype.
150
elments in defense-related gene promoters (Brown et al. 2003; Yang et al. 2005).
However, there is little known about the role of ERF transcription factors in resistance to
lepidopteran herbivores. Recent efforts to characterize ERFs have demonstrated their
roles in resistance to necrotrophic fungi, incompatible pathogens (McGrath et al. 2005;
Pre et al. 2008), as well as responsiveness to chitin application (Libault et al. 2007) and
insect attack (DeVos 2006). Bethke et al. (2009) recently showed that ERF104 serves as a
substrate for MAPK6 and is upstream of FLG22-induced ET signaling. The authors also
identified a number of pathenogenesis- related genes in ERF104 over-expression plants
including PDF1.2, PDF1.2b, Thi2.2, among others. Less is known about the functions of
ERF5, ERF6, and ERF105. Microarray experiments suggest that these three genes are
associated with carbon partitioning (Ferreira et al. 2008), changes in redox state (Giraud
et al. 2008) and sugar signaling (Veyres et al. 2006), all of which are important biological
triggers in plant-herbivore interactions (Maffei et al. 2009). Here, we examined the insect
resistance phenotypes in mutants of 4 highly homologous (Nakano et al. 2006) ERF
genes, namely, erf104, erf105, erf5 and erf6.
We examined the feeding behavior and growth of two insect species feeding on WT and
loss-of-function mutants. The results differed between the two. When corrected for
caterpillar size as Relative Consumption Rate (RCR), there were no differences in P.
rapae consumption among plant genotypes. However, RCR of S. exigua was
significantly less on erf104 and erf6, indicating a true impact of plant genotype on
consumption. These two mutants therefore exhibited true resistant phenotypes to S.
exigua, but not to P. rapae.
151
Consistent with the lack of variation in consumption rate, P. rapae grew about equally
well on all plant genotypes, and while its Efficiency of Conversion measures varied
among plant genotypes, those differences were not statistically significant. Overall, the
erf mutants had little or no impact on P. rapae feeding or growth.
The impact of erf mutants on S. exigua was quite different. Although S. exigua growth
rates also did not differ significantly, this was accomplished despite significant effects of
plant genotype on consumption. While S. exigua ate less mutant plant material, it was
able to compensate and utilize what it ate more efficiently than it used WT tissues.
Higher nutrition levels in mutant plants might explain this result, but no differences in
total protein levels were found among genotypes (see Supplementary Materials, Figure
1). It is also possible that S. exigua larvae shifted feeding to different leaves on the
mutant plants; we did not track individual leaf choice. Last, the increased residence time
of food in the gut under reduced consumption rates could lead to greater extraction
efficiency of nutrients from the food, leading to higher ECIs.
We expected that indolyl and aliphatic GS concentrations would explain variation in
insect feeding behavior and growth. GSs have been shown to decrease the performance of
generalist insects (Hansen et al. 2008; Traw & Dawson 2002). For example, feeding by
generalists was deterred when GS levels increased in two wild cabbage species
(Giamoustaris & Mithen 1995), while the same GS served as feeding stimulants to P.
rapae. Other studies have shown the role of GS as feeding stimulants and oviposition
152
cues for specialist insects, particularly P. rapae (Renwick et al. 2002; DeVos et al. 2008).
Still, GSs have also been found to be detrimental to specialists (Agrawal & Kurashige
2003). Furthermore, generalists have been shown to induce GS levels in brassicaecous
plants. Canola oilseed and wild mustard attacked by flea beetles had increased indolyl
GSs, but not aliphatics (Bodnaryk 1992; Bartlet 1999). According to Mewis et al. (2006),
S. exigua, but not P. rapae induced the production of aliphatic and indolyl GSs in WT
Arabidopsis. Taken together, these studies suggest that the impact of constitutive or
herbivory-induced indolyl or aliphatic glucosinolate levels may be insect- and plant-
dependent.
We found little clear evidence that glucosinolate concentrations were responsible for the
feeding and growth variation by the insects. Because our S. exigua and P. rapae
experiments were conducted on separate days and with a different set of plants, small
differences in growth conditions or plant sizes probably produced different initial GS
levels between experiments despite similar plant ages. Within each experiment, some
plant genotypes differed statistically in GS concentrations at the start of feeding
(constitutive).
Our measures of S. exigua performance and behavior were not related in any
straightforward way to our GS measurements of the youngest leaves. S. exigua avoided
feeding on erf104 and erf6 plants, and erf104 plants exhibited higher levels of post-
feeding (but not starting) indolyl and aliphatic glucosinolates, while erf6 GS
concentrations were not different from WT at either time. We had predicted that higher
153
GS concentrations would serve as feeding deterrents to S. exigua, and did find a weak but
significant correlation of RCR with indolyl GS. However, although erf104 plants had
greater induced aliphatic and indolyl GSs, we did not find a statistically significant
negative correlation between RCR and GS levels in any of the mutants individually. In
fact, we obtained several positive correlations between RCRs and GS across all the plants
in our study. The positive correlations with GS and RCR are not surprising for P. rapae.
However, it was surprising to see a positive correlation in erf5 plants and no relationship
on erf104 for S. exigua RCR and GS. These results suggest that the lower consumption
by S. exigua of erf104 did not arise simply because these plants had increased indolyl and
aliphatic GS levels after feeding. Furthermore, the positive relationship between RCR
and GS on erf5 plants suggests to us that the relationship between RCR and GS is more
complex than we first thought and may reflect two confounding effects. On the one hand,
we expect S. exigua preference for low GS to result in a RCR that is inversely
proportional to GS levels, leading to a negative relationship between RCR and GS. On
the other hand, because feeding increases GS levels, an inducible genotype can become
rapidly unpalatable as GS are induced to higher levels by feeding. Thus, although S.
exigua may have initially eaten in proportion to the amount of constitutive GS,
subsequent GS induction may have deterred feeding, with these effects cancelling each
other out in our measures of GS induction at the end of the experiment.
Nonetheless, we observed an obvious resistant phenotype in all erf mutants presented to
S. exigua that could not be explained by GS profiles. It is likely that other defense
mechanisms such as proteinase inhibitor activity (Kessler et al. 2006) or phenolics
154
(Morenoa et al. 2009) can account for the increased resistance in erf plants. Efforts are
currently underway in our laboratory to investigate this.
The range of glucosinolates or other traits found in the plant genotypes we studied had no
impact on P. rapae feeding or growth. We did find that several individual types of GS
were positively correlated with feeding, but as in the Spodoptera experiment these
relationships were different for each plant genotype and no consistent picture emerged
(data not shown). This is perhaps not surprising, since P. rapae specializes on
brassicaceous plants and has been shown to use GSs as feeding cues (Miles et al. 2005).
Hence we could not reliably associate biological impacts on either of these two insects
with variation in GS concentrations. In a similar study, Poehlman et al. (2008) found
that high GS levels in Brassica cultivars did not correlate with overall insect performance
of specialist or generalist insects.
The apparent lack of a direct link between amount of plant consumed, measured via
RCR, and GS levels may be related to the tissue we used for analysis. GSs were
harvested from the youngest, inner-most rosette tissue, which is typically not eaten by
insects and is the most GS inducible (H. Appel & J Schultz, personal communication).
Additionally, while conducting food choice bioassays with both insects, our lab has
shown that P. rapae prefers higher GS, younger leaves, while S. exigua chooses to eat
older leaves with lower GS content (McCartney & LoVerde, unpublished data).
Therefore, when measuring GS induction, harvesting the tissue remaining after insect
feeding creates a bias in the data. If P. rapae consumes mostly young leaves, then only
155
low GS older leaves remain and the reverse is true for S. exigua. Thus, whole plant
harvesting after feeding by either insect may reveal an incorrect result; shifting whole-
plant GS concentrations towards lower levels subsequent to P. rapae feeding and towards
higher levels in response to S. exigua. To overcome this problem, we harvested portions
of the plant that are rarely eaten by either insect, but are good indicators of GS induction.
Our results indicate that Arabidopsis plants impaired in erf signaling have different insect
resistance phenotypes and glucosinolate profiles compared to WT plants. In particular,
erf104, and to a lesser extent erf6, exhibited significantly resistant phenotypes with
respect to S. exigua attack. S.exigua larvae consumed less of those two plant genotypes,
but grew as well as on WT, evidently compensating physiologically. In previous studies
(Chapters 1 and 3), we found that the transcription of ERF104 is highly responsive to S.
exigua feeding, but not to P. rapae feeding, signifying a divergence in signaling
pathways following herbivory by each insect species. ERF104 is actively involved in
signaling after FLG22 and chitin elicitation (Bethke et al. 2009; Libault et al. 2007). It
may be that S. exigua triggers ERF104-mediated responses related to pathogens. And,
traits other than glucosinolates are evidently responsible for resistance of the erf mutants.
Our results suggest that ERF104, whose expression is strongly elicited by S. exigua, may
suppress a resistance phenotype in WT plants. On the other hand, none of the plant
genotypes exhibited resistance or susceptibility phenotypes to P. rapae. Similarly, P.
rapae feeding and growth were unaffected by the range of plant traits found in these plant
genotypes. It should be noted that we only tested the insect resistance phenotypes of
these plants using one knock-out allele. It could be that the differences we saw in the
156
genotypes were due to mutations in sites other than the ERF open reading frames.
Analysis with additional knock-out alleles for each mutation would need to be done to
eliminate this possibility.
We found no consistent relationships between GS and insect behavior or performance,
calling into question the effectiveness of this well-studied “defense” in Arabidopsis.
ERF function does not appear to be required for insect-induced GS increases, at least in
response to P. rapae attack. Differences in feeding behavior are critical to understanding
such plant-insect interactions, especially when they differ so greatly among insect
species.
157
References
1. Agrawal AA, Kurashige NS (2003) A Role for Isothiocyanates in plant resistance
against the specialist herbivore Pieris rapae. Journal of Chemical Ecology 29: 1403-1415
2. Aldea M, Hamilton JG, Resti JP, Zangerl AR, Berenbaum MR, DeLucia EH (2005)
Indirect effects of insect herbivory on leaf gas exchange in soybean. Plant, Cell & Environment 28: 402-411
3. Bartlet E, Kiddle G, Williams I, Wallsgrove R (1999) Wound-induced increases in
the glucosinolate content of oilseed rape and their effect on subsequent herbivory by a crucifer specialist. Entomologia Experimentalis et Applicata 91: 163-167
4. Beekwilder J, van Leeuwen W, van Dam NM, Bertossi M, Grandi V, Mizzi L,
Soloviev M, Szabados L, Molthoff JW, Schipper B, Verbocht H, de Vos RC, Morandini P, Aarts MG, Bovy A (2008) The impact of the absence of aliphatic glucosinolates on insect herbivory in Arabidopsis. PLoS ONE 3: e2068
5. Berrocal-Lobo M, Molina A, Solano R, Plant J (2004) Constitutive expression of
ETHYLENE-RESPONSE-FACTOR1 in Arabidopsis confers resistance to several necrotrophic fungi. Molecular Plant-Microbe Interactions 29: 23-32
6. Bethke G, Unthan T, Uhrig JF, Poschl Y, Gust AA, Scheel D, Lee J (2009) Flg22
regulates the release of an ethylene response factor substrate from MAP kinase 6 in Arabidopsis thaliana via ethylene signaling. Proceedings of the National Academy of Sciences 106: 8067-8072
7. Bodnaryk RP, Boter M, Ruiz-Rivero O, Abdeen A, Prat S, Conserved MYC (2004)
Effects of wounding on the glucosinolates in the cotyledons of oilseed rape and mustard. Phytochemistry 18: 2671-2267
158
8. Brader G, Tas E, Palva ET (2001) Jasmonate-dependent induction of indole glucosinolates in Arabidopsis by culture filtrates of the nonspecific pathogen Erwinia carotovora. Plant Physiology 126: 849-860
9. Brown PD, Tokuhisa JG, Reichelt M, Gershenzon J (2003) Variation of
glucosinolate accumulation among different organs and developmental stages of Arabidopsis thaliana. Phytochemistry 62: 471-481
10. Brown RL, Kazan K, McGrath KC, Maclean DJ, Manners JM (2003) A role for the
GCC-box in jasmonate-mediated activation of the PDF1.2 gene of Arabidopsis. Plant Physiology 132: 1020-1032
11. Buchner R (1987) Approach to determination of HPLC response factors for
glucosinolates. In BJ-PW Wathelet, ed, Glucosinolates in rapeseeds: Analytical Aspects, Vol 50-58. Kluwer Academic Publishers, Gembloux, Belgium
12. Celenza JL, Quiel JA, Smolen GA, Merrikh H, Silvestro AR, Normanly J, Bender J
(2005) The Arabidopsis ATR1 Myb transcription factor controls indolic glucosinolate homeostasis. Plant Physiology 137: 253-262
13. Champion A, Hebard E, Parra B, Bournaud C, Marmey P, Tranchant C, Nicole M
(2009) Molecular diversity and gene expression of cotton ERF transcription factors reveal that group IXa members are responsive to jasmonate, ethylene and Xanthomonas. Molecular Plant Pathology 10: 471-485
14. Chen W, Provart NJ, Glazebrook J, Katagiri F, Chang H-S, Eulgem T, Mauch F,
Luan S, Zou G, Whitham SA, Budworth PR, Tao Y, Xie Z, Chen X, Lam S, Kreps JA, Harper JF, Si-Ammour A, Mauch-Mani B, Heinlein M, Kobayashi K, Hohn T, Dangl JL, Wang X, Zhu T (2002) Expression profile matrix of Arabidopsis transcription factor genes suggests their putative functions in response to environmental stresses. The Plant Cell 14: 559-574
15. DeVos M (2006) Signal signature, transcriptomics, and effectiveness of induced
pathogen and insect resistance in Arabidopsis. Ph.D. University of Utrecht, Utrecht, Netherlands
159
16. DeVos M, Kriksunov KL, Jander G (2008) Indole-3-acetonitrile production from
indole glucosinolates deters oviposition by Pieris rapae. Plant Physiology 146: 916-926
17. DeVos M, Van Oosten VR, Van Poecke RMP, Van Pelt JA, Pozo MJ, Mueller MJ,
Buchala AJ, Metraux J-P, Van Loon LC, Dicke M, Pieterse C, M. J. (2005) Signal dignature and transcriptome changes of Arabidopsis during pathogen and insect attack. Molecular Plant-Microbe Interactions 18: 923-937
18. Dorn N, Cronin G, Lodge DM (2001) Feeding preferences and performance of an
aquatic lepidopteran on macrophytes: plant hosts as food and habitat. Oecologia 128: 406-415
19. Farrar RR, Barbour JD, Kennedy GG (1989) Quantifying food consumption and
growth in insects. Annals of the Entomological Society of America 82: 593-598
20. Ferreira FJ, Guo C, Coleman JR (2008) Reduction of plastid-localized carbonic
anhydrase activity results in reduced Arabidopsis seedling survivorship. Plant Physiology 147: 585-594
21. Foss LK, Rieske LK (2003) Species-specific differences in oak foliage affect
preference and performance of gypsy moth caterpillars Entomologia Experimentalis et Applicata 87-93. The Plant Journal 54:108: 81-92
22. Giamoustaris A, Mithen R (1996) The effect of flower colour and glucosinolates on
the interaction between oilseed rape and pollen beetles. Entomologia Experimentalis et Applicata 80: 206-208
23. Giraud E, Ho LHM, Clifton R, Carroll A, Estavillo G, Tan Y-F, Howell KA,
Ivanova A, Pogson BJ, Millar AH, Whelan J (2008) The Absence of ALTERNATIVE OXIDASE1a in Arabidopsis results in acute sensitivity to combined light and drought stress. Plant Physiology 147: 595-610
160
24. Gu Y-Q, Wildermuth MC, Chakravarthy S, Loh Y-T, Yang C, He X, Han Y, Martin GB (2002) Tomato transcription factors pti4, pti5, and pti6 activate defense responses when expressed in Arabidopsis. The Plant Cell 14: 817-831
25. Hansen BG, Kerwin RE, Ober JA, Lambrix VM, Mitchell-Olds T, Gershenzon J,
Halkier BA, Kliebenstein DJ (2008) A novel 2-oxoacid-dependent dioxygenase Involved in the formation of the goiterogenic 2-hydroxybut-3-enyl hlucosinolate and generalist insect resistance in Arabidopsis. Plant Physiology 148: 2096-2108
26. Hirai MY, Sugiyama K, Sawada Y, Tohge T, Obayashi T, Suzuki A, Araki R,
Sakurai N, Suzuki H, Aoki K, Goda H, Nishizawa OI, Shibata D, Saito K (2007) Omics-based identification of Arabidopsis Myb transcription factors regulating aliphatic glucosinolate biosynthesis. Proceedings of the National Academy of Sciences 104: 6478-6483
27. Hopkins RJ, van Dam NM, van Loon JJA (2009) Role of glucosinolates in insect-
plant relationships and multitrophic interactions. Annual Review of Entomology 54: 57-83
28. Kessler A, Baldwin IT (2002) Plant responses to insect herbivory: the emerging
molecular analysis. Annual Review of Plant Biology 53: 299-328
29. Kessler A, Halitschke R, Diezel C, Baldwin I (2006) Priming of plant defense
responses in nature by airborne signaling between Artemisia tridentata and Nicotiana attenuata. Oecologia 148: 280-292
30. Libault M, Wan J, Czechowski T, Udvardi M, Stacey G (2007) Identification of 118
Arabidopsis transcription factor and 30 Ubiquitin-Ligase genes responding to chitin, a plant-defense elicitor. Molecular Plant-Microbe Interactions 20: 900-911
31. Lorenzo O, Piqueras R, Sanchez-Serrano JJ, Solano R (2003) ETHYLENE
RESPONSE FACTOR1 integrates signals from ethylene and jasmonate pathways in plant defense. The Plant Cell 15: 165-178
32. Maffei ME, Mithofer A, Boland W (2007) Before gene expression: early events in
plant-insect interaction. Trends in Plant Science 12: 310-316
161
33. McGrath KC, Dombrecht B, Manners JM, Schenk PM, Edgar CI, Maclean DJ,
Scheible W-Rd, Udvardi MK, Kazan K (2005) Repressor- and activator-type ethylene response factors functioning in jasmonate signaling and disease resistance identified via a genome-wide screen of Arabidopsis transcription factor gene expression. Plant Physiology 139: 949-959
34. Mewis I, Appel HM, Hom A, Raina R, Schultz JC (2005) Major signaling pathways
modulate Arabidopsis glucosinolate accumulation and response to both phloem-feeding and chewing insects. Plant Physiology 138: 1149-1162
35. Mewis I, Tokuhisa JG, Schultz JC, Appel HM, Ulrichs C, Gershenzon J (2006) Gene
expression and glucosinolate accumulation in Arabidopsis thaliana in response to generalist and specialist herbivores of different feeding guilds and the role of defense signaling pathways. Phytochemistry 67: 2450-2462
36. Mikkelsen MD, Naur P, Halkier BA (2004) Arabidopsis mutants in the C-S lyase of
glucosinolate biosynthesis establish a critical role for indole-3-acetaldoxime in auxin homeostasis. The Plant Journal : for Cell and Molecular Biology 37: 770-777
37. Mikkelsen MD, Petersen BL, Glawischnig E, Jensen ABg, Andreasson E, Halkier
BA (2003) Modulation of CYP79 genes and glucosinolate profiles in Arabidopsis by defense signaling pathways. Plant Physiology 131: 298-308
38. Miles CI, Campo MLd, Renwick JAA (2005) Behavioral and chemosensory
responses to a host recognition cue by larvae of Pieris rapae. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology 191: 147-155
39. Moreno JE, Tao Y, Chory J, Ballare CL (2009) Ecological modulation of plant
defense via phytochrome control of jasmonate sensitivity. Proceedings of the National Academy of Sciences 106: 4935-4940
40. Nakano T, Suzuki K, Fujimura T, Shinshi H (2006) Genome-wide analysis of the
ERF gene family in Arabidopsis and rice. Plant Physiology 140: 411-432
162
41. Nakano T, Suzuki K, Ohtsuki N, Tsujimoto Y, Fujimura T, Shinshi H (2006) Identification of genes of the plant-specific transcription-factor families cooperatively regulated by ethylene and jasmonate in Arabidopsis thaliana. Journal of Plant Research 119: 407-413
42. Ohme-Takagi M, Shinshi H (1995) Ethylene-inducible DNA binding proteins that
interact with an ethylene-responsive element. The Plant Cell 7: 173-182
43. Ohta M, Matsui K, Hiratsu K, Shinshi H, Ohme-Takagi M (2001) Repression
domains of class II ERF transcriptional repressors share an essential motif for active repression. The Plant Cell 13: 1959-1968
44. Orians C (2005) Herbivores, vascular pathways, and systemic induction: facts and
artifacts. Journal of Chemical Ecology 31: 2231-2242
45. Poelman EH, Galiart RJFH, Raaijmakers CE, van Loon JJA, van Dam NM (2008)
Performance of specialist and generalist herbivores feeding on cabbage cultivars is not explained by glucosinolate profiles. Entomologia Experimentalis et Applicata 127: 218-228
46. Pre M, Atallah M, Champion A, De Vos M, Pieterse CMJ, Memelink J (2008) The
AP2/ERF domain transcription factor ORA59 integrates jasmonic acid and ethylene signals in plant defense. Plant Physiology 147: 1347-1357
47. Renwick JAA, Lopez-Vidriero I (1999) Food consumption by larvae of Pieris rapae:
addiction to glucosinolates? . Entomologia Experimentalis et Applicata 91: 51-58
48. Reymond P, Bodenhausen N, Van Poecke RMP, Krishnamurthy V, Dicke M,
Farmer EE (2004) A conserved transcript pattern in response to a specialist and a generalist herbivore. The Plant Cell 16: 3132-3147
49. Schmelz EA, Alborn HT, Tumlinson JH (2003) Synergistic interactions between
volicitin, jasmonic acid and ethylene mediate insect-induced volatile emission in Zea mays. Physiologia Plantarum 117: 403-412
163
50. Slansky F, Scriber JM (1985) Food consumption and utilization. In GA Kerkut, LI Gilbert, eds, Comprehensive Insect Physiology, Biochemistry and Pharmacology, Vol 4: Regulation: Digestion Nutrition Excretion. Pergamon Press, N.Y., pp 87-163
51. Smallegange RC, van Loon JJA, Blatt SE, Harvey JA, Agerbirk N, Dicke M (2007)
Flower vs. leaf feeding by Pieris brassicae: glucosinolate-rich flower tissues are preferred and sustain higher growth rate. Journal of Chemical Ecology 33: 1831-1844
52. Thompson GA, Goggin FL (2006) Transcriptomics and functional genomics of plant
defence induction by phloem-feeding insects. Journal of Experimental Botany 57: 755-766
53. Traw MB, Dawson TE (2002) Differential induction of trichomes by three
herbivores of black mustard. Oecologia 131: 526–532
54. van der Fits L, Memelink J (2001) The jasmonate-inducible AP2/ERF-domain
transcription factor ORCA3 activates gene expression via interaction with a jasmonate-responsive promoter element. The Plant Journal: for Cell and Molecular Biology 25: 43-53
55. Van Oosten VR, Bodenhausen N, Reymond P, Van Pelt JA, Van Loon LC, Dicke M,
Pieterse CMJ (2008) Differential effectiveness of microbially induced resistance against herbivorous insects in Arabidopsis. Molecular Plant-Microbe Interactions 21: 919-930
56. Veyres N, Danon A, Aono M, Galliot S, Karibasappa YB, Diet A, Grandmottet F,
Tamaoki M, Lesur D, Pilard S, Boitel-Conti M, Sangwan-Norreel BS, Sangwan RS (2008) The Arabidopsis sweetie mutant is affected in carbohydrate metabolism and defective in the control of growth, development and senescence. The Plant Journal 55: 665-686
57. von Dahl CC, Winz RA, Halitschke R, Kuhnemann F, Gase K, Baldwin IT (2007)
Tuning the herbivore-induced ethylene burst: the role of transcript accumulation and ethylene perception in Nicotiana attenuata. The Plant Journal 51: 293-307
164
58. Wittstock U, Agerbirk N, Stauber EJ, Olsen CE, Hippler M, Mitchell-Olds T,
Gershenzon J, Vogel H (2004) Successful herbivore attack due to metabolic diversion of a plant chemical defense. Proceedings of the National Academy of Sciences 101: 4859-4864
59. Yan X, Chen S (2007) Regulation of plant glucosinolate metabolism. Planta 226:
1343-1352
60. Yang Z, Tian L, Latoszek-Green M, Brown D, Wu K (2005) Arabidopsis ERF4 is a
transcriptional repressor capable of modulating ethylene and abscisic acid responses. Plant Molecular Biology 58: 585-596
61. Zhu-Salzman K, Salzman RA, Ahn J-E, Koiwa H (2004) Transcriptional regulation
of sorghum defense determinants against a phloem-feeding aphid. Plant Physiology 134: 420-431
165
Supplementary Materials
Protein Analysis
Total crude protein was isolated from frozen tissue ground in liquid nitrogen. Frozen
powder was weighed and placed into a microcentrifuge tube and suspended in 200uL
protein extraction buffer similar to that described in Conlon and Salter (2007) (125mM
Tris-HCL pH 8.8, 1% w/v SDS, 10% v/v glyercol, and 50mM Na2S2O5). Samples were
vortexed for 30 seconds and placed on ice until all samples where prepared. Samples
were then centrifuged at maximum speed (14,000 X G) for 10 minutes and the
supernatant was pipetted into a fresh tube. We used 5uL of a 2X diluted supernatant for
the Bio-Rad DC (Detergent Compatible) Protein Assay. Samples were pipetted into a 96
well plate and a standard curve was generated using molecular grade Bovine Serum
Albumin (Invitrogen, Carlsbad, CA). Colorimetric differences were read using a Synergy
HT plate reader at 750nm.
Supplementary Figure 4.1: Total Protein Levels in WT and erf plants
Prot
ein
Leve
ls m
g/g
grou
nd ti
ssue
pow
deaa
r
166
Supplementary Figure 4.2: Initial Plant Size for each Bioassay
Supplementary Figure 4.3: Initial Insect Size for each Bioassay
167
Chapter 5:
The role of ERF Transcription Factors in defense against the specialist and generalist caterpillars in
Arabidopsis thaliana
168
Research Summary
Upon insect attack, plants must carefully coordinate molecular, biochemical, and
physiological events in order to induce the appropriate defense responses to continue to
grow, reproduce, and survive. Signals released by wounding (Dellasert et al. 2004),
trichome damage, (Clauss et al. 2006) and elicitors from salivary deposits (Alborn et al.
1997, Mattiaci et al. 2004; Schmelz et al. 2006, 2009; DeVos et al. 2009) released during
chewing come in contact with wounded tissue and activate entire signaling cascades
aimed at thwarting insect attackers. Although no known receptors for insect elicitors
have been specifically identified, several studies suggest that components in insect saliva
react with plant cells with ligand-receptor kinetics (Truitt et al. 2004; Schmelz et al.
2007). Shortly following perception/reception, the rapid elicitation and synthesis of
several stress-related hormones occurs, including ethylene (ET) (see review by vonDahl
et al. 2007), jasmonic acid (JA) (Thaler et al. 2002; see review by Kessler & Baldwin
2002), and salicylate (SA) (see review by Beckers & Spoel 2006). Often, the synthesis of
these hormones, particularly ET, requires enzyme activation via MAP Kinase
phosphorylation (Liu & Zhang 2004). MAP Kinase action during insect herbivory has
gained increasing attention and plants impaired in MAPK signaling are more susceptible
to insect attack (Wu et al. 2008; Kandoth et al. 2007) Next in the cascade is the activation
of transcription factors (TFs), which can up- or down-regulate downstream defense-
related genes by binding to specific cis-elements in insect-elicited gene promoters. Once
these genes are transcribed, they are quickly translated into proteins that may serve as
enzymes in secondary metabolite production (such as glucosinolate biosynthesis) or
169
feeding deterrents (ie. proteinase inhibitors) (see review by Van Poecke 2007), thus
changing the overall chemical composition of the plant. Ultimately, it is change in plant
chemistry that affects whether an insect will be deterred from feeding or attacked by
summoned parasitoids. Overall changes in chemistry may have no major effect on the
insect as some herbivores have developed methods to avoid detection (stealth) (Schultz
2002) and tolerate secondary metabolites (Wittstock et al. 2994; Agrawal & Kurashige
2003). Thus, there are many levels—molecular, biochemical, physiological, and
ecological--- at which plant-herbivore interactions can be investigated. By measuring
hormone release, TF activation, defense gene expression, secondary metabolite
production, and insect performance, the research presented in my thesis examined plant-
insect interactions at all of these levels. The overall goal of my research project was to
help elucidate the mechanisms utilized by plants when responding to generalist vs.
specialist herbivores and based on the results, I believe that part of this goal was
achieved.
All of this research was conducted using Arabidopsis thaliana (Columbia ecotype), a
member of the Brassicacae family. Two different caterpillar species were used: the
dietary specialist Pieris rapae L. (Cabbage White Butterfly) which feeds only on
members of the Brassicaceae [of which Arabidopsis thaliana is a member], and the
dietary generalist, Spodoptera exigua Hubner (Beet Army Worm). For the microarray
project, which is discussed next, we also examined changes in gene expression caused
after feeding by 2 phloem-feeding insects (aphids) including the specialist, Brevicoryne
170
brassicae, and the generalist, Myzus perscicae (green peach aphid). However, the
majority of the research presented here was conducted using lepidopteran herbivores.
To begin this project, I analyzed patterns in insect-induced gene transcription in
Arabidopsis using a whole genome microarray. Although I was not initially part of the
insect assay or RNA preparation for the array, I spent a significant amount of time on
data analysis and gene characterization. To do this, I conducted on-line searches of GO
annotations and the current literature to thoroughly curate all of the TF that were
differentially regulated in our microarray. This search resulted in the identification of
193 unique transcription factors from 34 different gene families. All of the TF genes
regulated by insect feeding can be seen in Chapter 1, Table 1.2.
Transcription factors are proteins that bind to specific sites in gene promoters called cis-
elements to either activate or repress their expression. A seminal paper written in 2000
(Reichman) suggested that there are over 1500 TFs in Arabidopsis. The functions of
many of these are known, and several TF microarray studies have been conducted using
various abiotic and biotic stresses such as cold, drought, salinity (Chen et al. 2002), chitin
(Libault et al. 2007). However, ours was the first study to specifically examine TF
profiles after insect herbivore treatment.
Arabidopsis TF gene profiles differed greatly among the insect (and wounding)
treatments. The overall pattern that we observed was that generalist insects elicited the
greatest number of genes and transcription factors and that specialists were more
171
“stealthy” (Chapter 1, Supplemental Table 1.2). This is interesting because Myzus
persicae (the generalist aphid) caused the least amount of damage seen on our plants
(Heidi Appel, personal communication). DeVos et al. (2005) reported similar findings
when conducting a whole-genome microarray with M. persciae and 4 other pathogen or
insect treatments. Numerous elicitors have been identified in salivary components of
generalist insects, including volicitin (Alborn et al. 1997), inceptins (Schmelz et al.
2005), caeliferins (REF), and a 3-10kDa protein in M. persicae saliva (DeVos et al.
2008), while few elicitors have been found in specialists (Mattiaci et al. 2004). One
hypothesis for why this could be happening is that the plants are slightly ahead in the
evolutionary “arms-race” with generalists and are able to perceive elicitor signals to
induce defenses or attract parasitoids. Furthermore, because many specialists can tolerate
their hosts’ defenses (Wittstock et al. 2004), inducing complex chemistries after attack is
not an efficient use of a plant’s energy. In this way, specialists might then be ahead in the
biological arms race with plants.
In addition to indentifying TF genes that were important to insect-induced signaling, we
sought to indentify cis-elements in co-expressed genes critical to differential responses.
Numerous cis-elements have been found by identifying co-expressed genes in
Arabidopsis, including Abscisic Acid (ABA) Responpsive Binding Elements (ABREs)
(Chen et al. 2002) and Rapid Wound Response Elements (RWREs) (Walley et al. 2005).
However, to date, no papers have been published that have identified insect-specific
motifs in Arabidopsis gene promoters. Using several bioinformatics tools, I performed a
very thorough search for all known and uncharacterized consensus motifs in co-expressed
172
genes from each insect treatment, tissue type (local/systemic), and time points. We then
conducted principal component analysis on an entire matrix containing all genes,
treatments, and the number of elements found in gene promoters regulated by each insect
treatment. Unfortunately, no specific “cis-element” fingerprints were found. This
suggested that there are no specific elements or groups of elements that are indicative of a
specific insect treatment. This was supported by large numbers of elements that were
shared among genes in all treatments. The identification of an insect-specific promoter
would have been a significant finding towards the development of herbivore-resistant
crops. Nonetheless, searches of cis-element databases using co-expressed gene
promoters as data sets presented several motifs that were enriched in genes differentially
regulated by insect treatments (Chapter 1, Table 1.4). These included: ABRE-like
elements and G-Boxes, which are important in water-related stresses, ABA responses,
and drought tolerance (Abe et al. 2003). Enriched Evening Elements and IBOXES, which
are circadian rhythm and light-related elements (Harmer et al. 2000), suggested that
feeding by insects utilized similar signaling pathways as critical diurnally-mediated
physiological responses. Although several studies have shown that reduced light or
shading can decrease insect resistance in plants (see Ballere 2008 for review), it would be
interesting to further investigate if these motifs mediate this response.
One of the most dramatic and interesting patterns that emerged from the array data was
the differential expression patterns of ERF transcription factors by the generalist S.
exigua and the specialist P. rapae. ERF transcription factors were first identified based
on their similar sequence homology to APETALA-type TFs which regulate flower
173
development (see review by Gutterson & Rueber 2004). Much of what we know about
ERF signaling has been revealed by examining ERF1 transcription, which is highly ET
responsive and requires functional ETR1 and COI1 genes (Lorenzo et al. 2004). Besides
being ET and JA responsive, many ERFs play critical roles in response to chitin (Libault
et al 2007, necrotrophic fungi (McGrath et al. 2005), bacterial pathogens (Onate-Sanchez
et al. 2007), and flg22 (Bethke et al. 2009). However, little is known about their role in
insect resistance.
The ERF microarray results were the hypothetical basis for the remainder of my thesis
project. Initially, microarray analysis detected 20 different ERF Transcription Factors
transcriptionally regulated by insect feeding. To confirm these findings, I developed
gene-specific primers for 17 of these genes and conducted Real Time Reverse
Transcription- PCR on the same RNA used in the array. Because RT-PCR is more
sensitive than a microarray, I found that some ERFs were also regulated by P. rapae
feeding. Furthermore, I had a 92% confirmation rate and discovered 5 genes that were
uniquely expressed to only one insect treatment, including TINY2 (P. rapae only),
SIMRAP2.4, DREBb, ERF104, and ERF11 (S. exigua only). This led us to ask the
question “why did S. exigua cause increased expression of ETHYLENE Response Factor
TF genes, but not P. rapae? Of course, the logical hypothesis was ethylene.
A role for ethylene signaling in plant-herbivore relations is well-documented, yet there is
no generalized model for ET action as this appears to be insect and plant-dependent (for
review see vonDahl et al. 2007a). Several studies have shown increases in secondary
174
metabolite, anti-feeding agents, and parasitoid-attracting volatiles after ET treatment in
plants (Hudgins & Franceschi 2004; Harfouche et al. 2006; Schmelz et al. 2003). Still,
there is a debate within the literature as to whether induction of ET signaling is a
generalist-beneficial adaptation or a specialist “trick”. For example, ET released after M.
sexta (a specialist in Nicotiana) feeding decreased JA-inducible nicotine production in
tobacco (Kahl et al. 2000). Similarly, larvae of the specialist, Pieris rapae, grew poorly
on Arabidopsis ethylene signaling mutants (Mewis et al. 2005, 2006). Recently, Diezel et
al. 2009 found that M. sexta but not S. exigua induced ET production in tobacco. In
contrast with this, Stotz et al. (2000) found that ethylene signaling increased
susceptibility to the generalist Spodoptera littoralis in Arabidopsis, and ET induction by
S. exigua feeding has been repeatedly shown in corn (Schmelz et al. 2003, 2006, 2007,
2009). Using path analysis Vogel et al. (2007) found that transcriptional responses in
Boechera divaricarpa to the specialist Plutella xylostella were mediated through ET and
SA pathways, while responses to the generalist Trichoplusia ni were mediated through
ET and JA signaling. This supports a third explanation that ET helps to “fine-tune” SA
vs. JA signaling in plants after insect attack (vonDahl et al. 2007b); however, to date, no
studies have compared directly ET, JA, and SA induction by specialist vs. generalist
feeding in Arabidopsis.
Based on our microarray expression patterns as well as evidence provided in the
literature, I hypothesized that plant defense after generalist (S. exigua) attack requires
ethylene signaling through ERF transcription factors. My hypothesis was that S. exigua
induces ethylene and JA production, thus activating ERF transcription factor expression
175
as well as the down-stream PR genes, PR3, PR4 and PDF1.2 (from Lorenzo et al. 2004).
However, I predicted that P. rapae would only induce JA production and activate
signaling pathways through AtMYC2 to promote the JA-responsive defense genes VSP2,
LOX3, and Thi2.1 (Lorenzo et al. 2004, Lorenzo & Solano 2005). The overall
hypothetical model for my thesis is seen here in Figure 5.1.
Figure 5.1: Hypothetical model for differential expression patterns by generalist and specialist insects.
176
To test this hypothesis, we measured ethylene released by Arabidopsis plants after insect
treatments at numerous time points and collected tissue to measure JA, JA-conjugates,
and SA via HLPC-MS and ERF and defense gene expression using RT-PCR. Our results
show that feeding by both insects triggers increased production and emission of ethylene
but does so at different times. After only 15 minutes, S. exigua elicited a rapid ET burst,
which continued until 6 hours. The ET burst elicited by P. rapae occurred at 2 hours and
stayed above control levels for most of the remaining time course to 72 hours. Thus, our
data suggest that ET signaling might be an important component for responses to both
insects, but that timing may be equally critical. One could hypothesize that by initially
“evading” or “suppressing” an ET burst, P. rapae avoids a potential “window of
opportunity” for the plant to rapidly trigger ERF activation and other ET-related defense
responses.
JA is an important signal in response to both insects. Biotic or herbivore stress (Schmelz
et al. 2003a; see review by Halitschke & Baldwin 2004) has been repeatedly shown to
induce JA production in plants. Both P. rapae feeding in Arabidopsis (DeVos et al.
2005, Reymond et al. 2000) and S. exigua feeding in maize increased the production of
JA. However, there is little research available on the induction of JA by generalist insects
in Arabidopsis. We also measured JA-Isoleucine levels because two seminal papers
(Chini et al. 2007; Thines et al. 2007) simultaneously reported that JA-Isoleucine (JA-IL)
is the biologically active form of JA in the plant by binding to the SCFCOI1 protein
complex. In this study we find that both insects increase the production of JA and JA-IL
after feeding but S. exigua elicited almost twice the concentration of JA, and about 1.5X
177
as much JA-IL as P. rapae. Our results suggest that the down-stream responses to S.
exigua and P. rapae in Arabidopsis are a function of both JA and ET signaling pathways,
but timing and concentrations may be equally important. However, because SA levels did
not increase compared to controls by either insect at any time point in my experiments, I
found no evidence for SA involvement in plant responses to these two insects.
RT-PCR can be a powerful tool in measuring gene expression after different insect
treatments, but efforts must be taken to ensure that expression values are determined
correctly. Using RT-PCR we measured gene expression of over 20 ERF TF genes,
defense marker genes (including PR3, PR4, PDF1.2, LOX3, Thi2.1, and VSP2), and
control genes for this experiment. However, alternative controls for RT-PCR were used
to avoid normalization errors caused by unstably expressed control genes. As described
in Chapter 2, I encountered a technical hurdle in our first RT-PCR analysis of the
microarray RNA in that no adequately stable “house-keeping” or reference gene could be
identified among all of our treatments. For the microarray data described in Chapter 1,
gene expression was normalized to cDNA levels measured after reverse transcription.
This method sufficed for our initial experiment but was a less than optimum
normalization strategy moving forward. After failing to find invariable expression in 12
commonly used reference genes including 18S, EFa-1, ACT7, TUB2, and others
suggested by Czechowski et al. (2005), I developed a method using an exogenous
Luciferase mRNA “spike” and gene-specific primers as an RT-PCR control. Thus, for all
of our RT-PCR reactions for our ERF time course, gene expression was normalized to
both starting RNA levels and exogenous Luciferase mRNA levels that were added to a
178
master mix (to avoid pipetting errors) immediately before the reverse transcription step.
Overall, this method helped normalize our data successfully (Chapter 2, Table 2.3) and
reduced the time and cost of conducting real-time PCR on numerous reference genes.
Some of the gene expression results supported my original hypothesis, while others
negated it. Overall, ERF transcriptional responses to P. rapae were muted or down-
regulated compared to controls. Conversely, responses to S. exigua were dramatic and
highly up-regulated relative to controls. Three of the five insect-specific ERFs (ERF104,
ERF11, DREBb), identified in the array project showed similar expression profiles at 6
and 24 hours (which were the 2 times points also measured in the array). However, the
expression of SimRAP2.4 was erratic and a signficant increase in gene expression in
TINY2 after P. rapae feeding did not occur. ERF104 and DREBb expression were
consistent with what we observed in the array. Transcription of these two genes was
significantly different between insect treatments, with S. exigua increasing expression
and P. rapae decreasing them. For example, AtERF1 expression increased (although not
significantly) from controls after 30 minutes and then tapered off after S. exigua feeding.
However, the expression of AtMYC2 was quickly increased by S. exigua, possibly
through JA-signaling, but then after 6 hours, AtMYC2 expression induced by P. rapae
was significantly up-regulated, as expected. Only S. exigua significantly up-regulated the
ERF-controlled genes, PR3 and PR4, but not PDF1.2. Although in our study, P. rapae
slightly induced the expression of PDF1.2 (albeit, not significantly) at 6 hours, DeVos
(2006) reported a suppressive effect of P. rapae feeding or regurgitant on PDF1.2
expression via AtMYC2 signaling. Here, after 24 hours, PDF1.2 was slightly down-
179
regulated; however, these data were not statistically significant. The slight elicitation of
PDF1.2, but not PR3 or PR4 by P. rapae suggests an ERF-independent mechanism for
PDF1.2 activation. Recently, Zander et al. (2009) indentified TGA transcription factors
that induce PDF1.2 expression in a JA-dependent/ ERF-independent manner under high
ET signaling conditions. This may be an additional mechanism utilized by plants to
induce JA/ET-related genes through by-passing the antagonistic ERF and AtMYC2
pathways. JA-responsive LOX3 and Thi2.1 gene expression were slightly up-regulated
by both insects, but not significantly. This is slightly unexpected as both insects
significantly increased JA and JA-IL levels. In general, the results from ET
measurements and RT-PCR partially support the originally hypothesis, but suggest a
more complicated model, which includes additional TF pathways and careful timing of
hormone release and subsequent gene expression.
Although we noticed significant differences in gene expression between insects with
respect to ERF TF genes, the question still remained as to whether these genes played an
important role in insect resistance in Arabidopsis. To assess this, we used a genetics
approach. In previous experiments, the transcription of ERF104 was consistently S.
exigua specific, thus one hypothesis is that this gene is critical for resistance (or
susceptibility) to the generalist insect. To test this hypothesis, we assessed the insect-
resistance phenotypes of erf104 as well as erf105, erf5, and erf6 knock-out plants..
Mutant seeds were provided by Dr. Shuqun Zhang and were either homozygous GABI or
Salk lines with mutations in ORFs (S. Zhang, personal communication). We then
180
measured the insect growth, tissue consumption, glucosinolate levels, and protein
concentrations in WT and each of these mutant lines
I used a non-destructive digital photography method to calculate the amount of plant
material present before and after insect feeding (Green et al. 2010, in prep). Using
differences in growth of non-treated plants, we calculated an average growth factor for
each genotype. Based on our measures of leaf densities, we were able to get an objective,
accurate determination of the total specific leaf mass consumed by each insect during the
feeding trial. The overall method for conducting our insect bioassays is seen in Chapter
4, Figure 4.1.
Relative growth rates, relative consumption rates, total tissue eaten, and feeding indices
revealed insect-resistance phenotypes in several of the mutants. Relative growth rates
(RGR) were not affected by P. rapae or S. exigua in any genotype. However, when
growth rates were compared to how much plant material was eaten, an interesting result
emerged. S. exigua ate less of all mutant genotypes than it did of WT, and least of erf104.
S. exigua relative consumption rates (RCR), which correct for initial insect size, were
lower in erf6 and erf104 plants relative to WT. Food quality indices (ECI), which use
both weight gain and amount eaten as variables, suggested erf104 plants were used more
efficiently for growth by S. exigua, but no effect of plant genotype on food quality was
seen for P. rapae (Chapter 4, Figure 4.2).
181
Our results are in contrast with those of Mewis et al. (2005) and others (Stotz et al. 2000,
2002; Winz & Baldwin 2001) who found that disruption in ET signaling using the ET-
blocker, MCP, enhanced food quality, thus allowing insects to eat more tissue. Unlike
JA-signaling impaired coi1 plants which are readily consumed by S. exigua (Reymond et
al. 2004; Mewis et al. 2005) and M. sexta (Paschold et al. 2007), erf mutant plants show
an opposite phenotype. Insects were able to eat less of erf without having a negative
effect on their growth rates. Several phenomena could explain these results. First, erf
plants may be more nutritious, allowing insects to sustain normal growth rates despite
eating less tissue. However when we measured total crude protein among genotypes, we
did not observe any significant differences (Chapter 4, Supplemental Figure 4.1).
Second, relative growth weights (RGR) as determined by weighing insects before and
after feeding may not be an adequate measure of insect performance. Although this
method is broadly used in the literature (Chung et al. 2008; Cipollini et al. 2004;
Halitschke & Baldwin 2003, to list a few), other metrics, such as days to pupation and
pupal mass may be better indicators (Van Oosten et al. 2008). If this experiment were to
be redesigned, these two variables should be measured. Finally, erf plants may be more
resistant to (less eaten by) insects because they are better chemically defended. The latter
suggests a negative role of ERFs in plant defense.
In Arabidopsis, the secondary metabolites glucosinolates, are known to be important in
plant defenses against herbivory (see review by Hopkins et al. 2009). To evaluate
changes in defense responses between genotypes, we measured constitutive and induced
glucosinolates levels in control and insect-treated plants. Glucosinolates are secondary
182
metabolites produced by members of the Brassicaceae that contain a sulfur and glucose
moiety bonded to an indolyl or aliphatic side chain (see review by Hopkins et al. 2009).
Glucosinolates are feeding deterrents to many generalist insects (Beekwilder et al. 2008;
Agrawal & Kurashige 2003), yet also serve as oviposition cues and feeding cues for
specialists (DeVos et al. 2008; Smallgange et al. 2007). The role of ERF transcription
factors in glucosinolate production and regulation is unclear. The regulation of GS-
related genes is largely under control of MYB family TFs (Celenza et al. 2005; Hirai et
al. 2007), thus this is the first study to propose a link between glucosinolate metabolism
and ERF signaling in Arabidopsis. Furthermore, there is limited evidence for the role of
ethylene in GS production. In one study, Mewis et al. (2005) found that elicitation of GS
after insect attack by S. exigua required intact ET signaling through ETR1. Therefore, we
predicted that intact ET signaling is required for GS elicitation and plants impaired in erf
responses would also have lower GS production after S. exigua herbivory. As a
consequence, we expected higher growth rates and relative consumption rates of S.
exigua on erf mutants. Because P. rapae can detoxify GS and uses GSs as feeding cues,
we expected any impaired GS elicitation in erf plants to have little to no effect on growth
rates and possibly have a negative effect on the relative consumption of tissue.
Our results showed an increased production of induced indolyl glucosinolates from basal
levels by both insects in all genotypes (Chapter 4, Figure 4.3). However, erf104 plants
had higher levels of induced aliphatic and indolyl GSs after S. exigua feeding, which is
the opposite of what we expected. Mewis et al. (2006) found Arabidopsis indolyl GS
levels were significantly increased relative to controls after S. exigua feeding, but not P.
183
rapae feeding. In our Pieris bioassay, the glucosinolate chemistry was highly erratic in
both constitutive and induced plants; erf104 plants had lower basal levels of indolyls,
while erf104, erf105 and erf5 had lower levels of aliphatics. Pieris-induced aliphatic GS
were significantly reduced in erf5 and erf6 plants and were no different from basal levels
in erf6 plants.
We conducted regression analyses to determine the correlation between total aliphatic or
indolyl levels and relative consumption rates. A negative relationship (p=0.0280) existed
across all genotypes between total indolyl GS levels and RCR during S. exigua feeding,
however these were not significant by genotype, except for erf5, which actually had a
positive correlation (Chapter 4, Table 4.1). Although induced indolyl and aliphatic GS
levels were higher in erf104 plants, we did not see a statistically significant correlation of
S exigua RCR and GS induction, which we would have seen if S. exigua ate less of
erf104 because these plants had increased GS production. The lack of a significant
correlation may arise from two opposing effects on the relationship between feeding and
GS: although S. exigua may have eaten in proportion to the amount of GS induced, GS
production may also have increased in response to feeding, thus cancelling each other
out. In the P. rapae bioassay, a positive relationship between RCR and GS levels in WT
and erf104 plants was seen. Also, significantly positive correlation among all genotypes
between the amount eaten by P.rapae and induced indolyl levels was observed.
Our results are in line with those of Poehlman et al. (2007) who also found that
performance of generalist herbivores on cruciferous plants was independent of
184
glucosinolate profiles. The authors speculated that other defense-related factors, such as
proteinase inhibitors or phenolic compounds may be more important. Thus, one of the
next lab projects is to use some of the remaining plant tissue from both feeding assays to
measure phenolic compounds in constitutive and induced plants. Additional experiments
would need to be performed to measure either PI gene expression or protein levels via
Western blotting.
Overall Conclusions
There are two major important conclusions that I can draw from this body of work.
First, JA production may be a general stress response during herbivory, but ET
production may mediate additional down-stream responses, and the relative timing of
these events may be critical. ET was induced by both specialist and generalist insects but
the timing of ET release differed. In some studies (Diezel et al. 2009; vonDahl et al.
2007; Schmelz et al. 2006 ), ET levels by insect feeding was only measured at one time
point. Similarly, gene expression should also be examined at multiple time points. Our
results show that levels can fluctuate over time and choosing an arbitrary single time
point to measure hormone production or gene expression can lead to different
conclusions. For example, our gene expression studies suggest that after 6 hours, P.
rapae can induce the expression of PDF1.2. In his thesis DeVos (2006) found that P.
rapae feeding suppressed PDF1.2 production through AtMYC2. Here, P. rapae slightly
suppressed PDF1.2 production after 24 hours, which is in agreement with DeVos 2006.
At earlier time points our data suggest a different signaling pathway after P. rapae
185
feeding; one that may include the induction of PDF1.2 through different TFs such as
TGAs (Zander et al 2009).
Second, ERF104 seems particularly and differentially important to interactions between
Arabidopsis and the two insects we studied. ERF104 is phylogenetically related to other
members of the ERF Group IX such as ERF105, ERF5 and ERF6 (Nakano et al. 2006).
This gene is very responsive to chitin (Libault et al. 2007), yet whether S. exigua is also
using chitin-related signaling is unclear. Also, ERF104 is induced by the Flg22 bacterial
elicitor demonstrating its role as part of the innate immunity response in plants (Bethke et
al. 2009a). Bethke et al. (2009a) found that ERF104 is a MPK6 substrate and its
activation is most likely upstream of ET production because MPK6 transcription and
activity are not affected by ET or ACC (its precursor). ET production then occurs via
post-translational stabilization of ACC Synthase by MPK6 (Liu & Zhang et al. 2004).
However, our results show a rapid increase in ET after S. exigua feeding (after 30
minutes), yet ERF104 activation peaked at 6 hours. Therefore our results do not fit this
current model of ERF104 activation and there are currently unanswered questions
regarding ERF104 function and timing (Bethke et al 2009b). Our results also show that
erf104 plants were more resistant (less eaten) and had increased levels of both indolyl and
aliphatic glucosinolates. This suggests that intact ERF104 genes may be involved in the
negative regulation of defenses specifically after insect attack by S. exigua.
Still, the larger question remains as to what is the source of the differential perception and
response to these two insects. Components in saliva may be part of the answer. To date
186
only one elicitor in Pieris oral secretions, B-glucosidase, has been identified (Mattiaci et
al. 2004). However, responses to S. exigua may originate from several elicitors found in
the saliva of caterpillars in the genus Spodoptera. Fatty-acid-JA conjugates like volicitin,
as well as ATPase derived peptides called inceptins, rapidly trigger JA and ET production
in corn (Schmelz et al. 2003) and cowpea (Schmelz et al. 2007). Recently, Schmelz et
al. 2009 found that neither volicitin or inceptins from S. exigua increased JA or ET levels
in Arabidopsis that were greater than damaged controls. Only caeliferins from
grasshopper oral secretions elicited an ET or JA response. Using SAR- (ndr1, npr1)
plants, Weech et al. (2008) suggested that a component in S.exigua saliva negatively
regulates JA signaling, but this is independent of SA. Our data also support this as
neither insect induced the production of SA. So what in S. exigua saliva might be
triggering ET production and differential gene expression compared to P. rapae feeding?
One suggestion could be bacterial endosymbionts from S. exigua oral secretions,
however, there is little evidence in the literature to support this We had our insect species
tested for Wolbachia infection, but the PCR results came back negative for both species
(Kristen Leach, personal communication).
Another probable trigger is glucose oxidase (GOX), an H2O2 generating enzyme found in
S. exigua saliva (Diezel et al. 2009; Bede et al. 2006; Musser et al. 2005; Felton and
Eichenseer, 1999). Rapid ROS signaling, possibly generated from H2O2, has been
demonstrated after S. exigua feeding (Maffei et al. 2006; 2007). Furthermore, oxidative
product generation after S. exigua but not P. rapae feeding also explains the patterns of
C2H2 Transcription Factors that we observed in Chapter 1 (see Table 1.1). These were
187
highly induced by S. exigua but not P. rapae. These TFs have well-documented functions
in responses to oxidative stress signaling (Devletova et al. 2005; Englbrecht et al. 2004).
To my knowledge, no papers have been published that examined the effects of GOX in
Arabidopsis, therefore, this should also be a future experiment for our lab to conduct. If
the results from these studies are comparable to S. exigua feeding assays, then we are one
step closer to identifying GOX as a major differential signal between S. exigua and P.
rapae feeding.
My research shows that signaling after feeding by two different caterpillars does not
entail a “one size fits all” response. The initial perception of each insect may start at the
feeding site, but the downstream release of hormones, activation of transcription factors,
transcription of defense genes, and production of secondary metabolites indicate how the
coordinated timing of early signaling events channel signals into two distinct responses.
Our observations of insect resistance phentoypes in erf mutants and insect-specific gene
expression patterns clearly demonstrate that ERFs are playing a crucial role in the
differential responses to specialist and generalist insects in Arabidopsis.
188
References
1. Abe H, Urao T, Ito T, Seki M, Shinozaki K, Yamaguchi-Shinozaki K (2003)
Arabidopsis AtMYC2 (bHLH) and AtMYB2 (MYB) function as transcriptional activators in abscisic acid signaling. The Plant Cell 15: 63-78
2. Agrawal AA, Kurashige NS (2003) A Role for Isothiocyanates in plant resistance
against the specialist herbivore Pieris rapae. Journal of Chemical Ecology 29: 1403-1415
3. Alborn HT, Hansen TV, Jones TH, Bennett DC, Tumlinson JH, Schmelz EA, Teal
PEA (2007) Disulfooxy fatty acids from the American bird grasshopper Schistocerca americana, elicitors of plant volatiles. Proceedings of the National Academy of Sciences 104: 12976-12981
4. Ballare CL (2009) Illuminated behaviour: phytochrome as a key regulator of light
foraging and plant anti-herbivore defence. Plant, Cell & Environment 32: 713-725 5. Beckers GJ, Spoel SH (2006) Fine-tuning plant defence signalling: salicylate versus
jasmonate. Plant Biology 8: 1-10 6. Bede J, Musser R, Felton G, Korth K (2006) Caterpillar herbivory and salivary
enzymes decrease transcript levels of Medicago truncatula genes encoding early enzymes in terpenoid biosynthesis. Plant Molecular Biology 60: 519-531
7. Beekwilder J, van Leeuwen W, van Dam NM, Bertossi M, Grandi V, Mizzi L,
Soloviev M, Szabados L, Molthoff JW, Schipper B, Verbocht H, de Vos RC, Morandini P, Aarts MG, Bovy A (2008) The impact of the absence of aliphatic glucosinolates on insect herbivory in Arabidopsis. PLoS ONE 3: e2068
8. Bethke G, Scheel D, Lee J (2009) Sometimes new results raise new questions: The
question marks between mitogen-activated protein kinase and ethylene signaling. Plant Signaling and Behavior 4: 672-674
9. Bethke G, Unthan T, Uhrig JF, Poschl Y, Gust AA, Scheel D, Lee J (2009) Flg22
regulates the release of an ethylene response factor substrate from MAP kinase 6 in Arabidopsis thaliana via ethylene signaling. Proceedings of the National Academy of Sciences 106: 8067-8072
10. Brunner AM, Yakovlev IA, Strauss SH (2004) Validating internal controls for
quantitative plant gene expression studies. BMC Plant Biology 4: 14
189
11. Celenza JL, Quiel JA, Smolen GA, Merrikh H, Silvestro AR, Normanly J, Bender J (2005) The Arabidopsis ATR1 Myb transcription factor controls indolic glucosinolate homeostasis. Plant Physiology 137: 253-262
12. Chen W, Provart NJ, Glazebrook J, Katagiri F, Chang H-S, Eulgem T, Mauch F,
Luan S, Zou G, Whitham SA, Budworth PR, Tao Y, Xie Z, Chen X, Lam S, Kreps JA, Harper JF, Si-Ammour A, Mauch-Mani B, Heinlein M, Kobayashi K, Hohn T, Dangl JL, Wang X, Zhu T (2002) Expression profile matrix of Arabidopsis transcription factor genes suggests their putative functions in response to environmental stresses. The Plant Cell 14: 559-574
13. Chung HS, Koo AJK, Gao X, Jayanty S, Thines B, Jones AD, Howe GA (2008)
Regulation and function of Arabidopsis JASMONATE ZIM-domain genes in response to wounding and herbivory. Plant Physiology 146: 952-964
14. Cipollini D, Enright S, Traw MB, Bergelson J (2004) Salicylic acid inhibits jasmonic
acid-induced resistance of Arabidopsis thaliana to Spodoptera exigua. Molecular Ecology 13: 1643-1653
15. Clauss MJ, Dietel S, Schubert G, Mitchell-Olds T (2006) Glucosinolate and trichome
defenses in a natural Arabidopsis lyrata population. Journal of Chemical Ecology 32: 2351-2373
16. Czechowski T, Stitt M, Altmann T, Udvardi MK, Scheible W-R (2005) Genome-
wide identification and testing of superior reference genes for transcript normalization in Arabidopsis. Plant Physiology 139: 5-17
17. Davletova S, Schlauch K, Coutu J, Mittler R (2005) The zinc-finger protein Zat12
plays a central role in reactive oxygen and abiotic stress signaling in Arabidopsis. Plant Physiology 139: 847-856
18. Delessert C, Wilson IW, Van Der Straeten D, Dennis ES, Dolferus R (2004) Spatial
and temporal analysis of the local response to wounding in Arabidopsis leaves. Plant Molecular Biology 55: 165-181
19. DeVos M (2006) Signal signature, transcriptomics, and effectiveness of induced
pathogen and insect resistance in Arabidopsis. Ph.D. University of Utrecht, Utrecht, Netherlands
20. DeVos M, Kriksunov KL, Jander G (2008) Indole-3-acetonitrile production from
indole glucosinolates deters oviposition by . Plant Physiology 146: 916-926 21. DeVos M, Van Oosten VR, Van Poecke RMP, Van Pelt JA, Pozo MJ, Mueller MJ,
Buchala AJ, Metraux J-P, Van Loon LC, Dicke M, Pieterse C, M. J. (2005) Signal
190
dignature and transcriptome changes of Arabidopsis during pathogen and insect attack. Molecular Plant-Microbe Interactions 18: 923-937
22. Diezel C, von Dahl CC, Gaquerel E, Baldwin IT (2009) Different lepidopteran
elicitors account for cross-talk in herbivory-induced phytohormone signaling. Plant Physiology 150: 1576-1586
23. Durrant WE, Dong X (2004) Systemic Acquired Resistance. Annual Review of
Phytopathology 42: 185-209 24. Englbrecht C, Schoof H, Bohm S (2004) Conservation, diversification and expansion
of C2H2 zinc finger proteins in the Arabidopsis thaliana genome. BMC genomics 5: 39
25. Felton GW, Korth KL (2000) Trade-offs between pathogen and herbivore resistance.
Current Opinion in Plant Biology 3: 309-314 26. Gutterson N, Reuber TL (2004) Regulation of disease resistance pathways by
AP2/ERF transcription factors. Current Opinion in Plant Biology 7: 465-471 27. Halitschke R, Baldwin IT (2003) Antisense LOX expression increases herbivore
performance by decreasing defense responses and inhibiting growth-related transcriptional reorganization in Nicotiana attenuata. The Plant Journal: for Cell and Molecular Biology 36: 794-807
28. Harfouche AL, Shivaji R, Stocker R, Williams PW, Luthe DS (2006) Ethylene
signaling mediates a maize defense response to insect herbivory. Molecular Plant-Microbe Interactions 19: 189-199
29. Harmer SL, Hogenesch JB, Straume M, Chang H-S, Han B, Zhu T, Wang X, Kreps
JA, Kay SA (2000) Orchestrated transcription of key pathways in Arabidopsis by the circadian clock. Science 290: 2110-2113
30. Hirai MY, Sugiyama K, Sawada Y, Tohge T, Obayashi T, Suzuki A, Araki R,
Sakurai N, Suzuki H, Aoki K, Goda H, Nishizawa OI, Shibata D, Saito K (2007) Omics-based identification of Arabidopsis Myb transcription factors regulating aliphatic glucosinolate biosynthesis. Proceedings of the National Academy of Sciences 104: 6478-6483
31. Hopkins RJ, van Dam NM, van Loon JJA (2009) Role of glucosinolates in insect-
plant relationships and multitrophic interactions. Annual Review of Entomology 54: 57-83
191
32. Hudgins JW, Franceschi VR (2004) Methyl jasmonate-induced ethylene production is responsible for conifer phloem defense responses and reprogramming of stem cambial zone for traumatic resin duct formation. Plant Physiology 135: 2134-2149
33. Kahl J, Siemens DH, Aerts RJ, Gäbler R, Kühnemann F, Preston CA, Baldwin IT
(2000) Herbivore-induced ethylene suppresses a direct defense but not a putative indirect defense against an adapted herbivore. Planta 210: 336-342
34. Kandoth PK, Ranf S, Pancholi SS, Jayanty S, Walla MD, Miller W, Howe GA,
Lincoln DE, Stratmann JW (2007) Tomato MAPKs LeMPK1, LeMPK2, and LeMPK3 function in the systemin-mediated defense response against herbivorous insects. Proceedings of the National Academy of Sciences 104: 12205-12210
35. Kessler A, Baldwin IT (2002) Plant responses to insect herbivory: the emerging
molecular analysis. Annual Review of Plant Biology 53: 299-328 36. Libault M, Wan J, Czechowski T, Udvardi M, Stacey G (2007) Identification of 118
Arabidopsis transcription factor and 30 Ubiquitin-Ligase genes responding to chitin, a plant-defense elicitor. Molecular Plant-Microbe Interactions 20: 900-911
37. Liu Y, Zhang S (2004) Phosphorylation of 1-aminocyclopropane-1-carboxylic acid
synthase by MPK6, a stress-responsive mitogen-activated protein kinase, induces ethylene biosynthesis in Arabidopsis. The Plant Cell 16: 3386-3399
38. Lorenzo O, Chico JM, Sanchez-Serrano JJ, Solano R (2004) JASMONATE-
INSENSITIVE1 encodes a MYC transcription factor essential to discriminate between different jasmonate-regulated defense responses in Arabidopsis. The Plant Cell 16: 1938-1950
39. Lorenzo O, Solano R (2005) Molecular players regulating the jasmonate signalling
network. Current Opinion in Plant Biology 8: 532-540 40. Maffei ME, Mithofer A, Arimura G-I, Uchtenhagen H, Bossi S, Bertea CM, Cucuzza
LS, Novero M, Volpe V, Quadro S, Boland W (2006) Effects of feeding Spodoptera littoralis on lima bean leaves. III. Membrane depolarization and involvement of hydrogen peroxide. Plant Physiology 140: 1022-1035
41. Maffei ME, Mithofer A, Boland W (2007) Before gene expression: early events in
plant-insect interaction. Trends in Plant Science 12: 310-316 42. Mattiacci L, Dicke M (2004) b-Glucosidase: an elicitor of herbivore-induced plant
odor that attracts host-searching parasitic wasps. Proceedings of the National Academy of Sciences 92: 12837-12842
192
43. Mewis I, Appel HM, Hom A, Raina R, Schultz JC (2005) Major signaling pathways modulate Arabidopsis glucosinolate accumulation and response to both phloem-feeding and chewing insects. Plant Physiology 138: 1149-1162
44. Mewis I, Tokuhisa JG, Schultz JC, Appel HM, Ulrichs C, Gershenzon J (2006) Gene
expression and glucosinolate accumulation in Arabidopsis thaliana in response to generalist and specialist herbivores of different feeding guilds and the role of defense signaling pathways. Phytochemistry 67: 2450-2462
45. Musser RO, Cipollini DF, Hum-Musser SM, Williams SA, Brown JK, Felton GW
(2005) Evidence that the caterpillar salivary enzyme glucose oxidase provides herbivore offense in solanaceous plants. Archives of Insect Biochemistry and Physiology 58: 128-137
46. Onate-Sanchez L, Anderson JP, Young J, Singh KB (2007) AtERF14, a member of
the ERF family of transcription factors, plays a non-redundant role in plant defense. Plant Physiology 143: 400-409
47. Paschold A, Halitschke R, Baldwin IT (2007) Co(i)-ordinating defenses: NaCOI1
mediates herbivore- induced resistance in Nicotiana attenuata and reveals the role of herbivore movement in avoiding defenses. The Plant Journal 51: 79-91
48. Poelman EH, Galiart RJFH, Raaijmakers CE, van Loon JJA, van Dam NM (2008)
Performance of specialist and generalist herbivores feeding on cabbage cultivars is not explained by glucosinolate profiles. Entomologia Experimentalis et Applicata 127: 218-228
49. Reymond P, Bodenhausen N, Van Poecke RMP, Krishnamurthy V, Dicke M, Farmer
EE (2004) A conserved transcript pattern in response to a specialist and a generalist herbivore. The Plant Cell 16: 3132-3147
50. Riechmann JL, Heard J, Martin G, Reuber L, -Z. C, Jiang, Keddie J, Adam L, Pineda
O, Ratcliffe OJ, Samaha RR, Creelman R, Pilgrim M, Broun P, Zhang JZ, Ghandehari D, Sherman BK, -L. Yu G (2000) Arabidopsis transcription factors: Genome-wide comparative analysis among eukaryotes. Science 290: 2105-2110
51. Schlaeppi K, Bodenhausen N, Buchala A, Mauch F, Reymond P (2008) The
glutathione-deficient mutant pad2-1 accumulates lower amounts of glucosinolates and is more susceptible to the insect herbivore Spodoptera littoralis. The Plant Journal : for Cell and Molecular Biology 55: 774-786
52. Schmelz EA, Alborn HT, Banchio E, Tumlinson JH (2003) Quantitative relationships
between induced jasmonic acid levels and volatile emission in Zea mays during Spodoptera exigua herbivory. Planta 216: 665-673
193
53. Schmelz EA, Carroll MJ, LeClere S, Phipps SM, Meredith J, Chourey PS, Alborn HT, Teal PEA (2006) Fragments of ATP synthase mediate plant perception of insect attack. Proceedings of the National Academy of Sciences 103: 8894-8899
54. Schmelz EA, Engelberth J, Alborn HT, Tumlinson JH, Teal PEA (2009)
Phytohormone-based activity mapping of insect herbivore-produced elicitors. Proceedings of the National Academy of Sciences 106: 653-657
55. Schmelz EA, LeClere S, Carroll MJ, Alborn HT, Teal PE (2007) Cowpea
chloroplastic ATP synthase is the source of multiple plant defense elicitors during insect herbivory. Plant Physiology 144: 793-805
56. Schultz JC (2002) Biochemical ecology: how plants fight dirty. Nature 416: 267 57. Stotz HU, Koch T, Biedermann A, Weniger K, Boland W, Mitchell-Olds T (2002)
Evidence for regulation of resistance in Arabidopsis to Egyptian cotton worm by salicylic and jasmonic acid signaling pathways. Planta 214: 648-652
58. Stotz HU, Pittendrigh BR, Kroymann J, Weniger K, Fritsche J, Bauke A, Mitchell-
Olds T (2000) Induced plant defense responses against chewing insects. Ethylene signaling reduces resistance of Arabidopsis against Egyptian cotton worm but not diamondback moth. Plant Physiology 124: 1007-1018
59. Thaler JS, Farag MA, Pare PW, Dicke M (2002) Jasmonate-deficient plants have
reduced direct and indirect defences against herbivores. Ecology Letters 5: 764-774 60. Truitt CL, Wei H-X, Pare PW (2004) A plasma membrane protein from Zea mays
binds with the herbivore elicitor volicitin. The Plant Cell 16: 523-532 61. Van Oosten VR, Bodenhausen N, Reymond P, Van Pelt JA, Van Loon LC, Dicke M,
Pieterse CMJ (2008) Differential effectiveness of microbially induced resistance against herbivorous insects in Arabidopsis. Molecular Plant-Microbe Interactions 21: 919-930
62. Van Poecke Remco MP (2009) Arabidopsis-Insect Interactions. In The Arabidopsis
Book. The American Society of Plant Biologists, pp 1-34 63. Vogel H, Kroymann J, Mitchell-Olds T (2007) Different transcript patterns in
response to specialist and generalist herbivores in the wild Arabidopsis relative Boechera divaricarpa. PLoS ONE 2: e1081
64. von Dahl C, Baldwin I (2007) Deciphering the role of ethylene in plant–herbivore
interactions. Journal of Plant Growth Regulation 26: 201-209
194
65. von Dahl CC, Winz RA, Halitschke R, Kuhnemann F, Gase K, Baldwin IT (2007) Tuning the herbivore-induced ethylene burst: the role of transcript accumulation and ethylene perception in Nicotiana attenuata. The Plant Journal 51: 293-307
66. Walley JW, Coughlan S, Hudson ME, Covington MF, Kaspi R, Banu G, Harmer SL,
Dehesh K (2007) Mechanical stress induces biotic and abiotic stress responses via a novel cis-element. PLoS Genetics 3: 1800-1812
67. Weech M, Chapleau M, Pan L, Ide C, Bede JC (2008) Caterpillar saliva interferes
with induced Arabidopsis thaliana defence responses via the systemic acquired resistance pathway. Journal of Experimental Botany 59: 2437-2448
68. Winz RA, Baldwin IT (2001) Molecular interactions between the specialist herbivore
Manduca sexta (Lepidoptera, Sphingidae) and its natural host Nicotiana attenuata. IV. Insect-Induced ethylene reduces jasmonate-induced nicotine accumulation by regulating putrescine N-methyltransferase transcripts. Plant Physiology 125: 2189-2202
69. Wittstock U, Agerbirk N, Stauber EJ, Olsen CE, Hippler M, Mitchell-Olds T,
Gershenzon J, Vogel H (2004) Successful herbivore attack due to metabolic diversion of a plant chemical defense. Proceedings of the National Academy of Sciences 101: 4859-4864
70. Yu D, Chen C, Chen Z (2001) Evidence for an important role of WRKY DNA
binding proteins in the regulation of NPR1 gene expression. The Plant Cell 13: 1527-1540
71. Zander M, La Camera S, Lamotte O, Metraux JP, Gatz C (2009) Arabidopsis thaliana
class-II TGA transcription factors are essential activators of jasmonic acid/ethylene-induced defense responses. The Plant Journal 61: 200-210
195
VITA
Erin MacNeal Rehrig was born in Poughkeepsie, NY and grew up outside of Scranton,
PA. She attended Dunmore High School in Dunmore, PA. She obtained a B.S. in
Biology and a minor in Chemistry from Bloomsburg University of Pennsylvania. She
completed a M.S. in Horticulture from the Pennsylvania State University. She also
received a M.Ed. in Science Education from the Pennsylvania State University. She
earned her Ph.D. in Plant, Insect, and Microbial Sciences in May 2010 from the
University of Missouri.