Genome-Wide Identification of Apple Ubiquitin SINA E3 Ligase and Functional Characterization of MdSINA2

SINA (Seven in absentia) proteins are a small family of ubiquitin ligases that play important roles in regulating plant growth and developmental processes as well as in responses to diverse types of biotic and abiotic stress. However, the characteristics of the apple SINA family have not been previously studied. Here, we identified 11 MdSINAs members in the apple genome based on their conserved, N‐terminal RING and C-terminal SINA domains. We also reconstructed a phylogeny of these genes; characterized their chromosomal location, structure, and motifs; and identified two major groups of MdSINA genes. Subsequent qRT-PCR analyses were used to characterize the expression of MdSINA genes in various tissues and organs, and levels of expression were highest in leaves. MdSINAs were significantly induced under ABA and carbon- and nitrate-starvation treatment. Except for MdSINA1 and MdSINA7, the other MdSINA proteins could interact with each other. Moreover, MdSINA2 was found to be localized in the nucleus using Agrobacterium-mediated transient expression. Western-blot analysis showed that MdSINA2 accumulated extensively under light, decreased under darkness, and became insensitive to light when the RING domain was disrupted. Finally, ABA-hypersensitive phenotypes were confirmed by transgenic calli and the ectopic expression of MdSINA2 in Arabidopsis. In conclusion, our results suggest that MdSINA genes participate in the responses to different types of stress, and that MdSINA2 might act as a negative regulator in the ABA stress response.


INTRODUCTION
Ubiquitination is one of the most important post-translational modifications in plant development and the responses of plants to environment (Devoto et al., 2003;Xu and Jaffrey, 2013;Stone, 2014;Zhang et al., 2015). The ubiquitin-proteasome system consists of ubiquitin (Ub), ubiquitin activating enzyme (E1), ubiquitin-binding enzyme (E2), ubiquitin ligase (E3), and the 26S proteasome, and functions in the degradation of target proteins in a three-enzyme cascade manner (Sadanandom et al., 2012;Jansen et al., 2014). Ubiquitin molecules are small spherical (76 residues) and highly conserved proteins that can covalently modify target proteins to alter their stability, function, and localization (Schimke, 1973;Callis et al., 1995;Dametto et al., 2015). Initially, E1 activates Ub to form the E1-Ub complex with ATP. The activated Ub is transferred to the cysteine residue of the E2 ubiquitin-binding enzyme, forming an intermediary. The E3 ubiquitin ligase then recognizes the E2-Ub intermediary and the substrate protein, and transfers the Ub from the E2-Ub intermediary to the target protein. Finally, the substrate protein labeled by ubiquitin can be transported to the 26S proteasome for degradation (Santner and Estelle, 2010).
Ubiquitin ligases bind to specific ubiquitin-binding enzymes and target proteins, and the E3 ligase determines the diversity and specificity of the target protein. In the plant kingdom, ubiquitin ligase can be classified into three types based on its reaction mechanism and special conserved domains: HECT (Homology to E6-Associated Carboxy-Terminus) ubiquitin ligases, RING (Really Interesting New Gene) ubiquitin ligases, and U-box ubiquitin ligases (Duplan and Rivas, 2014). Among ubiquitin ligases, RING type E3 ligases have been the most extensively studied. RING type E3 ligases contain a conserved RING domain that consists of histidine and cysteine and can bind to zinc ions (Stone et al., 2005;Metzger et al., 2014). RING ubiquitin ligases can be further divided into two groups: single subunit RING ubiquitin ligases and multi-subunit RING ubiquitin ligases. Multi-subunit RING ubiquitin ligases can be subdivided into four types: Skp1-Cullin-F-box, VHL-ELONGIN-CUL2/5, BTB-CUL3 and DDB-CUL4 (Hua and Vierstra, 2011;Sadanandom et al., 2012). In recent years, RING finger ubiquitin E3 ligases have been extensively studied and have been shown to play important roles in plant growth and development as well as in response to stress. SINA (seven in absentia) proteins, RING-finger E3 ligases that were first identified in Drosophila (Carthew and Rubin, 1990), are involved in regulating the differentiation of light receptors (Li et al., 1997). Most SINA family members possess a highly conserved N-terminalRINGfingerdomainandaC-terminalSINAdomain.TheN-terminaldomainisforbindingtoE2,andtheSINAdomainrecognizes the target protein which is subsequently degraded by the 26S proteasome (Hu and Fearon, 1999;Den Herder et al., 2012). In addition, studies have found that SINA homologs can regulate their target proteins to adapt to different developmental stages and environmental changes. For example, SINAT5 Ler (Landsberg ecotype) can mediate the degradation of the transcriptional activator NAC1 which is involved in the auxin signaling pathway, thereby regulating lateral root development in Arabidopsis (Xie et al., 2002). Further study has shown that SINAT5 Ler could interact with FLC (Flowering Locus C), LHY (LATE ELONGATEDHYPOCOTYL), and DET1 (DE-ETIOLATED1) to regulate flowering time in Arabidopsis by promoting the degradation of FLC and LHY Park et al., 2010). SINAT2 is involved in carotenogenesis by interacting with RAP2.2 (Welsch et al., 2007). All SINATs can interact with dephosphorylated BES1, which is one of the core transcription factors involved in BR signaling. However, only SINAT5 Ler is known to be able to negatively regulate BR signaling by mediating the degradation of BES1.
Other major findings include the accumulation of SINATs proteins in light and their degradation in the dark  as well as the diverse synergistic and antagonistic functions of SINA members. For example, SINAT1 and SINAT2 can negatively regulate starvation-induced autophagy by ubiquitinating ATG6 (AUTOPHAGY PROTEIN6) or ATG13 (AUTOPHAGY PROTEIN13). Conversely, SINAT6 promotes autophagy by increasing the number of autophagic puncta under nutrient-rich or nutrient-poorconditions (Qietal.,2017;Qietal.,2020).Arecentstudy has demonstrated that the ectopic expression of tomato SlSINA4 resulted in cell death in N. benthamiana leaves, but overexpression of any of the other five SlSINAs can suppress the hypersensitive response and cell death . Abscisic acid (ABA) plays a critical role in plant growth and stress adaptation (Knight and Knight, 2001). Overexpression of SINA2 (SEVEN IN ABSENTIA 2) increases tolerance to drought by inducing the closure of stomata in Arabidopsis (Bao et al., 2014). OsDIS1, a homologous protein of SINAinrice,isanegativeregulatorinthedroughtresponse (Ningetal., 2011). In addition, Siah1 and Siah2, human SINA homologs are involved in multiple processes such as synaptic transmission, apoptosis, tumor suppression, and stress response (Wheeler et al., 2002;Franck et al., 2006;Khurana et al., 2006;Fukuba et al., 2007).
Our knowledge of the SINA family in apple is limited. Here, we identified 11 MdSINA members in apple using bioinformatics analyses. We analyzed tissue expression patterns and the responses of these genes to abiotic stress. Additionally, we assessed the ability of these genes to form homodimers and heterodimers using yeast two-hybrid (Y2H) assays. Our results provide basic information on the function of SINA proteins in apple.
MEGA: X was used to construct a neighbor-joining evolutionary tree with 1,000 bootstrap replications and a Poisson model (Kumar et al., 2018). Six SINA proteins in rice (Oryza sativa) were obtained from the Phytozome database (https://phytozome.jgi.doe.gov/pz/portal.html) (Goodstein et al., 2012). In MEGA X, ClustalW method was selected for sequence alignment, partial deletion and 95% site coverage cutoff were used gaps/missing data treatment.

Synteny Analysis
The syntenic blocks information of the apple genome can be downloaded from the Plant Genome Duplication Database (http://chibba.agtec.uga.edu/duplication/). The genome sequence of the homologous SINA proteins in apples was evaluated using BLASTP. The MCScanX algorithm (Wang et al., 2012) was used to scan syntenic blocks containing the apple SINA genes.

Plant Materials and Treatments
Materials included various apple tissues, tissue-cultured apple seedlings (Malus × domestic "Royal Gala"), "Orin" apple calli, "Columbia" (WT) Arabidopsis seedlings (Col-0), and tobacco (Nicotiana benthamiana). The growth conditions of tissuecultured apple seedlings, "Orin" apple calli, Arabidopsis seedlings and tobacco were the same as those described previously . Three-week-old shoot tissue culture seedlings were used for rooting with 0.2 mg/L IAA. We collected the roots, stems, leaves, flowers, and fruits of a 7-year-old "Gala" apple tree, which were immediately frozen in liquid nitrogen and stored at 80°C for tissue expression analysis.
For ABA treatment, 3-week-old apple seedlings were treated with 100 mM ABA for 0, 1, 3, 6, 12, and 24 h following previously described methods  and were then immediately frozen in liquid nitrogen and stored at −80°C for subsequent analysis. For the carbon-deprivation treatment, 3-week-old apple seedlings with roots grown in MS medium were treated without sucrose and in the dark for 0, 6, 12, 24, 36, and 48 h. For nitratedeprivation treatment, 3-week-old seedlings grown on MS medium were transferred to nitrogen-deficient MS liquid medium and grown under normal growth conditions for 0, 6, 12, 24, 36, and 48 h.

RNA Extraction and qRT-PCR Analysis
Total RNAs of apple samples and calli was isolated using a RNA extraction kit (Tiangen, Beijing, China) and used for qRT-PCR analysis following previously described methods (Ren et al., 2019). Three technical and biological replicates were performed to detect the transcripts of MdSINA genes. Specific primers of the MdSINA genes were designed and used for qRT-PCR analysis (Supplementary Table 1).

Vector Construction and Genetic Transformation
The open reading frames of MdSINA2 was cloned into the pRI-GFP vector to construct the overexpression vector. The primer pairs MdSINA2-F (5′-GTCGAC ATGGACTTGGAAAGCAT CGAGTGT-3′)/−R and (5′-GGATCC GCTACACAGGTTT GGTATGCA-3′) were used to amplify the full-length MdSINA2. "Orin" apple calli were infected by the recombinant plasmids 35S::GFP-MdSINA2 and WT (35S::GFP) via the Agrobacterium-mediated transformation method as previously described . The transgenic Arabidopsis were obtained by using the floral dip transformation method (Clough and Bent, 1998).

Yeast Two-Hybrid (Y2H) Assays
Interactions among MdSINA proteins were characterized by Y2H assays. The full-length cDNA of MdSINAs was cloned and inserted into yeast vector pGBT9 and pGAD424. The recombinant plasmids of pGAD424-MdSINAs and pGBT9-MdSINAs were transformed into yeast "Y2H Gold." The yeast was grown on selection medium as described previously ; the primer pairs used are listed in Supplementary  Table 2.

Subcellular Localization of MdSINA2
The recombinant plasmid 35S::GFP-MdSINA2 and the control vector 35S::GFP were injected into the epidermal cells of N. benthamiana leaves via Agrobacterium-mediated transient expression. The fluorescent signals were observed by a laser confocal microscope (Zeiss LSM 510 META, Jena, Germany).

Protein Extraction and Western-Blotting
The total protein of 2-week-old MdSINA2-GFP transgenic calli was isolated as described . After the western blot assay, anti-GFP (Sigma-Aldrich) was used to detect protein abundance. Actin served as a protein loading control.

ABA-Sensitive Assays
Transgenic and WT calli were cultivated on MS medium containing 100 mM ABA. Fresh weight and superoxide anions content were measured after 18 days. The superoxide anions generation rates were determined by spectrophotometry using their corresponding assay kits (Keming, Suzhou, China).
Homozygous seeds of WT and transgenic Arabidopsis were plated on MS medium with or without either 1 or 3mM ABA and grown at 22°C under an 8/16 h dark/light photoperiod. Photographs were taken after 8 days.

Statistical Analysis
All experiments were performed in triplicate. Error bars showed the standard deviation of three biological replicates. GRAPHPAD PRISM 6.01 software was used to conduct a oneway ANOVA to detect significant differences (*, P < 0.05).

Genome-Wide Identification of Apple SINA Genes
Five SINATs were previously identified in the A. thaliana genome, including AtSINA1 (SINAT1), AtSINA2 (SINAT2), AtSINA3 (SINAT3), AtSINA4 (SINAT4), and AtSINA5 (SINAT5 Ler ) (Xie et al., 2002;Wang et al., 2008). Five SINAT protein sequences were used to blastp in the apple genome database. After software screening and manual detection based on the conserved domains, 11 MdSINAs were obtained ( Table 1). The MdSINA genes were named based on their chromosomal locations (MdSINA1-11). We found that the 11 MdSINA genes were unevenly distributed on eight chromosomes in the apple genome. Gene distribution density was highest on chromosome 7 (three genes), followed by chromosome 1 (two genes). The range of lengths of the MdSINA proteins was small, from 287 (MdSINA10/11) to 334 (MdSINA8) amino acids, and the molecular weights were predicted to be between 32,548-37,952 Da. The theoretical isoelectric point ranged from 6.49 (MdSINA7) to 8.55 (MdSINA5). In addition, the subcellular locations of the 11 MdSINA genes were predicted, suggesting that the MdSINA proteins were localized in either the cytoplasm or nucleus ( Table 1).

Phylogenetic and Synteny Analyses of Apple SINA Genes
We performed a phylogenetic analysis of SINA proteins in apple, Arabidopsis and rice to gain additional insight into the evolutionary relationships of apple SINA proteins (Figure 1). The 22 SINA proteins used in the analysis were divided into two groups in these three species. Groups I contained five MdSINA proteins (MdSINA1, MdSINA2, MdSINA5, MdSINA7, and MdSINA8), and Group II contained six proteins (MdSINA3, MdSINA4, MdSINA6, MdSINA9, MdSINA10, and MdSINA11). We noticed that all genes except MdSINA5 appeared in pairs based on the phylogenetic tree of the apple SINA family, implying possible gene duplication have occurred during the evolution of the SINA gene family. Subsequently, we used MCScanX to detect gene duplication events in MdSINAs in the apple genome. Five pairs, MdSINA1/7, MdSINA2/8, MdSINA3/6, MdSINA4/9, and MdSINA10/11, were localized in duplicated genomic regions (Figure 2A). And those genes which are flanked the MdSINA genes are also present in the syntenic blocks (data not shown), indicating that segmental duplication events occurred. Moreover, 11 apple SINA genes distributed across eight chromosomes, including 1, 2, 3, 6, 7, 11, 12, 14 ( Figure 2B), which have highlighted strong collinearity between large segments of chromosomes 3 and 11, and between shorter segments of chromosomes 1 and 7, 2 and 7, 12 and 14 (Velasco et al., 2010).

Gene Structure and Multiple Sequences Alignment Analyses of MdSINA Genes
The exon-intron structure of the MdSINA genes was examined using the online server GSDS to provide more insight into the evolution of these genes ( Figures 3A, B). Gene structures were highly similar within each of the two group in terms of exon number. Furthermore, a similar exon-intron structure was observed within the same group of MdSINA. For example, Group I genes contained two introns and three exons except for MdSINA1. However, Group II genes possessed three introns and three exons. Although the protein sequence similarity between MdSINA1 and MdSINA7 was over 95% (data not shown), the structures of their associated gene differed. These differences implied that these two genes possessed divergent functions during their evolution. The conserved motifs of apple SINA proteins were confirmed by MEME, and 15 distinct motifs were identified ( Figure 3C). Motifs 1-6 were observed in all MdSINA proteins and contained conserved RING and SINA domains. Motif 4 was identified as the RING domain, while motifs 1, 2, 3, 5, and 6 were components of the SINA domain. Most members of Group I had motif 10 at the C-terminus, while all Group II members lacked this motif. Five gene pairs had the similar, unique motif structure. For example, motifs 11 and13 were the unique motifs in MdSINA1/7 and MdSINA2/8, respectively. The conservation of SINA genes during evolution was substantiated by the distribution of motifs. The uniqueness of the motif distribution in different gene pairs reflects the conservation and differentiation of SINA genes function during evolution. Moreover, the consensus sequences of motifs is stacked, suggested that all MdSINA genes shared highly conserved or completely conserved sites ( Figure 3D).
We performed multiple sequences alignment to detect the conserved domain of MdSINA proteins (Figure 4). In the BLASTP analysis, the highest percentage of amino acid sequence identity obtained was between pairs of MdSINA genes, and these genes were also closely grouped in the phylogenetic tree. Our results suggest that an N-terminal C3HC4-type (Cys-X2-Cys-X-Cys-X-His-X-Cys-X2-Cys-X-Cys-X2-Cys) RING domain and a Cterminal SINA domain occurred in all members. The RING domain is generally composed of eight conserved cysteine (Cys) and histidine (His) residues, and these eight conserved amino acid residues can bind two zinc ions to form a zinc-finger structure.

Expression Patterns of MdSINAs in Different Tissues
SINAs, from both animals and plants, have been found to play vital roles in different physiological processes. In order to further investigate the roles of MdSINAs in apple growth and development, testing the expression level of MdSINAs in different tissues is necessary. We performed qRT-PCR analysis to examine the expression levels of MdSINAs in a variety of tissues, including root, stem, leaf, flower, and fruit. The five gene pairs exhibited A B  similar spatial expression patterns ( Figure 5). MdSINA1/7, MdSINA2/8, MdSINA4/9, and MdSINA10/11 showed relatively high expression levels in leaves and flowers, suggesting that these genes may be involved in leaf and flower development. In addition, MdSINA3/6 showed similar expression patterns with higher expression levels in stems, leaves, flowers, and fruit, indicating that they may play an important role in the growth and development of apples. MdSINA5 was only highly expressed in leaves. Overall the distinct spatial expression patterns imply that these genes play diverse roles in apple growth and development.

Expression Patterns of MdSINAs in Response to ABA and Nutrient Starvation
A previous study has shown that SINAT2 promotes drought tolerance in an ABA-dependent manner in Arabidopsis (Bao et al., 2014). To explore the potential response to ABA, the expression levels of MdSINAs were detected under ABA (100 mM) treatment. The expression levels of MdSINAs were highest at 3 h (approximately3-6-fold), and then returned to normal levels ( Figure 6A). In addition, some of these genes had a higher expression level at 24 h, implying that MdSINAs were involved in ABA response and may act as important regulators of drought resistance as has been described previously.
Recently, studies have shown that SINAT1/2/6 are involved in starvation-induced autophagy in Arabidopsis Qi et al., 2017;Qi et al., 2020). To further characterize the potential roles of MdSINA genes in nutrient starvation, the expression levels of MdSINAs were examined under carbon-and nitrogen-starvation conditions. Under carbon starvation, MdSINA4/9 were slightly upregulated, and MdSINA5 was significantly up-regulated ( Figure  6B). Under nitrate-starvation conditions, the expression levels of MdSINA3/5/6 changed little; however, the expression of other MdSINAs was markedly up-regulated ( Figure 6C). Thus, MdSINAs may be involved in the process of autophagy and function in different stress responses.

Interactions Among MdSINA Proteins
Previous studies have suggested that SINA proteins can form both homodimers and heterodimers to regulate their own stability via the proteasome degradation pathway and perform different biological functions (Hu and Fearon, 1999;Xie et al., 2002;Depaux et al., 2006;Wang et al., 2018). The homo-and hetero-dimerization of MdSINA proteins was confirmed by Y2H experiments. First, the full-length cDNA of each MdSINA gene was cloned into the bait vector pGBT9 and the prey vector pGAD424. The fusion plasmids were then cotransformed into yeast cells (the "Y2H Gold" strain) grown on selection medium. All MdSINA proteins except MdSINA1 and MdSINA7 can interact with each other to form homodimers and heterodimers, implying that MdSINA1/7 might function as a single E3 ligase (Figure 7). Specifically, MdSINA4/9-BD show weak autonomous activation. However, MdSINA4/9-AD can interact with other MdSINAs proteins except for MdSINA1 and MdSINA7. Thus MdSINAs can form homodimers and heterodimers.

Subcellular Localization of MdSINA2 Proteins
The predicted subcellular localization of MdSINA proteins is shown in Table 1, and MdSINA2 was selected for verification.
Considering that ubiquitination can occur in the cytoplasm and nucleus to regulate the stability of cytoplasmic and nuclear proteins, the localization of ubiquitin ligase must localize in the cytoplasm, nucleus, or both (Xie et al., 2002;Heck et al., 2010;Yoo et al., 2013;Wang et al., 2018). To determine the subcellular localization of MdSINA2, we constructed the fusion plasmid 35S::MdSINA2-GFP and injected it into the epidermal cells of N. benthamiana leaves via Agrobacterium-mediated transient expression. After 48 h, we observed the GFP signal under confocal laser scanning microscopy. The fluorescent GFP signal was distributed throughout the cells or only in the nucleus of the epidermal cells of N. benthamiana leaves, which were injected with either 35S::GFP or 35S::GFP-MdSINA2, respectively ( Figure S1). Thus MdSINA2 was localized in the nucleus, which was consistent with our initial prediction.

Light Signaling Mediates MdSINA2 Stability in a RING Domain-Dependent Manner
To gain further insight into the roles of MdSINA2, we cloned the full-length cDNA of MdSINA2 into plant expression vector with GFP label and transformed it into apple calli ( Figure S2). SINA proteins are E3 ubiquitin ligases that have important functions in plants and animals, and understanding the factors that affect the stability of MdSINA proteins is important. In A. thaliana, levels of SINA proteins increase under light and decrease in darkness. However, SINAT5 protein without a RING domain, which represents the Columbia ecotype (Col-0), is insensitive to dark and light treatment . Therefore, we determined whether the effect of light on the stability of the SINA protein in apple was the same as that observed in Arabidopsis. Two-weekold MdSINA2 transgenic calli were grown on MS medium for 24 h under continuous light or dark conditions. We then transferred them to the dark or light for 1, 3, 6, 12, or 24 h. The protein level of MdSINA2 dramatically increased under light but decreased under dark conditions ( Figure 8A). Disruption of the ubiquitination site of the SINA protein is known to result in the loss of self-ubiquitination activity (Xie et al., 2002;Wang et al., 2018). To test whether the degradation of MdSINA2 protein depends on the RING domain, we obtained a mutant protein, Cys74!Ser (C74S), in which the RING domain was disrupted following the methods of a previous study (Xie et al., 2002). Although the MdSINA2 protein level in MdSINA2-GFP calli increased and decreased under light and darkness, respectively, there were no other noticeable differences between MdSINA2 C74S -GFP calli ( Figure 8B). Thus, the stability of lightmediated MdSINA2 protein is dependent on its RING domain.

MdSINA2 Improves Sensitivity to ABA Treatment in Transgenic Apple Calli and Seed Germination of Arabidopsis
MdSINA2 transcripts were significantly induced by ABA treatment. To evaluate the role of MdSINA2 in the ABA response, 2-week-old MdSINA2 transgenic calli and WT were grown on MS medium with 100 µM ABA. MdSINA2-OX calli grew much slower than WT (Figures 9A, B). Moreover, MdSINA2-OX calli had lower fresh weight and accumulated more superoxide anions than WT ( Figures 9C, D).
In addition, MdSINA2 was successfully transformed into wildtype Arabidopsis ( Figure S2). Three independent lines were selected for seed germination experiments. The transgenic Arabidopsis seeds were sown on MS medium supplemented with different concentrations of ABA. The germination rate was negatively correlated with ABA concentration (Figure 10). There was no significant difference in germination rates between WT and transgenic seeds on MS medium. However, the germination rate of transgenic seeds was significantly lower on MS medium supplemented with 1 mM or 3 mM ABA compared with WT. Moreover, the transgenic Arabidopsis seeds hardly germinated on medium with 3 mM ABA even after 8 days. Therefore, overexpression of MdSINA2 results in an ABA-hypersensitive phenotype.

DISCUSSION
The function of SINA (Seven in absentia) homologous proteins have been extensively studied in animals, especially in the medical field, and these studies have revealed that SINA proteins are associated FIGURE 7 | Y2H assay to test pair-wise interactions among MdSINA proteins. The yeast vector pGAD424 and pGBT9 were used as negative controls. The arrow indicates that BD-SINA4/9 showed weak autonomous activation.  with resistance to various diseases such as cancer (Liu et al., 2008;Cai et al., 2015;Siswanto et al., 2018). Recently, an increasing number of studies investigating the function of SINA homologs have been conducted in plants . SINA E3 ligase regulates plant growth and the response to different stresses via degradation of multiple targeted proteins. In this study, 11 MdSINA genes were identified in apple by genome-wide analysis ( Table 1). MdSINA genes were found to be unevenly distributed across eight chromosomes through chromosomal localization. Five SINA homologous proteins are known in A. thaliana (SINAT1-5), 10 homologous proteins in Populus trichocarpa, six homologous proteins in O. sativa, at least six homologous proteins in Zea mays, and six homologous proteins in both Medicago truncatula and Solanum lycopersicum (Den Herder et al., 2008;Wang et al., 2008;Wang et al., 2018). However, the 11 SINA genes in apple may reflect an evolutionary expansion of the SINA family. Multiple sequence alignments indicated that all MdSINA proteins contained two conserved domains: a RING domain and a SINA domain. Phylogenetic analysis was conducted to characterize the evolutionary relationships of SINA genes in apple, Arabidopsis, and rice ( Figure 1). The results showed that SINA genes could be divided into two groups (Groups I and II) that differed in gene structure and motif composition between Group I and II. For instance, genes in the same group, such as MdSINA2 and MdSINA8 or MdSINA4 and MdSINA9, exhibited similar exon-intron structures and motif compositions. The main mechanism driving the evolution of SINA genes appears to be gene duplication followed by divergence (Ohno et al., 1968;Flagel and Wendel, 2009). A previous study has shown multiple segmental duplication events in Arabidopsis, poplar, and rice . Our results suggest that segmental duplication events have also occurred in apple. Five gene pairs showing similar gene structures and protein sequences imply that these proteins show some degree of functional redundancy, such as MdSINA2/8. A more comprehensive characterization of the MdSINA gene family is necessary given the high functional diversity of SINA homologs that has been documented in different species. Therefore, qRT-PCR analyses of the expression patterns of MdSINA genes were conducted. These 11 MdSINA genes were differentially regulated at the transcriptional level in different apple tissues. The transcript levels of most MdSINA family members were significantly higher in leaves and lower in roots ( Figure 5), which is in contrast with the transcriptional patterns of these genes documented in tomato , implying that they might play an important role in leaf physiology in apple. Given that ABA signaling plays a vital role in regulating the response to different types of stress (Cutler et al., 2010), such as drought, cold, and salt, we analyzed the expression profiles of MdSINA genes under ABA treatment. The expression levels of MdSINA genes were altered under different degrees of ABA induction, suggesting that MdSINAs might participate in the stress response as a functional component of the ABA signaling pathway.
Furthermore, increasing evidence has demonstrated that autophagy can be induced by nutrient starvation and transport needless proteins or damaged organelles into vacuoles for degradation and circulation (Lee et al., 2010;Avin-Wittenberg et al., 2015;Chen et al., 2017;Islam et al., 2019). A large number of autophagy-related proteins (ATGs) have been documented in plants; these ATGs play an important role in regulating core autophagy mechanisms (Soto-Burgos et al., 2018;Zhuang et al., 2018). To date, a few studies have shown that SINA homologs are involved in the autophagy process in Arabidopsis, and SINAT-RNAi plants exhibit a starvation hypersensitivity phenotype. SINAT1/2 were negative regulators in autophagy process under nutrient starvation condition by mediating the ubiquitylation and degradation of ATG6/13. Here, we observed different expression patterns of MdSINAs under carbon-and nitrogenstarvation treatments ( Figures 6B, C), implying that MdSINAs may play different functions in different periods and under different starvation treatments.
The stability and activity of SINA ligases in vivo depend on their dimerization (Hu and Fearon, 1999;Polekhina et al., 2002;Depaux et al., 2006). Hence, we detected the dimerization of MdSINA proteins. The SINA proteins in apple are surprisingly different from those in tomato and M. truncatula (Den Herder et al., 2008;Wang et al., 2018), all of which can interact with each other. We found that MdSINA1/7 neither formed homodimers nor heterodimers with any of the other MdSINA proteins. Thus, MdSINA proteins may function via the formation of homodimers and heterodimers or in another manner.
SINA ubiquitin ligases have a wide range of functions in plants and animals. They can regulate the stability of many important proteins in plants and animals. Previous studies have shown that the ABA response factor can regulate the transcription level of SINA2, but it remains unclear how the protein levels of SINATs are regulated (Bao et al., 2014). The identification of key factors regulating the stability of SINA proteins could enhance our understanding of the function of SINA proteins. We found that the MdSINA2 degraded in the dark, and it became insensitive to light after its RING domain had been destroyed. Based on this result and those of previous studies Qi et al., 2020), we speculate that the degradation of MdSINA2 in the dark may stem from the lack of energy available under dark conditions. ABA plays an important biological role in stomatal closure, stress resistance, seed germination, and senescence (Finkelstein et al., 2002;Assmann, 2003;Zhu, 2016;Sussmilch et al., 2017). The observation that MdSINA2 is significantly induced by ABA ( Figure 5) and is highly expressed in leaves suggests that MdSINA2 may be a functional component of ABA signaling and may be involved in the tolerance of different types of stress by regulating leaf development. Indeed, when we compared the germination rate of wild type (Col-0) and 35S::MdSINA2 on MS media supplemented with different concentrations of ABA, significant differences were observed in seed germination ( Figure 10). Our results differed from those documented for SINA2 in transgenic Arabidopsis (Bao et al., 2014) under different ABA concentration treatments, where no significant differences were observed among wild-type and experimental groups. Although the specific mechanism of MdSINA2 in the responses to ABA and different types of stress remains unclear, our results suggest that altering MdSINA2 expression can significantly improve sensitivity to ABA in transgenic calli and Arabidopsis. Additional screening of possible targets with which MdSINA2 could interact in ABA signaling pathways will help elucidate the role that it plays in ABA signaling.
In conclusion, 11 MdSINA genes were identified in the apple genome. Bioinformatics analyses and analyses of gene expression were conducted to characterize their roles of these genes in apple growth and development. The functional characterization suggested that apple SINA members may function by responding to ABA signaling. These findings provided new insight into the ability of plants to resist abiotic stress.

DATA AVAILABILITY STATEMENT
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 in the article/ Supplementary Material.

AUTHOR CONTRIBUTIONS
H-LL and Y-JH conceived the experiments. H-LL, XW, X-LJ, and Z-WQ performed the research. H-LL, XW, and X-LJ analyzed the data. H-LL and XW wrote the manuscript. X-LJ and Y-JH revised the manuscript. All authors contributed to the article and approved the submitted version.