Identification of candidate genes associated with less-photosensitive anthocyanin phenotype using an EMS mutant (pind) in eggplant (Solanum melongena L.)

Eggplant (Solanum melongena L.) is a highly nutritious and economically important vegetable crop. However, the fruit peel of eggplant often shows poor coloration owing to low-light intensity during cultivation, especially in the winter. The less-photosensitive varieties produce anthocyanin in low light or even dark conditions, making them valuable breeding materials. Nevertheless, genes responsible for anthocyanin biosynthesis in less-photosensitive eggplant varieties are not characterized. In this study, an EMS mutant, named purple in the dark (pind), was used to identify the key genes responsible for less-photosensitive coloration. Under natural conditions, the peel color and anthocyanin content in pind fruits were similar to that of wildtype ‘14-345’. The bagged pind fruits were light purple, whereas those of ‘14-345’ were white; and the anthocyanin content in the pind fruit peel was significantly higher than that in ‘14-345’. Genetic analysis revealed that the less-photosensitive trait was controlled by a single dominant gene. The candidate gene was mapped on chromosome 10 in the region 7.72 Mb to 11.71 Mb. Thirty-five differentially expressed genes, including 12 structural genes, such as CHS, CHI, F3H, DFR, ANS, and UFGT, and three transcription factors MYB113, GL3, and TTG2, were identified in pind using RNA-seq. Four candidate genes EGP21875 (myb domain protein 113), EGP21950 (unknown protein), EGP21953 (CAAX amino-terminal protease family protein), and EGP21961 (CAAX amino-terminal protease family protein) were identified as putative genes associated with less-photosensitive anthocyanin biosynthesis in pind. These findings may clarify the molecular mechanisms underlying less-photosensitive anthocyanin biosynthesis in eggplant.


Introduction
Anthocyanins are natural flavonoid pigments, which are responsible for blue, purple and red colors in plant tissues, such as flowers, fruits, seed coats, leaves, and stems.In addition to their role in coloring, anthocyanins attract pollinators and seed disperser and protect plants from biotic and abiotic stresses (Fan et al., 2016;Sivankalyani et al., 2016;Sun et al., 2018;Naing and Kim, 2021).Anthocyanins also show beneficial effects on human health, including antioxidation, antimutagenicity, cardiovascular disease prevention, liver protection, and the inhibition of tumor cell metastasis (Pojer et al., 2013;Liu et al., 2016).
Eggplant (Solanum melongena L.) is a globally cultivated vegetable crop with high economic benefits (Saini and Kaushik, 2019;Oladosu et al., 2021).The color of eggplant fruit is an important quality trait, with purple varieties exhibiting high levels of anthocyanins in the fruit peel (Nayanathara et al., 2016;Nino-Medina et al., 2017).However, low light during cultivation often leads to poor coloration, which reduces their visual quality and commercial value (Li et al., 2018).Therefore, eggplant genotypes with less-and non-photosensitive anthocyanin biosynthesis are valuable for the breeding of low-light tolerant varieties.The purple peel under the calyx is a good indicator of less-and nonphotosensitive coloration (Xiang et al., 2015;He et al., 2019).Although some quantitative trait loci (QTL) that are responsible for purple peel under calyx have been identified in eggplant (Chen, 2015;Toppino et al., 2016;Mangino et al., 2022) and markers have been developed (Xiang et al., 2015;Zhang et al., 2021b), the underlying causal genes conferring less-photosensitive biosynthesis of anthocyanin in eggplant remain unknown.
In this study, a less-photosensitive anthocyanin biosynthesis mutant purple in the dark (pind) in eggplant was discovered.Under bagging conditions, the pind fruit color was light purple, whereas the wild type was white, indicating that anthocyanin biosynthesis in the pind mutant was less dependent on light.Phenotypic and genetic analysis, gene mapping, and transcriptomic analysis were performed to identify the causative genes conferring lessphotosensitive anthocyanin biosynthesis in the fruit peel of pind.Taken together, our results provided novel insight into lessphotosensitive anthocyanin accumulation in eggplant fruit peel.

Materials and population construction
The mutant was identified from M 1 line in an ethyl methanesulfonate (EMS) mutagenized population of eggplant variety '14-345', which is white under the calyx.After two generations of self-fertilization, we obtained a homozygous mutant and named it "pind" (Figure 1A).The pind mutant and wildtype '14-345' were crossed to produce F 1 plants.Subsequently, F 2 and BC 1 populations were generated from F 1 plants via self-and backcross with '14-345' or pind, respectively.They were used as segregating populations for genetic analysis and mapping the candidate gene.In addition, pind and a cultivated eggplant '18-305' (Figure S1), which is white under the calyx, were crossed to obtain another F 2 population ('305F 2 ').The '305F 2 ' population was also used for gene mapping.All plants were grown in a plastic greenhouse in the experimental fields of Hebei Agricultural University, Baoding, China.During the cultivation, the temperature inside the greenhouse was ranged from 18 to 32°C, the air humidity ranged from 50 to 70%, and the daytime light intensity ranged from 40k to 55k lx.

Phenotypic investigation and measurement of anthocyanin content and color index
The fruit of the plants were bagged on the 5 th day after flowering and the fruit color and anthocyanin content were investigated on the 14 th day under bagging condition.Color differences were assessed by the color index of red grapes (CIRG).A CR-400 colorimeter (Konica Minolta, Chroma Meter, Osaka, Japan) was used to measure the values of L*, a*, and b*.CIRG was calculated with the equation CIRG = [(180-H)/(L* + C*)], H=arctan (b*/a*), C=(a* 2 + b* 2 ) 0.5 (Carreno et al., 1995;Zhang et al., 2008).
The total anthocyanin content in the peel was quantified using the pH differential spectroscopic method (Cheng and Patrick, 1991).
One-way analysis of variance (ANOVA) was conducted on the CIRG value and anthocyanin content, and significant differences between groups were assessed by Duncan's multiple range tests (p < 0.05) using SPSS 16.0 Statistics (SPSS Inc., Chicago, IL, USA).

Whole genome sequence of bulked DNA
In the F 2 population under bagging conditions, 30 plants with purple fruit and 30 plants with white fruits were selected to construct the mixed pools 'P' and 'G', respectively.Genome DNA was extracted using the CTAB method.
Bulk segregant analysis (BSA)-based sequencing and MutMap analysis were used to map the candidate genes.The qualified DNA was randomly broken into fragments with a length of 350bp, and libraries were built using a TruSeq Library Construction Kit, followed by Illumina PE150 sequencing.BWA software was used to map the clean data to the eggplant reference genome (Li et al., 2021).The UnifiedGenotyper module of GATK3.8 software (Mckenna et al., 2010) was used to detect SNPs, and VariantFiltration was used to filter SNP detection.The SNP-index value with '14-345' as the reference was calculated to identify the key genes.The candidate intervals were determined with the threshold value: SNP-index (P) > 0.67, SNP-index (G) < 0.1, and D(SNP-index) > 0.6.

Obtaining recombinants
To narrow down the candidate region, the genotype of individuals in the F 2 and '305F 2 ' populations was detected using Kompetitive allele specific PCR (KASP) technology to screen recombinants.The primers used for genotype verification of the individual plants are listed in Table S1.

RNA extraction, library construction, and RNA-sequencing
Total RNA was extracted using an Eastep Super Kit (Shanghai Promega, Shanghai, China).The extracted RNA was treated with DNase to remove the genome DNA and the integrity and quantity were examined using 1% agarose gels and a Nanodrop 2000c Spectrophotometer (Thermo Nanodrop Technologies, Wilmington, DE, USA).A total of 1 mg RNA per sample was used to construct RNA-seq libraries.After qualification, mRNA was isolated from total RNA using oligo-magnetic beads.The mRNA was interrupted, reverse transcribed into cDNA, and then the cDNA was purified with an AMpure XP system (Beckman Coulter, Beverly, MA, USA).Sequencing libraries were generated using NEBNext ® Ultra ™ RNA Library Prep Kit for Illumina ® (NEB, Boston, MA, USA).Twentyfour libraries were sequenced on the Illumina Hiseq platform (Illumina, San Diego, CA, USA).The library construction and sequencing were completed by Novogene (Beijing, China).

RNA sequencing data analysis
Raw data (raw reads) in fastq format were filtered to obtain clean reads, and the Q20, Q30, and GC content of the clean data Effect of bagging on the peel color of '14-345' (wild type) and pind (mutant) fruits.were calculated.The clean data were aligned to the eggplant reference genome (Li et al., 2021) using Hisat2 v2.0.5.software (Mortazavi et al., 2008).The featureCounts v.5.0-p3 tool in the subread software was used to count the reads mapped to each gene (Liao et al., 2014) and the expected number of fragments per kilobase of transcript sequence per millions base pairs sequenced (FPKM) value was calculated based on the read count and the length of the gene.The differential expression analysis was performed using the DESeq2 R package (1.20.0), and the genes with the criteria |log 2 (fold change)| > 1 and padj≤ 0.05 were considered DEGs (Love et al., 2014).The relative expression level log 2 (ratios) of all DEGs was clustered by the K-means method.Plant TFs were predicted by hmmscan based on iTAK software (Perez-Rodriguez et al., 2010).Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment was analyzed with clusterProfile software and a padj of less than 0.05 was considered significantly enriched (Yu et al., 2012).

Quantitative real-time PCR analysis
Fourteen genes were chosen for the validation of RNA-seq using qRT-PCR.Primer Premier 5.0 software was used to design the primers, which are listed in Table S2.A total of 1 mg RNA per sample was reverse transcribed using a PrimeScript ™ reagent Kit with gDNA Eraser (TaKaRa, Beijing, China) in 20 mL of reaction mixture.The qRT-PCR was performed on a LightCycler ® 96 instrument (Roche, Basel, Switzerland), using THUNDERBIRD SYBR qPCR Mix (TOYOBO, Shanghai, China) with the following reactions: 95°C 2 min; 95°C 30s, 60°C 10s, and 68°C 10s for 40 cycles.The PCR products were quantified by 2 −△△CT method (Livak and Schmittgen, 2001) with normalization to the expression level of SmGAPDH (EGP1067575).Significant differences between groups were assessed by Student's t-test (p < 0.05) using SPSS 16.0 Statistics (SPSS Inc., Chicago, IL, USA).

Phenotypic characterization of pind
The pind mutant was obtained from eggplant '14-345' mutagenized population with EMS.The coloration of the '14-345' fruit peel is photosensitive with white-colored fruit under the calyx.Interestingly, a light purple pigmentation was observed under the calyx of pind, indicating that the coloration of pind was less photosensitive.To further characterize the less-photosensitive coloration in pind, the fruit was bagged to mimic dark conditions.Compared to the white fruit of '14-345', the pind fruit was light purple (Figure 1A).Under natural conditions, the color index, CIRG value, of '14-345' was similar to that of pind, whereas the CIRG value of the pind mutant was significantly higher than that of '14-345' under bagging conditions (Figure 1B).
The anthocyanin contents in the peel of '14-345' and pind were also quantified.The anthocyanin content in the peel of '14-345' and pind was comparable under natural conditions.However, the level of anthocyanin in the bagged eggplants decreased significantly in both '14-345' and pind compared with natural conditions.Anthocyanin content in the peel of bagged pind (21.1 mg/100 g FW) was significantly higher than that in the bagged '14-345' (4.5 mg/100 g FW) (Figure 1C).These results indicated that light was required for the coloration of fruit peel in '14-345', but less so for pind.

Genetic analysis of pind
To investigate the inheritance of pind, segregating populations were constructed.All the F 1 plants from the crosses of pind and '14-345' had purple fruits under bagging conditions, which was phenocopied pind.Out of the F 2 plants, 116 had purple bagged fruits, while 37 had white bagged fruits.A chi-squared test revealed that this segregation pattern agreed with the 3:1 Mendelian segregation ratio (P=0.82>0.05,c 2 = 0.05<c 2 0.05,1 = 3.84).In the W-BC 1 population (the backcross progeny of F 1 with '14-345'), 30 plants displayed white wild type fruit, whereas the remaining 23 plants produced purple fruit under bagging conditions.The segregation pattern fitted the ratio 1:1 (P=0.34>0.05,c 2 = 0.93<c 2 0.05,1 = 3.84).All M-BC 1 plants, the backcross progeny of F 1 with pind, exhibited purple fruits under bagging conditions (Table 1).These results suggested that the mutant gene was a single dominant allele.

Identification of the candidate genomic region
To identify the genomic region responsible for lessphotosensitive anthocyanin biosynthesis in the fruit peel of pind, BSA-based genome resequencing and MutMap analysis were performed.Quality control resulted in 96,777,013 and 90,196,644 high-quality reads and 98.67% and 98.64% of them were mapped on the eggplant genome (Tables S3, S4), from which 1421190 SNPs were detected in total.Based on the D(SNP-index), a single significant interval in a 12.2 Mb region (from 2.6 to 14.8 Mb) on chromosome 10 was identified as a candidate region of the mutation (Figure 2A).There were 2,058 SNPs, including 22 non-synonymous mutation sites.
To identify the causal mutation in this region, several molecular markers were designed to uniformly cover the preliminary mapping interval.Fifteen recombinant plants were identified and the candidate gene was mapped to an 8.2 Mb region (from 6.6 Mb to 14.8 Mb) between markers M5 and M10 (Figure 2B).
To further map the SNP related to less-photosensitive anthocyanin biosynthesis, another F 2 population, '305F 2 ' (pind × '18-305'), was used to obtain recombinants.The '305F 2 ' population consisted of 408 plants, of which 305 had purple fruit and 103 had white fruit under bagging conditions (Table S5).Thirty-four recombinants were identified and the candidate gene was finally mapped to a 4.0 Mb region between the markers M12 and M9 (Figure 2C), from 7.7 Mb~11.7 Mb on chromosome 10.A total of 224 SNPs that were associated with 75 genes (Table S6), were detected in the candidate interval.

Transcriptome comparison between pind and '14-345' in response to dark treatment
To identify the candidate gene responsible for lessphotosensitive anthocyanin biosynthesis in pind, RNA-seq analysis was performed on the fruits of pind (MT) and '14-345' (WT) under bagging (D) and natural (L) conditions for two stages, 2days (2d) and 5 days (5d) after bagging.For both stages, the fruits of the '14-345' bagged were white, whereas the fruits of pind were light purple under bagged conditions.The fruit peel color of '14-345' and pind at the two stages was almost the same under natural conditions (Figure 3A).

Analysis of DEGs for lessphotosensitive anthocyanin biosynthesis in pind
Due to the difference in fruit color between '14-345' and pind under bagged conditions (MT_D vs. WT_D) and the differences in the color of '14-345' fruit between bagged and natural conditions (WT_D vs. WT_L), we screened for common DEGs in these combinations.The Venn diagram clearly illustrated the relationships of DEGs between different comparisons, and 238 Overview of DEGs identified by RNA-seq analysis in the peel of '14-345' and pind fruits under bagging conditions at two developmental stages.DEGs (Figure 4A; Table S7) and 145 DEGs (Figure 4B; Table S8) were commonly shared by 'MT_D vs. WT_D' and 'WT_D vs. WT_L' on 2d and 5d, respectively.Moreover, 35 DEGs were commonly shared by these four combinations (Figure 4C; Table S9), which may participate in anthocyanin biosynthesis in pind under dark conditions.Therefore, these 35 DEGs, including 12 structural genes and three TFs, were used for further analysis.Under natural conditions, there was no significant difference in the expression of structural genes between '14-345' and pind fruit peel.Bagging inhibited the transcripts of most of the structural genes, including 4CL, CHS, CHI, F3H, F3′5′H, DFR, and ANS in '14-345' while not in pind, as the transcriptional level of structural genes were extremely higher in pind than in '14-345' (Figure 4D).These data suggest that '14-345' and pind have different responses to bagging in terms of anthocyanin biosynthesis.The highly expressed structural genes in pind under dark conditions showed pind was less photosensitive.
TFs can regulate the expression of structural genes by directly binding to cis-regulatory elements in the promoter of the genes, which play important roles in plant growth and development (Chen et al., 2019;Jiang et al., 2022;Liu et al., 2023).In this study, MYB TF SmMYB113 (EGP21875), WRKY TF SmTTG2 (EGP21679), and bHLH TF SmGL3 (EGP16978) were identified from the 35 DEGs as the putative main regulators of the anthocyanin biosynthesis in a less-photosensitive manner in the fruit peel of pind.Under natural conditions, the expression of SmMYB113, SmTTG2, and SmGL3 was comparable between '14-345' and pind.Under bagging conditions, SmMYB113, SmTTG2, and SmGL3 was downregulated both in '14-345' and pind compared to natural conditions (Figure 4E).Notably, the expression levels of MYB113, TTG2, and GL3 were significantly increased in the peel of pind compared to '14-345' under bagged conditions (Figure 4E).
To confirm the reliability of the RNA-seq data, six structural genes (C4H, CHI, F3H, DFR, ANS, and 5GT), five TFs (MYB113, GL2, GL3, TT8, and TTG2), and three genes involved in light signaling (COP1, UVR8, and CRY3) were selected to analyze expression profiles by qRT-PCR.The transcript abundances of these genes determined using qRT-PCR agreed with those determined from transcriptome sequencing (Figure 5), indicating the reliability of the RNA-seq results.

Candidate gene analysis
Based on the gene mapping analysis, there were 75 genes in the candidate region (Table S6), of which only nine (Figure 6) were expressed in the eggplant peel according to the transcriptome analysis.Twenty-seven SNPs were detected to be associated with these nine genes, and only one non-synonymous SNP (within EGP21953) was identified.The gene EGP21953 (Table 2) had an amino acid residue substitution at aa-262 (Ala-to-Tyr mutation).According to the annotated information, EGP21953 encodes a CAAX amino-terminal protease family protein.However, there was no significant difference in EGP21953 expression between '14-345' and pind, neither in natural nor in bagged conditions.Among the other eight genes, whose related SNPs were detected in intergenic regions or introns, only EGP21875, EGP21950, and EGP21961, were differentially expressed between the peel of pind and '14-345' (Figure 6; Table 2).EGP21875 encodes an MYB TF, "SmMYB113", which has been shown to participate in the regulation of anthocyanin biosynthesis (Zhang et al., 2014;Jiang et al., 2016;He et al., 2019;Yang et al., 2022).The expression of SmMYB113 was inhibited in the fruit peel of '14-345' under bagged conditions.Interestingly, SmMYB113 showed significantly higher expression in the peel of pind than in '14-345' at both 2d and 5d stages, pinpointing the activation of SmMYB113 in pind in darkness.There was one SNP 22.9 kb upstream of the start codon and one SNP 13 kb downstream of the stop codon of SmMYB113, respectively.Because the intergenic SNPs play a potentially important role in phenotype variation in plants by altering gene expression (Schwartz et al., 2009;Ding et al., 2012;Li et al., 2012;Wang et al., 2015), these two SNPs in the intergenic region associated with SmMYB113 may affect its expression.EGP21950 encoded an unknown functional protein, and there was one SNP 859 bp upstream of the start codon of EGP21950.The expression of EGP21950 was significantly up-regulated in the fruit peel of pind in comparison to '14-345' at 2d under bagged conditions, but its expression level showed no significant difference between '14-345' and pind under natural conditions and at 5d under bagged conditions.In addition to EGP21953, EGP21961 also encoded a CAAX amino-terminal protease family protein, and its associated SNP was about 7 kb upstream of the initiation codon.EGP21961 was down-regulated significantly in the peel of pind compared to '14-345', in both natural and bagged conditions.

Gene identification related to lessphotosensitive anthocyanin biosynthesis in eggplant
Anthocyanin is an important flavonoid type pigment in flowers and fruits, with light being an essential regulator of anthocyanin accumulation (Takos et al., 2006;Meng and Liu, 2015;Bai et al., 2017;Kim et al., 2021;Ma et al., 2021a).Proper anthocyanin accumulation in fruits or flowers can be hindered by low-light conditions (caused by weather and climate, the cultivation facility, or plant morphology) (Hong et al., 2015;Cao et al., 2016;Zhang et al., 2021a;Guo et al., 2022).Therefore, understanding the mechanism of less-and non-photosensitive anthocyanin biosynthesis and creating the less-and non-photosensitive cultivars will provide valuable resources for the breeding of lowlight tolerant varieties In this study, we discovered a less-photosensitive anthocyanin biosynthesis mutant (pind) in eggplant, and a candidate gene that was physically mapped to 7.7 ~11.7 Mb on chromosome 10 (Figure 2C).This result was consistent with previous reports, showing that the candidate genes that may correspond to loci controlling fruit peel color under calyx in eggplants were mapped to chromosome 10 ( Zhang et al., 2021b;Mangino et al., 2022;Qiao et al., 2022).In addition to chromosome 10, the candidate genes controlling the fruit peel color under calyx in eggplants were also found on other chromosomes, including chromosome 1, 3, 4, and 11 (Mangino et al., 2022;Qiao et al., 2022).However, the above studies did not distinguish materials as less-photosensitive or nonphotosensitive materials.When comparing the phenotypic differences between the less-photosensitive or non-photosensitive Expression pattern analysis of the genes in the candidate region.The number in the box indicates the FPKM value.1A), whereas there was no significant difference for non-photosensitive materials (He et al., 2019;He et al., 2022).
Recent studies have found that the candidate gene related to non-photosensitive anthocyanin biosynthesis in eggplant is SmFTSH10 (He et al., 2022), which was physically close to, but outside the candidate interval 7.7Mb-11.7Mb(Figure 3C) for less-photosensitive anthocyanin biosynthesis in this study, indicating less-and non-photosensitive anthocyanin biosynthesis in eggplant may be controlled by different locus.

Candidate genes related to lessphotosensitive anthocyanin biosynthesis in pind
Through MutMap and transcriptome analysis, we obtained four candidate genes: EGP21875, EGP21950, EGP21953, and EGP21961 (Table 2).EGP21953 and EGP21961 are homologs of the gene AT1G14270.1 in Arabidopsis, which is a CAAX amino-terminal protease family member.However, at present, there is no literature or report indicating that the CAAX amino-terminal protease family proteins under investigation exert regulatory control over anthocyanin biosynthesis.EGP21950 encodes a protein with an unknown function, that contained a DUF616 (Protein of unknown function) domain (Figure S3).Additionally, proteins encoded by structural genes and regulatory genes didn't possess CAAX prenyl endopeptidase-like domains or UDF616 domains (Figure S3), indicating their low likelihood to bind with CAAX aminoterminal protease family proteins or DUF616 proteins.Therefore, we speculate that EGP21950, EGP21953, or EGP21961 had minimal potential in contributing to the genetic basis of less-photosensitivity.
The gene EGP21875 encoded an MYB TF SmMYB113, homologous to AtMYB113 in Arabidopsis.In this study, SmMYB113 was hardly expressed in the fruit peel of wild type '14-345' under bagged conditions, whereas its expression was considerably up-regulated in the pind mutant compared to '14-345'.It was reported that overexpression of SmMYB113 (also known as SmMYB1) resulted in the up-regulation of SmCHS, SmCHI, SmF3H, SmANS, and other genes, and a high level of anthocyanin accumulation (Zhang et al., 2014;Yang et al., 2022).Thus, anthocyanin-pigmentation in pind in the dark may be related to the mutations of SmMYB113.Collectively, we speculated that EGP21875 (SmMYB113) was the best candidate gene for the regulation of less-photosensitive anthocyanin biosynthesis in the fruit peel of pind.
In addition, the SNPs related to EGP21875 (SmMYB113) were more than 13kb from the coding region.Some studies have shown that a non-coding region far from the gene promoter may change the activity of neighboring genes.For example, the Teosinte Branched 1 (TB1) gene is mainly responsible for a major-effect QTL that controls morphological differences in plant architecture between maize (Zea mays subsp.mays) and its wild relative, teosinte (Z.mays subsp.mexicana and subsp.parviglumis) (Clark et al., 2004;Studer et al., 2011); the expression of tb1 was altered by a sequence >41kb upstream of tb1 (Clark et al., 2006).In addition, booster1 (b1) was a regulator gene responsible for the biosynthesis of flavonoid pigments in maize; a 6-kb region at about 100kb upstream of the transcription start site of gene b1 was found to be a hepta-repeat enhancer, which may increase the transcription initiation rate (Stam et al., 2002).The non-coding region located distally from the gene promoter could be considered as a putative cis-regulatory element that modulates gene expression through recruitment of transcription factors (Stadhouders et al., 2014;Visser et al., 2014;Roberts et al., 2016).Studies have shown that natural variation in cis-regulatory regions of genes played important roles in phenotypic variations by altering gene expression (Frary et al., 2000;Schwartz et al., 2009;Muños et al., 2011;Van Der Knaap et al., 2014;Chu et al., 2019).Taken together, we speculated that the position of the SNPs in the intergenic region of EGP21875 (SmMYB113) might coincide with a regulatory element that controls the SmMYB113 expression level.

The activation of MYB113 may be necessary for anthocyanin accumulation in pind peel under dark conditions
Under bagging conditions, '14-345' fruits exhibited white coloration while the pind mutants displayed purple coloration.Consequently, under bagging conditions, the structural genes were expressed at a minimal level in the peel of '14-345' but showed higher expression in pind (Figure 4D).It is well known that the expression of structural genes is directly regulated by transcription factors, among which MYB TFs, bHLH TFs, and WD40 are key modulators (Shan et al., 2019;Li et al., 2020).In this study, expression of SmMYB113, SmGL3 (bHLH) and SmTTG2 (WRKY) was significantly higher in pind peel than '14-345' in darkness (Figure 4E).Moreover, bHLH TF SmTT8 was inactive in '14-345' under bagged conditions, but strongly activated in pind on 5d (Table S8; Figure 5).Therefore, under bagging conditions, the inactive transcription factors, such as SmMYB113, SmGL3, SmWRY44 and SmTT8, cannot stimulate the expression of structural genes, which is the underlying reason for the inability of ' 14-345' to produce anthocyanins under bagging conditions (Figure 7A).
In Arabidopsis, AtMYB113 interacts with AtTT8 or AtGL3 and AtTTG1, forming the MBW complex, to regulate the expression of late biosynthesis genes (LBGs) (Ramsay and Glover, 2005).The MYB TF is the main determinant of the MBW complex, and MYBs control the expression of early biosynthesis genes (EBGs) and LBGs (Stracke et al., 2007;Gonzalez et al., 2008;Tang et al., 2020).In tomato, after the overexpression of SlANT1 (a homolog of AtMYB113), the expression of the structural genes, and the bHLH gene SlAN1 (a homolog of AtTT8) was significantly up-regulated (Mathews et al., 2003;Kiferle et al., 2015).SlAN1 probably activates the expression of structural genes through the MYB-AN1-WD40 complex, whereas SlJAF13 (a homolog of AtGL3) regulates the transcription of SlAN1 through the MYB-SlJAF13-WD40 complex (Montefiori et al., 2015;Liu et al., 2018).Therefore, we speculated that under bagged conditions, the up-regulation of EBGs was mainly due to the activation of SmMYB113 in the peel of pind; the LBGs were directly regulated by the MBW complex composed of SmMYB113, SmTT8, and WD40 (SmTTG1), whereas the expression of SmTT8 was activated by the MBW complex SmMYB113-SmGL3-SmTTG1; and SmGL3 was activated by SmMYB113 (Figure 7B).
In addition, WRKY44 was reported to regulate anthocyanin accumulation.In tobacco, the coloration of flowers was observed to increase upon overexpression of TTG2, whereas a decrease in flower coloration was achieved through silencing TTG2 (Li et al., 2017).Anthocyanin accumulation was induced in tobacco leaves after transient overexpression of kiwifruit WRKY44 (Peng et al., 2020).WRKY44 was reported to interact with BMW to regulate the corresponding biological processes (Pesch et al., 2014;Gonzalez et al., 2016;Verweij et al., 2016;Lloyd et al., 2017).In summary, it was suggested that under bagging conditions, the expression of SmTTG2 was up-regulated by the MBW complex and SmTTG2 could bind to the MBW complex to promote the expression of structural genes in pind (Figure 7B).
Based on the expression patterns and putative function of the genes in other plants, we proposed a working model that describes the regulatory mechanism of anthocyanin biosynthesis in the peel of '14-345' and pind under dark conditions (Figure 7).It is indicated that the activation of SmMYB113, SmTT8, SmTTG2 and SmTT8 is the necessary factor for anthocyanin accumulation in the dark in eggplant.Considering that SmMYB113 was the most upstream in the pathway, it is postulated that SmMYB113 was the first transcription factor to be activated and serves as the most crucial regulator.Combined with gene mapping analysis, it is speculated that the activation of SmMYB113 in pind under bagged conditions may be related to the SNP of the intergenic region of EGP21875.
(A) Fruit color of '14-345' (WT) and pind (MT) under natural light (L) and dark (D) conditions.The peel color under the calyx is indicated with arrows.White bar = 4cm.(B) CIRG (color index of red grape) value and (C) total anthocyanin content of the peel of '14-345' and pind under natural and dark conditions.Data marked with *, ** and ns indicate P <0.05, P<0.01 and no significance, respectively.
FIGURE 4 Identification of the DEGs related to less-photosensitive anthocyanin biosynthesis in pind.(A-C) Venn diagram of the different comparisons of DEGs.(D) Expression pattern analysis of structural genes involved in less-photosensitive anthocyanin biosynthesis in fruit peel of pind.Note: Red gene ID represents candidate DEGs related to less-photosensitive anthocyanin biosynthesis in the fruit peel of pind.The size of the circle represents log 2 (FPKM+1); (E) Expression pattern analysis of TFs involved in less-photosensitive anthocyanin biosynthesis in the fruit peel of pind.The number in the box indicates the FPKM value.

FIGURE 5
FIGURE 5Validation of RNA-seq results using qRT-PCR.Y-axis: relative expression.The blue bar and red line represent the data from qRT-PCR and RNA-seq data, respectively.Asterisks indicate significant differences determined by the Student's t-test (*P < 0.05, **P < 0.01).ns, no significance.

TABLE 1
The segregation of the fruit color in dark condition.

TABLE 2
Candidate genes related to less-photosensitive anthocyanin synthesis in the fruit peel of pind.