During ripening, citrus fruits undergo a conspicuous color change in the peel from green to yellow, red, or orange (Iglesias et al., 2007). This color change, commonly referred to as peel degreening, is mainly caused by the successive degradation of two major green-colored chlorophyll (Chl) pigments, namely, Chl a and Chl b, to colorless compounds (Hörtensteiner, 2006). To enter into the degradation pathway, Chl b is first converted to Chl a by Chl b reductase, which is encoded by non-yellow coloring 1 (NYC1) and NYC1-like (NOL) (Horie et al., 2009). Key steps in the Chl a degradation pathway include the dephytilation by chlorophyllase (CLH) and pheophytinase (PPH) and the removal of the central Mg atom by Mg dechelatase, which is encoded by stay-green (SGR) (Shimoda et al., 2016). The resulting compound pheophorbide is then converted to red Chl catabolites (RCC) by pheophorbide a oxidase, and RCC is reduced to colorless compounds by RCC reductase (RCCR), which is encoded by accelerated cell death 2 (ACD2) (Mach et al., 2001). In addition, Chl degradation is also facilitated by the disaggregation of photosystem proteins, particularly the light-harvesting complex (LHC) and Chl a–b binding (CAB) proteins (Barry, 2009), which leads to the accumulation of free Chls and hence allows for easier access by degrading enzymes. Several transcription factors, including ethylene response factors (ERFs), directly influence Chl degradation-related genes (Yin et al., 2016), and their expression levels have been shown to negatively correlate with Chl content in citrus fruit peels (Xie et al., 2014).

It has been broadly demonstrated that exogenous ethylene application accelerates peel degreening in citrus fruits (Mayuoni et al., 2011), despite being classified as non-climacteric. In this regard, ethylene treatments are routinely used in the citrus fruit industry to hasten the degreening process or to ensure uniform color (Porat, 2008; Morales et al., 2020). Exogenous ethylene activates the expression of various genes encoding Chl degradation enzymes (Shemer et al., 2008; Yin et al., 2016), as well as transcription factors (Yin et al., 2016). However, the role of endogenous ethylene in natural peel degreening is still puzzling since mature citrus fruits produce trace ethylene levels (Katz et al., 2004).

Low temperature has also been shown to promote peel degreening in several citrus fruit species including lemons (Manera et al., 2012; Conesa et al., 2019; Mitalo et al., 2020), grapefruits (Manera et al., 2013), oranges (Carmona et al., 2012a), and mandarins (Tietel et al., 2012). However, there is a conflicting information regarding the regulatory mechanisms involved. On one hand, it is argued that the low ethylene amounts produced by mature citrus fruits are likely involved in low temperature-induced peel degreening (Goldschmidt et al., 1993; Carmona et al., 2012b). However, our recent study in lemons (Mitalo et al., 2020) showed that low-temperature enhancement of peel degreening is achieved by the activation of genes associated with Chl degradation, carotenoid metabolism, photosystem proteins, and transcription factors, independently of endogenous ethylene. Ethylene-independent stimulation of fruit ripening by low temperatures has also been reported in other fruit species such as kiwifruit (Asiche et al., 2018; Mitalo et al., 2019) and apples (Tacken et al., 2010). In citrus fruit, however, it remains unclear whether a similar phenomenon takes place in other species besides lemons or not, as some species/varieties are known to be more sensitive to ethylene than others (Petracek and Montalvo, 1997; John-Karuppiah and Burns, 2010; Alós et al., 2014).

In Japan, citrus cultivation is predominated by Satsuma mandarin (Citrus unshiu Marc.) fruit (Omura and Shimada, 2016), which comprises numerous cultivars that account for 62.5% of all citrus acreage¹ During fruit maturation, especially in the late-harvesting genotypes, peel degreening often begins in October (Takagi et al., 1994; Ikoma et al., 2001), in conjunction with the ambient low temperatures that are associated with autumn. Additionally, most early-maturing genotypes, which are usually harvested in August (during high summer temperatures), typically display poor peel coloration (personal observation). These findings suggest that low temperatures are required for normal color development in Satsuma mandarins, but the phenomenon is yet to be investigated objectively.

In this study, we examined whether low temperature had a direct role in peel degreening during natural maturation in Satsuma mandarin fruit. Postharvest peel color changes were monitored in response to propylene (a well-known ethylene analog; McMurchie et al., 1972) treatment at 25°C or during storage at 5°C, 10°C, 15°C, 20°C, and 25°C, as well as during on-tree maturation. RNA-Seq was then used to find genes that were differentially regulated by ethylene signaling and/or low temperature. Furthermore, gene expression changes during postharvest and on-tree peel degreening were compared to determine the roles of low temperature and ethylene in their regulation. Overall, data in this study demonstrate that low temperature is deeply involved in the regulation of peel degreening in Satsuma mandarins, possibly via an ethylene-independent mechanism.

Peel degreening is a critical part of fruit ripening in many citrus fruits (Iglesias et al., 2007), and there is an ongoing effort, both for physiological and commercial reasons, to deduce the regulatory mechanisms involved. In the present study, treatment with propylene, a well-known ethylene analog (McMurchie et al., 1972), triggered rapid peel degreening and the associated loss of Chl pigments in Satsuma mandarin fruit (Figure 1). These findings are consistent with previous reports on Satsuma mandarins (Morales et al., 2020) and other citrus fruits (Rodrigo and Zacarias, 2007; Shemer et al., 2008; Mayuoni et al., 2011), where the application of exogenous ethylene was shown to accelerate ripening-related peel color changes. The mechanisms and pathways involved in citrus peel degreening in response to exogenous ethylene sources have been well-characterized to date (Shemer et al., 2008; Alós et al., 2014; Yin et al., 2016). However, the causal factors of natural peel degreening remain elusive, as mature citrus fruit are non-climacteric, producing only minute endogenous ethylene amounts (Katz et al., 2004). Indeed, ethylene production levels were undetectable both in propylene-treated and stored samples (data not shown), confirming the non-climacteric nature of the Satsuma mandarin fruit used in this study.

Postharvest storage at 10°C and 15°C clearly stimulated green color loss and a reduction in peel Chl content, whereas no changes were observed at 20°C and 25°C (Figure 2), which indicated that low temperatures are required for peel degreening to occur. Previous studies have also demonstrated that low/intermediate temperatures promote peel degreening in various citrus fruit species (Zhu et al., 2011; Tietel et al., 2012; Mitalo et al., 2020). The regulatory mechanisms for low temperature-triggered peel degreening are still unclear, but it has been suggested that ethylene signaling is involved; that is, either trace levels of System I ethylene produced by mature citrus fruit are physiologically active (Goldschmidt et al., 1993; Carmona et al., 2012b) or the ethylene sensitivity of mature citrus fruit likely increases during maturation (Alós et al., 2014). However, we recently demonstrated that repeated treatments of lemon fruit with 1-methylclopropene (1-MCP) to inhibit ethylene signaling failed to abolish the accelerated peel degreening triggered during low-temperature storage. These findings indicate that ethylene signaling is not functional during natural peel degreening, at least in mature lemon fruit. In this study, we attempted to replicate this finding in Satsuma mandarins, but 1-MCP treatments surprisingly stimulated massive ethylene production and subsequent peel browning (Supplementary Figure 1).

To delve into the mechanism of how low storage temperatures induced peel degreening in Satsuma mandarins, we analyzed the transcriptomic changes involved relative to those induced by ethylene. This analysis enabled us to identify a distinct set of genes that were exclusively regulated by low temperature, and ethylene signaling triggered by propylene treatment had no influence on their expression (Figure 4C, Supplementary Table 3). These low temperature-specific genes provided us with the first indication that the mechanisms involved in low-temperature promotion of peel degreening are independent of ethylene influence. Additionally, there was a second distinct set of genes that were exclusively regulated by the ethylene signal triggered by propylene treatment (Figure 4C, Supplementary Table 2). The finding that these ethylene-specific genes showed insignificant changes in expression during storage further suggests that ethylene signaling was not functional in stored Satsuma mandarins. Thus, even though the third group of genes, including those associated with Chl degradation, responded to both propylene treatment and low storage temperatures (Figure 5, Supplementary Table 4), it appears that distinct and independent regulatory mechanisms are involved. Indeed, Mitalo et al. (2020) demonstrated that ClLHCB2 was downregulated by both ethylene treatment and low storage temperatures during degreeening in mature lemons; however, its downregulation by low temperature was not inhibited by repeated 1-MCP treatments, confirming that the observed expression changes were independent of ethylene.

On-tree peel degreening and the associated reduction in Chl content typically coincided with gradual autumnal drops in environmental temperature (Figure 3). The peel color changes and associated gene expression were particularly initiated when ambient temperatures in the orchard location had decreased to below 16°C (Figures 3, 6), which also agrees with previous findings in other citrus genotypes (Manera et al., 2012, 2013; Conesa et al., 2019), indicating that low ambient temperatures are required for normal degreening during on-tree maturation. Interestingly, the two ethylene-specific genes presented in this study (CuCOPT1 and CuPOX-A2) showed insignificant expression changes during on-tree peel degreening (Figures 7A,B), further alluding to a non-functional ethylene signaling pathway in mature Satsuma mandarin fruits. In lemons, however, ClCOPT1 expression levels were undetectable by RNA-Seq while ClPOX-A2 was upregulated highly by ethylene treatment and slightly by low storage temperatures (Mitalo et al., 2020), and this highlights fruit species-related differences in gene expression patterns during peel degreening. Nevertheless, the low temperature-specific expression pattern displayed by CuERF3 (Figure 7C) was also reported in lemon fruit during on-tree peel degreening as environmental temperatures decreased. These findings clearly suggest that low temperature plays a role in modulating on-tree peel degreening in Satsuma mandarins likely in an ethylene-independent manner.

The transcriptomic data further indicated that many genes were differentially expressed during storage at 5°C (Figure 4A), yet no peel degreening changes took place (Figure 2). A similar observation was also reported in our previous study of lemon fruit (Mitalo et al., 2020). While the cause of this anomaly is unclear, it could possibly be due to the slow protein synthesis rate and/or low activity of peel degreening-associated enzymes at 5°C (Yun et al., 2012).

Previous studies have demonstrated the correlation between Chl degradation and the expression of CLH genes during ethylene-induced degreening in various citrus genotypes (Jacob-Wilk et al., 1999; Yin et al., 2016). However, transcript levels of CLH do not increase during peel degreening in non-treated fruit both on and off the tree (Jacob-Wilk et al., 1999; Mitalo et al., 2020), raising uncertainties about its role. In this study, all the identified CLH genes, including CuCLH1, were downregulated both in response to ethylene and low storage temperatures and displayed minimal expression changes during on-tree peel degreening (Supplementary Figure 2, Supplementary Table 2, Supplementary Table 3). Therefore, it is possible that, in Satsuma mandarin fruit, CLH enzyme activity is either not involved in Chl degradation or regulated at a different level other than transcription.

Besides Chl degradation, the other important process that contributes to peel degreening is carotenoid metabolism (Kato, 2012). While the carotenoid content and profiles were not quantified in this study, RNA-Seq analysis identified many biosynthetic genes that were differentially expressed, and subsequent RT-qPCR analysis confirmed that most of them, including CuPSY1, CuLCYb2a, and CuNCED5, were upregulated by both propylene treatment and low storage temperatures (Supplementary Figure 3). Increased expression of carotenoid biosynthetic genes in response to either ethylene signaling, or low temperature storage has also been reported during peel degreening in various citrus fruits (Rodrigo and Zacarias, 2007; Matsumoto et al., 2009; Mitalo et al., 2020).

This study investigated the roles of ethylene and low temperature in fruit ripening-related peel degreening in Satsuma mandarin fruit during on-tree maturation and postharvest storage. Transcriptomic data unveiled distinct sets of genes that were regulated by ethylene signaling and/or low temperature. A comparison of the expression patterns of these genes provides a clear indication that low temperature promotes peel degreening in Satsuma mandarin fruit, most likely via an ethylene-independent mechanism.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: DNA Data Bank of Japan (DDBJ) database (DRR376678–DRR376696).

OM, WA, and YK conceived and designed the study. OM and WA did most of the experiments with close supervision from YK, TA, and KU. KU performed RNA-Seq analysis. SK and HE provided technical assistance. OM wrote the first draft of the manuscript which was substantially improved by SK and HE. All authors have read and approved the submitted version. All authors contributed to the article and approved the submitted version.

OM is a JSPS International research Fellow (Graduate School of Life and Environmental Sciences, University of Tsukuba). This study was supported in part by a Grant-in-Aid for Scientific Research (grant no. 20H02977) by the Japan Society for the Promotion of Science, Japan (JSPS).

WA is employed by Del Monte Kenya Ltd. The remaining authors declare that the study was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.