Introduction

The external color of fruit directly affects consumer preference and, consequently, fruit marketability. The color(s) of fruit surfaces are largely governed by an interplay between the biosynthetic pathways of the anthocyanins, carotenoids and chlorophylls in the fruit exocarp (Timberlake and Henry 1986; Lancaster et al. 1997). Apart from these pigments, studies have shown that epicuticular waxes deposited on the fruit surface also contribute to its color by establishing a white-grayish powdery coating (Lara et al. 2015). Epicuticular waxes are a rich mixture of very-long-chain fatty acids, with chain lengths > C20, and their derivatives, including primary and secondary alcohols, aldehydes, n-alkanes, ketones and wax esters (Samuels et al. 2008; Buschhaus and Jetter 2011; Arya et al. 2021). The change in surface color due to epicuticular wax accumulation is thought to be driven by these highly lipophilic, vibrant compounds’ effect on the optical attributes of the fruit surface, including brightness, glossiness and glaucousness (Ward and Nussinovitch 1996; Liu et al. 2012; Hen-Avivi et al. 2016; Loypimai et al. 2017; Trivedi et al. 2021).

The color versatility of fruit rinds is clearly portrayed in the various melon (Cucumis melo L.) varieties. A member of the cucurbits (Cucurbitaceae) family, melon is one of the most important, universally cultivated agronomic crops. Its widespread, long-term breeding and domestication in almost all the warm climate regions on Earth has generated immense phenotypic diversity, including in the melon fruit’s size, dimensions, sugar content, acidity, texture, aroma, and rind color (Burger et al. 2010). The external rind colors of current melon varieties span from white, through light yellow and orange, to dark green. The pigments responsible for the various rind colors are carotenoids (mainly β-carotene), chlorophylls, and flavonoids like naringenin chalcone; the flavonoids were shown to accumulate in melon fruit rinds independently from the biosynthetic pathways of carotenoids and chlorophylls (Tadmor et al. 2010).

Many of the Charentais melon varieties belonging to the Cantalupensis group produce a white-grayish color that is prevalent in immature fruit. The genetic inheritance of this color was first described by Kubicki (1962) as a dominant trait and is likely controlled by a single gene named Wi (Dogimont 2011). This dominant white rind trait was recently mapped to a ~ 1.6 Mb interval on chromosome 7 that most likely corresponds to the Wi gene (Pereira et al. 2018), however, the causative gene sequence is yet to be disclosed nor to be associated with the biosynthetic pathways of cuticular lipids. Another light rind phenotype is common across multiple melon varieties and displays a recessive inheritance. In a recent study, Oren et al. (2019) mapped this recessive gene for light rind color in immature Cucumis melo fruit to chromosome 4 and therefore confirmed that it is different from the dominant Wi gene. The authors then found that the causative gene is the melon homolog of the ARABIDOPSIS PSEUDO-RESPONSE REGULATOR2-LIKE (CmAPRR2) transcription factor, and demonstrated that mutations in this gene are associated with rind color intensity in both immature and mature fruits (Oren et al. 2019).

Previous evidence suggest that in blueberries, plums, grapes and apples, the accumulation of epicuticular waxes accounts for the white coating on their surfaces, often referred to as “wax blooms” (Belding et al. 1998; Saftner et al. 2008; Wisuthiphaet et al. 2014; Yan et al. 2022). Even though the dominant white rind trait in melon fruit was genetically mapped, no gene(s) involved in the biosynthesis of epicuticular waxes or cuticular lipids were identified in this genetic region of Wi. Hence, in the current paper we sought to determine whether the external white color in the melon fruit is associated with the accumulation of epicuticular waxes in its rind. We utilized a subset of the MelonCore25 set that we recently established for trait dissection in melon fruit. This multi-parental framework consists of a core subset of 25 diverse founder melon varieties representing the phenotypic divergence of melon fruit (Oren et al. 2022). Herein, we compared the fruit rind morphology and wax crystal deposition patterns of five selected melon accessions from the MelonCore25 set that form white and green rinds. These accessions were analyzed using light and electron microscopy, and profiled for their epicuticular wax composition via gas chromatography-mass spectrometry (GC–MS). These analyses were performed at four key stages of fruit development, from immaturity to full maturity, allowing us to dissect the developmental dynamics of epicuticular wax accumulation in the five varieties. Our data provides the first-ever inclusive metabolic repertoire of epicuticular waxes deposited on top of the melon fruit rind, highlighting the melon fruit as a rich source of these lipidic compounds. Moreover, it revealed that the external white rind color in melon fruit tightly associated with the accumulation of epicuticular wax monomers from various biochemical classes, particularly n-alkanes, fatty alcohols, aldehydes and wax esters. Based on these findings, we suggest that the dominant white rind color and the Wi gene responsible for this trait is likely to be associated directly or indirectly with the biosynthetic pathways of cuticular lipids.

Materials and methods

Plant material and growth conditions

The following five Cucumis melo spp. melo sub-species were characterized: the Cantalupensis “Védrantais” (VEP), NDD1 and “Noy Yizréel” (NY), the Inodorus “Tendral Verde Tardio” (TVT), and the Reticulatus “Dulce” (DUL). Plants were grown under natural field conditions and normal irrigation and fertilization regimes in Bene-Darom, Israel (31.815790, 34.693633).

Microscopic studies

The structure and morphology of fruit surfaces were studied in peel tissues gently isolated from mature (40 days after anthesis; daa) VEP, NDD1, NY, TVT and DUL fruit. Samples were immediately observed under an Olympus SZX7 binocular and images were captured using a Pixelink PL-D795CU digital camera. To visualize the fine structure and density of epicuticular waxes deposited on top of fruit surfaces at high resolution, peel discs were gently dissected from the fruit described above and instantly fixed in 100% methanol for 10 min. Samples were then transferred into clean 100% ethanol for 30 min and eventually dehydrated in 100% absolute ethanol overnight. Dehydrated, fixed samples were mounted on designed scanning electron microscope (SEM) holders and fully dehydrated by following the critical point dehydration (CPD) protocol, using a Quorum K850 critical point dry system and CO2, to maintain the original tissue structure (Halbritter 1998). Dry samples were then coated with a 2 nm layer of Au/Pd particles using a Quorum Q150TES turbo-molecular pumped sputter prior to observations under a JEOL 7800 FEG SEM microscope at 5–10 kV.

Chemical profiling of epicuticular waxes via GC–MS

Epicuticular waxes were isolated from 0.8 cm-diameter peel discs excised from the fruit of VEP, NDD1, NY, TVT and DUL melon at 10 daa, 20 daa, 30 daa and 40 daa following immersion once for 30 s in 5 ml chloroform solution (Mercury; for analysis EMSURE® ACS,ISO,Reag. Ph Eur) containing n-hexatriacontane (C36) alkane internal standard (Sigma-Aldrich; 98%). Chloroform extracts containing epicuticular waxes were allowed to evaporate fully under a stream of nitrogen in 1 ml reaction vials (Supelco, Merck). Next, epicuticular waxes were re-suspended in fresh 100 μl chloroform (Mercury; for analysis EMSURE® ACS,ISO,Reag. Ph Eur), followed by the addition of 20 μl pyridine (Sigma-Aldrich; 99.8%) and 20 μl BSTFA (N,O-bis(trimethylsilyl)trifluoroacetamide; Sigma-Aldrich; GC–MS purity grade; ≥ 99.0%), and incubation of samples at 70 °C for 1 h in a dry bath incubator (Hangzhou Ruicheng Instrument, China). The samples were then transferred into 100 μl inserts placed onto 2 ml GC–MS running vials. A sample volume of 1 μl was injected in a split-less mode on a GC–MS system (Agilent 7693 A Liquid Autoinjector, 8860 gas chromatograph equipped with a 5977B mass spectrometer; Agilent Technologies, Santa Clara, CA, USA). GC experiments were performed using Agilent J&W GC Columns (HP-5MS UI column; 30 m length, 0.250 mm diameter, and 0.25 μm film thickness) with an injection temperature of 270 °C, interface temperature of 250 °C, and ion source temperature of 200 °C. Helium was used as the carrier gas, at a constant flow rate of 1.2 ml min−1. The temperature program was: 0.5 min isothermal at 70 °C, 30 °C min−1 oven temperature ramp to 210 °C; 5 °C min−1 ramp to 330 °C, constant temperature for 21 min. Mass spectra were recorded with a 40–850 m/z scanning range. Chromatograms and mass spectra were evaluated using the MSD ChemStation software (Agilent Technologies). Integrated peaks of mass fragments were normalized for sample leaf surface diameter and the respective C36 n-alkane internal standard signal. For identification, the corresponding mass spectra and retention time indices were compared to the NIST20 library as well as in-house spectral libraries.

Statistical analyses

All graphs and statistical analyses were performed using the tools embedded in GraphPad Prism software v8.0.1. The significance was calculated according to the one-way ANOVA test, as described in the figure legends. Principal component analyses (PCAs) were generated using the statistical tools embedded in the MetaboAnalyst v6.0 web software; all metabolic datasets were normalized using the log10-transformtion and auto-scaling (mean-centered and divided by the standard deviation of each metabolite) methodologies (Pang et al. 2021; Lu et al. 2023; Ewald et al. 2024).

Results

The external white rind trait in melon fruit is connected to epicuticular wax accumulation

There is evidence in the literature that the white-grayish powdery coating in the rind of some fleshy fruit is associated with the deposition of epicuticular waxes on top of the fruit surface. We, therefore, speculated that melon fruit varieties that form white rinds might deposit higher levels of epicuticular waxes on their surface compared to varieties with green rinds. To establish a link between the appearance of white rind on melon fruit and the accumulation of epicuticular waxes, we utilized our MelonCore25 founder collection of melon varieties that were recently genotyped for high-resolution dissection of multiple fruit traits (Oren et al. 2022). We focused on five melon varieties belonging to the Cucumis melo spp. melo sub-species group: the Cantalupensis VEP, NDD1, and NY, the Inodorus TVT, and the Reticulatus DUL (Fig. 1). NDD1 and DUL rinds undergo reticulation during fruit maturation, while the other three varieties form smooth peels lacking reticulation. With respect to rind color, VEP and NDD1 develop distinct external white rinds; TVT and DUL feature green rinds, whereas NY displays mixed white and green colors in its rind (Fig. 1).

Fig. 1
figure 1

The rind phenotype of the five investigated Cucumis melo spp. melo fruit accessions. "Védrantais" (VEP), “NDD1” and "Noy Yizréel" (NY) belong to the Cantalupensis group, "Tendral Verde Tardio" (TVT) to the Inodorus group, and "Dulce" (DUL) to the Reticulatus group. Upper images represent immature fruit 10 days after anthesis (daa) while bottom images represent mature 40 daa fruit. Light microscopy images below mature fruit represent their rind colors. Scale bar for fruit images = 3 cm; in light microscopy images = 1 cm

Full size image

Due to their extremely small size and high dissolvability, epicuticular waxes can be detected on top of the rind surface via scanning electron microscopy (SEM). We analyzed rind morphology and epicuticular wax crystal ultrastructure of immature (10 days after anthesis, daa) and mature fruit (40 daa) formed by the five melon accessions under high-resolution SEM. In all five cases, the cells building the fruit surface had structures typical of a fleshy fruit peel at 10 daa that were enlarged, expanded and flattened toward maturity (Fig. 2). However, substantial differences were found in the patterns of the accumulated epicuticular wax crystals on the fruit surface between the melon accessions. Surprisingly, even though the dominant white rind was mostly associated with immature fruit, we found at 10 daa fruit a very low coverage of epicuticular waxes in the rinds of all five accessions. Nevertheless, we could still visualize some more wax crystals on the surfaces of VEP, NDD1 and NY fruit, which form white rinds (Fig. 2). However, mature fruit at 40 daa displayed much denser wax coverage compared to 10 daa fruit, yet at this developmental stage, the differences between the white and green rind fruit were substantial and clearly visualized. VEP fruit, which formed the most pronounced white rind, were massively covered by waxes, followed by NDD1 and NY (Fig. 2). We detected small portions of epicuticular wax crystals on the rind of TVT fruit, with only minute amounts of waxes deposited on the rind of DUL fruit (Fig. 2). Apart from key differences in epicuticular wax density, mature VEP, NDD1 and NY fruits deposited massive amounts of irregular and crystal-like platelets on their wax crusts, whereas the surfaces of mature TVT and DUL fruits were covered with relatively minute amounts of more bulbed-like wax structures (Fig. 2). Altogether, the microscopy observations qualitative link the extent of external white rind in melon fruit to the synthesis and accumulation of epicuticular waxes. Moreover, it demonstrates that these differences not only relate to the density of epicuticular waxes deposited on top of the fruit surface but also to the wax crystal ultrastructure. Finally, the findings demarcate that this linkage exists in immature fruit as previously assumed, but it is even more prevalent in later developmental stages and fruit maturity.

Fig. 2
figure 2

The extent of external white rind in melon fruit is linked to epicuticular wax accumulation. Scanning electron microscopy (SEM) of rind morphology and epicuticular wax crystal ultrastructure of immature (10 daa) and mature (40 daa) VEP, NDD1, NY, TVT and DUL fruit. White arrowheads indicate epicuticular wax crystals deposited on top of fruit rind. For each developmental stage the scale bar in upper panels = 1 mm; in bottom panels = 10 µm

Full size image

Melon fruit rind is a rich source of epicuticular waxes

To delineate further how epicuticular wax accumulation contributes to the formation of external white rind in melon fruit, we had to provide an inclusive monomeric repertoire of the epicuticular waxes, as there is a scarcity of studies on the biosynthesis of epicuticular waxes in melon fruit. Of the two GC–MS studies of epicuticular waxes in the melon fruit rind conducted to date, one compared Cucumis melo L. varieties with smooth or reticulated rinds (Cohen et al. 2019), and the other profiled the oriental melon Cucumis melo L. var. makuwa (Park et al. 2023). Both studies delineated the fatty acids, fatty alcohols, n-alkanes, aldehydes and triterpenoids present in the rinds. While these studies highlight the biochemical constituents of epicuticular wax in melon fruit rinds, they do not provide detailed monomeric contours. Consequently, we decided to first profile epicuticular wax composition in 40 daa VEP fruit, whose rind displayed the most significant external white color and the densest deposition of epicuticular waxes.

An analysis of these compounds using established GC–MS-based protocols (Sarkar et al. 2023) revealed that the melon fruit rind is a rich source of epicuticular waxes. We were able to positively annotate 48 metabolites belonging to the biochemical classes of fatty acids (8 metabolites), fatty alcohols (12), aldehydes (4), fatty amides (1), n-alkanes (13), tocopherols (2), triterpenoids (3), and wax esters (5) (Table 1). Calculations of relative abundances of each of the identified metabolites based on their biochemical classes revealed that the two prime classes are n-alkanes, accounting for almost 51% of the total epicuticular wax content, and fatty alcohols, representing 30%. The most abundant n-alkanes were C31 (n-hentriacontane; 14.1%), C33 (n-tritriacontane; 8.8%), C29 (n-nonacosane; 8.2%) and C27 (n-heptacosane; 6.1%), followed by C23 (n-tricosane; 3.4%), C25 (n-pentacosane; 3.3%) and C32 (n-dotriacontane; 3.1%). The fatty alcohols were dominated by C28 (1-octacosanol; 9.2%), C26 (1-hexacosanol; 6.9%) and C22 (1-docosanol; 4.8%). The fatty acids, aldehydes and tocopherols each contributed ~ 5% to the total epicuticular wax content. The predominant fatty acids were C16 (hexadecenoic acid), C18:1(7) (11-octadecenoic acid) and C28 (octacosanoic acid), each contributing ~ 1.5%. The major aldehydes were C32 (dotriacontanal; 2.6%) and C28 (octacosanal; 1.8%), with β-amyrin (4.2%) representing the key tocopherol. Lastly, the triterpenoids and wax esters contributed 2% and 1.6%, respectively, to the total epicuticular wax content, with glutinol (1.5%) being the major triterpenoid and docosyl acetate (0.8%) representing the most abundant wax ester.

Table 1 The melon fruit rind is a rich source of epicuticular waxes
Full size table

Developmental dynamics in epicuticular wax accumulation in melon fruit

Microscopy observations provided a qualitatively link between the extent of white rind to the accumulation of epicuticular waxes, both at the immature and mature fruit developmental stages. Thus, we next conducted similar GC–MS analyses of the epicuticular wax in 10, 20, 30 and 40 daa VEP, NDD1, NY, TVT and DUL fruit rinds (Suppl. Tables S1-S4). This comparative study enabled us to (i) understand the developmental dynamics in epicuticular wax accumulation in melon fruit accessions with white and green rinds; and consequently (ii) further clarify the association between the white rind trait and epicuticular waxes, and determine the relative contribution of specific wax monomers and/or biochemical classes to this trait.

Based on the 48 annotated wax monomers appearing in Table 1, we plotted all samples' metabolic datasets of the five melon varieties at the four developmental stages onto principal component analyses (PCAs), enabling us to distinguish differences in these datasets between varieties and/or developmental stages. The results showed that samples belonging to VEP and NDD1 fruit, which develop external white rind, and the TVT and DUL fruit, which feature green rind, were departed in all four developmental stages (Fig. 3). This trend suggests differences in total wax loads with possible variations in the metabolite levels within the different wax biochemical classes identified. Developing NY fruit, too, differed from all other samples, but by 40 daa, their samples were relatively closer to that of VEP and NDD1 fruit compared to TVT and DUL (Fig. 3). Calculating the variances explained by the first two principal components (PC1 and PC2) inferred increased percentages during fruit development, with 63%, 71%, 78% and 89% calculated for the PCAs of 10 daa, 20 daa, 30 daa and 40 daa, respectively (Fig. 3). This variance patterning thus suggests that total wax loads and/or wax monomer levels of VEP, NDD1 and NY fruit are unlike those of TVT and DUL fruit, with these differences becoming more pronounced as the fruit develop and mature.

Fig. 3
figure 3

Developmental dynamics in epicuticular wax accumulation in melon fruit. Principal component analyses (PCAs) of the spatial distribution of epicuticular wax profiles following GC–MS analyses in rind samples of VEP, NDD1, NY, TVT and DUL fruit at 10, 20, 30 and 40 daa. The variances explained by each of the first two components (PC1 and PC2) appear in parentheses

Full size image

Changes in total wax loads and compositional variances of epicuticular waxes in melon fruit varieties with external white and green rinds

Calculations of the total wax load in the five melon accessions during the four stages of development point to a general increase in the deposition of epicuticular waxes throughout fruit development; this trend was mainly detected in the VEP, NDD1 and NY verities (Fig. 4A). At immature 10 daa, VEP and NDD1 rinds had approx. 75% and 50% more waxes compared to NY, TVT and DUL rinds, respectively, in accord with the SEM observations at this developmental stage (Fig. 4A). At 20 daa, the amounts of accumulated waxes in VEP, NDD1 and DUL fruit was slightly higher than in NY and TVT fruit (Fig. 4A). At 30 daa, VEP and NDD1 fruit accumulated up to 2.6-fold more waxes compared to NY, TVT and DUL fruit (Fig. 4A). Nevertheless, the most prominent differences in total wax loads were at the mature stage of 40 daa, when VEP, NDD1 and NY fruit exhibited substantial inclines in their total wax loads, while TVT and DUL fruit displayed only minor changes in their total wax loads compared to 30 daa (Fig. 4A). Again, these patterns corroborate our microscopic observations at this developmental stage, and provide an explanation for the highest separation and variances observed in the 40 daa principal component analysis (PCA). They further demonstrate that the external white rind trait is indeed associated with the synthesis of epicuticular waxes, which seemingly accumulate at later stages of development, toward fruit maturity.

Fig. 4
figure 4

Changes in total wax loads and compositions in epicuticular waxes in melon fruit accessions with external white and green rinds. A Total wax loads in VEP, NDD1, NY, TVT and DUL fruit at 10, 20, 30 and 40 daa following GC–MS analyses. The y-axis represents peak areas following normalization to the C36 n-alkane internal standard. Averages ± S.E. of n = 3 (each replicate consists of peel discs isolated from at least three individual fruit). Significance was calculated according to a one-way ANOVA of P-value < 0.05 and the Tukey’s post-hoc method and presented in small letters above dots. B Bar graphs of compositional variances in epicuticular waxes according to their biochemical classes in mature (40 daa) VEP, NDD1, NY, TVT and DUL fruit following GC–MS analyses

Full size image

To determine whether the five accessions also differ in their composition of epicuticular waxes, i.e., if each biochemical class contributes differently to the overall total wax load, we calculated the relative contribution (%) of each of these classes in the metabolic datasets obtained from 40 daa fruit, where the highest loads were measured. The n-alkanes constituted slightly more than 50% of the total wax loads in both VEP and NDD1 fruit, and 47% in NY fruit (Fig. 4B). This figure dropped to 42% in TVT fruit, and in DUL fruit, it constituted only 21% of the total wax load. In contrast, fatty acids accounted for 40% of the total wax load in DUL fruit, whereas in all other varieties, their contribution ranged between 6 and 14% (Fig. 4B). At the same time, the abundance of fatty amides was very low in VEP, NDD1 and NY fruit, as opposed to the higher abundances in TVT and DUL. Lastly, aldehydes and wax esters were notably less abundant in DUL fruit (Fig. 4B). Together, these data suggest that the external white rind trait in melon fruit is not only associated with higher total wax loads but also with compositional variances in certain biochemical classes such as n-alkanes, apparently at the expense of fatty acids.

Melon fruit varieties with external white rinds accumulate high levels of n-alkanes, fatty alcohols, aldehydes and wax esters

As we detected quantitative and compositional differences in the epicuticular wax profiles of 40 daa melon fruit with external white and green rinds, we also compared the levels of each biochemical class in the five varieties. We found distinct correlations between the appearance of external white rind and the accumulation of n-alkanes, fatty alcohols, aldehydes, tocopherols, wax esters and triterpenoids. These correlations were most significant in VEP fruit, but they were also pronounced in NDD1 and NY fruit (Fig. 5). Large amounts of n-alkanes and wax esters were found in VEP and NDD1 fruit, moderate levels in NY fruit, with the lowest levels detected in TVT and DUL fruit (Fig. 5). For example, VEP fruit accumulated ninefold more C31 n-alkane—the most abundant alkane in the epicuticular wax profile in melon fruit—than did DUL fruit (Suppl. Tables S1-S4). The level of docosyl acetate, the predominant wax ester identified, was ~ 500-fold higher in VEP and NDD1 fruit than in DUL fruit, where it was virtually absent (Suppl. Tables S1-S4). VEP fruit accumulated 5.4-fold higher levels of fatty alcohols compared to TVT and DUL fruit, where NDD1 and NY exhibited 2.1- and 1.8-folds, respectively (Fig. 5). Similarly, aldehydes were 16-fold more abundant in the peel of VEP fruit, and slightly more than ninefold more abundant in NDD1 and NY fruit, compared to DUL fruit (Fig. 5). Here, although TVT fruit accumulated significantly fewer aldehydes compared to VEP, NDD1 and NY fruit, it still synthesized sixfold more aldehydes than did DUL fruit (Fig. 5).

Fig. 5
figure 5

Melon fruit varieties with external white rinds accumulate high levels of fatty alcohols, aldehydes, n-alkanes and wax esters. The accumulation patterns of fatty acids, fatty alcohols, aldehydes, fatty amides, n-alkanes, tocopherols, triterpenoids and wax esters, in mature (40 daa) VEP, NDD1, NY, TVT and DUL fruit. The y-axes represent peak areas following normalization to the C36 n-alkane internal standard. Averages ± S.E. of n = 3 (each replicate consists of peel discs isolated from at least three individual fruit). Significance was calculated according to a one-way ANOVA of P-value < 0.05 and the Tukey’s post-hoc method and presented in small letters above bars

Full size image

Unlike the biochemical classes listed above, fatty acids did not show a specific correlation with either white or green rind varieties, nor with the calculated total wax loads. For instance, NDD1, NY and DUL fruit accumulated the highest levels of fatty acids and TVT the lowest levels (Fig. 5). DUL fruit accumulated larger amounts of C16, C20 and C22 fatty acids than did all the other fruit varieties, but at the same time, deposited much less C28 fatty acid (Suppl. Tables S1–S4). NDD1 fruit accumulated the highest levels of C15:1(9), C18 and C28 fatty acids, while NY fruit synthesized significantly higher levels of the C18:1(7) fatty acid (Suppl. Tables S1–S4). Lastly, we detected only one fatty amide in our metabolic datasets—13-docosenamide, a minor wax monomer in the overall composition of epicuticular waxes in the melon fruit. Nevertheless, its levels portray an interesting pattern: its levels are highest in DUL and TVT fruit, the two varieties with green rind—up to 7.8-fold more compared to VEP, NDD1 and NY fruit, which featured higher total wax loads and external white rinds (Fig. 5).

Discussion

Fruit color, a dominant indicator of quality that directly influences consumer choice, is determined by the level and composition of pigments in the epidermal and sub-epidermal layers of the fruit exocarp (Lara et al. 2019). In some fruit, however, color variation does not predominantly depend on pigmentation. For instance, blueberries, which are white-blue in color, accumulate anthocyanins that produce red coloring. Their whitish-bluish appearance comes, in fact, from the accumulation of epicuticular waxes, with color variations in different blueberry genotypes closely related to the content and composition of cuticular waxes (Yan et al. 2022). A recent study showed that the chromatic blue-ultraviolet reflectance of the dark-pigmented blueberries, plums and juniper cones arises from the interaction of their randomly arranged wax crystals with light. Evidently, even though these species’ wax morphologies differ greatly, their spectral signatures are similar (Middleton et al. 2024), implying a pivotal role for epicuticular waxes in fruit surface coloration. This role was also reported in plums, grapes and apples, where these components account for the white coating on their surfaces, often referred to as “wax blooms” (Belding et al. 1998; Saftner et al. 2008; Wisuthiphaet et al. 2014).

Melon species reflect remarkable phenotypic variations in terms of fruit surface pigmentation. Like in many other fruits, these pigments are largely attributed to the synthesis of carotenoids, flavonoids and chlorophylls (Tadmor et al. 2010). Recently, a key recessive gene, CmAPRR2, was shown to account for a large part of the light rind types across melon diversity and is explained by fewer pigments due to a lower number of chloroplasts and chromoplasts (Oren et al. 2019). Similarly, the APRR2 gene was shown to account for the white immature rind color in cucumber fruit, a close relative of melon, and to associate with reduced chloroplast number and chlorophyll content (Liu et al. 2016). Yet, previous reports have suggested that some melon varieties may also produce a light color, reflected in their white-grayish coating. The dominant white rind color is likely controlled by the single Wi gene (Kubicki 1962; Dogimont 2011) that was mapped to an ~ 1.6 Mb interval on chromosome 7 (Pereira et al. 2018). The results obtained here provide the first qualitative and quantitative chemical evidence that the dominant white rind trait is apparently linked to the biosynthesis of epicuticular waxes in the melon fruit rind. Yet, thus far gene(s) within the corresponding ~ 1.6 Mb interval on chromosome 7 were not identified to be involved with the biosynthetic pathways of epicuticular waxes and/or cuticular lipids, urging for future efforts to map this genomic region.

We first demonstrate that the rind of melon fruit is a rich source of epicuticular waxes, including fatty acids, fatty alcohols, aldehydes, fatty amides, n-alkanes, tocopherols, triterpenoids, and wax esters. The dominant components in these waxes are the n-alkanes, followed by the fatty alcohols. This compositional trend was reported for other fleshy fruit including wild and cultivated tomatoes (Vogg et al. 2004; Leide et al. 2007; Yeats et al. 2012), cucumber (Wang et al. 2015a,b), pepper (Parsons et al. 2012), apple (Belding et al. 1998; Legay et al. 2017), Asian and European pear varieties (Wu et al. 2017, 2018), and eggplant (Bauer et al. 2005). In many of these fruit species, C29 and C31 were the predominant n-alkanes, in line with our observations for the melon fruit rind epicuticular wax coverage. Our data showed that the melon fruit rind accumulates glutinol, a pentacyclic triterpenoid. Park et al. (2020) also reported the existence of glutinol in the wax fractions isolated from the rind of Cucumis melo var L. cv “Smart”, and also as part of the epicuticular wax composition in the bell pepper fruit (Bauer et al. 2005). Glutinol is known for its anti-inflammatory activity and anti-oxidant properties (Sandhu et al. 2023), and thereby our data may suggest the melon fruit rind as a good source for extracting this compound for health-beneficial reasons.

Our microscopy observations delineated a dense coverage of epicuticular waxes on the rind of VEP and NDD1 fruit, marked by a distinct white rind. In contrast, a scattered distribution of waxes was detected on the surface of TVT and DUL fruit, which features green rind and lacks a white coating. Consistently, our metabolite profiling showed that the former accessions accumulated far greater total wax loads, attributed primarily to the accumulation of n-alkanes, fatty alcohols, aldehydes and wax esters. Thus, our data provides, for the first time, a possible link between that accrual of epicuticular waxes and the dominant white rind trait in melon fruit. The link between a white color appearance and epicuticular waxes has also been reported in the surfaces of other fruit and has been suggested to be species- and biochemical class-specific. Our finding that the dominant white color trait in melon is associated with the accumulation of n-alkanes has also been reported in grapes (Shin et al. 2009). In plums, a continuous layer of amorphous or crystalline waxes accounts for the white color that was mostly governed by the accumulation of cyclic triterpenoids (Mukhtar et al. 2014). Similarly, the white coating in blueberries seems to be affected by the accumulation patterns of triterpenoids, as well as β-diketones (Chu et al. 2017; Yan et al. 2022). Various watermelon cultivars also exhibit wax bloom and bloomless phenotypes, and several candidate genes for the white rind color were recently suggested based on QTL-Seq analyses (Lee et al. 2022), though the metabolic foundation underlying this phenotype or the profile of epicuticular waxes have yet to be reported. Interestingly, wax blooms, or white coloring of pumpkin fruit surfaces were accompanied by the appearance of silicon (Sakata et al. 2008). Another study used SEM and electron probe micro-analyzer (EPMA) tools to reveal that the surfaces of cucumber fruit deposit silicon particularly in the vicinity of warts, that are remnants of spines and/or trichomes (Tripathi et al. 2017). Nonetheless, we could not find any reports suggesting the existence of silicon in the exocarp surfaces of melon fruit verities, yet the fact that these element structures were observed in other Cucurbitaceae species raise the possibility that components other than epicuticular waxes may contribute to the white color rind phenotype in melon.

The ultrastructure of epicuticular wax crystals deposited on top of the white rind surfaces of VEP, NDD1 and NY fruits also differed than those found at TVT and DUL fruit surfaces. It is presumed that various structures of wax crystals like rodlets, tubules or platelets, depend on different chemical compositions of epicuticular waxes. Like in our case, platelet wax crystals were associated with higher portions of aldehydes and alkanes in the epicuticular wax compositions of 'Newhall' navel orange fruits (Wang et al. 2014). At the same time, a similar shape of wax crystals was associated with the accumulation of primary alcohols in wheat waxes (Koch et al. 2006), further suggesting that wax crystal ultrastructure can be influenced by the accumulation of different wax biochemical classes. Nevertheless, our data provide a link between the deposition of wax platelets at the surfaces of melon fruit cultivars to the accumulation of n-alkanes, alcohols, aldehydes and esters.

Surface waxes were shown to influence fruit postharvest quality and thus impact its shelf-life capacities. For instance, wax alkanes reduced fruit water loss in zucchini (Carvajal et al. 2021), tomato (Romero and Rose 2019) and guava (Huang et al. 2020), whereas triterpenoids were shown to affect similarly blueberries during storage (Yan and Castellarin 2022). In grapes, the epicuticular wax layer provides mechanical strength upon changes occur during post-harvest storage (Chang and Keller 2021). As above mentioned, to the best of our knowledge we disclosed here the first-ever inclusive metabolic repertoire of epicuticular waxes deposited on top of the melon fruit rind. Hence, it would be highly interesting to further investigate whether these rich wax repertoires play important roles during postharvest and whether manipulating wax profiles in melon may provide a means of improving fruit quality during storage.

Conclusions

In the current study, we show that melon fruit has a rich repertoire of epicuticular waxes from various biochemical classes. In addition, we show, for the first time, that, like in other fleshy fruit species, the white rind color phenotype in melon is associated with higher loads of epicuticular waxes, particularly from the biochemical classes of n-alkanes, fatty alcohols, aldehydes and wax esters. The accumulation of wax components belonging to these biochemical classes seem to be associated with the deposition of platelet-like wax crystals at the surfaces of mature fruits with white rinds. A major locus on chromosome 7 was already proposed to account for the dominant white rind color in melon fruit, and it will be interesting to link between these genomic data to the metabolic characteristics described herein for this type of rind. Together, this study offers new insight for melon breeding programs seeking to improve the fruit surface color to meet market preferences.