Article type : Original Article

Accepted Article
ETHQV6.3 is involved in melon climacteric fruit ripening and is encoded by a NAC

domain transcription factor

Pablo Ríos1$, Jason Argyris1, Juan Vegas1&, Carmen Leida2, Merav Kenigswald3,4, Galil

Tzuri3, Christelle Troadec5, Abdelhafid Bendahmane5, Nurit Katzir3, Belén Picó6, Antonio J.

Monforte2, Jordi Garcia-Mas1*

1
IRTA, Centre for Research in Agricultural Genomics CSIC-IRTA-UAB-UB, Barcelona,

Spain
2
Instituto de Biología Molecular y Celular de Plantas (IBMCP), Universitat Politècnica de

València (UPV)-Consejo Superior de Investigaciones Científicas (CSIC), Valencia, Spain

3
Department of Vegetable Research, Agricultural Research Organization (ARO), Newe Ya’ar

Reseach Center, Ramat Yishay, Israel

4
Robert H. Smith Faculty of Agriculture, Food and Environment, The Hebrew University of

Jerusalem, Israel
5
Institute of Plant Sciences Paris-Saclay, IPS2, INRA, CNRS, University of Paris-Sud,

University of Evry, University of Paris-Diderot, Sorbone Paris-Cité, University of Paris-

Saclay, Orsay, France

6
COMAV-UPV, Institute for the Conservation and Breeding of the Agricultural Biodiversity,

Universitat Politècnica de València, Valencia, Spain

$
Current position: Syngenta España S.A., 04710 El Ejido, Spain
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/tpj.13596
This article is protected by copyright. All rights reserved.
 &
Current position: Sesvanderhave N.V., 3300 Tienen, Belgium
Accepted Article
*
To whom correspondence should be addressed: Jordi Garcia-Mas, jordi.garcia@irta.cat,

IRTA, Centre for Research in Agricultural Genomics CSIC-IRTA-UAB-UB, Edifici CRAG,

Campus UAB, 08193 Cerdanyola, Barcelona, Spain

Running title: A NAC transcription factor is involved in melon fruit ripening

Keywords: non-climacteric ripening, ETHQV6.3, NAC transcription factor, TILLING

mutant, Cucumis melo, introgression line

e-mail addresses:

pablo.rios@syngenta.com

jason.argyris@irta.cat

juan.vegas@sesvanderhave.com

carmen.leida@gmail.com

merav.kenigswald@mail.huji.ac.il

galilt@volcani.agri.gov.il

christelle.troadec@ips2.universite-paris-saclay.fr

abdel.bendahmane@ips2.universite-paris-saclay.fr

katzirn@volcani.agri.gov.il

mpicosi@btc.upv.es

amonforte@ibmcp.upv.es

jordi.garcia@irta.cat

This article is protected by copyright. All rights reserved.

 Summary

Fruit ripening is divided into climacteric and non-climacteric types depending on the
Accepted Article
presence or absence of a transient rise in respiration rate and the production of autocatalytic

ethylene. Melon is ideal for the study of fruit ripening, as both climacteric and non-

climacteric varieties exist. Two introgressions of the non-climacteric accession PI 161375,

encompassed in the QTLs ETHQB3.5 and ETHQV6.3, into the non-climacteric “Piel de

Sapo” background are able to induce climacteric ripening independently. We report that the

gene underlying ETHQV6.3 is MELO3C016540 (CmNAC-NOR), encoding a NAC (NAM,

ATAF1,2, CUC2) transcription factor that is closely related to the tomato NOR (non-

ripening) gene. CmNAC-NOR was functionally validated through the identification of two

TILLING lines carrying non-synonymous mutations in the conserved NAC domain region. In

an otherwise highly climacteric genetic background, both mutations provoked a significant

delay in the onset of fruit ripening and in the biosynthesis of ethylene. The PI 161375 allele

of ETHQV6.3 is similar to that of climacteric lines of the cantalupensis type, and when

introgressed into the non-climacteric “Piel de Sapo”, partially restores its climacteric ripening

capacity. CmNAC-NOR is expressed in fruit flesh of both climacteric and non-climacteric

lines, suggesting that the causal mutation may not be acting at the transcriptional level. The

use of a comparative genetic approach in a species with both climacteric and non-climacteric

ripening is a powerful strategy to dissect the complex mechanisms regulating the onset of

fruit ripening.

Introduction

Fruit ripening is the last stage of the fruit developmental program in which fruit

undergoes a series of physiological and metabolic changes that protect the seeds from

environmental conditions and promote their dispersion (Giovannoni, 2001). Two types of

This article is protected by copyright. All rights reserved.

 ripening have been defined with respect to the role of the plant hormone ethylene: climacteric

ripening, which is characterized by the autocatalytic biosynthesis of ethylene and the increase
Accepted Article
in respiration at the onset of ripening, and non-climacteric ripening, in which both ethylene

production and respiration rate remain low throughout the process (McMurchie et al., 1972,

Lelièvre et al., 1997). Ethylene is involved in many plant developmental processes, including

flower development and sexual determination, abscission and plant organ senescence, and

biotic and abiotic stress responses (Abeles et al., 1992). It plays a primary role in the

regulation of climacteric fruit ripening, by acting as a triggering signal initiating the

biochemical and physiological processes that lead to the characteristics of a ripe fruit

(McMurchie et al., 1972). These usually include the formation of an abscission layer,

changes in fruit color, development of aroma, fruit softening and a short postharvest life.

Conversely, non-climacteric fruits do not typically display these characteristics. Despite the

physiological differences between climacteric and non-climacteric ripening several common

features exist, suggesting that common molecular and regulatory processes may underlie both

types of ripening (Giovannoni, 2004).

Ripening has been a major focus of plant breeding in fleshy fruits, with special effort

on the improvement of organoleptic quality and post-harvest durability (Handa et al., 2014).

Tomato is the model species for studying climacteric ripening, and important advances in the

elucidation of the ethylene biosynthetic pathway (Alexander and Grierson, 2002), as well as

its signalling (Klee, 2004) and transduction components (Adams-Phillips et al., 2004) have

been achieved. The availability of ripening-impaired mutants in tomato allowed the

identification of three main transcription factors involved in its regulation: RIN (ripening-

inhibitor), CNR (Colorless non-ripening) and NOR (non-ripening) (Vrebalov et al., 2002,

Manning et al., 2006, Giovannoni, 2007). The rin, Cnr and nor mutants produce completely

developed fruits with fertile seeds that are unable to initiate fruit ripening and remain in a

This article is protected by copyright. All rights reserved.

 mature green stage. This phenotype is due to the inhibition of autocatalytic ethylene

biosynthesis and respiration, and absence of flesh softening, aroma volatiles biosynthesis,
Accepted Article
chlorophyll degradation and pigment biosynthesis (Robinson and Tomes, 1968, Tigchelaar et

al., 1973, Thompson et al., 1999, Kovács et al., 2009). Although fruits from mutants in any

of these three genes fail to ripen in response to exogenous ethylene, the expression of

ethylene-responsive genes is not impaired in fruits or in other plant tissues (Giovannoni,

2007). It has been suggested that RIN, CNR and NOR may belong to a highly conserved

ripening regulation system that controls not only ethylene biosynthesis, but the overall

ripening process, and that this system is common to climacteric and non-climacteric species

alike (Klee and Giovannoni, 2011). Additional transcription factors involved in the regulation

of fruit ripening include the positive regulators TAGL1 (Itkin et al., 2009), LeHB-1 (Lin et al.,

2009) and SlNAC4 (Zhu et al., 2014), and the negative regulators LeAP2a (Chung et al.,

2010) and LeERF6 (Lee et al., 2012). Recent studies also suggest the involvement of

miRNAs (Gao et al., 2015) and epigenetic regulation (Zhong et al., 2013, Liu et al., 2015) in

fruit ripening. Despite these recent advances, the full complexity of the ethylene-dependent

and independent regulation of fruit ripening remains to be resolved.

Melon (Cucumis melo L.) has emerged as an interesting model for fruit ripening

studies due to the existence of both climacteric and non-climacteric genotypes within the

species (Ezura and Owino, 2008). The cantalupensis (e.g. “Védrantais”) and reticulatus

(“Dulce”) varieties show climacteric ripening and short shelf-life, whereas inodorus varieties

like “Piel de Sapo” (PS), are non-climacteric and show long shelf-life (Saladié et al., 2015).

The role of ethylene in melon fruit ripening regulation was demonstrated by reducing its

biosynthesis in antisense CmACO1 “Védrantais” plants (Ayub et al., 1996, Pech et al., 2008).

These experiments showed that the development of an abscission layer, the rind color change

and the production of aroma volatiles were processes strictly ethylene-dependent, while flesh

This article is protected by copyright. All rights reserved.

 softening only partially so. Conversely, carotenoid biosynthesis, sugar and organic acid

accumulation were ethylene-independent.

Accepted Article
The genetic basis of melon fruit ripening was first studied in a RIL population

generated from the cross between the climacteric variety “Védrantais” (cantalupensis) and

the non-climacteric exotic accession PI 1611375 (SC, conomon) (Perin et al., 2002). Al-3 and

Al-4 in chromosomes 8 and 9, respectively, were found to be involved in the development of

an abscission layer and the autocatalytic ethylene biosynthesis and four QTLs in

chromosomes 1, 2, 3 and 11 were involved in the amount of ethylene produced. More

recently, the near-isogenic line SC3-5-1, originated from the cross between PS and SC,

showed climacteric ripening despite both parents being non-climacteric (Eduardo et al.,

2005). SC3-5-1 contains two QTLs, ETHQB3.5 and ETHQV6.3 in chromosomes 3 and 6,

respectively, involved in the regulation of climacteric ripening (Moreno et al., 2008, Vegas et

al., 2013). Both QTLs are capable of inducing climacteric ripening in the non-climacteric

background of PS individually, but they also interact to increase ethylene biosynthesis and

intensity of the ripening-associated processes (Vegas et al., 2013). Interestingly, there was no

commonality between the QTLs from this study and Perin et al. (2002), suggesting that the

genetic basis of fruit ripening in melon is complex and variety-specific.

In previous work, ETHQV6.3 was mapped to a 4.5 Mb region of melon LG VI (Vegas

et al., 2013). In this study we identified, characterized, and functionally validated

MELO3C016540 (CmNAC-NOR) as the causal gene for ETHQV6.3.

This article is protected by copyright. All rights reserved.

 Results

Positional cloning of ETHQV6.3

Accepted Article
The 2008-F2 mapping population, obtained after crossing the near isogenic line SC3-

5-1 (carrying both ETHQV6.3 and ETHQB3.5) to PS, was used to map ETHQV6.3 in a 4.5

Mb centromeric region of melon chromosome 6 (Vegas et al., 2013). We obtained the 2012-

F4 segregating population from 7M80-11.4, an individual of the 2008-F2 mapping population,

heterozygous for ETHQV6.3 and fixed for the PS alleles for ETHQB3.5 (Figure S1). The

genotyping of 1,131 2012-F4 individuals with flanking markers SNP-64658 and SNP-

2826073 allowed for the identification of 27 recombinants in the interval (Figure S2a).

Twenty-four SNPs polymorphic between PS and SC and evenly distributed in the SNP-

64658/SNP-2826073 interval (Table S1) were used to delimit the recombination point in each

recombinant. A progeny test was performed with 15 informative recombinants, where 20

individuals per family were phenotyped for climacteric ripening after recording fruit

abscission (Figure S3), which allowed the mapping of ETHQV6.3 between markers SNP-

2691690 and SNP-2826073 in a 139 kb interval (Figure S2a; Table S2).

This interval contains 5 annotated genes in the reference genome v3.5 (Garcia-Mas et

al., 2012) (Table S3), two of which are transcription factors of the NAC-domain family,

MELO3C016536 and MELO3C016540. Recombinants R24, R25 and R26 were genotyped

with 6 additional SNPs between SNP-2691690 and SNP-2826073 (SEQ-1 to SEQ-6, Table

S1), which allowed a reduction of the interval to 80.7 kb between markers SEQ-3 and SNP-

2826073 containing MELO3C016538, MELO3C016539 and MELO3C016540 (Figure S2b-

c). MELO3C016538 and MELO3C016539 encode short mRNAs of 247 and 396 bp,

respectively, with no homologies or reported expression in sequence databases. Thus

MELO3C016540 (CmNAC-NOR), identified as a member of the NAC-domain transcription

factor family, was considered a good candidate for ETHQV6.3.

This article is protected by copyright. All rights reserved.

 A QTL for climacteric ripening from a different genetic background maps to an identical
Accepted Article
genomic interval as CmNAC-NOR

An F3 population obtained from the cross between the “Noy-Amid” (non-climacteric,

inodorous type) and “Dulce” (climacteric, reticulatus type) was phenotyped for ethylene

emission at harvest. Parental lines, F1 and 131 F3 plants representing the whole scale of

ethylene emission were selected for genotyping from 700 F3 plants previously evaluated for

ethylene emission in 2013. The same individuals were genotyped with 76,988 SNPs

identified by RAD-seq and used to map QTLs for ethylene emission at harvest. One of the

QTLs (LOD 5.3, r2 = 0.16) collocated with ETHQV6.3 in chromosome 6. The QTL interval

included 160 annotated genes and a bin of 17 genes at the LOD peak (MELO3C016523 to

MELO3C016539) (Figure S4; Table S4). CmNAC-NOR was located adjacent to the peak,

with only one SNP in the intergenic region separating them, strengthening the findings

presented above and suggesting that the allele of CmNAC-NOR for ETHQV6.3 may be

common in climacteric/non-climacteric melon germplasm.

CmNAC-NOR belongs to the melon NAC domain transcription factor family and is

phylogenetically related to the tomato SlNAC-NOR

Transcription factors of the NAC family are plant specific. They contain a conserved

domain NAC (NAM, ATAF1,2, CUC2) distributed with subdomains A-E in the N-terminal

region, which is involved in DNA binding (Puranik et al., 2012). We identified 81 genes of

the NAC-domain family in the melon genome, putatively encoding 92 proteins and evenly

distributed in the 12 chromosomes. CmNAC-NOR is 1,771 bp in length, contains three exons

(184, 314 and 564 bp) and two introns (89 and 183 bp) and encodes a predicted 352 aa

protein (Figure 1, Figure 2a).

This article is protected by copyright. All rights reserved.

 CmNAC-NOR was aligned with 37 NAC proteins of known function from different

plant species (Table S5) and the melon NAC-domain family (Figure 3). The alignment
Accepted Article
showed a highly conserved N-terminal region of approximately 200 aa containing the NAC

domain. The proteins in the cladogram clustered according to their biological function: group

1 mainly includes NAC proteins involved in growth and development, but also cell wall

metabolism and senescence; group 2a contains NAC proteins involved in stress response, and

group 2b contains NAC proteins involved in senescence, but is more heterogeneous. Group

2b contains the tomato SlNAC-NOR (non-ripening) involved in fruit ripening, which clusters

close to CmNAC-NOR. Another tomato NAC protein also involved in fruit ripening,

SlNAC4, is clustered in group 2a. The phylogenetic analysis indicates that CmNAC-NOR is a

closely related homologue of the tomato SlNAC-NOR, a regulator of climacteric fruit ripening

(Giovannoni, 2004), reinforcing its potential as the candidate gene for ETHQV6.3.

Association of CmNAC-NOR with climacteric ripening

In a previous work, Leida et al. (2015) studied the association of candidate genes with

climacteric behaviour in a panel of 175 melon accessions that included wild relatives, feral

types, landraces and breeding lines, representing the diversity of the species. The accessions

were phenotyped for fruit ripening behaviour and genotyped with a set of 251 SNPs, of which

60 were located in 34 candidate genes involved in ethylene and cell wall pathways. Two

SNPs on chromosomes 11 and 12 were associated with ripening traits, but no association was

detected with SNPs on chromosome 6, as none located near CmNAC-NOR were assayed.

A non-synonymous SNP G411T in CmNAC-NOR (Table S6) was previously

identified after re-sequencing eight pools of accessions that represented the main melon

botanical groups and was polymorphic between climacteric and non-climacteric groups of

melons (snv26555 available at www.melogene.net) (Blanca et al., 2012). Climacteric

This article is protected by copyright. All rights reserved.

 cantalupensis and momordica melons had the T allele, the non-climacteric inodorus melons

had the G allele, and both alleles were present in the group of African agrestis showing
Accepted Article
variable climacteric behaviour (Leida et al., 2015). We genotyped SNP G411T in the panel of

175 melon accessions used in Leida et al. (2015) and its association with ripening related

traits was assessed (Table S7). SNP G411T was found to be highly associated with ripening

type (p= 4.35x10-5) and abscission layer formation (p= 6.48x10-4), further supporting the

implication of CmNAC-NOR in climacteric fruit ripening.

Sequence diversity of CmNAC-NOR in melon germplasm

We selected a group of 54 melon accessions representing 11 of the 16 botanical

groups (Pitrat, 2008) of the two subspecies melo and agrestis (Table S8) from the above-

mentioned panel to characterize the genetic variability of CmNAC-NOR. We identified 12

SNP and 5 indel in 54 accessions, distributed in the promoter region (2), 5’UTR (3), exons

(6), introns (4), 3’UTR (1) and the terminator region (1) (Figure 1; Table S8). A phylogenetic

analysis of 47 of these sequences showed a clear separation of the cantalupensis and inodorus

groups, although five cantalupensis accessions were clustered in the inodorus group (Figure

1). Interestingly, the allele of SC clustered close to the climacteric cantalupensis group. As

there was a strong correlation between the three variables that were phenotyped in the

accessions panel, ripening type was used in modelling the effects of the sequence differences.

Seven polymorphisms in CmNAC-NOR were significantly related to ripening type (Figure 1;

Tables S6 and S8). G411T and T533A produced non-synonymous amino acid changes

A108S and S236N, respectively, although located outside the NAC domain region and

predicted as neutral (Figure S5). The polymorphisms showing the strongest significance were

INDEL-282 and INDEL-126, located in the promoter and the 5’UTR, respectively. INDEL-

126 is particularly interesting as it shows a 26 bp indel in the 5’UTR (Figure S6). The

This article is protected by copyright. All rights reserved.

 INDEL-126 analysis in the accessions resulted in the identification of 9 alleles, structured in

4 blocks with a polyA track (A), and three repeats GAGAAAA (B), GAAAAAA (C) and
Accepted Article
GAAATAA (D). The SC allele (CON) is similar to the cantalupensis one (CAN), containing

only block A, whereas PS (INO) contains the block structure ABCD.

Functional validation of CmNAC-NOR through the characterization of TILLING mutants

In order to validate CmNAC-NOR as the candidate gene for ETHQV6.3, we screened

for mutants using the climacteric “Charentais Mono” TILLING platform (Dahmani-Mardas

et al., 2010). We screened 6,200 M2 families with two overlapping amplicons, A1 and A2 of

920 and 807 bp, respectively, which covered the 1,334 bp ORF of CmNAC-NOR. We

identified 21 families containing 20 mutations (Table S9). Family 5388 was discarded as it

contained three mutations (T411G, A533T and A978G) that are not the expected G:C to A:T

change produced by EMS. This resulted in 20 mutant M2 families containing 17 mutations.

We identified 12 mutations in exons, 3 in introns and 2 in the 3’UTR, with eight of them

producing non-synonymous amino acid changes (Figure 2a; Table S9). Three of the non-

synonymous mutations in the CmNAC-NOR protein were located between residues 15 and

178, corresponding to the NAC domain (Figure S7). E59K and P129L were located in

subdomains B and D, respectively, and S164F was placed near subdomain E. We used

PROVEAN (Choi et al., 2012) to predict the effect of the mutations, which suggested E59K,

P129L and S164F as deleterious mutations.

The mutant families were phenotyped in two consecutive seasons, after discarding

families 228, 4978 and 503 that shared the same mutation as 2923, 4321 and 502 (Table S9).

The first season the number of seed for some of the eight mutant families was limited, which

resulted in the availability of a low number (< 5) of homozygous wild type (W) and

homozygous mutant (M) individuals per family. We used days from pollination to abscission

This article is protected by copyright. All rights reserved.

 to phenotype the mutant families, and families 246 (E59K) and 502 (P342L) significantly

increased the time to abscission in M plants in 6.4 and 11.7 days, respectively. However, the
Accepted Article
low number of individuals phenotyped for each class in each family, the loss of several fruits

that were severely affected by a fungal disease, and the absence of fruit abscission in some

control “Charentais Mono” plants, made that the number of replicates were low to apply a

powerful statistics analysis. A new phenotyping assay was performed in a second season,

where we chose to phenotype days from pollination to external color change as a more robust

measure of ripening. The external color change is a good approximation of the peak of

ethylene production in climacteric fruits, as we have observed in a RIL population from the

cross of “Védrantais” x PS. Six mutant families were phenotyped (246, 432, 4933, 3717,

2503 and 502), using a higher number of W and M allelic groups per family (between n=7

and n=18). In two families, both allelic groups showed statistically significant differences in

ripening behaviour: 246 (E59K; p-value=4.4 x e-8) and 432 (P129L; p-value=1.6 x 10-6),

with 7.2 and 5.6 additional days to external color change, respectively, compared to the

controls (Table S10; Figure 2b).

We applied a method for measuring ethylene fruit production, based in non-invasive

ethylene quantification in attached fruit with chromatography-mass spectrometry (Pereira et

al., in press), in the fruits of the mutant families 246 and 432. We observed a significant

increase in the days from pollination to the production of the ethylene peak in both families

(37.3 days W vs. 45.7 days M in family 246; 38 days W vs. 42 days M in family 432) (Figure

4a, Table 1), coinciding with 6.1 and 5.7 additional days to external color change. The values

for days from pollination to abscission, although not significant, were also increased in both

families (Table 1). However, we could not see a significant difference in the amount of

ethylene produced in both mutant families (Figure 4b), probably due to other genes in the

“Charentais Mono” genetic background that also control the ripening process.

This article is protected by copyright. All rights reserved.

 The ripening delay observed in the mutant families 246 (E59K) and 432 (P342L)

confirmed that CmNAC-NOR is involved in the control of climacteric fruit ripening. Both
Accepted Article
mutations are located in subdomains B and D of the NAC domain, in residues that are

conserved in NAC proteins of known function (Figure S7).

CmNAC-NOR is expressed primarily in fruit

In order to know if CmNAC-NOR plays a role only in fruit, or it is also expressed in

other organs, we performed qPCR in the non-climacteric PS, and in the NILs containing

ETHQB3.5 (GF35), ETHQV6.3 (GF40) or both of them (GF31) (Figure 5a). CmNAC-NOR is

highly expressed in fruit tissue of all four genotypes, both climacteric and non-climacteric,

during fruit development at 20 DAP, 30 DAP and harvest, whereas the expression in leaves

and roots is very low. We also tested the expression of another NAC-domain containing gene,

MELO3C016536, which is also located in the original ETHQV6.3 interval (Figure 5e).

MELO3C016536 is expressed in fruit tissue in GF31, GF35 and GF40, but it is not expressed

in fruit of the non-climacteric PS. We also tested the expression of three genes known to be

involved in ethylene biosynthesis in melon fruit: CmACO1, CmACS1 and CmACS5 (Saladié

et al., 2015) (Figure 5b-c-d, Table S11). All three showed the highest expression in GF31 and

GF35 at harvest, and much lower expression in GF40. CmACO1 and CmACS1 expression

was also detected in root tissue.

Discussion

The map-based cloning of the ripening QTL ETHQV6.3, identified in the PI 161375

(SC) x “Piel de Sapo” (PS) genetic background, revealed that the underlying gene is CmNAC-

NOR, which encodes a transcription factor of the NAC (NAM, ATAF1,2 and CUC2) family.

A QTL for climacteric ripening in a mapping population derived from distinct parental lines

This article is protected by copyright. All rights reserved.

 (“Noy Amid” x “Dulce”) also maps to the identical genome interval containing CmNAC-

NOR. Furthermore, a genome wide association analysis showed that SNP G411T, present in
Accepted Article
CmNAC-NOR, is strongly associated with ripening behaviour in a panel of 175 melon

accessions. Taken together, these findings support the involvement of CmNAC-NOR in the

climacteric ripening process.

The confirmation of CmNAC-NOR as the causal gene of ETHQV6.3 was demonstrated

after characterizing several TILLING mutants in the highly climacteric “Charentais Mono”

genetic background (Dahmani-Mardas et al., 2010). Two mutant families containing the non-

synonymous mutations E59K and P232L showed a significant delay in the onset of the

climacteric ripening, both at the level of external color change and presence of an ethylene

peak, when compared to controls. The delay of the ripening process is therefore compatible

with non-synonymous amino acid changes in subdomains B and D, respectively, of the

highly conserved NAC domain region causing an alteration of gene function (Figure S7).

The NAC domain transcription factors (TF) constitute one of the largest families of

plant TFs (Puranik et al., 2012). A phylogenetic analysis of the melon NAC gene family

including NAC proteins of known function from different plant species suggests that

CmNAC-NOR is a closely related homologue of the tomato Nor gene, which is involved in

fruit ripening (SlNAC-NOR, Figure 3). Both proteins are included in a clade that contains

other NAC proteins involved in stress response and senescence processes (Zhu et al., 2014).

The tomato nor (non-ripening) mutant (Tigchelaar et al., 1973) produces fruit with mature

seed. However the characteristic respiration and ethylene peaks, the degradation of

chlorophylls, and the biosynthesis of carotenes observed during ripening in wild type tomato

are absent (Klee and Giovannoni, 2011). A network analysis combining transcriptome,

proteome and metabolome data using the tomato mutants nor, rin (ripening-inhibitor) and Nr

(Never-ripe) reported that nor exerts a global effect on ethylene-related gene expression and

This article is protected by copyright. All rights reserved.

 may be acting upstream of rin in the regulation of ethylene biosynthesis (Osorio et al., 2011).

Interestingly, the Spanish “de Penjar” traditional tomato type is well known for its
Accepted Article
extraordinarily long shelf life. At least part of this characteristic is attributed to the alcobaça

(alc) mutant, which is allelic to nor (Casals et al., 2012). Similarly to the E59K and P232L

melon mutants, the alc mutant is due to a non-synonymous V106D amino acid change in the

NAC subdomain C region, producing a fruit with delayed ripening and long shelf life. On the

other hand, the nor mutant is due to a 2-bp deletion in the third exon, producing a non-

functioning protein and an extreme non-ripening phenotype (Casals et al., 2012). The NAC

gene ppa008301m has also been proposed as the candidate gene for a major locus controlling

maturity date in peach (Pirona et al., 2013), and although no functional validation has yet

been reported, ppa008301m is also phylogenetically close to Nor. Other members of the

NAC family have been involved in the ripening process, as MaNAC1 and MaNAC2 in banana

(Shan et al., 2012). These data suggest an important role of NAC genes in the control of fruit

ripening among different plant clades.

Similar to the nor tomato mutant, the inodorus melon type PS does not show a peak of

ethylene during ripening, fruit abscission is absent, the exocarp color remains green through

maturation and fruit softening is reduced (Table S12). More interestingly, an exogenous

ethylene treatment does not induce the onset of ripening in PS nor in the tomato nor mutant

(Vegas et al., 2013, Saladié et al., 2015). The SC allele of ETHQV6.3 introgressed into the

PS non-climacteric type in line GF40 shows a moderate climacteric type (Table S12). The SC

allele of ETHQB3.5, another QTL in chromosome 3, is capable of independently rescuing the

climacteric ripening capacity of PS in line GF35 (Table S12), suggesting that at least two

genes may be impaired in the non-climacteric phenotype of PS.

The sequence of CmNAC-NOR in a panel 54 melon accession belonging to different

botanical groups revealed 17 polymorphisms (SNP and indel), of which 7 were strongly

This article is protected by copyright. All rights reserved.

 associated with the climacteric phenotype (Figure 1). Two features of the sequence diversity

analysis of CmNAC-NOR deserve further attention. First, the CmNAC-NOR haplotypes of the
Accepted Article
non-climacteric SC (Con-SC) and three other conomon types (Con-Paul, Con-Pat81 and Con-

FreeC) were included in the climacteric cantalupensis cluster. Second, all 15 inodorus

haplotypes and other non-climacteric accessions where clustered together. Nine out of 14

cantalupensis and reticulatus climacteric accessions were in the climacteric cantalupensis

cluster but 5 (Can-Pres, Can-Y, Can-PS, Can-CA and Can-Ef) were included in the inodorus

group. A genetic analysis in a RIL population obtained from the cross between the

climacteric variety “Védrantais” (cantalupensis) and the non-climacteric SC, revealed that

the development of an abscission layer and the autocatalytic ethylene biosynthesis were

controlled by Al-3 and Al-4, and four additional QTLs were involved in regulating the

amount of ethylene (Perin et al., 2002). Interestingly none of these QTL map in the same

genomic intervals than ETHQB3.5 and ETHQV6.3, suggesting that the non-climacteric

phenotype of SC should be attributed to mutations in different QTLs alleles than ETHQB3.5

and ETHQV6.3. This would also explain why the ETHQV6.3 allele of SC, which is almost

identical to that of the climacteric cantalupensis accessions, is able to partially rescue the

climacteric phenotype when introgressed into the non-climacteric PS. The non-climacteric

phenotypes of SC and PS are different (Table S12), as accumulation of carotenoids in the

flesh and the induction of a set of ethylene biosynthetic genes are observed in SC (Vegas et

al., 2013, Saladié et al., 2015). Recently, QTLs that delay fruit ripening of the climacteric

“Védrantais” containing introgressions of the exotic “Ginsen Makuwa” line (makuwa type)

have been reported in chromosomes 7 and 10 (Perpiñá et al., 2016). The complexity of the

climacteric phenotype, with at least 10 QTLs reported in melon, suggests that the division of

ripening behaviour into just two classes may be revised into a more complex scenario that

envisions ripening in a continuous spectrum with non-climacteric and highly climacteric

This article is protected by copyright. All rights reserved.

 types at the extremes. Thus, a group of climacteric cantalupensis accessions, which contain

the PS allele of CmNAC-NOR, still show climacteric behaviour, probably due to the presence
Accepted Article
of the climacteric alleles for other QTLs involved in ripening. Similarly, the delayed ripening

phenotype observed in the “Charentais Mono” mutants E59K and P232L would also support

this hypothesis (Table S12). The recent availability of a non-invasive method for the ethylene

quantification in attached fruits will help classifying melon accessions according to their

ripening behaviour in a more precise manner (Pereira et al., in press).

Our current data does not allow the identification of the causal polymorphism of the

climacteric phenotype among the 7 polymorphisms highly associated with ripening behaviour

identified in CmNAC-NOR. CmNAC-NOR is expressed in flesh at different stages of fruit

development in both climacteric and non-climacteric types, peaking around 30 DAP (Figure

5a) when the ripening process starts, and it shows very low expression in leaves and roots.

The same pattern of expression in fruit tissue has also been reported in the climacteric

“Védrantais” and “Dulce” and the non-climacteric SC (Saladié et al., 2015). The lack of

differential expression of CmNAC-NOR in fruit flesh among distinct ripening phenotypes

suggests that its regulation may occur through other mechanisms. Two of the natural

polymorphisms found in CmNAC-NOR produce non-synonymous changes A108S and

S236N, which are located outside the NAC subdomains, but still may affect interaction with

other proteins or binding to DNA. INDEL-126 is particularly interesting as it is located in the

5’UTR of the gene, the conomon and cantalupensis alleles being different from the non-

climacteric inodorus types. The possible effect of INDEL-126 in the translation of CmNAC-

NOR in both melon types deserves further attention.

Finally, the tomato NAC gene SlNAC4 has a role in abiotic stress response and is a

positive regulator of fruit ripening, affecting ethylene synthesis and carotenoid accumulation

(Zhu et al., 2014). SlNAC4 probably interacts with NOR and RIN and it emerges as a new

This article is protected by copyright. All rights reserved.

 player in the complex regulatory network of fruit ripening in tomato (Zhu et al., 2014).

Among the clusters of NAC proteins 2a and 2b, which include CmNAC-NOR and SlNAC-
Accepted Article
NOR, other melon NAC proteins are also found (Figure 3). MELO3C016536, which is in the

same genomic interval contained in ETHQV6.3, is phylogenetically related to SlNAC4 and

other NAC proteins involved in stress responses and shows a clear differential expression in

fruit flesh of climacteric and non-climacteric lines (Figure 5e). It would not be surprising that,

as in tomato, other NAC genes are also involved in regulating melon fruit ripening.

The use of comparative physiology and genetics in melon, a species that contains both

climacteric and non-climacteric genotypes, has begun to help to elucidate the differences

between these two types of ripening behaviours. It has also provided a link to a common

mechanism of ripening shared with tomato, the classical climacteric model species for

studying fruit ripening. Our results and other current genetic data suggest that several factors

are involved in the regulation of fruit ripening in melon, and ETHQV6.3, the first one

characterized, shows similarities with the well-studied tomato Nor. Further investigation of

other melon QTLs involved in fruit ripening is required to complete the complex picture of

this important process. It should however be noted that the ripening-associated changes

observed in PS could lead to considering it a ripening mutant instead of a “true” non-

climacteric fruit, and that the mechanisms regulating fruit ripening in non-climacteric species

may be different from those operating in melon non-climacteric types.

Experimental procedures

Plant material

A mapping population originated from the cross SC3-5-1 x PS (Figure S1) was used

for the positional cloning of ETHQV6.3. SC3-5-1 (GF31) is a near-isogenic line (NIL) that

harbours two homozygous introgressions (carrying both ETHQB3.5 and ETHQV6.3) in

This article is protected by copyright. All rights reserved.

 chromosomes 3 and 6 from the accession PI 161375 (C. melo var. conomon, SC) in the “Piel

de Sapo” (C. melo var. inodorous, PS) genetic background (Vegas et al., 2013). NILs GF35
Accepted Article
and GF40 contain ETHQB3.5 and ETHQV6.3, respectively. All plants were grown in a

greenhouse in coco-fibre bags and all flowers were self-pollinated manually allowing only

one fruit per plant.

A melon germplasm collection from the COMAV-UPV, which includes 175 melon

varieties (Leida et al., 2015) (Table S7), was used to study the association of the candidate

gene with ripening behaviour. A subset of 54 accessions from this collection was selected for

a detailed analysis (Table S8).

Four hundred and eighty F2 plants from a cross between “Noy-Amid” (C. melo var.

inodorous, Yellow Canary type) and “Dulce” (C. melo var. reticulatus, cantaloupe type)

(Harel-Beja et al., 2010) (NA x Dul) were phenotyped for ethylene emission at harvest in

2011. Twenty F3 plants of each of 32 F2 plants with extreme ethylene levels (16 plants >7.5

and 16 plants <0.5 µg/kg fresh fruit/hr) were grown in two repetitions in a greenhouse at Beit

Elazari, Israel in 2013. Flowers were manually pollinated and tagged at anthesis and 1-2

fruits were allowed to develop per plant.

DNA extraction and genotyping

Genomic DNA was extracted from young leaves according to CTAB method with

some modifications to improve quality (Garcia-Mas et al., 2000).

Eight SSRs and one Cleaved Amplified Polymorphic Sequence (CAPS) (Table S1)

were used to genotype the 2008-F2 population (Vegas et al. 2013) to identify 7M80-11.4. The

2012-F4 population was screened with TaqMan probes (Thermo Scientific, Waltham, USA)

SNP-64658 and SNP-2826073, designed by the Custom TaqMan Assay Design Tool

(www.lifetechnologies.com/snpcadt) using two flanking SNPs between PS and SC (Table S1)

This article is protected by copyright. All rights reserved.

 (Sanseverino et al., 2015). PCR reactions were prepared in a final volume of 5μl: 2.5 μl

2xTaqMan Universal PCR Master Mix (Thermo Scientific, Waltham, USA), 2.375 μl
Accepted Article
genomic DNA (40 ng/μl) and 0.125 μl TaqMan probes mix. Amplification was performed in

a Light Cycler 480 (Roche, Basel, Switzerland) with an initial cycle at 95ºC for 1 min, 10

cycles of temperature gradient consisting in 90ºC for 20 s and 61ºC for 1 min diminishing the

temperature from 61ºC to 57ºC at 0.4ºC per cycle, and 26 cycles at 95ºC for 20 s and 57ºC for

1 min. Fluorescence was measured at 37ºC. Twenty-four SNPs were genotyped using KASP

chemistry (LGC, Teddington, UK) in a BiomarkTM system (San Francisco, CA, USA). SNP

primers were designed with Kraken (Table S1). SNPs SEQ-1 to SEQ-6 were genotyped by

Sanger sequencing (Table S1). Sequences were analysed using Sequencher 5.0 (Gene Codes

Corporation, Ann Arbor, MI, USA). Amplicons PRO40.1, CDS40.1, CDS40.2 and CDS40.3

(Table S1) were designed to sequence CmNAC-NOR and to genotype the mutant families and

the melon germplasm collection.

Restriction-site-associated DNA sequencing (RAD-seq, Miller et al., 2007) and QTL

analyses were performed by NRgene LTD (Nes Ziyyona, Israel) using 131 F3 plants of the

NA x Dul population, representing the whole scale of ethylene emission.

Fruit phenotyping

Fruits were collected at abscission or harvested when fully ripe (between 65 and 70

DAP). Fruit ripening behaviour was assessed with traits closely associated to melon

climacteric ripening (Vegas et al., 2013). The development of an abscission layer was

measured using a scale from 0 to 4 (0: no abscission layer; 1: no-slip; 2: half-slip; 3: full-slip;

4: abscission) and days from pollination to abscission were recorded. External color change

was evaluated visually in fruit after abscission and harvested fruits. Days from pollination to

external color change were also measured for the phenotyping of the mutants. The production

This article is protected by copyright. All rights reserved.

 of characteristic climacteric aroma volatiles was detected by olfactory evaluation of fruit after

abscission and harvested fruits.

Accepted Article
The production of ethylene in the fruits of the mutant families was measured using a

method based in non-invasive ethylene quantification in attached fruit headspace by gas

chromatography-mass spectrometry (Pereira et al., in press).

Ripe fruits from the melon germplasm collection were phenotyped for fruit firmness

and abscission layer development by COMAV (Leida et al., 2015). The variable “ripening

type” represents the overall intensity of the climacteric ripening according to the germplasm

collection curators. Scores range from 0 (non-climacteric as PS) to 4 (very climacteric as

“Védrantais”).

NA × Dul fruits were sampled at ripening, determined by abscission layer

development and/or change of rind color. Ripening was verified by BRIX values. Evaluation

of ethylene emission was performed on the day after harvest. Ripe detached fruits were

enclosed for three hours in containers, under controlled atmosphere conditions. Headspace

gases were sampled by syringe through a septum in the lid. Ethylene was measured with a

gas chromatograph equipped with flame ionization detector (Varian 3300: Varian, Palo Alto,

CA, USA) and alumina column (HayeSep T Mesh- 100/120, Sciences, Deerfield, IL, USA).

Data analysis

DNA and protein multiple sequence alignments were obtained with Clustal Omega

(ClustalO, Sievers et al., 2011). The alignments were represented with Jalview 2.8

(Waterhouse et al., 2009). Phylogenetic analysis were performed using the Neighbor-joining

method in MEGA 6.06 (Tamura et al., 2013) with 1,000 Bootstrap iterations. Cladograms

were represented with the ape package for R (Paradis et al., 2004).

This article is protected by copyright. All rights reserved.

 Association analysis of climacteric behaviour with SNP G411T in the germplasm

collection was performed as Leida et al. (2015). Mixed Linear Models (MLM) implemented
Accepted Article
in TASSEL v.5.0 (Bradbury et al., 2007); www.maizegenetics.org) were used with a kinship

matrix to adjust for genetic structure using the full SNP data set as cofactors. Association

analysis of the polymorphisms in CmNAC-NOR with the ripening type score for each

accession were performed with ANOVA-GLM (aov and glm functions in R 3.2.1 (R

Development Core Team, 2016)).

For phenotyping each mutant family, plants homozygous for each of the two alleles

(M=mutant; W=wild type) were selected. The mean values obtained for each class were

compared with a t-Student test (t.test function in R 3.2.1 (R Development Core Team, 2016)),

or a Tukey HSD test in JMP 8.0.1 (SAS Institute Inc., NC).

Identification of TILLING mutants in CmNAC-NOR

Mutant identification in CmNAC-NOR consisted on the screening of 6,200 M2 families

using a nested PCR technique in the TILLING platform “Charentais Mono” (Dahmani-

Mardas et al., 2010). PCR amplification and mutation detection were carried out as

previously described (Dahmani-Mardas et al., 2010) using specific primers for the

amplification of regions A1 and A2 (Figure 2a, Table S1). Additional primers were designed

to validate the mutations by Sanger sequencing (Table S1). PROVEAN (Protein Variation

Effect Analyzer, Choi et al., 2012) was used to predict the impact of the mutation on the

protein function. Seed from M2 mutant families was obtained from URGV.

This article is protected by copyright. All rights reserved.

 qPCR expression

RNA from three biological replicates for GF31, GF35, GF40 and PS was isolated
Accepted Article
from mesocarp, root, and leaf tissue. RNA was isolated from 100 mg frozen sample and

ground using TriZOL® reagent (Ambion®, Life Technologies, Inc.). RNA samples were

purified with RNeasy® spin columns (Qiagen, Hilden, Germany) and treated with RNAse

free TURBO-DNase I (Turbo DNA-freeTM Kit; Applied Biosystems, Ambion®, USA) for

60 min at 37ºC. RNA quality was as in Saladié et al. (2015).

Gene expression analysis by qPCR was performed on a LightCycler® 480 Real-Time

PCR System using SYBR® Green I Mix (Roche Applied Science, USA). The relative

amounts of specific transcripts were determined using cyclophilin (CmCYP7) as a reference

gene (Saladié et al., 2015) and then normalized to PS expression in leaves. Primers were

designed with Primer3 (http://primer3.wi.mit.edu/) and checked for the presence of secondary

structures with NetPrimer (http://www.premierbiosoft.com/netprimer/) (Table S11).

Calculation of intra-assay variation, primer efficiencies, and amplification specificity of the

PCR by melting curve analysis, were as described previously (Saladié et al., 2015).

Acknowledgements

We thank Dr. J. Burger for providing the NAxDul population, Dr. E. Falik for

ethylene analysis of this population and both for their critical inputs. This work was

supported by the Spanish Ministry of Economy and Competitivity grant AGL2015-64625-

C2-1-R, Centro de Excelencia Severo Ochoa 2016-2020, and the CERCA

Programme/Generalitat de Catalunya to JGM; Spanish Ministry of Economy and

Competitivity/FEDER grant AGL2015-64625-C2-2-R to AJM; EU Framework Program

Horizon 2020 COST Action FA1106 Quality Fruit for networking activities to CL; European

This article is protected by copyright. All rights reserved.

 Research Council grant ERC-SEXYPARTH to AB; Chief Scientist of the Ministry of

Agriculture of Israel grant No. 261-1049-13 to NK.

Accepted Article
Conflicts of interest:

The authors declare no conflicts of interest

Short legends for supporting information

Figure S1. Scheme with the plant material used to identify ETHQV6.3.

Figure S2. High-resolution physical map of the ETHQV6.3 interval.

Figure S3. Distribution of the fruit abscission dates of the progenies of 15 recombinants,

expressed in days after pollination (DAP).

Figure S4. Integrative Genomics Viewer (IGV) snapshots of the genomic location of the

QTL for ethylene levels in chromosome 6 in the “Noy-Amid” x “Dulce” population.

Figure S5. Mutations in the CmNAC-NOR sequence.

Figure S6. Sequence of INDEL-126 in the collection of melon accessions.

Figure S7. Mutations E59K (family 246) and P129L (family 432) in the NAC domain region

of CmNAC-NOR and other NAC domain containing proteins.

Table S1. Sequences of the markers and primers used during the high-resolution mapping of

ETHQV6.3 and for the TILLING screening.

Table S2. Phenotyping and genotyping of 15 informative recombinants and fine mapping of

ETHQV6.3.

Table S3. Candidate genes annotated in the 139 kb interval between SNP-2691690 and SNP-

2826073.

This article is protected by copyright. All rights reserved.

 Table S4. QTL peak for ethylene measured at harvest in the RIL population of “Noy-Amid”

x “Dulce”.
Accepted Article
Table S5. NAC-domain containing proteins of different plant species with known function.

Table S6. Polymorphisms in CmNAC-NOR associated with the climacteric behaviour.

Table S7. Genotyping of SNP G411T in a panel of 175 melon accessions.

Table S8. Melon germplasm used for assessing the variation analysis of CmNAC-NOR.

Table S9. Mutations identified in CmNAC-NOR.

Table S10. Phenotyping of external color change in the mutants.

Table S11. Primer sequences for qPCR.

Table S12. Phenotypic information for the main genotypes discussed in the manuscript: PS,

SC, Védrantais, Charentais Mono, both TILLING mutants, and the introgression lines

containing ETHQV6.3 and ETHQB3.5.

References

Abeles, F.B., Morgan, P.W. and Saltveit, M.E. (1992) Ethylene in plant biology. San

Diego: Academic Press.

Adams-Phillips, L., Barry, C. and Giovannoni, J. (2004) Signal transduction systems

regulating fruit ripening. Trends Plant Sci, 9, 331-338.

Alexander, L. and Grierson, D. (2002) Ethylene biosynthesis and action in tomato: a model

for climacteric fruit ripening. J Exp Bot, 53, 2039-2055.

Argyris, J.M., Ruiz-Herrera, A., Madriz-Masis, P., Sanseverino, W., Morata, J., Pujol,

M., Ramos-Onsins, S.E.E. and Garcia-Mas, J. (2015) Use of targeted SNP

selection for an improved anchoring of the melon (Cucumis melo L.) scaffold genome

assembly. BMC Genomics, 16, 4.

This article is protected by copyright. All rights reserved.

 Ayub, R., Guis, M., Ben Amor, M., Gillot, L., Roustan, J.P., Latche, A., Bouzayen, M.

and Pech, J.C. (1996) Expression of ACC oxidase antisense gene inhibits ripening of
Accepted Article
cantaloupe melon fruits. Nat Biotechnol, 14, 862-866.

Blanca, J., Esteras, C., Ziarsolo, P., Perez, D., Ferna Ndez-Pedrosa, V., Collado, C.,

Rodra Guez de Pablos, R., Ballester, A., Roig, C., Canizares, J. and Pico, B.

(2012) Transcriptome sequencing for SNP discovery across Cucumis melo. BMC

Genomics, 13, 280.

Bradbury, P.J., Zhang, Z., Kroon, D.E., Casstevens, T.M., Ramdoss, Y. and Buckler,

E.S. (2007) TASSEL: Software for association mapping of complex traits in diverse

samples. Bioinformatics, 23, 2633-2635.

Casals, J., Pascual, L., Cañizares, J., Cebolla-Cornejo, J., Casañas, F. and Nuez, F.

(2012) Genetic basis of long shelf life and variability into Penjar tomato. Genet

Resour Crop Ev, 59, 219-229.

Choi, Y., Sims, G.E., Murphy, S., Miller, J.R. and Chan, A.P. (2012) Predicting the

functional effect of amino acid substitutions and indels. PLoS one, 7.

Chung, M.Y., Vrebalov, J., Alba, R., Lee, J., McQuinn, R., Chung, J.D., Klein, P. and

Giovannoni, J. (2010) A tomato (Solanum lycopersicum) APETALA2/ERF gene,

SlAP2a, is a negative regulator of fruit ripening. Plant J, 64, 936-947.

Dahmani-Mardas, F., Troadec, C., Boualem, A., Leveque, S., Alsadon, A.A., Aldoss,

A.A., Dogimont, C. and Bendahmane, A. (2010) Engineering melon plants with

improved fruit shelf life using the TILLING approach. PLoS One, 5, e15776.

Eduardo, I., Arus, P. and Monforte, A.J. (2005) Development of a genomic library of near

isogenic lines (NILs) in melon (Cucumis melo L.) from the exotic accession

PI161375. Theor Appl Genet, 112, 139-148.

This article is protected by copyright. All rights reserved.

 Esteras, C., Formisano, G., Roig, C., Diaz, A., Blanca, J., Garcia-Mas, J., Gomez-

Guillamon, M.L., Lopez-Sese, A.I., Lazaro, A., Monforte, A.J. and Pico, B.
Accepted Article
(2013) SNP genotyping in melons: genetic variation, population structure, and linkage

disequilibrium. Theor Appl Genet, 126, 1285-1303.

Ezura, H. and Owino, W.O. (2008) Melon, an alternative model plant for elucidating fruit

ripening. Plant Sci, 175, 121-129.

Gao, C., Ju, Z., Cao, D., Zhai, B., Qin, G., Zhu, H., Fu, D., Luo, Y. and Zhu, B. (2015)

MicroRNA profiling analysis throughout tomato fruit development and ripening

reveals potential regulatory role of RIN on microRNAs accumulation. Plant Biotech

J, 13, 370-382.

Garcia-Mas, J., Benjak, A., Sanseverino, W., Bourgeois, M., Mir, G., Gonzalez, V.M.,

Henaff, E., Camara, F., Cozzuto, L., Lowy, E., Alioto, T., Capella-Gutierrez, S.,

Blanca, J., Canizares, J., Ziarsolo, P., Gonzalez-Ibeas, D., Rodriguez-Moreno, L.,

Droege, M., Du, L., Alvarez-Tejado, M., Lorente-Galdos, B., Mele, M., Yang, L.,

Weng, Y., Navarro, A., Marques-Bonet, T., Aranda, M.A., Nuez, F., Pico, B.,

Gabaldon, T., Roma, G., Guigo, R., Casacuberta, J.M., Arus, P. and

Puigdomenech, P. (2012) The genome of melon (Cucumis melo L.). Proc Natl Acad

Sci USA, 109, 11872-11877.

Garcia-Mas, J., Oliver, M., Gomez-Paniagua, H. and de Vicente, M.C. (2000) Comparing

AFLP, RAPD and RFLP markers for measuring genetic diversity in melon. Theor

Appl Genet, 101, 860-864.

Giovannoni, J.J. (2001) Molecular biology of fruit maturation and ripening. Annu Rev Plant

Phys, 52, 725-749.

Giovannoni, J.J. (2004) Genetic regulation of fruit development and ripening. Plant Cell, 16

Suppl, S170-180.

This article is protected by copyright. All rights reserved.

 Giovannoni, J.J. (2007) Fruit ripening mutants yield insights into ripening control. Curr

Opin Plant Biol, 10, 283-289.

Accepted Article
Handa, A.K., Anwar, R. and Mattoo, A.K. (2014) Biotechnology of fruit quality. In Fruit

Ripening: Physiology, Signalling and Genomics (Nath, P., Bouzayen, M., Mattoo, A.

and Pech, J.C. eds). Boston, MA: CABI, pp. 259-290.

Harel-Beja, R., Tzuri, G., Portnoy, V., Lotan-Pompan, M., Lev, S., Cohen, S., Dai, N.,

Yeselson, L., Meir, A., Libhaber, S.E., Avisar, E., Melame, T., van Koert, P.,

Verbakel, H., Hofstede, R., Volpin, H., Oliver, M., Fougedoire, A., Stalh, C.,

Fauve, J., Copes, B., Fei, Z., Giovannoni, J., Ori, N., Lewinsohn, E., Sherman, A.,

Burger, J., Tadmor, Y., Schaffer, A.A. and Katzir, N. (2010) A genetic map of

melon highly enriched with fruit quality QTLs and EST markers, including sugar and

carotenoid metabolism genes. Theor Appl Genet, 121, 511-533.

Itkin, M., Seybold, H., Breitel, D., Rogachev, I., Meir, S. and Aharoni, A. (2009)

TOMATO AGAMOUS-LIKE 1 is a component of the fruit ripening regulatory

network. Plant J, 60, 1081-1095.

Klee, H.J. (2004) Ethylene signal transduction. Moving beyond Arabidopsis. Plant Physiol,

135, 660-667.

Klee, H.J. and Giovannoni, J.J. (2011) Genetics and control of tomato fruit ripening and

quality attributes. Annu Rev Genet, 45, 41-59.

Kovács, K., Fray, R.G., Tikunov, Y., Graham, N., Bradley, G., Seymour, G.B., Bovy,

A.G. and Grierson, D. (2009) Effect of tomato pleiotropic ripening mutations on

flavour volatile biosynthesis. Phytochemistry, 70, 1003-1008.

Lee, J.M., Joung, J.-G.G., McQuinn, R., Chung, M.-Y.Y., Fei, Z., Tieman, D., Klee, H.

and Giovannoni, J. (2012) Combined transcriptome, genetic diversity and metabolite

This article is protected by copyright. All rights reserved.

 profiling in tomato fruit reveals that the ethylene response factor SlERF6 plays an

important role in ripening and carotenoid accumulation. Plant J, 70, 191-204.

Accepted Article
Leida, C., Moser, C., Esteras, C., Sulpice, R., Lunn, J.E., de Langen, F., Monforte, A.J.

and Picó, B. (2015) Variability of candidate genes, genetic structure and association

with sugar accumulation and climacteric behavior in a broad germplasm collection of

melon (Cucumis melo L.). BMC Genetics, 16, 28.

Lelièvre, J.-M., Latchè, A., Jones, B., Bouzayen, M. and Pech, J.-C. (1997) Ethylene and

fruit ripening. Physiol Plantarum, 101, 727-739.

Lin, Z., Zhong, S. and Grierson, D. (2009) Recent advances in ethylene research. J Exp

Bot, 60, 3311-3336.

Liu, R., How-Kit, A., Stammitti, L., Teyssier, E., Rolin, D., Mortain-Bertrand, A., Halle,

S., Liu, M., Kong, J., Wu, C., Degraeve-Guibault, C., Chapman, N.H., Maucourt,

M., Hodgman, T.C., Tost, J., Bouzayen, M., Hong, Y., Seymour, G.B.,

Giovannoni, J.J. and Gallusci, P. (2015) A DEMETER-like DNA demethylase

governs tomato fruit ripening. Proc Natl Acad Sci USA, 112, 10804-10809.

Manning, K., Tor, M., Poole, M., Hong, Y., Thompson, A.J., King, G.J., Giovannoni,

J.J. and Seymour, G.B. (2006) A naturally occurring epigenetic mutation in a gene

encoding an SBP-box transcription factor inhibits tomato fruit ripening. Nat Genet,

38, 948-952.

McMurchie, E.J., McGlasson, W.B. and Eaks, I.L. (1972) Treatment of fruit with

propylene gives information about the biogenesis of ethylene. Nature, 237, 235-236.

Miller, M.R., Dunham, J.P., Amores, A., Cresko, W.A. and Johnson, E.A. (2007) Rapid

and cost-effective polymorphism identification and genotyping using restriction site

associated DNA (RAD) markers. Genome Res, 17, 240-248.

This article is protected by copyright. All rights reserved.

 Moreno, E., Obando, J.M., Dos-Santos, N., Fernandez-Trujillo, J.P., Monforte, A.J. and

Garcia-Mas, J. (2008) Candidate genes and QTLs for fruit ripening and softening in
Accepted Article
melon. Theor Appl Genet, 116, 589-602.

Osorio, S., Alba, R., Damasceno, C., Lopez-Casado, G., Lohse, M., Zanor, M., Tohge,

T., Usadel, B., Rose, J., Fei, Z., Giovannoni, J.J. and Fernie, A.R. (2011) Systems

biology of tomato Fruit development: combined transcript, protein, and metabolite

analysis of tomato transcription factor (nor, rin) and ethylene receptor (Nr) mutants

reveals novel regulatory interactions. Plant Physiol, 157, 405-425.

Paradis. E., Claude. J. and Strimmer, K. (2004) APE: analyses of phylogenetics and

evolution in R language. Bioinformatics, 20, 289-90.

Pech, J.C., Bouzayen, M. and Latche, A. (2008) Climacteric fruit ripening: Ethylene-

dependent and independent regulation of ripening pathways in melon fruit. Plant Sci,

175, 114-120.

Pereira, L., Pujol, M., Garcia-Mas, J. and Phillips, M.A. Non-invasive ethylene

quantification in attached fruit headspace at 1 ppb by gas chromatography – mass

spectrometry. Plant J, in press.

Perin, C., Gomez-Jimenez, M., Hagen, L., Dogimont, C., Pech, J.C., Latche, A., Pitrat,

M. and Lelievre, J.M. (2002) Molecular and genetic characterization of a non-

climacteric phenotype in melon reveals two loci conferring altered ethylene response

in fruit. Plant Physiol, 129, 300-309.

Perpiñá, G., Esteras, C., Gibon, Y., Monforte, A.J. and Picó, B. (2016) A new genomic

library of melon introgression lines in a cantaloupe genetic background for dissecting

desirable agronomical traits. BMC Plant Biol, 16, 154.

Pirona, R., Eduardo, I., Pacheco, I., Da Silva Linge, C., Miculan, M., Verde, I.,

Tartarini, S., Dondini, L., Pea, G., Bassi, D. and Rossini, L. (2013) Fine mapping

This article is protected by copyright. All rights reserved.

 and identification of a candidate gene for a major locus controlling maturity date in

peach. BMC Plant Biol, 13, 166.

Accepted Article
Pitrat, M. (2008) Melon (Cucumis melo L.). In Handbook of Crop Breeding Vol I:

Vegetables (Prohens, J. and Nuez, F. eds). New York: Springer, pp. 283–315.

Puranik, S., Sahu, P.P., Srivastava, P.S. and Prasad, M. (2012) NAC proteins: regulation

and role in stress tolerance. Trends Plant Sci, 17, 369-381.

R Development Core Team (2016) R: A language and environment for statistical

computing. Vienna, Austria: R Foundation for Statistical Computing.

Robinson, R. and Tomes, M. (1968) Ripening inhibitor: a gene with multiple effects on

ripening. Rep Tomato Genet Coop, 18, 36-37.

Saladié, M., Cañizares, J., Phillips, M.A., Rodriguez-Concepcion, M., Larrigaudière, C.,

Gibon, Y., Stitt, M., Lunn, J.E. and Garcia-Mas, J. (2015) Comparative

transcriptional profiling analysis of developing melon (Cucumis melo L.) fruit from

climacteric and non-climacteric varieties. BMC Genomics, 16, 440.

Sanseverino, W., Hénaff, E., Vives, C., Pinosio, S., Burgos-Paz, W., Morgante, M.,

Ramos-Onsins, S.E.E., Garcia-Mas, J. and Casacuberta, J.M. (2015) Transposon

insertions, structural variations, and SNPs contribute to the evolution of the melon

genome. Molecular Biol Evol, 32, 2760-2774.

Shan, W., Kuang, J.-f.F., Chen, L., Xie, H., Peng, H.-h.H., Xiao, Y.-y.Y., Li, X.-p.P.,

Chen, W.-x.X., He, Q.-g.G., Chen, J.-y.Y. and Lu, W.-j.J. (2012) Molecular

characterization of banana NAC transcription factors and their interactions with

ethylene signalling component EIL during fruit ripening. J Exp Bot, 63, 5171-5187.

Sievers, F., Wilm, A., Dineen, D., Gibson, T.J., Karplus, K., Li, W., Lopez, R.,

McWilliam, H., Remmert, M., Soding, J., Thompson, J.D. and Higgins, D.G.

This article is protected by copyright. All rights reserved.

 (2011) Fast, scalable generation of high-quality protein multiple sequence alignments

using Clustal Omega. Mol Syst Biol, 7, 539.

Accepted Article
Thompson, A.J., Tor, M., Barry, C.S., Vrebalov, J., Orfila, C., Jarvis, M.C.,

Giovannoni, J.J., Grierson, D. and Seymour, G.B. (1999) Molecular and genetic

characterization of a novel pleiotropic tomato-ripening mutant. Plant Physiol, 120,

383-390.

Tamura, K., Stecher, G., Peterson, D., Filipski, A. and Kumar, S. (2013) MEGA6:

Molecular evolutionary genetics analysis version 6.0. Mol Biol Evol, 30, 2725-9.

Tigchelaar, E., Tomes, M., Kerr, E. and Barman, R. (1973) A new fruit ripening mutant,

nonripening (nor). Rep Tomato Genet Coop, 23, 33-34.

Vegas, J., Garcia-Mas, J. and Monforte, A.J. (2013) Interaction between QTLs induces an

advance in ethylene biosynthesis during melon fruit ripening. Theor Appl Genet, 126,

1531-1544.

Vrebalov, J., Ruezinsky, D., Padmanabhan, V., White, R., Medrano, D., Drake, R.,

Schuch, W. and Giovannoni, J. (2002) A MADS-Box Gene Necessary for Fruit

Ripening at the Tomato Ripening-Inhibitor (Rin) Locus. Science, 296, 343-346.

Waterhouse, A.M., Procter, J.B., Martin, D.M., Clamp, M. and Barton, G.J. (2009)

Jalview Version 2-a multiple sequence alignment editor and analysis workbench.

Bioinformatics, 25, 1189-91.

Zhong, S., Fei, Z., Chen, Y.-R.R., Zheng, Y., Huang, M., Vrebalov, J., McQuinn, R.,

Gapper, N., Liu, B., Xiang, J., Shao, Y. and Giovannoni, J.J. (2013) Single-base

resolution methylomes of tomato fruit development reveal epigenome modifications

associated with ripening. Nature Biotechnol, 31, 154-159.

Zhu, M., Chen, G., Zhou, S., Tu, Y., Wang, Y., Dong, T. and Hu, Z. (2014) A new tomato

NAC (NAM/ATAF1/2/CUC2) transcription factor, SlNAC4, functions as a positive

regulator of fruit ripening and carotenoid accumulation. Plant Cell Physiol, 55, 119-

135.
This article is protected by copyright. All rights reserved.
ccepted Articl
Tables

Table 1. Phenotyping of ethylene production in the mutant families 246 and 432. Ethylene production during fruit ripening was measured in

mutant families 246 and 432 and the CharMono line. The external color change and the abscission dates were also recorded. W: homozygote for

the wild type allele, M: homozygote for the mutant allele. SD: standard deviation. Asterisks indicate the level of significance after a Tukey HSD

test. *: p-value < 0.05; **: p-value < 0.01; ***: p-value < 0.001.

mean ± SD (DAP) Mean differences in days

M2 Family W M M-W W-CharMono M-Char Mono

Days after 246 37,3 ± 1,1 45,7 ± 0,6 8,4 *** 0 8,4 ***
pollination to peak
ethylene production 432 38 ± 1,6 42 ± 1,7 4,0 * 0,3 4,7 *
CharMono 37,3 ± 1,5 - - - -
246 38,6 ± 2,6 44,8 ± 1,9 6,1 *** 1,5 7,7 ***
External color
432 36,3 ± 1,6 42 ± 2,2 5,7 *** -0,7 4,9 ***
change
CharMono 37,1 ± 1,9 - - - -
246 42 ± 2,8 47,3 ± 2,9 5,3 -2,3 3
Abscission 432 38,7 ± 2,1 43,8 ± 3 5,1 -5,6 * -0,5
CharMono 44,3 ± 2,7 - - - -

This article is protected by copyright. All rights reserved.

 Figure legends

Figure 1. Sequence diversity of CmNAC-NOR. Left panel: Cladogram representing the

Accepted Article
sequence of CmNAC-NOR in 47 melon accessions. The scale indicates genetic distance.

Colors for each accession represent the melon botanical classification: green, inodorus; dark

blue, ameri and other European traditional varieties; red, cantalupensis and reticulatus; light

blue, flexuosus; grey, dudaim; purple, momordica; orange, conomon; pink, agrestis. Top

panel: The structure of CmNAC-NOR with the position of SNPs (triangles) and indels

(arrows), numbered from 1 to 17. Solid arrows and triangles marked with an asterisk indicate

a significant association of each variation with the type of fruit ripening. Right panel:

genotyping of the collection for the 7 variations significantly associated with the type of fruit

ripening. Colors indicate the observed alleles for each SNP/indel. In green the PS (inodorus)

allele and in red the cantalupensis allele. For indels, additional alleles are represented with

different colors as in Table S8. The ripening type score for each accession (0 = non-

climacteric as PS, 4 = highly climacteric as “Védrantais”) is included in the last column.

Figure 2. Mutants identified for CmNAC-NOR. A. Structure of CmNAC-NOR. A1 and A2

represent the regions amplified for TILLING. Red boxes represent UTRs, blue boxes

represent exons and blue lines represent introns. The NAC domain is represented with a

purple line under exons 1 and 2. Red triangles represent non-synonymous mutations; green

triangles represent synonymous mutations; blue triangles represent mutations in non-coding

regions; grey triangles represent discarded mutations corresponding to family 5388. B.

Phenotypic differences according to external color change in M2 families of mutant families

246, 432, 4933, 3717, 2503 and 502. In the Y-axis, days between pollination and external

color change are represented. W (red) is homozygous for the wild type allele; M (green) is

homozygous for the mutated allele. Asterisks indicate statistically significant differences

between each group after a t-Student test. Significance level ***: p-value < 0.001.

This article is protected by copyright. All rights reserved.

 Figure 3. Cladogram containing the melon NAC family and NAC proteins of known function

of other plant species. The zoom shows the clade that contains MELO3C016540 (CmNAC-
Accepted Article
NOR) and tomato SlNAC-NOR and SlNAC4. The prefix for each protein sequence indicates

the plant species (Table S5). Colors indicate protein function: red, stress response; green, cell

wall metabolism; blue, plant growth and development; purple, senescence; orange, fruit

ripening. The lower scale represents the relative genetic distance. Group 1 contains proteins

involved in growth, development and cell wall metabolism. Group 2a contains proteins

involved in stress response. Group 2b contains proteins involved in senescence and fruit

ripening.

Figure 4. A. Box plots for days after pollination to peak ethylene production (DTP) in (n=4)

fruits of “Charentais Mono” (MONO), and (n=3) fruits in each of two homozygous wild type

(WT), and two homozygous mutant (MU) families of CmNAC-NOR. Asterisks indicate

significant differences between WT and MU families connected by horizontal bars at

p<0.001 (***) and p<0.05 (*) with Tukey HSD. B. Three day interval including the peaks of

ethylene production in the MONO and WT (closed symbols) and MU families (open

symbols) according to days after pollination (DAP). Means are plotted ± SD (n=4) for

MONO and (n=3) for WT and MU families.

Figure 5. CmNAC-NOR qPCR expression. CmNAC-NOR expression was measured in GF31,

GF35, GF40 and PS (A), CmACO1 (B), CmACS1 (C), CmACS5 (D), and MELO3C016536

(E). Gene expression was plotted relative to PS expression in leaves and measured in

developing fruit at 20 and 30 days after pollination (DAP), and at harvest, and in leaf and root

tissue. Means are plotted ± SE (n=3).

This article is protected by copyright. All rights reserved.

Accepted Article

This article is protected by copyright. All rights reserved.

Accepted Article

This article is protected by copyright. All rights reserved.

Accepted Article

This article is protected by copyright. All rights reserved.