See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/326498467

STAYGREEN, STAY HEALTHY: a loss-of-susceptibility mutation in the

STAYGREEN gene provides durable, broad-spectrum disease resistances for

over 50 years of US cucumber production

Article  in  New Phytologist · July 2018

DOI: 10.1111/nph.15353

CITATIONS

0

READS

24

15 authors, including:

Some of the authors of this publication are also working on these related projects:

Watermelon RNA-Seq View project

Screening Watermelon germplasm for markers and resistance to Anthracnose View project

Yuhui Wang

University of Wisconsin–Madison

7 PUBLICATIONS   37 CITATIONS   

SEE PROFILE

Todd C. Wehner

North Carolina State University

208 PUBLICATIONS   2,427 CITATIONS   

SEE PROFILE

Junsong Pan

Shanghai Jiao Tong University

68 PUBLICATIONS   601 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Yuhui Wang on 24 July 2018.

The user has requested enhancement of the downloaded file.

STAYGREEN, STAY HEALTHY: a loss-of-susceptibility mutationin the STAYGREEN gene provides durable, broad-spectrum dis-ease resistances for over 50 years of US cucumber production

Yuhui Wang1, Junyi Tan1, Zhiming Wu1,2, Kyle VandenLangenberg3, Todd C. Wehner3, Changlong Wen4,

Xiangyang Zheng5, Ken Owens5, Alyson Thornton6, Hailey H. Bang6, Eric Hoeft6, Peter A. G. Kraan7,

Jos Suelmann7, Junsong Pan1,8 and Yiqun Weng1,9

1Horticulture Department, University of Wisconsin, Madison, WI 53706, USA; 2Institute of Cash Crops, Hebei Academy of Agriculture & Forestry Sciences, Shijiazhuang, Hebei 050051,

China; 3Horticultural Science Department, North Carolina State University, Raleigh, NC 27695, USA; 4Beijing Vegetable Research Center, Beijing Academy of Agricultural and Forestry

Sciences, Beijing 100097, China; 5Magnum Seeds Inc., Dixon, CA 95620, USA; 6HM Clause Seed Company, Davis, CA 95618, USA; 7Bayer Vegetable Seeds, 6083 AB Nunhem, the

Netherlands; 8Department of Plant Science, School of Agriculture and Biology, Shanghai Jiao Tong University, Shanghai 200241, China; 9USDA-ARS Vegetable Crops Research Unit,

Madison, WI 53705, USA

Author for correspondence:YiqunWengTel: +1 608 262 0028

Email: yiqun.weng@ars.usda.gov

Received: 2 April 2018Accepted: 13 June 2018

New Phytologist (2018)doi: 10.1111/nph.15353

Key words: broad-spectrum resistance,cucumber (Cucumis sativus), durable resis-tance, loss-of-susceptibility, R gene, stay-green.

Summary

� The Gy14 cucumber (Cucumis sativus) is resistant to oomyceteous downy mildew (DM),

bacterial angular leaf spot (ALS) and fungal anthracnose (AR) pathogens, but the underlying

molecular mechanisms are unknown.� Quantitative trait locus (QTL) mapping for the disease resistances in Gy14 and further

map-based cloning identified a candidate gene for the resistant loci, which was validated and

functionally characterized by spatial-temporal gene expression profiling, allelic diversity and

phylogenetic analysis, as well as local association studies.� We showed that the triple-disease resistances in Gy14 were controlled by the cucumber

STAYGREEN (CsSGR) gene. A single nucleotide polymorphism (SNP) in the coding region

resulted in a nonsynonymous amino acid substitution in the CsSGR protein, and thus disease

resistance. Genes in the chlorophyll degradation pathway showed differential expression

between resistant and susceptible lines in response to pathogen inoculation. The causal SNP

was significantly associated with disease resistances in natural and breeding populations. The

resistance allele has undergone selection in cucumber breeding.� The durable, broad-spectrum disease resistance is caused by a loss-of-susceptibility muta-

tion of CsSGR. Probably, this is achieved through the inhibition of reactive oxygen species

over-accumulation and phytotoxic catabolite over-buildup in the chlorophyll degradation

pathway. The CsSGR-mediated host resistance represents a novel function of this highly con-

served gene in plants.

Introduction

Plants have evolved multi-layered resistance mechanisms to pro-tect themselves against pathogen infection. In basal immunity,elicitors, or pathogen/microbe/damage-associated molecular pat-terns (PAMP/MAMP/DAMPs), are recognized by plant patternrecognition receptors (PRRs) resulting in pattern-triggeredimmunity (PTI). PTI commences a cascade of physiologicalchanges in the cellular environment, including the rapid produc-tion of reactive oxygen species (ROS), phosphorylation of mito-gen-activated protein kinases, synthesis of defense hormones andinduction of defense genes (Boller & Felix, 2009). Pathogens cansecrete effectors to counter the effects of PRRs, and plants haveevolved host resistance (R) gene-mediated effector-triggered

immunity (ETI), which is often associated with a hypersensitiveresponse (HR) involving programmed cell death (PCD) (Jones& Dangl, 2006; Thomma et al., 2011; Cui et al., 2015) as well assalicylic acid (SA) and ROS-regulated autophagy (Yoshimotoet al., 2009; Hofius et al., 2011; Zhou et al., 2014).

Most R genes cloned so far encode cell surface or intracellularreceptors, such as receptor-like proteins/kinases (RLPs/RLKs) ornucleotide-binding and leucine-rich repeat receptors (NLRs).Such perception-based ETI often provides strong and fast resis-tance on pathogen infection, but the resistance can also be easilydefeated during the host–pathogen battle (Boutrot & Zipfel,2017; Kourelis & van der Hoorn, 2018). Some R genes rely onthe loss-of-susceptibility mechanism for host resistance; muta-tions in these genes disrupt biological processes that are critical

No claim to original US government works

New Phytologist� 2018 New Phytologist Trust

New Phytologist (2018) 1www.newphytologist.com

Research

for pathogen infection. These loss-of-susceptibility R genes areusually recessive and confer durable, but often partial, resistanceagainst a broad range of pathogens. A prime example of the loss-of-susceptibility mechanism is the barley mlo (mildew locus o)gene that encodes a defective, integral membrane protein whichis important for infection of the powdery mildew pathogen(B€uschges et al., 1997; Humphry et al., 2006, 2010). The mlo-mediated powdery mildew resistance has been found in wheat,grapevine, rice, tomato and cucumber (Elliott et al., 2002; Baiet al., 2008; Nie et al., 2015; Pessina et al., 2016; Acevedo-Garciaet al., 2017). The wheat Lr34, Lr67 and Yr36 also belong to theloss-of-susceptibility R genes. Lr34 and Lr67 encode an ABC(ATP-binding cassette) transporter and a hexose transporter,respectively, that confer durable resistances against different rustand powdery mildew pathogens (Krattinger et al., 2009; Lagudahet al., 2009; Risk et al., 2013; Moore et al., 2015). Yr36 encodes aprotein fused with a serine/threonine kinase and STARTdomain, and confers partial resistance to stripe rusts by reducingthe ability of chloroplast thylakoid ascorbate peroxidases todetoxify peroxides (Fu et al., 2009; Gou et al., 2015). Thedurable and broad-spectrum disease resistances conferred by loss-of-susceptibility R genes are of obvious advantages in crop breed-ing, but they have only been identified in a few plant species (rice,wheat, barley, maize and Arabidopsis). Here, we report a loss-of-susceptibility mutation in the cucumber STAYGREEN (CsSGR)gene that results in durable resistance against multiple pathogensin the field.

The degradation of chlorophyll (Chl) by Chl catabolicenzymes (CCEs) causes the loss of green color that typicallyoccurs during leaf senescence. Mutations in CCE genes mayresult in a delay in foliar senescence, and thus the staygreen phe-notype (Thomas & Ougham, 2014; Kuai et al., 2018). For exam-ple, the STAYGREEN (SGR) gene encoding the magnesiumdechelatase is a key regulator in the Chl degradation pathway(Shimoda et al., 2016). It is the causal gene underlying Mendel’sgreen cotyledon trait in pea (Armstead et al., 2007; Sato et al.,2007). The Arabidopsis mutant nye1-1 (nonyellowing 1) exhibitsthe staygreen phenotype during leaf senescence because of a loss-of-function point mutation in AtSGR1/NYE1 (Cha et al., 2002;Park et al., 2007; Ren et al., 2007). SGR mutants have also beenidentified in rice (Jiang et al., 2007), tomato (Barry et al., 2008),bell pepper (Barry et al., 2008), tall fescue (Wei et al., 2011),Medicago truncatula (Zhou et al., 2011) and soybean (Fang et al.,2014). Other than its basic function in Chl degradation, SGRseems to play roles in root nodule senescence in legumes (Zhouet al., 2011), as well as lycopene and b-carotene biosynthesis intomato fruits (Luo et al., 2013). In Arabidopsis, increased anddecreased AtSGR expression, respectively, accelerated and sup-pressed HR cell death (Mur et al., 2010). The Arabidopsis noc1(nonchlorosis1) mutant is caused by a mutation in the AtSGRgene; the noc1 mutant plants exhibited reduced disease symptomsin response to infection by both the bacterial pathogenPseudomonas syringae pv. tomato (Pst) DC3000 and the fungalpathogen Alternaria brassicicola, but pathogen growth did notshow a significant difference relative to that on the wild-typeplant (Mecey et al., 2011). Thus, the pathogen-induced

expression of AtSGR is a critical step underlying the developmentof plant disease chlorosis (Mecey et al., 2011). In Medicago, theM. truncatula sgr mutant and alfalfa SGR-RNAi lines showedHR-like enhanced cell death on inoculation with Phakopsorapachyrhizi, the causal pathogen of Asian soybean rust, which sug-gested a possible role of SGR in nonhost resistance (Ishiga et al.,2015). However, the role of SGR as an R gene for disease resis-tance has not been reported. Here, we show that CsSGR is thecandidate gene conferring host resistance against the bacterialangular leaf spot (ALS), fungal anthracnose (AR) and oomyce-teous downy mildew (DM) pathogens in cucumber.

DM, ALS and AR are amongst the most important diseases ofcucumber and other cucurbit crops worldwide. The casual agentsare the obligate biotrophic oomycete Pseudoperonospora cubensis,bacterial Pseudomonas syringae pv. lachrymans and the hemi-biotrophic fungal Colletotrichum orbicular (syn. Colletotrichumlagenarium), respectively. Interestingly, the cucumber accessionPI 197087 from India has been known for a long time to be resis-tant to all three pathogens (Barnes, 1966; Peterson, 1975); itsresistances have been used extensively in cucumber breeding inthe USA. Several studies have also investigated the genetic basisof these resistances. For example, the widely used cucumbergermplasm lines Gy14 and WI 2757 are resistant to DM, ALSand AR with the resistances derived from PI 197087 (Petersonet al., 1982; Wyszogrodzka et al., 1987). Barnes & Epps (1952,1954) found the nearly immune resistance to AR of PI 197087in field tests. Fanourakis & Simon (1987) discovered the linkageof DM and AR resistances in WI 2757. ALS resistance in Gy14was controlled by a single recessive gene, psl (Dessert et al., 1982;Olczak-Woltman et al., 2007; Słomnicka et al., 2016). PI197087-derived DM resistance was controlled by a single reces-sive gene (dm1) with a classical HR (Barnes & Epps, 1954; vanVliet & Meysing, 1974, 1977; Fanourakis & Simon, 1987; Ken-nard et al., 1994; Horejsi et al., 2000). The dm1-conferred DMresistance has been widely deployed in commercial cucumbervarieties, which have provided effective protection to cucumberproduction in the USA for over 50 years until 2004, when newDM strain(s) (post-2004 strains) emerged in the cucumber fieldrendering dm1 resistance less effective (Holmes et al., 2006;Thomas et al., 2017). Nevertheless, dm1 still exhibits moderateresistance to the prevailing post-2004 DM strains in the USAand sufficient resistance in many other countries. We haverecently identified CsSGR as the candidate gene for the AR resis-tance locus cla in both Gy14 and WI 2757 (Pan et al., 2018).Here, we report quantitative trait locus (QTL) mapping andcloning of the dm1 and psl loci in Gy14 and WI2757. We showthat dm1, cla and psl are the same locus, and that CsSGR is thecandidate gene for dm1/psl/cla.

Materials and Methods

Plant materials, collection and statistical analysis ofphenotypic data

Gy14 and WI2757 are two inbred lines of cucumber (Cucumissativus L.), both of which have high and moderate resistance to

New Phytologist (2018) No claim to original US government works

New Phytologist� 2018 New Phytologist Trustwww.newphytologist.com

Research

NewPhytologist2

pre- and post-2004 field DM strains, respectively (Call et al.,2012a,b), and high/moderately high resistance to AR and ALS.Three populations were used in QTL mapping for DM and ALSresistance, including 504 recombinant inbred lines (RILs) fromGy149 9930 (G9RIL hereafter), and 87 RILs (WTRIL) and132 F2:3 families (WTF23) from WI27579 True Lemon cross(details in Supporting Information Table S1).

For initial QTL analysis, phenotypic data of the inoculationresponse to DM infection in G9RIL and WTRIL were collectedfrom five experiments (or environments) (NC2013, NC2014,NL2014.G9, NL2014.WT and NC2015) at two locations(NC =Clinton, North Carolina, USA; NL =Nunhems, theNetherlands) across 3 yr (Table S1). At the fine mapping stage,18 recombinant G9RILs were phenotyped in five environmentsin 2 yr (FL2015, IT2015, NC2015, NL2016 and WI2016) (FL,Florida, USA; IT, Italy; WI, Wisconsin, USA). All open fieldexperiments were conducted with natural infection of the DMpathogen, which was presumed to be a mixture of multiplestrains (Wang et al., 2016). For experiments conducted in con-trolled environments, plants were artificially inoculated on one-true-leaf-stage seedlings with sporangial suspension at a concen-tration of 59 104 to 19 105 spores ml�1 by misting the adaxialside of the leaf. Scoring of inoculated and control plants (inocu-lated with water) took place at 10–14 d post-inoculation (dpi).Three criteria were employed in scoring plants: yellowing (Yel),collapsing (Col) and general impression (GI) (Wang et al.,2016). In all NC or NL field trials, disease severity in the RILpopulations was evaluated two to three times, 1 wk apart, whereasone rating was performed for all other experiments.

Phenotyping of ALS pathogen inoculation responses was per-formed in four glasshouse or climate control room experiments(CA2011, WI2011, WI2015 and WI2017) (Table S1). TheP. syringae pv. lachrymans isolate LMG 5456 was used in theCA2011 test (source: A. Zitter, Cornell University, Ithaca, NY,USA). A more virulent P. syringae pv. lachrymans strain main-tained by HM.Clause Seed Company (Sun Prairie, WI, USA)was used in the other three screening tests (WI2011, WI2015and WI2017). Plants were scored at 6–8 dpi qualitatively aseither resistant or susceptible (WI2015) or quantitatively usingthe 1–9 or 1–5 rating scales.

Statistical analysis of phenotypic data was performed in R fol-lowing Wang et al. (2016). Briefly, means of disease scores ofeach F2:3 family or RIL were used in QTL analysis. In replicatedtrials with G9RILs, the means of each RIL were calculated forrating time, replication, trait and experiment; analysis of variance(ANOVA) was performed with the R/LME4 package to estimategenetic and environmental effects. Heritability estimates were cal-culated from variance components. Correlations among traitsacross environments were evaluated with Spearman’s rank corre-lation coefficients (rs).

QTL analysis of DM and ALS resistances and fine mappingof target QTLs

Two linkage maps with 458 single nucleotide polymorphism(SNP) and 240 simple sequence repeat (SSR) loci, developed

previously with 129 G9RILs and 132 WTF23, respectively (Heet al., 2013; Weng et al., 2015), were employed for QTL analysisin this study. A linkage map was developed using SLAF-Seq(specific length amplified fragment sequencing) and 87 WTRILsfollowing Wang et al. (2018). QTL analysis of DM/ALS resis-tances was performed in the R/QTL package with the multiple-QTL (MQM) model (Broman et al., 2003). The logarithm ofthe odds (LOD) threshold for declaring significant QTLs wasestablished using 1000 permutations (P < 0.05). The supportintervals for the detected QTLs were calculated using a 1.5-LODdrop interval.

Molecular marker discovery, fine mapping and candidategene identification for the dm1 and psl loci followed Pan et al.(2017). In brief, DNA sequences in the 1.5-LOD interval ofthe target QTL were extracted from the Gy14 draft genome(V1.0) (Yang et al., 2012); new markers inside this region weredeveloped to genotype a large segregating population to identifyrecombinants, which were then phenotyped in either controlledenvironments or open fields. This process was iterated until thetarget region was narrowed down to < 100 kb, which was thenannotated for gene prediction. Information on all markers usedin mapping, cloning or gene expression analysis is provided inTable S2.

Phylogenetic analysis of SGR homologs

The phylogenetic relationships of 23 SGR protein sequencesfrom 12 species were analyzed, including cucumber CsSGR andSTAYGREEN-LIKE (CsSGRL) and their homologs in 11 otherspecies (Table S3). Sequence alignment and clustering were per-formed with MEGA 7.0 (http://www.megasoftware.net/) usingnearest-neighbor joining with 1000 bootstrap replications.

Local association analysis of DM/AR/ALS resistances

We conducted local association analysis for DM, ALS and ARresistances at the dm1/psl/cla locus with 152 cucumber inbredlines (Table S4). Whole-genome resequencing reads of 113 lineswere downloaded from the National Center for BiotechnologyInformation (NCBI) database (Qi et al., 2013); 39 lines wereresequenced with Illumina platforms. SNPs within the 100-kbCsSGR target region were called with the BWA-GATK4.0 work-flow using 9930 V2.0 as the reference; the SNPs were filteredwith maximum missing count of five lines and minor allele countof five lines in VCFtools. The linkage disequilibrium (LD) blockwas evaluated in Haploview (Barrett et al., 2005). The SNPs inthe LD block of the candidate gene were selected for hierarchicalclustering by the function HCLUST in R. R/PHANGORN (Schliep,2011) was used for bootstrapping with 1000 repeats.

In silico bulked segregant analysis (BSA) (Li et al., 2016) wasperformed to validate the candidate for the dm1/cla/psl locus byidentifying consensus SNPs (haplotypes) inside the target regionin the R and S bulks consisting of six resistant and six susceptiblelines, respectively. In addition, 41, 50 and 82 cucumber lineswere employed for local association analysis using the RidgeRegression algorithm in the R/RRBLUP package (Endelman,

No claim to original US government works

New Phytologist� 2018 New Phytologist TrustNew Phytologist (2018)

www.newphytologist.com

NewPhytologist Research 3

2011). Phenotypic data for these lines to the pre-2004 DMstrains, and the AR and ALS inoculation responses, wereobtained from historical data (Wehner & Shetty, 1997), and ourprevious (Pan et al., 2018) and present studies, respectively.

The association of the CsSGR locus with field DM resistancewas also examined in breeding populations. Phenotypic data forresponses to natural DM infection of 763 breeding lines withPoinsett 76 as dm1 donor were collected in multiple-year fieldtrials. All lines were genotyped with the SNP marker inside thedm1 locus. Association between DM resistance and SNPs wasperformed by single marker analysis.

Expression analysis of genes in the Chl degradationpathway

We examined the expression dynamics of CsSGR, together withthe genes in the Chl catabolism pathways, after inoculation withthe DM and ALS pathogens in Gy14, WI2757 and 9930 cucum-ber lines. For quantitative polymerase chain reaction (qPCR), leafsamples from inoculated and mock seedlings were collected at 0,3, 5 and 7 dpi for the DM pathogens; qPCR procedures followedLi et al. (2016) with the cucumber ubiquitin extension gene asthe reference. The relative gene expression levels were determinedby 2�DDCT. Each sample was run with three biological and threetechnical replicates.

Results

Delay of chlorosis is characteristic of DM resistance in Gy14and WI2757

We phenotyped responses to DM pathogen infection in theG9RIL and WTRIL populations in five environments withthree criteria (Yel, Col and GI). Typical HR-type necrosis wasnot obvious in either population, and delay of yellowing wasmore evident in resistant plants (Fig. S1). In the G9RIL popu-lation, Gy14 exhibited moderate resistance, although it showedslightly better performance in the NL trials than in the NCtrials (Table S5), probably as a result of the differential viru-lence structure of the P. cubensis populations at the two loca-tions (Thomas et al., 2017). With disease progress, the meandisease scores for Yel, Col and GI of the population increased(in NC trials) or decreased (in NL trials), suggesting greateroverall severity of symptoms of plants in the whole populationwith increased exposure to the DM inoculum (note that thescoring scales used in NC and NL were opposite) (Table S5;Fig. S2). Interestingly, although the overall range of Yel diseasescores of the population remained largely the same for bothNC2013_Yel and NL2014_Yel datasets, many plants in thepopulations remained low in Yel disease scores (Table S5;Fig. S2). The frequency distribution of GI disease scores in allenvironments was largely normal, whereas that for Yel wasmore bi-modular, indicating a single gene underlying the delayof chlorosis in Gy14 (Fig. S3). Mean Yel and GI scores of thefirst rating time across different environments were significantlycorrelated (rs = 0.45–0.73, P < 0.001) (Fig. S3). These data

suggest that Yel is a major component of DM disease symp-tom development.

The trends in symptom development and frequency distribu-tion of mean disease scores in the WTRIL population were simi-lar to those in the G9RIL population (Table S5; Fig. S4).However, WI2757 performed slightly better than Gy14 for DMresistance in all trials, which could be evidenced from the rela-tively slower increase in mean disease scores across rating times(Table S5; Fig. S4a). Interestingly, although the mean diseasescores of GI across three ratings in the NC trial were highly corre-lated (rs = 0.81–0.98), correlations among those of Yel and Colin NL and GI in NC trials for the WTRIL population were lowto moderate (rs = 0.23–0.49, P < 0.001) (Fig. S4b), suggesting adifferent genetic basis for anti-chlorosis and anti-necrosis inWI2757.

Gy14 and WI2757 share dm1-conferred DM resistance

As ANOVA indicated significant environmental effects (Table S6),QTL analysis of DM resistance in G9RIL was performed withGI means from individual locations, rating times and rating crite-ria (Table 1; Fig. S5a). Remarkably, QTLs detected with datafrom various environments and different scoring criteria were allmapped to the same location in chromosome 5, indicating thatonly one QTL underlies DM resistance in Gy14. Based on its1.5-LOD interval (c. 3.2 Mbp, Table 1), this QTL should beidentical to the previously mapped dm1 in Gy14 (Kennard et al.,1994; Horejsi et al., 2000). Interestingly, when GI was used, theQTL could be detected only from the first rating, indicatingthat dm1 played its role at the early stages of infection (Fig. S5a).Moreover, the QTL detected by the Yel data had a much strongereffect (higher LOD score and larger additive effects) than theQTL detected with GI in both NC2013 and NL2014 trials, sug-gesting that Yel is a more accurate measurement of DM resistancein Gy14.

As compared with the findings from the G9RIL population,dm1-dependent anti-chlorosis was more evident from QTL anal-ysis in the WTRIL population (Fig. S5b). Linkage analysis withSLAF markers in 87 WTRILs resulted in a genetic map with1618 SNP loci (Table S7) for QTL analysis. When the Yel datawere used, a QTL co-localized with dm1 in Gy14 was detected;when the Col (necrosis) data were employed, another linkedQTL, dm5.2, was identified; when GI data, an evaluation of bothchlorosis and necrosis, were used, both QTLs were found(Fig. S5b), suggesting that dm1 and dm5.2 in WI2757 may playimportant roles in the delay of chlorosis and anti-necrosis, respec-tively. The second QTL was named dm5.2 because its locationoverlapped with dm5.2 identified previously in WI7120 and PI197088 cucumbers (Wang et al., 2016, 2018).

The ALS resistance locus psl in Gy14 and WI2757 is co-localized with dm1

Gy14 exhibited high to moderately high resistance to P. syringaepv. lachrymans inoculation with anti-chlorosis and some degreeof HR, whereas 9930 developed large chlorotic and necrotic spots

New Phytologist (2018) No claim to original US government works

New Phytologist� 2018 New Phytologist Trustwww.newphytologist.com

Research

NewPhytologist4

Tab

le1Su

mmaryofquan

titative

traitloci(Q

TLs)iden

tified

from

G9RIL,W

TRIL

andW

TF2

3populationsfordownymildew

(DM)an

dan

gularleaf

spot(ALS)resistan

cesfrom

multipleen

vironmen

ts

Map

ping

populations

Phen

otyping

environmen

tsRating

time*

QTL

detected

Chr

location

Peak

marker

Peak

(cM)

Peak

LOD

Chrposition

(9930V2.0)

Gy1

4_sca

V1.0

Gy1

4_sca

(V1.0)Pos

1.5-LODleft

marker

1.5-LODright

marker

Additive

effects

Phen

otypic

variation

(%)

DM.G9RIL

NC2013_Y

el1

dm1

Chr5

SNP.176869

12.0

22.63

5040372

scaffold02951

1153575

SNP.74621

SNP.111257

�1.84

44.35

2dm1

Chr5

SNP.176869

12.0

16.23

5040372

scaffold02951

1153575

SNP.74621

SNP.111257

�1.44

33.48

NC2013_G

I1

dm1

Chr5

SNP.176869

11.3

10.62

5040372

scaffold02951

1153575

SNP.74621

SNP.111257

�0.56

26.02

NC2014_G

I1

dm1

Chr5

SNP.176869

11.3

7.29

5040372

scaffold02951

1153575

SNP.74621

SNP.73809

�0.39

20.35

NL2

014_Y

el1

dm1

Chr5

SNP.111257

14.0

18.03

6306083

scaffold01028

487360

SNP.74621

SNP.146681

0.89

38.76

2dm1

Chr5

SNP.176869

12.0

18.50

5040372

scaffold02951

1153575

SNP.74621

SNP.111545

0.77

39.49

NL2

015_G

I1

dm1

Chr5

SNP.176869

12.0

6.45

5040372

scaffold02951

1153575

SNP.74621

SNP.146681

0.56

16.37

DM.W

TRIL

NL2

014_Y

el1

dm1

Chr5

SNP5_4

972558

3.0

20.20

4972558

scaffold02951

1221402

SNP5_4

735130

SNP5_6

432618

1.24

64.24

NL2

014_C

ol

1dm5.2

Chr5

SNP5_1

6720220

79.5

14.10

16720220

scaffold00438

509862

SNP5_1

6104657

SNP5_1

7384531

1.66

39.19

NC2015_G

I1

dm1

Chr5

SNP5_5

154518

3.1

3.46

5154518

scaffold02951

1041703

SNP5_3

374580

SNP5_6

536439

�0.39

9.59

1dm5.2

Chr5

SNP5_1

6720220

80.1

3.51

16720220

scaffold00438

509862

SNP5_1

5371270

SNP5_1

7384531

�0.44

12.28

2dm5.2

Chr5

SNP5_1

6720220

81.0

4.12

16720220

scaffold00438

509862

SNP5_1

5371270

SNP5_1

7384531

�0.47

15.79

3dm5.2

Chr5

SNP5_1

6720220

81.0

4.31

16720220

scaffold00438

509862

SNP5_1

5371270

SNP5_1

7384531

�0.47

15.96

ALS.G9RIL

WI2015

1psl

Chr5

SNP.176869

12.0

5.01

5040372

scaffold02951

1153575

SNP.74621

SNP.66261

�0.96

19.27

WI2017

1psl

Chr5

SNP.176869

12.0

8.11

5040372

scaffold02951

1153575

SNP.74621

SNP.146681

�0.38

27.08

ALS.W

TF2

:3WI2011

1psl

Chr5

SSR14876

3.0

3.38

4729262

scaffold02951

1468890

UW084473

UW038767

�0.55

15.83

1psl1.1

Chr1

SSR05817

85.2

3.70

17913978

scaffold01543

1736677

UW084541

SSR05817

0.52

15.37

CA2011

1psl

Chr5

UW063468

7.0

8.22

5947245

scaffold02951

245920

SSR12467

UW038767

�0.85

28.11

1psl3.1

Chr3

SSR00311

122

3.9

185645

scaffold03080

2644063

SSR15043

SSR00311

�0.45

7.6

*NoQTLs

weredetectedwithgen

eralim

pression(G

I)datain

NC2013(2

ndrating),NC2014an

dNL2

015(both

2ndan

d3rdratings).

No claim to original US government works

New Phytologist� 2018 New Phytologist TrustNew Phytologist (2018)

www.newphytologist.com

NewPhytologist Research 5

quickly (Table S5; Fig. S6). The frequency distribution of themean disease scores in the G9RIL population was normal(WI2017 data; Fig. S7a). Despite the different methods used ininoculation and scoring in WI2015 and WI2017 experiments,one QTL on Chr5 was consistently detected (Table 1; Fig. S7b),which was designated as psl according to Dessert et al. (1982).

QTL analysis with WI2011 and CA2011 data (Fig. S7c) in theWTF23 population identified three QTLs, including the major-effect QTL, psl, which was shared with Gy14, and two minor-effect QTLs, psl1.1 (WI2011) and psl3.1 (CA2011) (Table 1;Fig. S7d). Based on the 1.5-LOD intervals and associated peakmarker with each QTL, dm1 and psl appeared to be the samelocus. Indeed, we performed joint QTL analysis with data fromDM.2013.Yel, DM.2013.GI and ALS.WI2017 of the G9RILpopulation; the LOD profiles for QTLs from the three datasetswere completely overlapped with the same peak marker (Fig. 1).As DM and ALS resistances in Gy14 originated from PI 197087,it is evident that dm1 and psl belong to the same locus.

Dm1/Psl encodes the cucumber STAYGREEN (CsSGR)protein

To confirm the identity of dm1 and psl, we performed finegenetic mapping for the two loci independently. Two SSR mark-ers (SSR001243 and SSR00398) flanking the 1.5-LOD intervalof dm1/psl were used to genotype 375 G9RILs (Fig. 2a). Eighteenrecombinants were identified, which were tested for DM inocula-tion responses in five environments (Table S1). As Gy14 onlyshowed moderate resistance, the multi-location extensive pheno-typing of these recombinants turned out to be critical to preciselylocate the candidate gene region. Nine new markers were devel-oped in this region. All 11 markers were used to genotype the 18recombinants, and nine haplotypes were identified. The meanDM disease scores of the nine recombinants are shown inFig. 2(a), three and six of which were resistant and susceptible,respectively. From these data, the dm1 locus was reliably delim-ited into a 93.7-kb region defined by UW063781 and SNP05.

The 18 recombinants were also examined for inoculationresponses with the ALS pathogen, which phenotypically showedexactly the same responses (R or S) as DM inoculation (Fig. 2a),indicating that dm1 and psl were the same locus.

In the 93.7-kb candidate gene region, 12 genes were pre-dicted, including the cucumber STAYGREEN gene (CsSGR)(Fig. 2b,c; Table S8). We performed in silico BSA in thisregion using 12 re-sequenced lines, six (Gy14, WI2757, Gy8,H19, G421 and 2A) of which carried dm1 with high resis-tance to pre-2004 DM strains, and six (9930, Straight 8,Coolgreen, PI 249561, WI7120 and PI 197088) did not carrydm1 (Wang et al., 2016, 2018). There were 50–75 SNPs/in-dels between the 9930 reference and any of the 11 lines(Table S4), but only one SNP in the coding region of CsSGR(SNP08 in Fig. 2a) showed complete consistency betweenmarker genotype and pre-2004 DM resistance (Fig. 2d), sug-gesting that CsSGR is the most likely candidate gene for thedm1/psl locus.

CsSGR was predicted to contain four exons and its codingregion was 771 bp encoding a putative protein with 256 aminoacid residues. The complete genomic DNA sequence ofCsSGR has been deposited in the NCBI database (accessionno. MH493893). The intron-exon structure of CsSGR was val-idated by cloning its cDNA from Gy14, 9930 and WI2757.Alignment of whole-length genomic sequences of the CsSGRgene among the three lines revealed only one transition fromA in 9930 to G in Gy14/WI2757 at position 323(SNP.A323G) (Fig. S8), and consistent results were obtainedamong the 12 re-sequenced cucumber lines (Fig. 2d), whichresulted in an amino acid substitution of glutamine (Q) witharginine (R) at position 108 of the CsSGR protein (Q108R)(Fig. 2e).

Local association analysis provides further evidence ofCsSGR as the candidate gene

Our previous work (Pan et al., 2018) has shown that CsSGRis a candidate gene for the AR resistance locus cla in PI197087, suggesting that CsSGR in Gy14 is responsible forsimultaneous resistances against DM, ALS and AR. To furtherconfirm this, we conducted local association analysis in naturalpopulations using sequencing data from the 93.7-kb regionand phenotypic data from a subset of lines (presented inTable S4). After SNP calling and filtering, 318, 677 and 623high-quality SNPs were used in association analysis with 41,50 and 82 accessions for DM, AR and ALS resistances,respectively. As shown in the resulting dot plots (Fig. S9), thecausal SNP.A323G inside CsSGR had the highest �log10(P)value and thus the strongest association with DM/AR/ALSresistances, further confirming the identity between CsSGRand the dm1/psl/cla locus.

SGR is highly conserved across dicot and monocot plants

The SGR gene family has two to three members in mostplant genomes that could be classified into two subfamilies:

dm1

dm1

psl

LOD

scor

e

Fig. 1 Logarithm of the odds (LOD) profiles of quantitative trait loci(QTLs) for cucumber downy mildew (DM) and angular leaf spot (ALS)resistance. Only representative datasets of the cucumber Gy149 9930recombinant inbred line (G9RIL) population using NC2013 (for DM) andWI2017 (for ALS) data are presented. The x-axis is the cM position inchromosome 5, and the y-axis is the LOD support score; the horizontaldashed line is the threshold at P < 0.05 based on 1000 permutations; Yel,yellowing (chlorosis); GI, general impression.

New Phytologist (2018) No claim to original US government works

New Phytologist� 2018 New Phytologist Trustwww.newphytologist.com

Research

NewPhytologist6

SGR and SGR-like (SGRL) (Sakuraba et al., 2015). Arabidopsishas three members (AtSGR1, AtSGR2, AtSGRL), whereascucumber has one copy each (CsSGR and CsSGRL). Tounderstand the structural and functional relationship amongstSGR homologs, we constructed a phylogenetic tree using 23SGR/SGRL protein sequences from 12 species (Table S3). Inthe resulting neighbor-joining (NJ) tree (Fig. S3a), the SGRand SGRL clades were well separated. Within the SGR sub-clade, the clustering was largely consistent with the taxonomicposition in the life tree. For example, melon CmSGR wasthe closest to CsSGR.

We examined the domain structure of SGR homologs inArabidopsis, rice, pepper, tomato and melon, as well as 9930

and Gy14 cucumbers (Fig. 3b). All SGR proteins had threetypical domains, including the chloroplast transit peptidedomain, the highly conserved SGR domain and the variableC-terminal region. The Q108R substitution in Gy14 waslocated in the SGR domain. We predicted the protein struc-ture of the SGR domain encoded by CsSGR with the RAP-

TORX program (http://raptorx.uchicago.edu/). The predictedthree-dimensional model of the SGR domain with high-lighted substituted amino acid residues (Fig. S10) suggeststhat the Q108R substitution would cause a shortened b-foldsheet that connects to the a-helix at the SGR domain,which might affect its catalytic efficiency during Chla degra-dation.

UW

0639

35

S NP 1

0

UW

063 7

81

SNP 0

8S N

P 07

SNP 0

5

Inde

l92-

21

UW

085 2

28

5 .1 6

5.40

5 .4 4

5.46

5 .47

5 .7 0

5.94

Chr5(Mb)

Marker

qtl-dm1(~93.7 kb)

9930Gy14

7292

5.44 ± 1.30

7236 7.89 ± 1.30

6.50 ± 0.717.00 ± 1.41

7304

7.28 ± 1.47

HM115HM0197093

7128

7177

7.22 ± 1.57

5.59 ± 0.464.65 ± 0.79

7.00 ± 1.41

5.50 ± 0.507225

7.75 ± 1.50

5.53

UW

0633

526.

19

SSR0

0398

6.62

SSR0

0124

34.

79

R

S

SS

SR

RR

S

S

S

DMMean ± SD

Q108RATG TAG

(a)

(d)

(b)

(c)

(e)

100 bp

2.33 ± 0.474.67 ± 0.47

4.50 ± 0.505.00 ± 0.004.67 ± 0.473.75 ± 0.433.75 ± 0.434.00 ± 0.002.43 ± 0.492.33 ± 0.473.00 ± 0.00

ALSMean ± SD

ID Sample 323

1 9930 T A T A A T T G G C T T C A A A G G G A T G A A G T2 Coolgreen T A T A A T T G G C T T C A A A G G G A T G A A G T3 Straight8 T A T A A T T G G C T T C A A A G G G A T G A A G T4 WI7120 T A T A A T T G G C T T C A A A G G G A T G A A G T5 PI197088 T A T A A T T G G C T T C A A A G G G A T G A A G T6 PI249561 T A T A A T T G G C T T C A A A G G G A T G A A G T7 2A T A T A A T T G G C T T C G A A G G G A T G A A G T8 G421 T A T A A T T G G C T T C G A A G G G A T G A A G T9 GY8 T A T A A T T G G C T T C G A A G G G A T G A A G T10 GY14 T A T A A T T G G C T T C G A A G G G A T G A A G T11 H19 T A T A A T T G G C T T C G A A G G G A T G A A G T12 WI2757 T A T A A T T G G C T T C G A A G G G A T G A A G T

ID Sample 108

1 9930 I N N S Q L Q G W Y N W L Q R D E V V G E W K K V K2 Coolgreen I N N S Q L Q G W Y N W L Q R D E V V G E W K K V K3 Straight8 I N N S Q L Q G W Y N W L Q R D E V V G E W K K V K4 WI7120 I N N S Q L Q G W Y N W L Q R D E V V G E W K K V K5 PI197088 I N N S Q L Q G W Y N W L Q R D E V V G E W K K V K6 PI249561 I N N S Q L Q G W Y N W L Q R D E V V G E W K K V K7 2A I N N S Q L Q G W Y N W L R R D E V V G E W K K V K8 G421 I N N S Q L Q G W Y N W L R R D E V V G E W K K V K9 GY8 I N N S Q L Q G W Y N W L R R D E V V G E W K K V K10 GY14 I N N S Q L Q G W Y N W L R R D E V V G E W K K V K11 H19 I N N S Q L Q G W Y N W L R R D E V V G E W K K V K12 WI2757 I N N S Q L Q G W Y N W L R R D E V V G E W K K V K

Class

qtl-cla (~32 kb)

Fig. 2 Map-based cloning of the cucumber dm1/psl locus. Fine mapping with 504 cucumber Gy149 9930 recombinant inbred lines (RILs) (G9RILs)delimits the dm1 locus into a 93.7-kb region (a). Closed and open rectangles indicate alleles derived from resistant Gy14 and susceptible 9930, respectively.The nine haplotypes from 18 recombinants in the candidate gene region showed the same inoculation responses to downy mildew (DM) and angular leafspot (ALS) pathogens, as reflected by the mean disease scores from multiple phenotyping experiments. Amongst the 12 predicted genes in this region (b),the third one is the best candidate for dm1, which has four exons and encodes a Staygreen domain. A single nucleotide polymorphism (SNP) in the thirdexon results in a Q108R amino acid substitution (c). Alignment of genomic DNA (d) and protein (e) sequences among 12 cucumber lines reveals completeconsistency of the SNP alleles or Q108R amino acid substitution with downy mildew resistance to the pre-2004 DM strain. The six lines carrying the PI197087-derived dm1 resistance allele are 2A, G421, GY8, GY14, H19 andWI2757.

No claim to original US government works

New Phytologist� 2018 New Phytologist TrustNew Phytologist (2018)

www.newphytologist.com

NewPhytologist Research 7

The Chl degradation pathway may be involved in CsSGR-conferred disease resistance

We investigated the expression dynamics of CsSGR in Gy14,WI2757 and 9930 at 0, 3, 5 and 7 dpi with the DM pathogen(Fig. 4). We first evaluated the expression level of the P. cubensisinternal transcribed spacer (ITS) gene (Tian et al., 2011) forquantitative assessment of pathogen growth on the host plants.The abundance of P. cubensis on susceptible 9930 plants showeda steady increase at 5 and 7 dpi as compared with 3 dpi. By con-trast, pathogen growth on resistant Gy14 and WI2757 remainedrelatively low, and never exceeded two-fold of that at 3 dpi(Fig. 4a), indicating suppression of pathogen growth by dm1.Within 5 dpi, CsSGR expression was elevated in all three lines;however, at 7 dpi, it kept increasing in 9930, but started todecline in the two resistant lines (Fig. 4b). This coincided withdisease symptom development, which was visually distinctivebetween 9930 (yellowing) and Gy14 or WI2757 (staygreen)(Fig. 4c). Similar trends were also observed for CsSGR expressionin Gy14 and 9930 after inoculation with the ALS and ARpathogens (Pan et al., 2018), supporting a direct link betweenCsSGR expression and host resistance against these pathogens.

In Arabidopsis, AtSGR1 is a key regulator of Chl degradation.To understand the roles of CsSGR in Chl degradation in responseto P. cubensis inoculation, we investigated the expression of ninekey genes in the cucumber Chl degradation pathway, includingCsCAO, CsNOL, CsNYC1, CsHCAR, CsCLS, CsSGRL, CsPPH,CsPaO and CsRCCR (enzyme encoded by each gene is providedin Fig. 5a). For each gene, its expression in Gy14, WI2757 and9930 was examined at 0, 3, 5 and 7 dpi. CsCAO catalyzes theconversion of Chla to Chlb. At 0–7 dpi, its expression was

downregulated in 9930, but relatively steady in Gy14/WI2757(Fig. 5b). The expression levels of CsNOL, CsNYC1 andCsHCAR, which convert Chlb and Chla, were all transiently ele-vated at 3–5 dpi and returned to baseline at 7 dpi (Fig. 5c–e).CsCLS and CsSGRL convert chlorophyllide a to Chla andpheophorbide a, respectively, and their expression was largelysteady in the susceptible line and slightly increased or exhibitedno change in the resistant lines (Fig. 5f,g). The three genes,CsPPH, CsPaO and CsRCCR, acting downstream of CsSGR,exhibited an expression pattern that was largely similar to that ofCsSGR (Fig. 4b). That is, although the expression in all threelines increased on infection, the fold increase in the susceptible9930 was significantly higher than that in the resistant lines(Fig. 5h–j). We also examined the expression of CsCHLI, whichencodes the magnesium chelatase subunit I that plays an impor-tant role in Chl biosynthesis (e.g. Gao et al., 2016). The expres-sion of CsCHLI was slightly increased in the resistant lines, butdecreased in 9930 (Fig. 5k), indicating reduced Chl biosynthesisin 9930 after DM inoculation. Taken together, these data suggestthat the Chl degradation pathway may play a critical role inCsSGR-regulated disease resistance against DM in Gy14 andWI2757.

The dm1/psl/cla locus is of Indian origin and has under-gone diversifying selection

The donor of dm1/psl/cla was PI 197087 originating from India.To confirm this and to investigate the distribution of dm1/psl/claalleles in natural populations, we conducted clustering analysis in152 cucumber accessions using resequencing data (Table S4).Among them, 119 were cultivated cucumbers (CSS), including

SGR-like

(a) (b)

Fig. 3 (a) Phylogenetic relationships and (b) domain conservation of cucumber CsSGR and CsSGRL homologs with other selected plant species. (a) Aneighbor-joining tree for STAYGREEN (SGR) and STAYGREEN-LIKE (SGRL) protein sequences with MEGA 7.0 (http://www.megasoftware.net/). Thenumbers at the branch points represent bootstrap values (%) of 1000 replications. (b) Domain structure of SGR homologs. Blue line, chloroplast transitpeptide domain; red line, conserved SGR domain; green line, variable C-terminal region; red rectangle, amino acid substitution in Gy14 CsSGR protein.

New Phytologist (2018) No claim to original US government works

New Phytologist� 2018 New Phytologist Trustwww.newphytologist.com

Research

NewPhytologist8

73 from East/West Asia and Europe (Eurasian1–73), 22 fromSouth Asia (SAsia1–22), four from Africa (Africa1–4) and 20from the USA. In addition, 13 and 20 belonged to the wild(CSH1–13) and semi-wild Xishuangbanna (XIS1–20), respec-tively. We first evaluated LD in the 75-kb region harboring theCsSGR locus, which revealed an LD block spanning only 17 kb(Fig. S11), suggesting that this is a highly recombinogenic region.As such, 72 SNPs in a 20-kb vicinity of the CsSGR locus wereused for clustering analysis to construct a phylogenetic tree forthe 152 accessions (Fig. S12).

Among the 19 lines of US origin, nine (2A, G421, Gy14, Gy3,Gy8, H19, Poinsett 76, SC50 and WI2757) with DM/ALS/ARresistance from PI 190787 have been widely used in US cucum-ber breeding programs, and the remaining 10 are susceptible. Inthe resulting phylogenetic tree, all nine resistant lines plus PI190787 were in the same clade, whereas the 10 susceptible lineswere scattered in different clades. This was in sharp contrast withthe phylogenetic trees built with whole-genome marker-basedclustering, in which natural populations could be roughly classi-fied into East Asia, Eurasia and India/XIS groups (Qi et al.,2013). These data revealed the fact that the region in PI 197087harboring the dm1/psl/cla locus has been under diversifying selec-tion during breeding for disease resistances. Indeed, we examined

the SNP-based haplotypes in this region among 17 of the 19 USlines. The ALS, AR and pre-2004 DM resistance phenotypeswere in complete agreement with the marker-based haplotypes inthis region (Fig. S13).

The low LD in this region also implies that linkage drag shouldbe minimal and that this gene should be relatively easy to intro-gress into recipient cucumber lines through regular crosses. Thiswas verified in real cucumber breeding schemes. Over the last sev-eral years, many breeding lines have been generated for DM resis-tance by crossing with dm1-carrying Poinsett 76. These lines weretested for natural infection of P. cubensis in disease nurseries atFondi, Italy, and genotyped with SNP markers in the dm1 region.We tested the association of these SNPs with DM resistanceamong the 734 advanced breeding lines, which were selected onthe basis of phenotypic screening only. A single marker analysisrevealed that the causal SNP inside the CsSGR gene(SNP.A323G, or SNP08 in Fig. 2a) had the highest associationwith DM resistance (�log10(P) = 45.96). These data not only val-idated the candidacy of CsSGR for the dm1 locus, but also provedthe utility of this diagnostic marker in marker-assisted selection.In addition, this association also implies that dm1 still performswell in Europe cucumber fields, despite the fact that it confersonly moderate resistance to prevailing DM strains in the USA.

0

10

20

30

40

0 dpi 3 dpi 5 dpi 7 dpi 10 dpi 15 dpi

Rel

ativ

eex

pres

sion

leve

lR

elat

ive

expr

essi

onle

vel

CsSGRGy14WI27579930

**

(a)

(c)

Gy1

4W

I275

799

30(b)

**

*

**

0

2

4

6

8

10

0 dpi 3 dpi 5 dpi 7 dpi 10 dpi 15 dpi

PsITSGy14WI27579930

Fig. 4 Downy mildew pathogen growth, cucumber CsSGR gene expression and symptom development in Gy14, WI2757 and 9930 plants atdifferent days post-inoculation (dpi). (a) Relative expression level of ITS gene of Pseudoperonospora cubensis at 0, 3, 5 and 7 dpi, reflectingpathogen growth on the three host plants. (b) Relative expression level of CsSGR at 0, 3, 5 and 7 dpi. The values in (a, b) are given asmeans� SD from three biological replicates. Significant level at (F-test): *, P < 0.05; **, P < 0.01. (c) Symptoms on the first true leaves of Gy14,WI2757 and 9930 seedlings at 0, 3, 5, 7, 10 and 15 dpi.

No claim to original US government works

New Phytologist� 2018 New Phytologist TrustNew Phytologist (2018)

www.newphytologist.com

NewPhytologist Research 9

Discussion

CsSGR is a loss-of-susceptibility host resistance gene

A number of R genes have been cloned with durable resistanceagainst a broad range of pathogens. Often, they are loss-of-susceptibility mutations resulting in host reprogramming of bio-logical processes that are critical for effective pathogen infection(Wiesner-Hanks & Nelson, 2016; Kourelis & van der Hoorn,2018). For example, the barley mlo is a nonfunctional allele ofMlo encoding a membrane protein required for susceptibility to

powdery mildew (B€uschges et al., 1997). The wheat rust/powderymildew resistance gene Lr67 encodes a hexose transporter; theloss-of-function Lr67 may disturb the balance of sugars associ-ated with sugar signaling and partitioning, which may be neces-sary for pathogen invasion (Moore et al., 2015; Dodds &Lagudah, 2016). The rice multiple-pathogen-resistance locusGH3-2 encodes an indole-3-acetic acid (IAA)-amido synthetasesuppressing pathogen-induced IAA accumulation (Fu et al.,2011). Amazingly, transgenic expression of the wheat Lr34 orLr67 genes in barley, rice, sorghum and maize also confers resis-tance to multiple adapted pathogens, suggesting their roles in

Time post-inoculation (dpi)

Rel

ativ

e ex

pres

sion

leve

l

Rel

ativ

e ex

pres

sion

leve

l

Rel

ativ

e ex

pres

sion

leve

l

Rel

ativ

e ex

pres

sion

leve

l

Rel

ativ

e ex

pres

sion

leve

l

Rel

ativ

e ex

pres

sion

leve

l

Rel

ativ

e ex

pres

sion

leve

l

Rel

ativ

e ex

pres

sion

leve

lR

elat

ive

expr

essi

on le

vel

Rel

ativ

e ex

pres

sion

leve

l

Time post-inoculation (dpi)

Time post-inoculation (dpi) Time post-inoculation (dpi)

Time post-inoculation (dpi)Time post-inoculation (dpi)

Time post-inoculation (dpi)

Time post-inoculation (dpi)

Time post-inoculation (dpi) Time post-inoculation (dpi)

(a) (b) (c)

(d) (e)

(g) (h)

(j)

(f)

(i) (k)

Fig. 5 Expression dynamics of key genes in chlorophyll catabolic pathways in Gy14, WI2757 and 9930 cucumbers at different days after downy mildew(DM) pathogen inoculation (dpi). (a) Major steps in the chlorophyll degradation pathway. (b–k) Relative expression levels of selected genes in thedegradation pathway at 0, 3, 5 and 7 dpi. The values are given as means� SD from three biological replicates. The genes and the encoded enzymes/proteins are as follows: CLS, chlorophyll synthase; CAO, chlorophyllide a oxygenase; NYC1/NOL, chlorophyll b reductase; HCAR, 7-hydroxymethylchlorophyll a reductase; SGR, staygreen/Mg-dechelatase; SGRL, staygreen-like; PPH, pheophytinase; PaO, pheophorbide a oxygenase; RCCR, redchlorophyll catabolite reductase; CHLI, Mg-chelatase subunit I.

New Phytologist (2018) No claim to original US government works

New Phytologist� 2018 New Phytologist Trustwww.newphytologist.com

Research

NewPhytologist10

infection are conserved in these crops (e.g. Risk et al., 2013;Rinaldo et al., 2017; Schnippenkoetter et al., 2017; Sucher et al.,2017; Boni et al., 2018).

We have shown that CsSGR is the candidate gene for thedm1/psl/cla locus and that the Q108R amino acid substitu-tion in the CsSGR protein is responsible for the diseaseresistance in Gy14 (Figs 2, S8, S10). CsSGR clustered withother SGR homologs, such as AtSGR1 (Fig. 3), suggestingpossible conserved functions of SGR homologs in differentplant species. Arabidopsis AtSGR1 encodes a magnesiumdechelatase and plays important regulatory roles in Chldegradation (Shimoda et al., 2016; Kuai et al., 2018). Duringsenescence, AtSGR1 physically interacts with light-harvestingcomplex subunits of photosystem II (LHCII) to trigger Chland Chl-binding protein degradation (Park et al., 2007;H€ortensteiner, 2009; Sakuraba et al., 2012, 2015). In addi-tion, AtSGR1 and Medicago MtSGR have been found to benecessary for disease symptom development and HR-relatedcell death in response to pathogen infections (Mur et al.,2010; Mecey et al., 2011; Ishiga et al., 2015). Here, we haverevealed that CsSGR is the candidate for the dm1/psl/cla resis-tance locus, which may represent a novel function for thishighly conserved protein in plants. From the discussionsbelow, it is clear that the recessive mutation inside theCsSGR gene for durable resistance against multiple pathogensmay well be considered as a loss-of-susceptibility type Rgene.

Two points are worth mentioning with regard to this CsSGR-mediated durable resistance. First, loss-of-susceptibility R genesare often considered to be nonrace specific. Responses to ARpathogen infection were typical HRs (Pan et al., 2018), but Gy14showed high to moderate resistances to pre- and post-2004 DMstrains and two isolates of the P. syringae pv. lachrymanspathogen, respectively (Call et al., 2012a; Table S5). These obser-vations indicate that differential responses may exist for differentraces or pathotypes of the DM and ALS pathogens. It is notknown whether this indicates that recognition-based defenseresponses may exist during pathogen–host interactions. Second,the durability of loss-of-susceptibility-type broad-spectrum resis-tance sometimes carries a cost for the host plant. For example,mlo barley is associated with necrotic flecking and yield loss, aswell as increased susceptibility to several nonbiotrophic fungalpathogens (Kjær et al., 1990; Brown & Rant, 2013; McGrannet al., 2014). Similarly, Lr34- and Lr67-mediated resistancecauses leaf tip necrosis associated with accelerated senescence.However, no obvious negative traits were observed in cucumberlines carrying the sgr resistance allele from PI 197087. Further,the CsSGR locus is located in a region with low LD (Fig. S11),which was probably a reason for its wide use in the US cucumberbreeding programs.

CsSGR confers multi-disease resistance through modulationof the Chl degradation pathway

We have shown that CsSGR confers resistances to the biotrophicbacterial P. syringae pv. lachrymans and oomyceteous P. cubensis,

as well as the hemibiotrophic fungal C. orbicular. Plants activatedistinct defense responses to combat various pathogens of differ-ent lifestyles (Thomma et al., 1998; Glazebrook, 2005; Spoelet al., 2007; Mengiste, 2012). Regardless of the type of attackingpathogen, some downstream net outcome from the pathogen–host plant interactions seems common in host defense responses,which may include the generation of ROS, HR-related cell deathand the production of antimicrobial proteins or secondarymetabolites. Several studies have indicated the involvement ofSGR in host defense responses. Mur et al. (2010) found thatincreased expression of AtSGR accelerated HR-related PCD inpathogen-challenged resistant plants, which was probably causedby SGR-mediated disruption of LHCs, resulting in the buildupof phototoxic Chl catabolites and light-dependent generation ofROS (H2O2), although this process did not seem to contributeto plant resistance (Zurbriggen et al., 2009; Mur et al., 2010). Bycontrast, Ishiga et al. (2015) found that the Mtsgr mutant ofM. truncatula and the SGR-RNAi line of alfalfa showed HR-likeenhanced cell death on inoculation with the nonhost Asian soy-bean rust (ASR), suggesting a negative role of Medicago SGRgenes. As compared with the wild-type, the Mtsgr mutant exhib-ited enhanced defense responses to ASR by greater ROS accumu-lation and higher expression of defense-related genes. BothMecey et al. (2011) and Ishiga et al. (2015) found that pathogen-induced expression of SGR is a critical step in disease symptomdevelopment (chlorosis). Our study also supported the role ofCsSGR in chlorosis development, as evidenced by the elevatedexpression of CsSGR in the susceptible 9930 cucumber in com-parison with that in the resistant lines (Fig. 4b). Interestingly, nodifference in bacterial multiplication was observed between theArabidopsis noc1 mutant and its wild-type (Mecey et al., 2011).This was in contrast with results from the present study, in whichgrowth of the DM pathogen was significantly suppressed in resis-tant plants (Fig. 4a), suggesting a true host resistance mediatedby CsSGR.

The SGR gene plays a critical regulatory role in the Chl degra-dation pathway. Chl catabolic genes, such as SGR, PAO andPPH, have been reported to be regulated in response to infectionby various pathogens (e.g. Greenberg & Ausubel, 1993; Meceyet al., 2011; Ishiga et al., 2015). In Arabidopsis, it is known thatAtSGR1 interacts with all other CCEs, such as NYC1/NOL,HCAR, PPH, PAO and RCCR (Fig. 5a); it also interacts withLHCII proteins to facilitate Chl degradation during senescence(Sakuraba et al., 2012, 2013; Shimoda et al., 2016). The coloredintermediates of Chl breakdown, such as pheophorbide a (Pheidea) and red Chl catabolite (RCC), are potentially phototoxic,which may directly induce cell death or generate ROS that func-tion as signaling molecules to play a role in defense responsesagainst pathogen infection (Mach et al., 2001; Hirashima et al.,2009; Mur et al., 2010; Ishiga et al., 2015; Serrano et al., 2016).AtSGR1 may play a role in the detoxification of these phototoxiccatabolites by interacting with CCEs and LHCII proteins(Sakuraba et al., 2012, 2015; Shimoda et al., 2016). Indeed, theAtSGR RNAi line exhibits a staygreen phenotype with greatlyreduced accumulation of phytotoxic Chl catabolites (Armsteadet al., 2007).

No claim to original US government works

New Phytologist� 2018 New Phytologist TrustNew Phytologist (2018)

www.newphytologist.com

NewPhytologist Research 11

Based on the above discussion, we propose a working model toexplain the possible mechanisms of CsSGR-mediated host resis-tance in cucumber (Fig. 6), in which the mutant CsSGR proteininteracts with other CCEs and LHCII to mitigate the damagecaused by the production of ROS and buildup of phototoxic Chlcatabolites in resistant plants, and which result in disease in sus-ceptible plants. This model emphasizes the differential abilities ofthe resistant and susceptible alleles of the CsSGR gene to modu-late the dual roles (inducer of cell death and signal molecule) ofROS and catabolites of Chl breakdown during plant–pathogeninteractions. CsSGR in the wild-type (susceptible) is necessaryfor disease symptom development. In susceptible plants,pathogen infection will result in disassembly of the LHC,reduced photosynthesis capacity and increased expression ofgenes for CCEs (CsSGR, CsPPH, CsPAO and CsRCCR), which,together with decreased Chl biosynthesis (Fig. 5k), will result inChl reduction and thus chlorosis (yellowing). The senescing-likeprocesses generate ROS and phototoxic catabolites that acceleratecell death (necrosis). Meanwhile, the end products of Chl break-down may be re-mobilized as nutrition to promote pathogengrowth (Chen et al., 2010). In resistant plants, the Q108R aminoacid residue substitution in the CsSGR protein may alter theinteraction of CsSGR with other CCEs and LHCII. Thus, theexpression of CCE-encoding genes is largely steady on pathogen

infection (Fig. 5); Chl degradation is limited (thus staygreen),resulting in limited ROS production and phototoxic catabolitebuildup; thus, a more defensive cellular environment is main-tained with a net outcome of resistance (Fig. 6). The elevatedexpression of CsSGR (but not as high as in the susceptible line,Fig. 4) may keep ROS at a level that serves as a molecular signalto invoke other defense mechanisms or result in localized celldeath (HR) to limit the invasion of pathogens. The limitednecrosis will not disrupt the balance between Chl synthesis anddegradation.

This rather primitive model highlights the passive, loss-of-susceptibility strategy employed by resistant cucumber plants forthe broad-spectrum disease resistance conferred by CsSGR. How-ever, many details are still unknown. As discussed earlier, thedefense responses of Gy14 to different strains/isolates of the DM,ALS and AR pathogens vary. Is there any specific recognitionduring the interaction of the pathogen with the resistant plant toinvoke active defense responses, such as the production of antimi-crobial proteins or secondary metabolites? Are additional regula-tors and signaling pathways (for example, other transcriptionfactors, hormone signaling) involved in the resistance? All ofthese merit further investigations to fully understand the novelfunction of the STAYGREEN gene for durable and broad-spectrum host resistance in cucumber.

Pathogen-CsSGRinteractions?

Pathogen attack

Susceptibility

Increased ROS accumulation

Phytotoxic catabolites buildup

Increased expressionof CCE genes

Nutrition remobilization,

pathogen growth

Reduced Chlbiosynthesis

Accelerated Chlbreakdown

HR-PCD, ROS signaling,

defense proteins?

No disease, STAYGREEN

Chlorosis, necrosisdisease symptom

development

Chlorophyll a

Resistance

Pathogen growth

inhibition

Less affectedChl degradation

pathway

CsSGR, CCEs, LHCII

interactions

CsSGR(Q108)

CsSGR(R108)

Active defense responses?

Loss-of-susceptibility(passive defense)

Fig. 6 A working model to explain the possible mechanisms of CsSGR-mediated durable and broad-spectrum disease resistance in cucumber. Theresponses on pathogen infection in the susceptible and resistant (green) lines are depicted to the left and right of the vertical dashed line,respectively. Cssgr is a loss-of-susceptibility mutation of CsSGR that is necessary for disease symptom development. The amino acid substitutionfrom glutamine in the susceptible line (Q108) to arginine (R108) in the resistant line in the protein may affect the interaction of CsSGR withchlorophyll catabolism enzymes (CCEs) and light-harvesting complex subunits of photosystem II (LHCII) proteins, thus inhibiting the normalfunction of the chlorophyll degradation pathway genes (upregulation). It is not known whether there are interactions between mutant CsSGRprotein and the pathogen to trigger active defense responses. Chl, chlorophyll; HR, hypersensitive response; PCD, programmed cell death; ROS,reactive oxygen species.

New Phytologist (2018) No claim to original US government works

New Phytologist� 2018 New Phytologist Trustwww.newphytologist.com

Research

NewPhytologist12

Acknowledgements

The authors thank Kristin Haider for technical help and DrNischit Shetty of Monsanto Seed Company for phenotypingthe G9RIL population in the FL2015 experiment. Thisresearch was supported by the Agriculture and FoodResearch Initiative Competitive Grants under award numbers2013-67013-21105 and 2015-51181-24285 from the USDepartment of Agriculture National Institute of Food andAgriculture (to Y.Weng), the National Science Foundation ofChina (Project no. 31772305) and the Doctoral Fund fromHebei Academy of Agriculture and Forestry Sciences (Projectno. F16E05) (to Z.W.). The authors declare no conflicts ofinterest.

Author contributions

Y.Wang and J.T. performed the majority of the research. K.V.,T.C.W. and Y.Wang collected the DM phenotypic data in NCexperiments. P.A.G.K. and J.S. collected the DM data in NLexperiments. K.O., H.H.B. and E.H. phenotyped and genotypedthe DM resistance in breeding populations. J.T., Z.W., A.T.,X.Z. and K.O. performed the ALS phenotyping. J.P. collectedAR phenotypic data. C.W. conducted genotyping of the WTRILpopulation. Y.Weng designed and supervised the experimentsand participated in data analysis. Y.Weng and Y.Wang wrote themanuscript with input from the other coauthors. All authorsreviewed and approved the final submission.

ORCID

Yiqun Weng http://orcid.org/0000-0001-9457-2234

References

Acevedo-Garcia J, Spencer D, Thieron H, Reinst€adler A, Hammond-Kosack K,

Phillips AL, Panstruga R. 2017. mlo-based powdery mildew resistance in

hexaploid bread wheat generated by a non-transgenic TILLING approach.

Plant Biotechnology Journal 15: 367–378.Armstead I, Donnison I, Aubry S, Harper J, Hortensteiner S, James C, Mani J,

Moffet M, Ougham H, Roberts L et al. 2007. Cross-species identification of

Mendel’s I locus. Science 315: 73.Bai Y, Pavan S, Zheng Z, Zappel NF, Reinst€adler A, Lotti C, De Giovanni C,

Ricciardi L, Lindhout P, Visser R et al. 2008. Naturally occurring broad-

spectrum powdery mildew resistance in a Central American tomato accession is

caused by loss ofMlo function.Molecular Plant–Microbe Interactions 21: 30–39.

Barnes W. 1966. Development of multiple disease resistant hybrid cucumber.

Proceedings of the American Society for Horticultural Science 89: 390–393.Barnes W, Epps W. 1952. Two types of anthracnose resistance in cucumbers.

Plant Disease Reporter 36: 479.Barnes W, Epps W. 1954. An unreported type of resistance to cucumber downy

mildew. Plant Disease Reporter 38: 620.Barrett JC, Fry B, Maller J, Daly MJ. 2005.Haploview: analysis and

visualization of LD and haplotype maps. Bioinformatics 21: 263–265.Barry CS, McQuinn RP, Chung M-Y, Besuden A, Giovannoni JJ. 2008. Amino

acid substitutions in homologs of the STAY-GREEN protein are responsible

for the green-flesh and chlorophyll retainer mutations of tomato and pepper.

Plant Physiology 147: 179–187.

Boller T, Felix G. 2009. A renaissance of elicitors: perception of microbe-

associated molecular patterns and danger signals by pattern-recognition

receptors. Annual Review of Plant Biology 60: 379–406.Boni R, Chauhan H, Hensel G, Roulin A, Sucher J, Kumlehn J, Brunner S,

Krattinger SG, Keller B. 2018. Pathogen-inducible Ta-Lr34res expression in

heterologous barley confers disease resistance without negative pleiotropic

effects. Plant Biotechnology Journal 16: 245–253.Boutrot F, Zipfel C. 2017. Function, discovery, and exploitation of plant pattern

recognition receptors for broad-spectrum disease resistance. Annual Review ofPhytopathology 55: 257–286.

Broman KW, Wu H, Sen S, Churchill GA. 2003. R/qtl: QTL mapping in

experimental crosses. Bioinformatics 19: 889–890.Brown JKM, Rant JC. 2013. Fitness costs and trade-offs of disease resistance and

their consequences for breeding arable crops. Plant Pathology 62: 83–95.B€uschges R, Hollricher K, Panstruga R, Simons G, Wolter M, Frijters A, van

Daelen R, van der Lee T, Diergaarde P, Groenendijk J et al. 1997. The barleyMlo gene: a novel control element of plant pathogen resistance. Cell 88: 695–705.

Call AD, Criswell AD, Wehner TC, Ando K, Grumet R. 2012a. Resistance of

cucumber cultivars to a new strain of cucurbit downy mildew. HortScience 47:171–178.

Call AD, Criswell AD, Wehner TC, Klosinska U, Kozik EU. 2012b. Screening

cucumber for resistance to downy mildew caused by Pseudoperonospora cubensis(Berk. and Curt.) Rostov. Crop Science 52: 577–592.

Cha KW, Lee YJ, Koh HJ, Lee BM, Nam YW, Paek NC. 2002. Isolation,

characterization, and mapping of the stay green mutant in rice. Theoretical andApplied Genetics 104: 526–532.

Chen LQ, Hou BH, Lalonde S, Takanaga H, Hartung ML, Qu XQ, Guo WJ,

Kim JG, Underwood W, Chaudhuri B et al. 2010. Sugar transporters forintercellular exchange and nutrition of pathogens. Nature 468: 527–532.

Cui H, Tsuda K, Parker JE. 2015. Effector-triggered immunity: from pathogen

perception to robust defense. Annual Review of Plant Biology 66: 487–511.Dessert JM, Baker LR, Fobes JF. 1982. Inheritance of reaction to Pseudomonaslachrymans in pickling cucumber. Euphytica 31: 847–855.

Dodds PN, Lagudah ES. 2016. Starving the enemy. Science 354: 1377–1378.Elliott C, Zhou F, Spielmeyer W, Panstruga R, Schulze-Lefert P. 2002.

Functional conservation of wheat and riceMlo orthologs in defense modulation

to the powdery mildew fungus.Molecular Plant–Microbe Interactions 15: 1069–1077.

Endelman JB. 2011. Ridge regression and other kernels for genomic selection

with R package rrBLUP. Plant Genome Journal 4: 250.Fang C, Li C, Li W, Wang Z, Zhou Z, Shen Y, Wu M, Wu Y, Li G, Kong LA

et al. 2014. Concerted evolution of D1 and D2 to regulate chlorophylldegradation in soybean. Plant Journal 77: 700–712.

Fanourakis NE, Simon PW. 1987. Analysis of genetic linkage in the cucumber.

Journal of Heredity 78: 238–242.Fu D, Uauy C, Distelfeld A, Blechl A, Epstein L, Chen X, Sela H, Fahima T,

Dubcovsky J. 2009. A Kinase-START gene confers temperature-dependent

resistance to wheat stripe rust. Science 323: 1357–1360.Fu J, Liu H, Li Y, Yu H, Li X, Xiao J, Wang S. 2011.Manipulating broad-

spectrum disease resistance by suppressing pathogen-induced auxin

accumulation in rice. Plant Physiology 155: 589–602.Gao M, Hu L, Li Y, Weng Y. 2016. The chlorophyll-deficient golden leafmutation

in cucumber is due to a single nucleotide substitution in CsChlI for magnesium

chelatase I subunit. Theoretical and Applied Genetics 129: 1961–1973.Glazebrook J. 2005. Contrasting mechanisms of defense against biotrophic and

necrotrophic pathogens. Annual Review of Phytopathology 43: 205–227.Gou J-Y, Li K, Wu K, Wang X, Lin H, Cantu D, Uauy C, Dobon-Alonso A,

Midorikawa T, Inoue K et al. 2015.Wheat stripe rust resistance protein

WKS1 reduces the ability of the thylakoid-associated ascorbate peroxidase to

detoxify reactive oxygen species. Plant Cell 27: 1755–1770.Greenberg JT, Ausubel FM. 1993. Arabidopsismutants compromised for the

control of cellular damage during pathogenesis and aging. Plant Journal 4:327–341.

He X, Li Y, Pandey S, Yandell BS, Pathak M, Weng Y. 2013.QTL mapping of

powdery mildew resistance in WI 2757 cucumber (Cucumis sativus L.).Theoretical and Applied Genetics 126: 2149–2161.

No claim to original US government works

New Phytologist� 2018 New Phytologist TrustNew Phytologist (2018)

www.newphytologist.com

NewPhytologist Research 13

Hirashima M, Tanaka R, Tanaka A. 2009. Light-independent cell death induced

by accumulation of pheophorbide a in Arabidopsis thaliana. Plant and CellPhysiology 50: 719–729.

Hofius D, Munch D, Bressendorff S, Mundy J, Petersen M. 2011. Role of

autophagy in disease resistance and hypersensitive response-associated cell

death. Cell Death and Differentiation 18: 1257–1262.Holmes GJ, Wehner TC, Thornton A. 2006. An old enemy re-emerges.

American Vegetable Grower 54: 14–15.Horejsi T, Staub JE, Thomas C. 2000. Linkage of random amplified

polymorphic DNA markers to downy mildew resistance in cucumber (Cucumissativus L.). Euphytica 115: 105–113.

H€ortensteiner S. 2009. Stay-green regulates chlorophyll and chlorophyll-binding

protein degradation during senescence. Trends in Plant Science 14: 155–162.Humphry M, Bednarek P, Kemmerling B, Koh S, Stein M, Gobel U, Stuber K,

Pislewska-Bednarek M, Loraine A, Schulze-Lefert P et al. 2010. A regulon

conserved in monocot and dicot plants defines a functional module in

antifungal plant immunity. Proceedings of the National Academy of Sciences, USA107: 21896–21901.

Humphry M, Consonni C, Panstruga R. 2006. mlo-based powdery mildew

immunity: silver bullet or simply non-host resistance?Molecular Plant Pathology7: 605–610.

Ishiga Y, Rao Uppalapati S, Gill US, Huhman D, Tang Y, Mysore KS. 2015.

Transcriptomic and metabolomic analyses identify a role for chlorophyll

catabolism and phytoalexin duringMedicago nonhost resistance against Asiansoybean rust. Scientific Reports 5: 13061.

Jiang H, Li M, Liang N, Yan H, Wei Y, Xu X, Liu J, Xu Z, Chen F, Wu G.

2007.Molecular cloning and function analysis of the stay green gene in rice.

Plant Journal 52: 197–209.Jones J, Dangl J. 2006. The plant immune system. Nature 444: 323–329.Kennard WC, Poetter K, Dijkhuizen A, Meglic V, Staub JE, Havey MJ. 1994.

Linkages among RFLP, RAPD, isozyme, disease-resistance, and morphological

markers in narrow and wide crosses of cucumber. Theoretical and AppliedGenetics 89: 42–48.

Kjær BJ, Jensen HP, Jensen J, Jørgensen JH. 1990. Associations between three

ml-o powdery mildew resistance genes and agronomic traits in barley. Euphytica46: 185–193.

Kourelis J, van der Hoorn RAL. 2018. Defended to the Nines: 25 years of

resistance gene cloning identifies nine mechanisms for R protein function.

Plant Cell 30: 285–299.Krattinger SG, Lagudah ES, Spielmeyer W, Singh RP, Huerta-espino J,

Mcfadden H, Bossolini E, Selter LL, Keller B. 2009. A putative ABC

transporter confers durable resistance to multiple fungal pathogens in wheat.

Science 323: 1360–1363.Kuai B, Chen J, H€ortensteiner S. 2018. The biochemistry and molecular biology

of chlorophyll breakdown. Journal of Experimental Botany 69: 751–767.Lagudah ES, Krattinger SG, Herrera-Foessel S, Singh RP, Huerta-Espino J,

Spielmeyer W, Brown-Guedira G, Selter LL, Keller B. 2009. Gene-specific

markers for the wheat gene Lr34/Yr18/Pm38 which confers resistance tomultiple fungal pathogens. Theoretical and Applied Genetics 119: 889–898.

Li S, Pan Y, Wen C, Li Y, Liu X, Zhang X, Behera TK, Xing G, Weng Y. 2016.

Integrated analysis in bi-parental and natural populations reveals CsCLAVATA3(CsCLV3) underlying carpel number variations in cucumber. Theoretical andApplied Genetics 129: 1007–1022.

Luo Z, Zhang J, Li J, Yang C, Wang T, Ouyang B, Li H, Giovannoni J, Ye Z.

2013. A STAY-GREEN protein SlSGR1 regulates lycopene and b-caroteneaccumulation by interacting directly with SlPSY1 during ripening processes in

tomato. New Phytologist 198: 442–452.Mach JM, Castillo AR, Hoogstraten R, Greenberg JT. 2001. The Arabidopsis

accelerated cell death gene ACD2 encodes red chlorophyll catabolite reductaseand suppresses the spread of disease symptoms. Proceedings of the NationalAcademy of Sciences, USA 98: 771–776.

McGrann GRD, Stavrinides A, Russell J, Corbitt MM, Booth A, Chartrain L,

Thomas WTB, Brown JKM. 2014. A tradeoff between mlo resistance topowdery mildew and increased susceptibility of barley. Journal of ExperimentalBotany 65: 1025–1037.

Mecey C, Hauck P, Trapp M, Pumplin N, Plovanich A, Yao J, He SY. 2011. A

critical role of STAYGREEN/Mendel’s I locus in controlling disease symptom

development during Pseudomonas syringae pv tomato infection of Arabidopsis.

Plant Physiology 157: 1965–1974.Mengiste T. 2012. Plant immunity to necrotrophs. Annual Review ofPhytopathology 50: 267–294.

Moore JW, Herrera-Foessel S, Lan C, Schnippenkoetter W, Ayliffe M, Huerta-

Espino J, Lillemo M, Viccars L, Milne R, Periyannan S et al. 2015. A recently

evolved hexose transporter variant confers resistance to multiple pathogens in

wheat. Nature Genetics 47: 1494–1498.Mur LAJ, Aubry S, Mondhe M, Kingston-Smith A, Gallagher J, Timms-

Taravella E, James C, Papp I, H€ortensteiner S, Thomas H et al. 2010.Accumulation of chlorophyll catabolites photosensitizes the hypersensitive

response elicited by Pseudomonas syringae in Arabidopsis. New Phytologist 188:161–174.

Nie J, Wang Y, He H, Guo C, Zhu W, Pan J, Li D, Lian H, Pan J, Cai R. 2015.

Loss-of-function mutations in CsMLO1 confer durable powdery mildew

resistance in cucumber (Cucumis sativus L.). Frontiers in Plant Science 6: 1–14.Olczak-Woltman H, Masny A, Bartoszewski G, Płucienniczak A, Niemirowicz-

Szczytt K. 2007. Genetic diversity of Pseudomonas syringae pv. lachrymansstrains isolated from cucumber leaves collected in Poland. Plant Pathology 56:373–382.

Pan J, Tan J, Wang Y, Zheng X, Owens K, Li D, Li Y, Weng Y. 2018.

STAYGREEN (CsSGR) is a candidate for the anthracnose (Colletotrichumorbiculare) resistance locus cla in Gy14 cucumber. Theoretical and AppliedGenetics 131: 1577–1587.

Pan Y, Liang X, Gao M, Liu H, Meng H, Weng Y, Cheng Z. 2017. Round fruit

shape in WI7239 cucumber is controlled by two interacting quantitative trait

loci with one putatively encoding a tomato SUN homolog. Theoretical andApplied Genetics 130: 573–586.

Park S-Y, Yu J-W, Park J-S, Li J, Yoo S-C, Lee N-Y, Lee S-K, Jeong S-W, Seo

HSS, Koh H-J et al. 2007. The senescence-induced staygreen protein regulates

chlorophyll degradation. Plant Cell 19: 1649–1664.Pessina S, Lenzi L, Perazzolli M, Campa M, Dalla Costa L, Urso S, Val�e G,Salamini F, Velasco R, Malnoy M. 2016. Knockdown ofMLO genes reduces

susceptibility to powdery mildew in grapevine. Horticulture Research 3: 16016.Peterson C. 1975. Plant introduction in improvement of vegetable cultivars.

HortScience 10: 575.Peterson C, Williams P, Palmer M, Louward P. 1982.Wisconsin 2757

cucumber. HortScience 17: 268.Qi J, Liu X, Shen D, Miao H, Xie B, Li X, Zeng P, Wang S, Shang Y, Gu X

et al. 2013. A genomic variation map provides insights into the genetic basis of

cucumber domestication and diversity. Nature Genetics 45: 1510–1515.Ren G, An K, Liao Y, Zhou X, Cao Y, Zhao H, Ge X, Kuai B. 2007.

Identification of a novel chloroplast protein AtNYE1 regulating chlorophyll

degradation during leaf senescence in Arabidopsis. Plant Physiology 144: 1429–1441.

Rinaldo A, Gilbert B, Boni R, Krattinger SG, Singh D, Park RF, Lagudah E,

Ayliffe M. 2017. The Lr34 adult plant rust resistance gene provides seedlingresistance in durum wheat without senescence. Plant Biotechnology Journal 15:894–905.

Risk JM, Selter LL, Chauhan H, Krattinger SG, Kumlehn J, Hensel G, Viccars

LA, Richardson TM, Buesing G, Troller A et al. 2013. The wheat Lr34 geneprovides resistance against multiple fungal pathogens in barley. PlantBiotechnology Journal 11: 847–854.

Sakuraba Y, Kim YS, Yoo SC, H€ortensteiner S, Paek NC. 2013. 7-

Hydroxymethyl chlorophyll a reductase functions in metabolic channeling of

chlorophyll breakdown intermediates during leaf senescence. Biochemical andBiophysical Research Communications 430: 32–37.

Sakuraba Y, Park S-Y, Paek N-C. 2015. The divergent roles of STAYGREEN

(SGR) homologs in chlorophyll degradation.Molecules and Cells 38: 390–395.Sakuraba Y, Schelbert S, Park S-Y, Han S, Lee B-D, Andres CB, Kessler F,

Hortensteiner S, Paek N-C. 2012. STAY-GREEN and chlorophyll catabolic

enzymes interact at light-harvesting complex II for chlorophyll detoxification

during leaf senescence in Arabidopsis. Plant Cell 24: 507–518.Sato Y, Morita R, Nishimura M, Yamaguchi H, Kusaba M. 2007.Mendel’s

green cotyledon gene encodes a positive regulator of the chlorophyll-degrading

pathway. Proceedings of the National Academy of Sciences, USA 104: 14169–14174.

New Phytologist (2018) No claim to original US government works

New Phytologist� 2018 New Phytologist Trustwww.newphytologist.com

Research

NewPhytologist14

Schliep KP. 2011. phangorn: phylogenetic analysis in R. Bioinformatics 27: 592–593.

Schnippenkoetter W, Lo C, Liu G, Dibley K, Chan WL, White J, Milne R,

Zwart A, Kwong E, Keller B et al. 2017. The wheat Lr34multipathogen

resistance gene confers resistance to anthracnose and rust in sorghum. PlantBiotechnology Journal 15: 1387–1396.

Serrano I, Audran C, Rivas S. 2016. Chloroplasts at work during plant innate

immunity. Journal of Experimental Botany 67: 3845–3854.Shimoda Y, Ito H, Tanaka A. 2016. Arabidopsis STAY-GREEN, Mendel’s green

cotyledon gene, encodes magnesium-dechelatase. Plant Cell 28: 2147–2160.Słomnicka R, Olczak-Woltman H, Korzeniewska A, Niemirowicz-Szczytt K,

Bartoszewski G. 2016. Evaluation of a cucumber RILs population for

resistance to angular leaf spot. In: Kozik EU, Paris HS, eds. Proceedings ofCucurbitaceae 2016, 11th EUCARPIA meeting on genetics and breeding ofCucurbitaceae. Poland: Warsaw, 231–233.

Spoel SH, Johnson JS, Dong X. 2007. Regulation of tradeoffs between plant

defenses against pathogens with different lifestyles. Proceedings of the NationalAcademy of Sciences, USA 104: 18842–18847.

Sucher J, Boni R, Yang P, Rogowsky P, B€uchner H, Kastner C, Kumlehn J,

Krattinger SG, Keller B. 2017. The durable wheat disease resistance gene Lr34confers common rust and northern corn leaf blight resistance in maize. PlantBiotechnology Journal 15: 489–496.

Thomas A, Carbone I, Choe K, Quesada-Ocampo LM, Ojiambo PS. 2017.

Resurgence of cucurbit downy mildew in the United States: insights from

comparative genomic analysis of Pseudoperonospora cubensis. Ecology andEvolution 7: 6231–6246.

Thomas H, Ougham H. 2014. The stay-green trait. Journal of ExperimentalBotany 65: 3889–3900.

Thomma BPHJ, Eggermont K, Penninckx IAM, Mauch-mani B, Vogelsang R,

Cammue BP, Broekaert WF. 1998. Separate jasmonate-dependent and

salicylate-dependent defense-response pathways in Arabidopsis are essential forresistance to distinct microbial pathogens. Proceedings of the National Academyof Sciences, USA 95: 15107–15111.

Thomma BPHJ, N€urnberger T, Joosten MHAJ. 2011.Of PAMPs and effectors:

the blurred PTI–ETI dichotomy. Plant Cell 23: 4–15.Tian M, Win J, Savory E, Burkhardt A, Held M, Brandizzi F, Day B. 2011. 454

Genome sequencing of Pseudoperonospora cubensis reveals effector proteins witha QXLR translocation motif.Molecular Plant–Microbe Interactions 24: 543–553.

van Vliet GJA, Meysing WD. 1974. Inheritance of resistance to

Pseudoperonospora cubensis Rost. in cucumber (Cucumis sativus L.). Euphytica23: 251–255.

van Vliet GJA, Meysing WD. 1977. Relation in the inheritance of resistance to

Pseudoperonospora cubensis Rost and Sphaerotheca fuliginea Poll. in cucumber

(Cucumis sativus L.). Euphytica 26: 793–796.Wang Y, VandenLangenberg K, Wehner TC, Kraan PAG, Suelmann J, Zheng

X, Owens K, Weng Y. 2016.QTL mapping for downy mildew resistance in

cucumber inbred line WI7120 (PI 330628). Theoretical and Applied Genetics129: 1493–1505.

Wang Y, VandenLangenberg K, Wen C, Wehner TC, Weng Y. 2018.QTL

mapping of downy and powdery mildew resistances in PI 197088 cucumber

with genotyping-by-sequencing in RIL population. Theoretical and AppliedGenetics 131: 597–611.

Wehner TC, Shetty NV. 1997. Downy mildew resistance of the cucumber

germplasm collection in North Carolina field tests. Crop Science 37: 1331–1340.

Wei Q, Guo Y, Kuai B. 2011. Isolation and characterization of a chlorophyll

degradation regulatory gene from tall fescue. Plant Cell Reports 30: 1201–1207.Weng Y, Colle M, Wang Y, Yang L, Rubinstein M, Sherman A, Ophir R,

Grumet R. 2015.QTL mapping in multiple populations and development

stages reveals dynamic quantitative trait loci for fruit size in cucumbers of

different market classes. Theoretical and Applied Genetics 128: 1747–1763.Wiesner-Hanks T, Nelson R. 2016.Multiple disease resistance in plants. AnnualReview of Phytopathology 54: 229–252.

Wyszogrodzka A, Williams P, Peterson C. 1987.Multiple-pathogen inoculation

of cucumber (Cucumis sativus) seedlings. Plant Disease 71: 275.

Yang L, Koo D-H, Li Y, Zhang X, Luan F, Havey MJ, Jiang J, Weng Y. 2012.

Chromosome rearrangements during domestication of cucumber as revealed by

high-density genetic mapping and draft genome assembly. Plant Journal 71:895–906.

Yoshimoto K, Jikumaru Y, Kamiya Y, Kusano M, Consonni C, Panstruga R,

Ohsumi Y, Shirasu K. 2009. Autophagy negatively regulates cell death by

controlling NPR1-dependent salicylic acid signaling during senescence and the

innate immune response in Arabidopsis. Plant Cell 21: 2914–2927.Zhou C, Han L, Pislariu C, Nakashima J, Fu C, Jiang Q, Quan L, Blancaflor

EB, Tang Y, Bouton JH et al. 2011. From model to crop: functional analysis

of a STAY-GREEN gene in the model legumeMedicago truncatula and effectiveuse of the gene for alfalfa improvement. Plant Physiology 157: 1483–1496.

Zhou J, Yu JQ, Chen Z. 2014. The perplexing role of autophagy in plant innate

immune responses.Molecular Plant Pathology 15: 637–645.Zurbriggen MD, Carrillo N, Tognetti VB, Melzer M, Peisker M, Hause B,

Hajirezaei MR. 2009. Chloroplast-generated reactive oxygen species play a

major role in localized cell death during the non-host interaction between

tobacco and Xanthomonas campestris pv. vesicatoria. Plant Journal 60: 962–973.

Supporting Information

Additional Supporting Information may be found online in theSupporting Information section at the end of the article:

Fig. S1 Typical symptoms on cucumber plants on infection withthe downy mildew (DM) pathogen under different environ-ments.

Fig. S2 Phenotypic distribution of downy mildew (DM) diseasescores in the G9RIL population from Gy149 9930 mating atthree rating times with two scoring criteria under three environ-ments.

Fig. S3 Frequency distribution and Spearman rank correlationsof mean downy mildew (DM) disease scores in the G9RIL popu-lation at three rating times with two criteria under three environ-ments.

Fig. S4 Phenotypic distribution of mean downy mildew (DM)disease scores, as well as frequency distribution and Spearmanrank correlations of disease scores, in the WTRIL populationfrom the cross between WI2757 and True Lemon at three ratingtimes with three scoring criteria under two environments.

Fig. S5 Logarithm of odds (LOD) profiles of downy mildew(DM) resistance quantitative trait loci (QTLs) detected in G9RILand WTRIL populations in multiple environments using differ-ent rating criteria and at different rating times.

Fig. S6 Typical symptoms on cucumber plants on infection withthe angular leaf spot (ALS) pathogen in controlled environments.

Fig. S7 Distribution of mean angular leaf spot (ALS) diseasescores in the G9RIL and WTF23 populations in different pheno-typing experiments.

Fig. S8 Alignment of CsSGR in cucumber lines 9930, Gy14 andWI2757.

No claim to original US government works

New Phytologist� 2018 New Phytologist TrustNew Phytologist (2018)

www.newphytologist.com

NewPhytologist Research 15

Fig. S9 Association analysis of single nucleotide polymorphisms(SNPs) within the 93.7-kb region harboring the CsSGR locuswith pre-2004 downy mildew (DM), angular leaf spot (ALS) andanthracnose (AR) resistances in 41, 82 and 50 cucumber acces-sions, respectively.

Fig. S10 Predicted three-dimensional model of theSTAYGREEN (SGR) domain of the CsSGR protein in suscepti-ble 9930 and resistant Gy14.

Fig. S11 Schematic diagram of linkage disequilibrium (LD)blocks at the dm1/psl/cla locus.

Fig. S12 Dendrogram of 152 cucumber accessions based onmarkers within a 20-kb region harboring the dm1/psl/clalocus.

Fig. S13 Haplotypes of selected cucumber lines in the CsSGR(dm1/psl/cla) gene region.

Table S1 Details of environments used for phenotypic screeningin different mapping populations for quantitative trait locus(QTL) mapping of downy mildew (DM) and angular leaf spot(ALS) resistances in Gy14 and WI2757

Table S2 Information on the primers used in genetic mappingand gene expression in this study

Table S3 Staygreen and staygreen-like proteins from differ-ent plant species used for phylogenetic analysis in this study

Table S4 Cucumber materials and single nucleotidepolymorphism (SNP) genotypes at dm1/psl/cla locus used forhaplotyping and association analysis in the present study

Table S5 Phenotypic means and standard deviation (SD) ofdowny mildew (DM) and angular leaf spot (ALS) mean diseasescores in parental lines, their F1, and recombinant inbred lines(RILs) or F2:3 families derived from crosses between Gy14 and9930, as well as WI2757 and True Lemon, across all environments

Table S6 Analysis of variance (ANOVA) and estimation of heri-tability (h2) for resistance to downy mildew (DM) using generalimpression (GI) data of the G9RIL population

Table S7 Summary of linkage map developed with 1618 singlenucleotide polymorphism (SNP) loci and 87 WTRILs

Table S8 Annotated genes in the 97.3-kb region of the dm1 locusbased on 9930 draft genome assembly V2.0

Please note: Wiley Blackwell are not responsible for the contentor functionality of any Supporting Information supplied by theauthors. Any queries (other than missing material) should bedirected to the New Phytologist Central Office.

New Phytologist is an electronic (online-only) journal owned by the New Phytologist Trust, a not-for-profit organization dedicatedto the promotion of plant science, facilitating projects from symposia to free access for our Tansley reviews and Tansley insights.

Regular papers, Letters, Research reviews, Rapid reports and both Modelling/Theory and Methods papers are encouraged. We are committed to rapid processing, from online submission through to publication ‘as ready’ via Early View – our average timeto decision is <26 days. There are no page or colour charges and a PDF version will be provided for each article.

The journal is available online at Wiley Online Library. Visit www.newphytologist.com to search the articles and register for tableof contents email alerts.

If you have any questions, do get in touch with Central Office (np-centraloffice@lancaster.ac.uk) or, if it is more convenient,our USA Office (np-usaoffice@lancaster.ac.uk)

For submission instructions, subscription and all the latest information visit www.newphytologist.com

New Phytologist (2018) No claim to original US government works

New Phytologist� 2018 New Phytologist Trustwww.newphytologist.com

Research

NewPhytologist16

View publication statsView publication stats