**1. Introduction**

Pentatricopeptide (PPR) proteins are one of the largest gene families in plants, and are usually characterized by an array of 2–30 tandem repeats of a degenerated unit consisting of 30–40 amino acid (aa) sequence motifs [1]. According to the domain architecture, PPR proteins are divided into subfamilies of P and PPR-like long and short (PLS), which are characterized by motifs without space and motifs with interspaced either of short (31 aa) or long (35–36 aa) PPR-like motifs, respectively. Based on the domain assembly in the C terminal of a PPR protein, the PLS subfamily is further

classified into five subgroups: PLS, E1, E2, E+, and DYW [1–3]. Since the discovery of PPR proteins in yeas<sup>t</sup> (*Saccharomyces cerevisiae* L.) [4], these have been reported in various terrestrial plants. Being a large gene family, more than 400 PPR proteins have been reported in various plants, including *Arabidopsis* (441), foxtail millet (486), poplar (626), maize (491), and rice (477) [1,3,5–7]. PPR proteins have been found to exhibit RNA-binding properties, which facilitate in mediating gene expression through posttranscriptional processes associated with transcripts in the mitochondria, chloroplast, and nucleus. Thus, PPR genes are thought to have a major impact on organelle stability, including biogenesis, and function through their involvement in various posttranscriptional processes such as RNA-editing [1,8], RNA-splicing [9], and RNA-processing [10].

The functions of PPR proteins have been reported to be associated with plant growth and development and organelle formation. In Arabidopsis, a chloroplast-localized PPR protein, called EMBRYO-DEFECTIVE175 (EMB175), has been found to influence embryo morphogenesis [11]. Similarly, in maize, a mitochondria-localized P-type PPR protein, EMPTY PERICARP12 (EMP12), has been reported to be essential for embryogenesis and endosperm development through trans-splicing of mitochondrial *nad2* introns [12]. Hsieh et al. [13] showed that the *SLOW GROWTH3* (*SLO3*) gene, encoding a PPR protein, was involved in the splicing of *nad7* intron 2 in Arabidopsis, and its mutant, *slo3,* exhibited a dysfunctional mitochondrion, which resulted in growth retardation and delayed development. In rice, few DYW-type PPR proteins, such as OPAQUE AND GROWTH RETARDATION 1 (OGR1) and PPS1, play important roles in C →U RNA editing of mitochondrial transcripts; upon silencing of these genes using T-DNA insertion and RNAi, mutant plants exhibited various pleiotropic phenotypes, including late seed germination, retarded growth, delayed development, dwarfing, and partial pollen sterility at both vegetative and reproductive stages [14,15]. In addition to plant developmental process, PPR proteins have been reported to be involved in the responses to biotic and abiotic stresses. For instance, in Arabidopsis, several PPR proteins such as *SOAR1* [16], *PGN* [17], *SLG1* [18], and *PPR96* [19], have been shown to participate in responses to abiotic stresses.

PPR proteins have also been characterized to be involved in cytoplasmic male sterility (CMS); it is an important intriguing issue in plants [20], as only the male gametes are impaired, resulting in a failure to produce functional pollens. Some of the PPR proteins encoded as restorers of fertility (*Rf*) genes mask the mitochondrial transcripts that cause CMS and thus restores fertility. The *Rf*-*PPR* genes have been reported in various terrestrial plants, including petunia (*RF952*) [21], radish (orf687) [22], pepper (*CaPPR6*) [23], and Arabidopsis (*RFL2*) [24]. Increasing molecular evidence has clearly emphasized the roles of PPR genes in fruit development, ripening, and flesh color of plants [25–27]. *GUN1,* which encodes a plastid-located PPR protein, has been reported to be involved in the plastid-to-nucleus retrograde signaling pathway during fruit development and ripening in tomato [26]. In relation to this, ripening impaired tomato mutants such as *Cnr* dramatically repressed expression of genes associated with ripening, cell wall-degrading enzymes and PPR repeat-containing proteins, resulting in mature fruits with colorless pericarp tissue, thereby indicating that PPR proteins play a significant role in fruit development [25]. Recently, in melon (*Cucumis melo* L.), the *white-flesh* gene, named *CmPPR1* (MELO3C003069), encoding a plastid-targeted P-type PPR protein, has been reported to be a candidate gene in one of the two major QTL, which determine flesh color intensity [27]. Furthermore, the polymorphic SNP markers in P-class motifs of *CmPPR1* have been found to contribute to genetic variation in orange, green, and white fruit flesh colors within the species. It has also been established that this *CmPPR1* is possibly involved in plastid-to-nucleus retrograde signaling, thereby a ffecting the expression of plastid-targeted genes, indicating the involvement of PPR proteins in fruit flesh color variation [27].

Watermelon (*Citrullus lanatus*) is an important fruit crop with overall annual production of more than 103 million tons worldwide (http://www.fao.org/statistics/en/. Watermelon exhibits diverse variation in fruit-quality traits, including soluble sugars, firmness, fruit size, shape, skin, and flesh color, along with functional factors such as lycopene and β-carotene [28,29]. This extensive polymorphic variation motivates researchers to investigate the genetics of watermelon fruit-quality traits. PPR genes related to fruit development and flesh color variation in watermelon have not been studied yet. Fortunately, the recently released watermelon (97103 v2) genome sequence (http://cucurbitgenomics.org) provides an excellent opportunity to perform a genome-wide analysis of important gene families, including the PPR gene family in watermelon. In the present study, we identified and characterized 464 putative *PPR* genes from 97103 watermelon genome. Furthermore, we investigated their intron-exon organization, chromosomal localization, types of PPR motifs, functional diversification, subcellular locations, and phylogenetic analysis. With a focus on the involvement of PPR genes in fruit development and flesh color variation in watermelon, we also examined their expression patterns through several RNA-seq analyses from the cucurbit expression atlas (http://cucurbitgenomics.org/rnaseq/home). Finally, diagnostic SNP-CAPS markers of PPR genes were developed to study their association with fruit flesh color variation. Thus, the findings of this study will contribute to the understanding of PPR gene distribution and functions in watermelon, and also improve our understanding of the relationship between PPR genes and flesh color variations.
