Science - USA (2022-02-04)

NaJAZi, andNaJAZjnegatively correlated
with theEmpoascanumbers and damage.
The expression of JAZ genes known to me-
diateM. sextadefense responses, such as
NaJAZh( 17 ), showed no significant correla-
tions. JA, but not the canonical mediator of
JA signaling, JA-Ile, was negatively correlated
with theEmpoascanumbers and damage, and

negative correlations were also observed for

the hydroxylated and carboxylated JAs (OH-JA,

OH-JA-Ile, and COOH-JA-Ile) and JA-valine

conjugates (JA-Val). These results provided

forward-genetics confirmation of the central

role of JA signaling and pointed to down-

stream components likely involved in the

Empoascaleafhopper resistance response.

An elicited JA-JAZi module regulates

Empoascaresistance

To further disentangle the intricacies of the

Empoasca-elicited JA signaling sector and

its regulated downstream metabolic signa-

tures responsible forEmpoascaresistance, we

adopted a reverse-genetics approach to exam-

ine the involvement of phenolamides. Isogenic

Baiet al.,Science 375 , eabm2948 (2022) 4 February 2022 2of9

A

NaLOX3

NaAOS

NaAOC

NaOPR3

NaACX

NaJAR4

NaJAR6

NaJIH1

NaCOI1

NaJAZa

NaJAZb

NaJAZc

NaJAZd

NaJAZe

NaJAZf

NaJAZg

NaJAZh

NaJAZi

NaJAZj

NaJAZk

NaJAZl

NaJAZn

NaMYB8

NaMYC2a

NaMYC2b

NaNINJA

NaSIPK

NaWIPK

#E

mpoas

ca

%Dam

age

PCC

-0.2 0.1

P<0.05

JA

JA-Ile

OH-JA

OH-JA-Ile

COOH-JA-Ile

JA-Val

ABA

SA

#Empoas

ca

%Da

mage

PCC

-0.1 0.05

P<0.05

Transcriptomes

Phytohormones

Metabolomes

Transcriptomes

350 samples

Phytohormones

JA-deficient line

Empoasca spp.

MAGIC Population in Field

Metabolites

Phenolamides

Alkaloids

Quinate conjugates

Flavonoid glycosides

HGL-DTGs

O-acyl sugars

Amino acids

JA-related genes

SNPs

Jasmonates

ABA

SA

Phytohormones

PCC

-0.2 0.2

P<0.05 P<0.05

Coumaroylputrescine

Caffeoylputrescine

Feruloylputrescine

Caffeoylspermidine

Coumaroyl,caffeoylspermidineFeruloylspermidine

Di-caffeoylspermidine

Caffeoyl,feruloylspermidine

Di-feruloyl-spermidine

Nicotine

Nicotinic acid

Chlorogenic acid

Coumaroylquinic acidFeruloylquinic acid

Kaempferol-glucose-rhamnose

Lyciumoside I

Lyciumoside II

Attenoside

Lyciumoside IV

Nicotianoside IINicotianoside I

Nicotianoside III

Nicotianoside IV

Nicotianoside V

Nicotianoside VI

Nicotianoside VII

O-acyl sugar #2, class 2

OO-acyl sugar #3, class 2-acyl sugar #4, class 2

O-acyl sugar #5, class 2

O-acyl sugar #6, class 2

O-acyl sugar #9, class 3

O-acyl sugar #10, class 3

OO-acyl sugar #11, class 3-acyl sugar #14, class 4

O-acyl sugar #15, class 4

Phenylalanine

Tryptophan

Tyrosine

#E

mpoasca

%Da

mage

Metabolomes

B

Empoasca

Phenotypes

1816 samples

1706 samples

C

(eQTL) (mQTL)

(mQTL)

Correlation (PCC)

Co

rrelation (PCC)

Correlation (PCC)

D

Caffeoylputrescine

NaMYC2b

NaAOC

Feruloylputrescine

Coumaroylputrescine

NaJAZi

NaMYB8

NaMYC2a

JA-Ile

674 lines

(1) Correlation between

metabolites/transcripts

and Empoasca numbers

(2) Correlation between

metabolites/transcripts

and Empoasca damage

(1)(2)

(1)(2)

(1)(2)

O-acyl sugar #8, class 3

Fig. 1. A native insectÐguided multi-omics atlas of forward-genetics lines of
a field-grownN.attenuataMAGIC population highlights deviations from
canonical JA signaling forEmpoascaleafhopper nonhost resistance.(A) Field
plantation of the MAGIC RIL population of nativeN. attenuatatobacco plants in
their native habitat at the WCCER field station in Arizona, USA. Native opportunistic
Empoascaleafhopper communities and their feeding damage on leaves are
illustrated. Leafhoppers attack these plants in a JA-dependent manner and
preferentially select JA-deficient plants as hosts (images: R. Halitschke, D. Kessler,
A. Kessler). (B) Schematic of high-throughput phenotyping of phytohormones
(1816 samples), transcriptomes (350 samples), and metabolomes (1706 samples)
andEmpoascaphenotypes (1907 observations ofEmpoascaabundance and
leaf damage) of 674 MAGIC RILs and parental lines in the field. To remove
confounding factors caused by the stochastic nature of insect attack in nature,
we mimicked herbivore feeding by immediately applying freshly collected oral
secretions ofM. sextalarvae to standardized puncture wounds in leaves at
standardized leaf positions. This procedure, referred to as W + OS treatment,

standardizes the elicitation by precisely timing the initiation of herbivory-elicited

responses. The phytohormone and transcriptome datasets were collected 1 hour

after W + OS treatment and the metabolome dataset after 72 hours. PCC,

Pearson correlation coefficient. (C) Coassociation network built from correlations

among metabolomes, transcriptomes, phytohormones, and SNPs (PCC cutoff,

P< 0.05), and the top five most significant SNPs for each gene (small gray

circles), metabolites, or phytohormones from eQTL or mQTL imputations were

retained. Phytohormones and metabolites of different compound classes and

JA-related genes are labeled with different colors. JA-Ile, JA signaling genes

(NaAOC,NaMYC2a,NaMYC2b, andNaJAZi),NaMYB8, and phenolamides

(N-coumaroylputrescine,N-caffeoylputrescine, andN-feruloylputrescine) form

a central cluster and are labeled. See fig. S3 for identities of the more peripheral

nodes. ABA, abscisic acid; SA, salicylic acid. (D) Heatmap of coexpressions

among metabolites, JA-related genes, and phytohormones withEmpoasca

phenotypes (numbers and area damaged) calculated as pairwise PCC (only

significant correlations withP< 0.05 are shown with colors).

RESEARCH | RESEARCH ARTICLE