Abscisic acid is essential for rewiring of jasmonic acid-dependent defenses during herbivory

TL;DR: Production of ABA induced in response to leaf-chewing Pieris rapae caterpillars is required for both the activation of the MYC-branch and the suppression of the ERF-branches during herbivory, indicating that upon feeding by P. rapae, ABA is essential for activating theMYC- Branch and suppressing the ERf-br branch of the JA pathway, which maximizes defense against caterpillar.

Abstract: Jasmonic acid (JA) is an important plant hormone in the regulation of defenses against chewing herbivores and necrotrophic pathogens. In Arabidopsis thaliana, the JA response pathway consists of two antagonistic branches that are regulated by MYC- and ERF-type transcription factors, respectively. The role of abscisic acid (ABA) and ethylene (ET) in the molecular regulation of the MYC/ERF antagonism during plant-insect interactions is still unclear. Here, we show that production of ABA induced in response to leaf-chewing Pieris rapae caterpillars is required for both the activation of the MYC-branch and the suppression of the ERF-branch during herbivory. Exogenous application of ABA suppressed ectopic ERF-mediated PDF1.2 expression in 35S::ORA59 plants. Moreover, the GCC-box promoter motif, which is required for JA/ET-induced activation of the ERF-branch genes ORA59 and PDF1.2, was targeted by ABA. Application of gaseous ET counteracted activation of the MYC-branch and repression of the ERF-branch by P. rapae, but infection with the ET-inducing necrotrophic pathogen Botrytis cinerea did not. Accordingly, P. rapae performed equally well on B. cinerea-infected and control plants, whereas activation of the MYC-branch resulted in reduced caterpillar performance. Together, these data indicate that upon feeding by P. rapae, ABA is essential for activating the MYC-branch and suppressing the ERF-branch of the JA pathway, which maximizes defense against caterpillars.

Summary

Results

• ABA-and ET-dependency of JA-dependent defense gene expression upon P. rapae feeding Here, the authors investigated whether ABA and ET have a role in the differential expression of the MYC-and the ERF-branch during induction of JA-dependent defense signaling by P. rapae feeding.
• Expression of the MYC-branch marker gene VSP2 and the ERFbranch marker gene PDF1.2 was monitored in wild-type Col-0, MYC2-impaired mutant jin1-7 (hereafter called myc2), MYC2, MYC3, MYC4 triple mutant myc2,3,4, ABA biosynthesis mutant aba2-1 and ET response mutant ein2-1. First-instar P. rapae caterpillars were allowed to feed for 24 h on the different Arabidopsis genotypes, after which they were removed.
• CC-BY-NC-ND 4.0 International license available under a not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• The copyright holder for this preprint (which was this version posted August 28, 2019.
• PDF1.2 transcript levels were very low in both Col-0 and ein2-1.

Hormone accumulation upon P. rapae feeding

• To study whether the mutants used in this study are affected in herbivore-induced levels of jasmonates (JAs; JA, the biologically highly active conjugate JA-Ile and the JA-precursor OPDA) and ABA the authors monitored their accumulation in response to P. rapae feeding.
• Subsequently, hormone levels were measured in caterpillar-damaged leaves at different time points after caterpillar removal.
• Figure 2 shows that P. rapae feeding induced the accumulation of JA, JA-Ile, OPDA and ABA in Col-0 wild-type plants, confirming previous findings (Vos et al., 2013b) .
• This indicates that the biosynthesis of JAs is not significantly affected by the myc2 mutation and only relatively late affected by the aba2-1 mutation.
• The positive control, infection with the necrotrophic fungus B. cinerea, showed strongly enhanced ET production , whereas P. rapae infestation did not lead to changes in ET production over a 72-h feeding period compared to non-treated control plants .

The role of ABA in regulation of MYC/ERF antagonism

• To further investigate the role of ABA in the regulation of the MYC/ERF antagonism upon feeding by P. rapae, the authors determined the effect of exogenously applied ABA on the P. rapae-induced expression levels of VSP2 and PDF1.2.
• On the other hand, ABA application diminished the high P. rapae-induced PDF1.2 transcript levels in myc2 and aba2-1 plants at 30 h. CC-BY-NC-ND 4.0 International license available under a not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• The copyright holder for this preprint (which was this version posted August 28, 2019.
• This indicates that ABA antagonizes the activation of the ERF-branch independently of the MYC2, MYC3 and MYC4 transcription factors.
• To investigate whether the preference of P. rapae caterpillars for the ERFbranch-expressing myc2 and aba2-1 mutant plants coincides with increased performance of the caterpillars on these genotypes, the authors assessed their growth in nochoice assays with Col-0, myc2, myc2,3,4, aba2-1, ein2-1, and JA-nonresponsive coi1-1 plants.

Discussion

• The complex plant immune regulatory network that is activated upon recognition of attackers is largely controlled by plant hormones (Pieterse et al., 2012) .
• CC-BY-NC-ND 4.0 International license available under a not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• ABA is required for P. rapae-induced activation of the MYC-branch and repression of the ERF-branch Also in maize and rice plants, an increased production of JAs and ABA has been demonstrated upon root herbivory (Erb et al., 2009; Lu et al., 2015) .
• The ABA treatment stimulated the herbivore-induced MYC-branch in Col-0 plants, while in myc2 and myc2,3,4 plants ABA treatment strongly inhibited the enhanced expression of the ERF-branch .

ABA antagonizes the ERF-branch downstream of ORA59 at the GCC-box

• Analysis of the 35S::ORA59 transgenic line showed that ABA is able to suppress PDF1.2 even when ectopic ORA59 expression levels are constitutively high .
• Previously, Van der Does et al. ( 2013) investigated the suppressive effect of SA on JA-induced PDF1.2 expression.
• They also found that SA could suppress activation of PDF1.2 in the 35S::ORA59 line.
• Moreover, they reported that the GCCbox, which is present in the promoter of PDF1.2, and required for the JA-responsive expression, is essential and sufficient for transcriptional suppression by SA.
• Together, these data point towards a similar mechanism for SA-dependent and ABA-dependent suppression of the expression levels of ORA59 and PDF1.2 at the level of transcriptional regulation at the GCC-box.

Strong activation of the ET pathway is necessary for suppression of the MYC-branch

• The production of JA-Ile, JA and especially ABA was enhanced in the ein2-1 plants compared with Col-0 upon P. rapae feeding , suggesting that in wild-type plants basal activity of the ET pathway can inhibit herbivory-induced production of JA and ABA, which tempers the activation of the MYC-branch.
• This ET treatment led to activation of the ERF-branch during P. rapae feeding, while the MYC-branch was suppressed .
• CC-BY-NC-ND 4.0 International license available under a not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• The feeding preference of P. rapae caterpillars for aba2-1 and myc2 plants was not obviously correlated with enhanced performance (weight gain) on these mutants in no-choice assays , which corresponds with the observation that the ERF-branch activating B. cinerea infection or ACC pretreatment did not affect caterpillar performance .
• Plants were cultivated in a growth chamber with a 10-h day and 14-h night cycle at 70% relative humidity and 21°C.

Chemical treatments

• For gene expression analysis, plants were treated with MeJA (Serva, Brunschwig Chemie, Amsterdam, the Netherlands) or ABA (Sigma, Steinheim, Germany) by dipping the rosettes in a solution containing either 100 µM MeJA, 100 µM ABA or a combination of both chemicals and 0.015% (v/v) Silwet L77 (Van Meeuwen Chemicals BV, Weesp, the Netherlands) 24 h before caterpillar feeding.
• MeJA and ABA solutions were diluted from a 1000-fold concentrated stock in 96% ethanol.
• CC-BY-NC-ND 4.0 International license available under a not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• Five-week-old plants were placed separately in the cuvettes and remained there for the duration of the experiment.
• For northern blot analysis, 15 µg of RNA was denatured using glyoxal and dimethyl sulfoxide (Sambrook et al., 1989) , electrophoretically separated on 1.5% agarose gel, and blotted onto Hybond-N + membranes (Amersham, 's-Hertogenbosch, the Netherlands) by capillary transfer.

Jasmonates and ABA analysis

• For JA, JA-Ile, OPDA and ABA concentration analysis, 50-100 mg of P. rapaeinfested damaged leaves as well as undamaged leaves from non-infested control plants were grinded.
• At the start of the extraction 1 ml of cold ethylacetate containing D6-SA (25 ng/ml) and D5-JA (25 ng/ml) was added to the samples as an internal standard in order to calculate the recovery of the hormones measured.
• Multiple reaction monitoring was performed for parent-ions and selected daughter-ions after negative ionization: JA 209/59 (fragmented under 12V collision energy), JA-Ile 322/130 (fragmented under 19V collision energy), OPDA 291/165 (fragmented under 18V collision energy) and ABA 263/153 (fragmented under 9V collision energy).
• Analytes were quantified using standard curves made for each individual compound.

Ethylene measurements

• ET production was measured in a laser-driven photoacoustic detection system (ETD-300, Sensor Sense, Nijmegen, the Netherlands) connected to a 6-channel valve control box in line with a flow-through system (Voesenek et al., 1990) .
• Five-week-old plants were placed in 2-l air-tight cuvettes (four plants per cuvette), which were incubated under growth chamber conditions.
• After an acclimation time of 2 h, the cuvettes were continuously flushed with air (flow rate: 0.9 l/h), directing the flowthrough air from the cuvettes into a photoacoustic cell for ET measurements.
• ET levels were measured over consecutive 0.5 h time intervals, after which the machine switched to the next cuvette (n=6).

GUS assays

• . CC-BY-NC-ND 4.0 International license available under a not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• Shown; two-way ANOVA (treatment x time point), LSD test for multiple comparisons; RT-qPCR analysis of VSP2 and PDF1.2 gene expression at 30 h in leaves of Col-0, myc2, myc2,3,4, aba2-1 and ein2-1 plants that were treated with a mock solution or with 100 µM ABA 24 h prior to infestation with P. rapae.
• Indications above the brackets specify whether there is an overall statistically significant difference between myc2 and Col-0 (two-way ANOVA (treatment x genotype), LSD test for multiple comparisons; *** = P<0.001).

Full text

1
Abscisic acid is essential for rewiring of jasmonic acid-
1
dependent defenses during herbivory
2
3
Irene A Vos
1
, Adriaan Verhage
1
, Lewis G Watt
1
, Ido Vlaardingerbroek
1
,
4
Robert C Schuurink
2
, Corné MJ Pieterse
1
, Saskia CM Van Wees
1
5
6
1
Plant-Microbe Interactions, Department of Biology, Faculty of Science, Utrecht
7
University, P.O. Box 80056, 3508 TB Utrecht, the Netherlands
8
2
Department of Plant Physiology, Swammerdam Institute for Life Sciences,
9
University of Amsterdam, P.O. Box 94215, 1090 GE Amsterdam, the Netherlands
10
11
Address correspondence to s.vanwees@uu.nl
12
13
Short title: ABA differentially affects JA signaling
14
15
The author responsible for distribution of materials integral to the findings presented
16
in this article in accordance with the policy described in the Instructions for Authors
17
(www.plantcell.org) is Saskia Van Wees (s.vanwees@uu.nl).
18
.CC-BY-NC-ND 4.0 International licenseavailable under a
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (which wasthis version posted August 28, 2019. ; https://doi.org/10.1101/747345doi: bioRxiv preprint

2
Abstract
19
Jasmonic acid (JA) is an important plant hormone in the regulation of defenses
20
against chewing herbivores and necrotrophic pathogens. In Arabidopsis thaliana, the
21
JA response pathway consists of two antagonistic branches that are regulated by
22
MYC- and ERF-type transcription factors, respectively. The role of abscisic acid
23
(ABA) and ethylene (ET) in the molecular regulation of the MYC/ERF antagonism
24
during plant-insect interactions is still unclear. Here, we show that production of ABA
25
induced in response to leaf-chewing Pieris rapae caterpillars is required for both the
26
activation of the MYC-branch and the suppression of the ERF-branch during
27
herbivory. Exogenous application of ABA suppressed ectopic ERF-mediated PDF1.2
28
expression in 35S::ORA59 plants. Moreover, the GCC-box promoter motif, which is
29
required for JA/ET-induced activation of the ERF-branch genes ORA59 and PDF1.2,
30
was targeted by ABA. Application of gaseous ET counteracted activation of the
31
MYC-branch and repression of the ERF-branch by P. rapae, but infection with the
32
ET-inducing necrotrophic pathogen Botrytis cinerea did not. Accordingly, P. rapae
33
performed equally well on B. cinerea-infected and control plants, whereas activation
34
of the MYC-branch resulted in reduced caterpillar performance. Together, these data
35
indicate that upon feeding by P. rapae, ABA is essential for activating the MYC-
36
branch and suppressing the ERF-branch of the JA pathway, which maximizes
37
defense against caterpillars.
38
.CC-BY-NC-ND 4.0 International licenseavailable under a
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (which wasthis version posted August 28, 2019. ; https://doi.org/10.1101/747345doi: bioRxiv preprint

3
Introduction
39
In nature plants are a food source for over one million herbivorous insect species
40
(Howe and Jander, 2008). The evolutionary arms race between plants and their
41
herbivorous insect enemies has led to a highly sophisticated defense system in
42
plants that can recognize wounding and oral secretion of the insects and respond
43
with the production of nutritive value-diminishing enzymes, toxic compounds, or
44
predator-attracting volatiles (Kessler and Baldwin, 2002; Lawrence and Koundal,
45
2002; Wittstock et al., 2004; Chen et al., 2005; Mithöfer and Boland, 2012; Dicke,
46
2016). Conversely, insects can estimate the quality and suitability of the plant as a
47
food source by contact chemoreceptors on the insect mouthparts, antennae and tarsi
48
(Howe and Jander, 2008; Appel and Cocroft, 2014; Dicke, 2016). Because plant
49
defenses are costly, they are often only activated in case of insect or pathogen
50
attack and not constitutively expressed (Walters and Heil, 2007; Vos et al., 2013a).
51
The induced immune response is shaped by the induced production of diverse plant
52
hormones. The quantity, composition and timing of the hormonal blend tailors the
53
defense response specifically to the attacker at hand, thereby prioritizing effective
54
over ineffective defenses and minimizing fitness costs (De Vos et al., 2005; Pieterse
55
et al., 2012; Vos et al., 2013a; Vos et al., 2015).
56
Infestation with chewing herbivores or infection with necrotrophic pathogens
57
triggers the production of the plant hormone jasmonic acid (JA), and its bioactive
58
derivative JA-Ile (Creelman et al., 1992; Penninckx et al., 1996). Binding of JA-Ile to
59
the JA receptor complex consisting of the F-box protein COI1 and a JAZ repressor
60
protein (Xie et al., 1998; Yan et al., 2009; Sheard et al., 2010), leads to degradation
61
of JAZ proteins via the 26S proteasome pathway (Chini et al., 2007; Thines et al.,
62
2007). Without JA, JAZ proteins repress JA-responsive gene expression by binding
63
to transcriptional activators, such as MYC2, EIN3 and EIL1 (Pauwels and Goossens,
64
2011; Song et al., 2014b; Caarls et al., 2015). When JA accumulates the JAZ
65
proteins are degraded thereby releasing transcription factors that can activate JA-
66
regulated genes.
67
Within the JA pathway, two distinct, antagonistic branches of transcriptional
68
regulation are recognized; the MYC-branch and the ERF-branch. Feeding by
69
chewing herbivores activates the MYC-branch (Verhage et al., 2011; Vos et al.,
70
2013b). This branch is controlled by the basic helix-loop-helix leucine zipper
71
transcription factors MYC2, MYC3 and MYC4 leading to transcription of hundreds of
72
.CC-BY-NC-ND 4.0 International licenseavailable under a
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (which wasthis version posted August 28, 2019. ; https://doi.org/10.1101/747345doi: bioRxiv preprint

4
JA-responsive MYC-branch regulated genes, including VSP1 and VSP2 (Anderson
73
et al., 2004; Lorenzo et al., 2004; Fernández-Calvo et al., 2011; Niu et al., 2011).
74
Furthermore, previous studies have indicated that ABA plays a co-regulating role in
75
the activation of the MYC-branch (Anderson et al., 2004; Bodenhausen and
76
Reymond, 2007; Sánchez-Vallet et al., 2012; Vos et al., 2013b). For example, in the
77
ABA-deficient mutant aba2-1, expression of the JA-responsive gene VSP1 was
78
reduced upon feeding by caterpillars of Pieris rapae (small cabbage white) compared
79
to wild-type Col-0 plants (Vos et al., 2013b). In contrast to the herbivore-induced
80
MYC-branch, the ERF-branch is activated upon infection with necrotrophic
81
pathogens. The transcription factors EIN3 and EIL1 and the ERF transcription
82
factors ERF1 and ORA59 activate a large set of JA-responsive ERF-branch
83
regulated genes, including PDF1.2 (Caarls et al., 2015). The expression of ERF1,
84
ORA59 and PDF1.2 is impaired in both JA- and ethylene (ET)-insensitive mutants,
85
indicating that joint activation of the JA and ET pathways is necessary for full
86
expression of the ERF-branch (Penninckx et al., 1998; Lorenzo et al., 2003; Pré et
87
al., 2008; Broekgaarden et al., 2015).
88
It has been shown that the ABA co-regulated MYC-branch and the ET co-
89
regulated ERF-branch of the JA pathway antagonize each other. For example, upon
90
infestation with P. rapae caterpillars, the MYC-branch is activated, while the ERF-
91
branch is suppressed (Verhage et al., 2011; Vos et al., 2013b). In myc2 mutant
92
plants, ORA59 and PDF1.2 expression was highly upregulated after feeding by P.
93
rapae, indicating that in wild-type plants, MYC2 represses ORA59 and PDF1.2
94
expression after feeding by P. rapae (Verhage et al., 2011; Vos et al., 2013b).
95
Additionally, exogenously applied ABA had a positive effect on expression of the
96
MYC-branch after feeding by P. rapae (Vos et al., 2013b) and caused suppression of
97
PDF1.2 induction after exogenous application of JA (Anderson et al., 2004).
98
Recently, it was shown that the MYC-branch transcription factors MYC2, MYC3 and
99
MYC4 interact with the ERF-branch transcription factors EIN3 and EIL1 and that they
100
repress each other’s transcriptional activity (Song et al., 2014a).
101
These antagonistic effects between the MYC- and ERF-branch on gene
102
expression levels also have an effect on plant resistance. ABA-deficient mutants
103
have been reported to be more susceptible to herbivory (Thaler and Bostock, 2004;
104
Bodenhausen and Reymond, 2007; Dinh et al., 2013) and more resistant to
105
necrotrophic pathogens (Anderson et al., 2004; Sánchez-Vallet et al., 2012).
106
.CC-BY-NC-ND 4.0 International licenseavailable under a
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (which wasthis version posted August 28, 2019. ; https://doi.org/10.1101/747345doi: bioRxiv preprint

5
Conversely, ET insensitive mutants are in general more susceptible to necrotrophic
107
pathogens and more resistant to herbivorous insects compared to wild-type plants
108
(Van Loon et al., 2006; Broekgaarden et al., 2015). Hence, the interplay between the
109
MYC- and the ERF-branch may allow the plant to activate a specific set of JA-
110
responsive genes that is required for an optimal defense against the attacker
111
encountered (Pieterse et al., 2012).
112
To study the role of ABA and ET in the molecular regulation of the MYC/ERF
113
balance in Arabidopsis thaliana (hereafter Arabidopsis) upon attack by P. rapae, we
114
analyzed hormone signaling mutants for their gene expression response, hormone
115
production and defense against P. rapae. We provide evidence that after P. rapae
116
infestation ABA accumulation plays an essential modulating role in the activation of
117
the MYC-branch, possibly by activating the MYC2, MYC3 and MYC4 transcription
118
factors. Concomitantly, ABA can suppress the ERF-branch independently of the
119
MYC transcription factors, by targeting the GCC-box, which is present in the
120
promoters of ORA59 and PDF1.2. Furthermore, activation of the MYC-branch, either
121
by application of JA or ABA or by using the ein2-1 mutant, resulted in a negative
122
effect on caterpillar performance, whereas activation of the ERF-branch by infection
123
with the necrotrophic pathogen Botrytis cinerea did not.
124
125
Results
126
ABA- and ET-dependency of JA-dependent defense gene expression upon P. rapae
127
feeding
128
The JA-dependent transcriptional response of Arabidopsis to P. rapae feeding is
129
predominantly regulated through activation of the MYC-branch of the JA pathway
130
and concomitant suppression of the ERF-branch (Verhage et al., 2011). Here, we
131
investigated whether ABA and ET have a role in the differential expression of the
132
MYC- and the ERF-branch during induction of JA-dependent defense signaling by P.
133
rapae feeding. Expression of the MYC-branch marker gene VSP2 and the ERF-
134
branch marker gene PDF1.2 was monitored in wild-type Col-0, MYC2-impaired
135
mutant jin1-7 (hereafter called myc2), MYC2, MYC3, MYC4 triple mutant myc2,3,4,
136
ABA biosynthesis mutant aba2-1 and ET response mutant ein2-1. First-instar P.
137
rapae caterpillars were allowed to feed for 24 h on the different Arabidopsis
138
genotypes, after which they were removed. Comparable to Col-0, ein2-1 plants
139
showed strong P. rapae-induced transcription of VSP2 at 24 h and 30 h (Figure 1).
140
.CC-BY-NC-ND 4.0 International licenseavailable under a
not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (which wasthis version posted August 28, 2019. ; https://doi.org/10.1101/747345doi: bioRxiv preprint