Skip to main content
NCBI home page
Horticulture Research logoLink to Horticulture Research
. 2024 Dec 12;12(3):uhae349. doi: 10.1093/hr/uhae349

Integrative analysis of genome-wide association studies of polyphenols in apple fruits identifies the MdDof2.4-MdPAT10 module that promotes procyanidin accumulation

Zhongxing Li 1,2,#, Cai Gao 3,#, Tianle Fan 4, Yilin Cui 5, Zeyuan Liu 6, Lei Li 7, Qian Qian 8, Mengjie Cheng 9, Xiangqiang Zhan 10, Chundong Niu 11, Fengwang Ma 12, Peizhi Yang 13, Qingmei Guan 14,✉
PMCID: PMC11890028  PMID: 40061802

Abstract

Polyphenols represent a significant class of nutrients in apples, contributing to human health and well-being. Among these, procyanidins stand out as the most prevalent polyphenolic compounds in apple fruits. These compounds are abundant in wild apples and generally low in modern apple cultivars. Therefore, it is crucial to identify and recover genetically lost genes that regulate polyphenol accumulation in order to improve the apple quality. To achieve this, we conducted a genome-wide association study (GWAS) on 15 traits related to polyphenol content, utilizing 134 Malus accessions. We identified 1204 marker–trait associations (MTAs) and 840 candidate genes, including known polyphenol biosynthetic and regulatory genes, such as MYB7, TT4, and HCT/HQT. Notably, we pinpointed a protein S-acyl transferase 10 (PAT10), which is significantly associated with procyanidin content. Through experiments with transgenic calli, we determined that apple (Malus domestica) MdPAT10 positively regulated procyanidin accumulation. Furthermore, we identified a 51-bp insertion variant (In-868) on the promoter of the PAT10, which influences its expression. Both a yeast one-hybrid (Y1H) assay and an electrophoretic mobility shift assay (EMSA) revealed that MdDof2.4 was able to bind to the promoter of MdPAT10 containing In-868 (MdPAT10proIn-868), but not to the promoter of MdPAT10 without In-868 (MdPAT10pro). Moreover, MdDof2.4 promoted MdPAT10 (with MdPAT10proIn-868) expression and increased procyanidin accumulation in fruits. Overall, our results enhance the understanding of the biosynthetic regulation of apple polyphenols and provide a theoretical foundation and genetic resources for breeding apple varieties with optimal polyphenol content.

Introduction

Apple (Malus domestica), a member of the Rosaceae family, is one of the most widely cultivated and consumed fruit crops worldwide. In 2022, global production of apples exceeded 95 million tonnes [1]. Apples are particularly recognized as a rich dietary source of antioxidants due to their high levels of polyphenols and other bioactive compounds that contribute to human health and well-being. Polyphenolic compounds, a class of secondary metabolites with multiple phenolic hydroxyl groups, exhibit significant antioxidant and free radical scavenging activities [2, 3]. However, while polyphenolic compounds are abundant in wild apples, their levels are generally lower in modern apple cultivars [4, 5].

The biosynthetic pathway of polyphenolic compounds is highly complex and involves intricate molecular mechanisms controlled by key enzymes and diverse transcription factors [6, 7]. For example, key enzymes like anthocyanidin synthase (ANS), anthocyanidin reductase (ANR), and leucoanthocyanidin reductase (LAR), are pivotal rate-limiting steps in polyphenol biosynthesis [8, 9]. The MYB–bHLH–WD40 (MBW) complex is a well-established transcriptional regulator that plays a crucial role in anthocyanin biosynthesis [10], as well as other families of transcription factors, such as WIP, NAC, and WRKY. Notably, the overexpression (OE) of MdWRKY11 can increase the accumulation of flavonoids in apple callus tissues [11]. Sun et al. [12] reported that NAC52 binds to the promoters of MYB9 and MYB11, directly regulating LAR to promote anthocyanin and procyanidin (PC) biosynthesis in apple. Furthermore, research indicates that the levels of PC are affected by post-translational modifications, such as histone methylation [13], ubiquitination [14], and SUMOylation [15].

Palmitoylation is a dynamic and reversible post-translational modification that regulates the localization, activity, stability, and interactions of proteins. It plays a key role in cellular signal transduction, metabolism, apoptosis, and other biological processes [16]. TIP1 (Tip Growth Defective 1), a member of the palmitoyltransferase family identified in plants, regulates protein hydrophobicity in Arabidopsis, affecting protein–membrane binding, signal transduction, and intracellular vesicle transport [17]. Previous studies have shown that AtPAT10, exhibiting S-acyl transferase activity, enhanced the reproductive capacity of Arabidopsis by regulating the cell enlargement and division [18, 19]. AtPAT24 can complement the phenotype of the yeast PAT-deficient mutant (akr1), thus restoring the morphological and temperature-sensitive defects of the akr1 mutant [20]. Li et al. [21] found that AtPAT14 is influenced by salicylic acid signaling with its mutants exhibiting a reduced biomass phenotype and negatively regulating leaf senescence. Furthermore, members of the C subfamily of palmitoyltransferases (PAT19, PAT20, and PAT22) interact with the key cytoplasmic receptor kinase BSK1 and catalyze its palmitoylation modification of BSK1 in Arabidopsis [22]. In apples, MdPAT16 can palmitoylate MdCBL1, thus stabilizing its protein level and regulating sugar accumulation and salt tolerance [23]. To date, the involvement of palmitoylation in the regulation of PC biosynthesis is largely unknown.

In this study, we performed genome-wide association study (GWAS) analyses of 15 traits of polyphenol content in 134 Malus accessions to determine the genetic basis of polyphenol accumulation and identified marker–trait associations and candidate genes. Additionally, we conducted a comprehensive analysis of the association between the identified protein S-acyltransferase 10 (MdPAT10) and PC accumulation. These findings enhance our understanding of the regulatory pathway of apple polyphenol biosynthesis and provide genetic resources for further breeding of apple varieties with optimal polyphenol content.

Results

Polyphenolic compound content in Malus accessions

We collected fruit flesh from 134 Malus accessions, comprising 23 wild accessions (Wild), 32 Malus sieversii accessions (M. sieversii), 52 cultivars (Cultivar), and 27 hybrids (Hybrid), and determined the content of 15 polyphenolic compounds by liquid chromatography–mass spectrometry (LC–MS) (Fig. 1a, Fig. ,S1, and Table S1). The content of these 15 polyphenols in apple flesh showed significant variation. Among them, apigenin 7-glucoside (API7G), procyanidin B2 (PCB2), and epicatechin (ECT) had the highest average contents of 19.5, 4.6, and 0.9 mg/g, respectively. In contrast, isorhamnetin 3-O neohesperidoside (I3ON) and methyl gallate (MG) had the lowest content (Fig. 1a and Table S2). Rutin (RT), quercetin (QT), prunin (PN), and 12 other polyphenolic compounds exhibited coefficients of variation (CV) >60%, with RT having the highest CV of 383%, followed by QT and PN with CVs of 269% and 248%, respectively (Table S2).

Figure 1.

Figure 1

Open in a new tab

(A) Distribution and (B) correlation of the content of 15 polyphenolic compounds from 134 Malus accessions. The left diagonal section of the rectangle illustrates the Pearson correlation between the polyphenolic compounds, while the right diagonal section depicts the significance of these correlations. Statistical analyses were performed using the Student’s t-test, with significant differences are indicated by *P < 0.05, **P < 0.01, or ***P < 0.001. N.s. indicates no significant difference. CT, catechin; ECT, epicatechin; PCB1, procyanidin B1; PCB2, procyanidin B2; API7G, apigenin 7-glucoside; RT, rutin; QT, quercetin; HN, hyperin; AG, astragalin; PN, prunin; CGA, chlorogenic acid; FA, ferulic acid; PZ, phloridzin; MG, methyl gallate; I3ON, isorhamnetin 3-O neohesperidoside.

Correlations were observed among the content of the 15 polyphenolic compounds. The content of catechins (CTs), ECTs, procyanidin B1 (PCB1), and PCB2 showed significant correlations (Fig. 1b). Additionally, significant correlations were found between the content of API7G, RT, QT, HN, astragalin (AG), and PN (Fig. 1b), which is consistent with the network of biosynthetic metabolism pathways of these compounds.

We observed significant differences in the content of the 15 polyphenolic compounds among the three groups (Wild, M. sieversii, and Cultivar; Fig. S2). Higher content of polyphenols was found in M. sieversii and Wild groups, while lower content of polyphenols was observed in the Cultivar group. Notably, the content of PCB2 and API7G showed significant difference in M. sieversii and Wild groups compared with the Cultivar group (Fig. S2), which indicates that the phenolic content is altered during the domestication and improvement of apples.

Population structure and linkage disequilibrium

We performed single-nucleotide polymorphism (SNP) analysis on whole-genome resequencing data from 134 Malus accessions, utilizing the Genome Analysis ToolKit (GATK) Best Practices Workflow. Through this analysis, we identified a total of 10 987 271 high-confidence SNPs, corresponding to 17.58 SNPs per kilobase (kb), for subsequent analyses of genetic variation and population structure (Table S1 and Table S3).

To further investigate the genetic structure of the 134 Malus accessions, we performed population structure analyses. Firstly, we constructed a phylogenetic tree based on the neighbor-joining (NJ) method with 1000 bootstraps. The phylogenetic topologies indicate that the Wild, M. sieversii, and Cultivar are three monophyletic taxa, while the hybrids are primarily located in the central region between Wild and M. sieversii (Fig. 2a). Among them, the Cultivar group was genetically closer to the M. sieversii than to the Wild group. Subsequently, we performed admixture analyses. When K = 4 (best K, Fig. S3), Wild, M. sieversi,i, and Cultivar groups were clearly segregated from each other. Malus sieversii has a relatively independent genetic background and contributed genetically to the formation of cultivars to a certain extent (Fig. 2b). This finding is consistent with the previously described origins of domestication and dispersal routes of apples [24, 25]. Ultimately, we examined these accessions by SNP-based principal component analysis (PCA). We similarly observed that the Cultivar is clearly separated from the Wild and M. sieversii, closer to M. sieversii (Fig. 2c). In addition, PCA demonstrates that hybrids were mainly distributed in the middle of the Wild and M. sieversii, with some hybrids found between M. sieversii and Cultivar, or located within one of these groups (Fig. 2c). These results are in consistency with those of phylogenetic relationships (Fig. 2a). The patterns of the population structure suggest that the genetic origins of Hybrid indeed derive from genetic mixing between different species of the Malus genus.

Figure 2.

Figure 2

Open in a new tab

Genomic relationship and population structure of 134 Malus accessions. (A) Clades and groups, including Wild, M. sieversii, Hybrid, and Cultivar, as illustrated by a phylogenetic tree. (B) Population structure of model-based clustering when K = 4. (C) PCA of Wild, M. sieversii, Hybrid, and Cultivar. Different color markers represent the groups as in (A).

Rapid linkage disequilibrium (LD) decay was observed in the Wild group, while the LD decay process in M. sieversii, Cultivar, and Hybrid groups was relatively slow (Fig. S4a).

In addition, we found differences in nucleotide diversity (π) and population fixation index (FST) among the four groups. The highest π was found in the Hybrid group, followed by the Wild group, whereas the lowest π was found in the M. sieversii and Cultivar groups (Fig. S4b). The FST values showed that the cultivars were genetically distant from the M. sieversii and Wild groups, which aligns with the findings from the phylogenetic tree and PCA (Fig. 2a and 2c). These results imply a degree of population differentiation among the Wild, M. sieversii, and Cultivar groups, which might contribute to the differences in phenolic content among Malus accessions.

Genome-wide association study on 15 polyphenols

Based on the dataset of 10 987 271 high-confidence SNPs from 134 resequenced Malus accessions, we conducted GWAS analyses of 15 traits of individual polyphenol content. Acknowledging the significant impact of population genetic structure on GWAS efficacy [26], we corrected the effect of population stratification on the GWAS using PCA and kinship matrix. Subsequently, we employed the mixed linear model (MLM) method, as implemented in the GEMMA software [27], and identified 1204 marker–trait associations (MTAs) and 840 candidate genes (Fig. 3, Table 1, Fig,. S5, and Table S4).

Figure 3.

Figure 3

Open in a new tab

Manhattan plot of GWAS for 15 traits in Malus accessions. CT, catechin; ECT, epicatechin; PCB1, procyanidin B1; PCB2, procyanidin B2; API7G, apigenin 7-glucoside; RT, rutin; QT, quercetin; HN, hyperin; AG, astragalin; PN, prunin; CGA, chlorogenic acid; FA, ferulic acid; PZ, phloridzin; MG, methyl gallate; I3ON, isorhamnetin 3-O neohesperidoside.

Table 1.

Lead SNPs and related known genes significantly associated with 15 polyphenolic traits according to the GWAS for the 134 Malus accessions

Trait Chromosome Lead SNP -log 10 (P) PVE (%) Major allele Minor allele Gene symbol
CT Chr04 2 589 849 7.07 30.62 G A WD40
Chr16 38 580 219 7.13 30.70 C T bHLH
ECT Chr15 19 343 501 6.93 15.22 C G DHHC
Chr16 2 107 668 6.96 30.10 T C MYB7
Chr16 3 334 171 6.82 32.40 T A AP2
Chr16 3 455 183 6.95 33.10 G C TCP12
PCB1 Chr16 2 107 668 6.82 29.10 T C MYB7
Chr16 3 404 850 6.57 22.51 T C LAR
PCB2 Chr09 5 720 628 7.25 34.17 A C WD40
API7G Chr09 27 442 887 7.79 22.30 C G WRKY40
Chr09 28 410 789 6.78 18.88 G A HCT
RT Chr03 15 265 020 5.3 15.53 T C WIP3
Chr07 26 738 882 5.27 16.06 C T TT5
QT Chr04 19 713 978 8.28 11.92 A G TT4
Chr05 18 136 875 8.84 13.42 C T UGT73B3
Chr14 21 450 469 8.39 12.13 A G Dof
Chr15 4 360 622 12.38 15.57 G A MYB84
HN Chr01 11 202 493 6.28 18.66 C T MYB93
AG Chr07 614 305 7.53 23.02 G T bHLH
Chr15 17 358 160 7.13 21.62 C A TT2
Chr15 21 422 536 8.58 27.55 T G bZIP44
Chr17 11 145 873 6.77 20.63 A T UGT85A3
PN Chr10 26 219 946 7.61 32.89 A T ABI3
Chr14 17 647 147 6.9 29.47 T C MYB103
Chr17 3 777 291 8.55 26.11 T C MYB94
CGA Chr16 29 456 547 7.17 31.19 G A WD40
Chr17 28 263 534 7.94 36.38 A G bZIP61
Chr17 27 376 190 6.21 26.87 G T HCT/HQT
Chr17 27 726 498 5.94 24.81 C T 4CL2
FA Chr16 34 045 701 7.27 17.71 C T MYB36
PZ Chr05 19 940 664 6.87 30.23 G A WD40
MG Chr09 27 197 923 7.75 16.12 A T CCR1
Chr09 27 454 208 9.46 27.66 T C WRKY40
Chr09 28 479 365 7.91 16.02 G A HCT/HQT
I3ON Chr16 20 082 085 6.22 15.49 T A C2H2
Open in a new tab

PVE, phenotypic variation explained; CT, catechin; ECT, epicatechin; PCB1, procyanidin B1; PCB2, procyanidin B2; API7G, apigenin 7-glucoside; RT, rutin; QT, quercetin; HN, hyperin; AG, astragalin; PN, prunin; CGA, chlorogenic acid; FA, ferulic acid; PZ, phloridzin; MG, methyl gallate; I3ON, isorhamnetin 3-O neohesperidoside.

We identified known key loci involved in polyphenol regulation at these significant MTAs and candidate genes. A notable signal on chromosome 4, linked to QT content, was located within the MD04G1111500 gene (Fig. 4a). This gene encodes chalcone synthase (CHS), a key enzyme in the biosynthesis of QT [28]. The lead SNP (Chr4:19713978, P = 5.20 × 10−9, PVE = 11.92%) was associated with QT content and strongly linked with MD04G1111500 (Fig. 4a). Furthermore, Malus accessions with the homozygous G/G allele had higher content of QT than those with the homozygous A/A allele or heterozygous A/G allele (Fig. 4b).

Figure 4.

Figure 4

Open in a new tab

GWAS-based dissection of polyphenol traits and causative genes. (A–F) Integrated analysis of GWAS signals, LD blocks, and polyphenolic content for quercetin (QT), isorhamnetin 3-O neohesperidoside (I3ON), and prunin (PN) were conducted. Local Manhattan plot (top) and LD heat map (bottom) illustrate the GWAS signal for QT (A), I3ON (C), and PN (E) traits. Boxplots for QT (B), I3ON (D), and PN (F) content are presented based on the lead SNPs. The dash line indicates the lead SNPs at Chr4:19713978 (A), Chr16:20082085 (C), and Chr17:27376190 (E). Statistical analyses were performed using the Student’s t-test, with statistically significant differences indicated by ***P < 0.001 in (B, D, and F). N.s. indicates no significant difference. (G) Local Manhattan plot of catechin (CT), epicatechin (ECT), procyanidin B1 (PCB1), and procyanidin B2 (PCB2) on Chromosome 16. (H) Genes located in polyphenol hotspot QTL, where text-labeled genes are known loci in the phenolic synthesis pathway. (I) LD heat map surrounding the peak signal for ECT and PCB1 traits. (J) Haplotype analysis based on GWAS significant sites (Chr16:2107657, Chr16:2107668, and Chr16:2107716) on the MdMYB7 promoter. (K) Boxplots for PCB1 and PCB2 content are presented according to the haplotypes in (J). (L) Content of total procyanidin (PC) in WT, MdMYB7 overexpression (OE), and RNA interference (RNAi) lines of apple calli. Error bars indicate SD (n = 15). One-way analysis of variance (ANOVA , Duncan’s test) was performed, with statistically significant differences indicated by lowercase letters in (K and F). (M) DMACA staining of apple calli (WT, MdMYB7 OE, and RNAi lines).

MD16G1212200, located near the GWAS signal of I3ON on chromosome 16, encodes a C2H2 zinc finger protein (Fig. 4c). The lead SNP (Chr16:20082085, P = 6.00 × 10−7, PVE = 15.49%) was located upstream of MD16G1212200 and linked with the gene body. Accessions with the homozygous A/A allele had higher content of I3ON than those with the homozygous T/T allele or heterozygous A/T allele (Fig. 4d). Moreover, a signal for PN on chromosome 17 is located downstream of MD17G1050900 (Fig. 4e), which encodes an R2R3-type MYB transcription factor. Accessions with the homozygous T/T allele had lower content of PN than those with the homozygous C/C allele or heterozygous T/C allele (Fig. 4f). Additionally, a nonsynonymous SNP (Chr17:27376190, P = 6.17 × 10−7, PVE = 26.87%) was strongly associated with chlorogenic acid (CGA) content and is located within the hydroxycinnamoyl-CoA shikimate/quinate hydroxycinnamoyl transferase (HCT/HQT) gene (Table 1), which is the key rate-limiting enzyme for CGA biosynthesis [29].

PCs are a class of polyphenolic compounds consisting of CT or ECT units bound in various amounts. In this study, we identified genetic markers associated with the content of CT, ECT, PCB1, and PCB2 with 30, 88, 54, and 20 significantly associated SNPs, respectively (Table 1 and Table S4). Specifically, within the interval of 1.0–4.5 Mb on chromosome 16, we identified a GWAS hotspot region strongly correlated with these four traits (CT, ECT, PCB1, and PCB2 content; Fig. 4g and 4h). This hotspot region coincided with a previously reported hotspot area [30, 31].

Remarkably, three SNPs (Chr16:2107657, Chr16:2107668, and Chr16:2107716) located in the promoter region of the MdMYB7 were significantly associated with PC content (Fig. 4i, 4j, and Table S4). MYB7 encodes an R2R3-type MYB transcription factor, and its homologous protein in Arabidopsis (AtMYB7) acts as a repressor of flavonol biosynthesis [32]. Moreover, haplotype analysis revealed three principal haplotype combinations: HapA as ATT (n = 80), HapB as GCC (n = 15), and HapA/HapB (heterozygous) as ATT/GCC (n = 21) (Fig. 4j). Accessions carrying HapA or HapA/HapB exhibited significantly higher content of CT, ECT, PCB1, and PCB2 than those carrying HapB (Fig. 4k and Fig. S6).

We investigated the role of MdMYB7 in apple PC biosynthesis by generating transgenic OE and RNA interference (RNAi) apple calli of MdMYB7 (Fig. S7). By measuring the total PC content, we found that the MdMYB7 OE lines had significantly lower PC than wild type (WT). Conversely, the MdMYB7 RNAi lines demonstrated significantly elevated PC levels relative to the WT (Fig. 4l). Furthermore, dimethylaminocinnamaldehyde (DMACA) staining confirmed these findings (Fig. 4m). These findings indicate that MdMYB7 acts as a negative regulator of PC accumulation in apples.

PAT10 plays a critical role in PC biosynthesis

Through GWAS of PCB2 and ECT, we pinpointed a significant peak associated with the PAT10 gene on chromosome 15 (Fig. 5a and 5b). Signal peaks also appeared near the PAT10 in the GWAS of CT and PCB1, but these peaks were not statistically significant (Fig. S8). Sequence homology analysis revealed that PAT10 encodes the protein S-acyl transferase, whose function is critical for plant metabolism [18, 19]. Furthermore, through the phylogenetic tree and protein sequence alignment, we established that PAT10s are highly conserved in flowering plants (Fig. S9), suggesting that PAT10 might be an ancient gene with fundamental roles in flowering plants. Importantly, we found that the expression level of PAT10 was significantly higher in high-PC accessions than low-PC accessions (Fig. 5c).

Figure 5.

Figure 5

Open in a new tab

The MdPAT10 enhances procyanidin accumulation in apple. (A) Local Manhattan plot surrounding the GWAS signal for procyanidin B2 (PCB2, top) and epicatechin (ECT, bottom). (B) LD heat map surrounding the peak signal for PCB2 and ECT traits. (C) qRT-PCR analysis of MdPAT10 in low- and high-PC content apples. (D) PC content in the transiently overexpressed or silenced MdPAT10 calli. 35S:MdPAT10 or pTRV2-MdPAT10 were transiently transformed in ‘Qin Guan’ fruit at 100 DAF. Gene expression of MdPAT10 was measured by qRT-PCR analysis (left). Statistical analyses were performed using the Student’s t-test, with statistically significant differences indicated by *P < 0.05, **P < 0.01, or ***P < 0.001 in (C and D). N.s. denotes no significant difference. Error bars indicate SD (n = 3 for qRT-PCR assay, and n = 20 for PC content assay). (E) PC content determination of WT and transgenic calli of MdPAT10 OE and RNAi. Error bars indicate SD (n = 15). Statistical analyses were performed using the one-way ANOVA, with statistically significant differences indicated by different lowercase letters. (F) DMACA staining of apple calli (WT, MdPAT10 OE, and RNAi lines).

Subsequently, we determined the function of PAT10 in apple PC biosynthesis. Initially, we found that the transiently overexpressed MdPAT10 in apple fruits resulted in a significant increase in the expression of MdAPT10 as well as in the total PC content, in comparison to control fruits (Fig. 5d). Additionally, virus-induced gene silencing (VIGS) of MdPAT10 in apple fruits showed a decrease in the expression of MdPAT10, along with a corresponding decrease in the total PC content compared to the control (Fig. 5d). Moreover, we generated MdPAT10 OE and RNAi lines of apple calli (Fig. S10). The total PC content was significantly higher in the MdPAT10 OE calli than that in the WT, while it was notably reduced in the MdPAT10 RNAi calli relative to the WT (Fig. 5e). Meanwhile, DMACA staining further verified the above results (Fig. 5f). Collectively, these findings indicate that MdPAT10 plays a positive regulatory role in the biosynthesis of PCs in apples.

Natural variation in the PAT10 promoter is associated with PC content in apples

Population genetic analyses revealed a footprint of domestication selection on the PAT10 gene. Nucleotide diversity (π) and Tajima’s D were calculated by using 97 wild and 106 cultivated samples that we previously documented [24]. Within the PAT10 gene and its promoter region, π was higher in wild samples than in cultivated samples (Fig. 6a), indicating a reduction in genetic diversity among cultivated apples. Moreover, in the promoter of the PAT10 gene, Tajima’s D value was greater than zero in wild samples, while in cultivated samples, it was less than zero. This disparity suggests the presence of selection pressure during the domestication process (Fig. 6a).

Figure 6.

Figure 6

Open in a new tab

In-868 on MdPAT10 promoter is significantly associated with PC content. (A) Nucleotide diversity (π) and Tajima’s D analyses of MdPAT10 and its promoter region within the wild and cultivated apples. (B) Sanger sequencing and (C) PCR analysis of the MdPAT10 promoter region in apple varieties. (D) Genotype distributions of In-868 in wild species and cultivated varieties. Source data in Fig. S11. (E) The content of CT, ECT, PCB1, and PCB2 of 134 Malus accessions with or without In-868. Statistical analyses were performed using the Student’s t-test, with statistically significant differences indicated by *P < 0.05, **P < 0.01, or ***P < 0.001. N.s. indicates no significant difference. In (D and E), +/+ represents accessions carrying homozygous In-868; −/− indicates accessions without homozygous In-868; −/+ denotes accessions carrying heterozygous In-868.

Subsequent analysis of the PAT10 promoter by Sanger sequencing and polymerase chain reaction (PCR) analysis identified a 51-bp insertion variant, designated In-868, located at −868 bp upstream of the MdPAT10 start codon. This variant was present in the ‘Jin Hong’ and ‘Orin’ varieties, but not in the ‘Golden Delicious’ and ‘Honeycrisp’ varieties (Fig. 6b and 6c). Furthermore, In-868 showed a heterozygous pattern in ‘Gan Hong’ and ‘Pound Sweet’ varieties (Fig. 6b and 6c). PCR analysis using additional 283 wild species and 96 cultivated varieties revealed notable differences in the distribution pattern of In-868 between wild species and cultivated varieties (Fig. 6d and Fig. S11). Specifically, the proportion of In-868 in wild species was 18.02%, whereas in cultivated varieties, it was only 5.21% (Fig. 6d and Fig. S11). This observation suggests that the In-868 genotype may have undergone negative selection during the domestication and improvement of apple varieties.

Remarkably, we found that accessions with In-868 had significantly higher levels of CT, ECT, PCB1, and PCB2 than those without In-868 (Fig. 6e). These results suggest that the In-868 variant is associated with PC accumulation in apples.

Interaction of MdDof2.4 with MdPAT10proIn-868 increases MdPAT10 expression and PC accumulation

Our analysis of the In-868 natural variant sequence using the PlantRegMap website revealed that it contains a typical binding site (5′-AAAAG-3′) for MdDof2.4 (Table S5). A yeast one-hybrid (Y1H) assay confirmed that MdDof2.4 had the ability to bind to the In-868 on the MdPAT10 promoter (MdPAT10proIn-868), while it did not bind to the MdPAT10 promoter without In-868 (MdPAT10pro) (Fig. 7a). Additionally, an electrophoretic mobility shift assay (EMSA) demonstrated that the recombinant MdDof2.4 protein binds to the MdPAT10proIn-868 fragment containing 5′-AAAAG-3′ (hot probe with biotin labeling). The binding of MdDof2.4 to the hot probe was competed off by unlabeled cold probe (150- and 250-fold) (Fig. 7b), indicating the specificity of the MdDof2.4-MdPAT10proIn-868 interaction. Moreover, a dual-luciferase reporter gene assay (Dual-LUC) showed that the combination of MdDof2.4 and MdPAT10proIn-868 could activate the expression of LUC (Fig. 7c and 7d), suggesting a strong binding affinity between MdDof2.4 and MdPAT10proIn-868. These results demonstrate that MdDof2.4 binds specifically to MdPAT10proIn-868, but not to MdPAT10pro.

Figure 7.

Figure 7

Open in a new tab

Interaction of MdDof2.4 with MdPAT10proIn-868 increases the expression of MdPAT10 and the accumulation of PC. (A) The Y1H assay shows the binding of MdDof2.4 to the insertion-containing promoter of MdPAT10 (MdPAT10proIn-868). (B) EMSA demonstrates the binding of MdDof2.4 to the 5′-AAAAG-3′ motif in MdPATA10proIn-868. The hot probe is a biotin-labeled fragment containing the 5′-AAAAG-3′ motif; while the cold probe is a nonlabeled competitive probe, prepared at concentrations 150- and 250-fold, respectively, of the hot probe; the mutant motif is 5’-CCCCC-3′. (C) Diagrams of the effector and reporter constructs. (D) Interaction between MdDof2.4 and MdPAT10proIn-868, as determined by dual luciferase reporter assays in Nicotiana benthamiana leaves. Data are presented as the mean ± SE (n = 9). Statistical analyses were performed using the one-way ANOVA, with statistically significant differences indicated by different lowercase letters. (E–F) qRT-PCR analysis of MdDof2.4 (E) and MdPAT10 (F) genes in WT, transiently overexpressed, or silenced MdDof2.4 fruits. 35S: MdDof2.4 or pTRV2-MdDof2.4 were transiently transformed in ‘Gan Hong’ fruit at 100 DAF. (G) Determination of the PC content in WT, transiently overexpressed, or silenced MdDof2.4 fruits. Error bars indicate SD (n = 20). Statistical analyses were performed using the Student’s t-test, with statistically significant differences indicated by *P < 0.05, **P < 0.01, or ***P < 0.001 in (E–G). N.s. denotes no significant difference. (H) A proposed working model of the regulation of PC accumulation by the MdDof2.4-MdPAT10 module in apples.

The transient OE of MdDof2.4 in the fruits of the cultivar apple ‘Gan Hong’ (harboring the In-868, Fig. 6b and 6c) promoted the expression of MdPAT10 (Fig. 7e and 7f) and increased the total PC content (Fig. 7g). Additionally, silencing MdDof2.4 in ‘Gan Hong’ apple fruits by the VIGS technique resulted in a significantly reduced expression of MdPAT10 and a significant reduction of the total PC content, compared to the fruits transformed with the empty vector (Fig. 7e–g). These findings suggest that MdDof2.4 is capable of promoting MdPAT10 expression and thereby enhancing PC accumulation (Fig. 7h).

Discussion

Polyphenolic metabolites in apples exhibit variation influenced by factors such as varieties, tissue types, growth periods, and environmental conditions [5, 33, 34]. Nevertheless, genetic variability remains the primary determinant of their content in plants [35, 36]. Previous studies have indicated that during the domestication and improvement of apples, there has been an increase in fruit size and sugar content, while the polyphenol content has decreased [4, 37]. In the study, we analyzed the content of 15 phenolic compounds across 134 Malus accessions and found that phenolic content was higher in wild accessions and relatively lower in domesticated apples, which is consistent with the trend observed during apple domestication and improvement [25, 37]. Through GWAS technology, we investigated the factors governing polyphenol biosynthesis, aiming to provide genetic resources for breeding apple varieties with enhanced health-promoting properties.

Association analyses of polyphenol content traits with genetic markers have revealed key regions involved in the regulation of polyphenol synthesis. For example, the 1.0- to 4.5-Mb interval on Malus domestica chromosome 16 has been identified as a hotspot quantitative trait locus (QTL) for polyphenol biosynthesis, containing several key genes closely related to this pathway, including LAR1 and MYB7 [30, 31]. Additionally, MBW complex has regulatory functions in the flavonoid biosynthetic pathway [10, 38]. For instance, in the ‘Yanfu 8’ apple variety, MdMYB114 and MdbHLH3/33 are involved in the formation of the MBW complex, promoting anthocyanin biosynthesis [39]. Furthermore, it has been demonstrated that in a variety of plants, including tomato, coffee, potato, and artichoke, HCT/HQT is a rate-limiting regulatory enzyme in the CGA biosynthetic pathway [40–42]. This finding was also confirmed in our study, where GWAS analyses based on CGA content similarly identified the HCT/HQT motif as a key site.

PCs represent a significant portion of the polyphenolic compounds found in apples and are predominantly synthesized through the flavonoid pathway. Studies have revealed that the flavonoid content can vary by 10 to 100 times among different apple varieties [35]. Although the biosynthetic pathway of PCs in apple has been largely elucidated, there remains a relative scarcity of studies focusing on their molecular regulatory mechanisms, particularly regarding post-translational modifications. Notably, palmitoylation, an important post-translational modification of proteins, plays a role in various localization and signaling pathways. Jiang et al. [23] demonstrated that MdPAT16 regulates sugar metabolism and salt tolerance in apple. However, the role of PATs in regulating the synthesis of phenolic compounds, such as PCs, has yet to be reported.

In this study, we found that the module of MdDo2.4-MdPAT10 regulates the PC accumulation. As a transcription factor, MdDof2.4 activates the expression of MdPAT10, which in turn increases the levels of PCs and their precursors in apple. Specifically, the In-868 variant in the MdPAT10 promoter region enhances the binding affinity of MdDof2.4, thereby promoting the expression of MdPAT10. Conversely, the absence of the In-868 results in the loss of the binding site for MdDof2.4, leading to decreased expression of MdPAT10. However, the specific molecular mechanisms remain to be investigated, particularly the effect of MdPAT10-mediated palmitoylation on PC biosynthesis.

Conclusion

In summary, our study revealed the distributional characteristics of polyphenol content within the Malus genus and identified key enzymes and transcription factors involved in regulating polyphenol accumulation through GWAS. Additionally, we confirmed the significant role of MdDof2.4-MdPAT10 module in the regulation of PC biosynthesis. These findings enhance our understanding of the regulatory network of apple PC biosynthesis and provide gene resources for breeding of new apple varieties.

Materials and methods

Plant material collection

In this study, we collected the fruits of 134 Malus accessions from the Apple Germplasm Resource Garden at the Horticultural Farm of the Northwest A&F University (NWAFU, Yangling, China). The accessions included 23 wild (Wild), 32 M. sieversii (M. sieversii), 52 cultivars (Cultivar), and 27 hybrids (Hybrid) (Table S1). These accessions had been subjected to whole-genome resequencing based on the Illumina platform in a previous study [24], and the average sequencing depth of the samples was ~18.62-fold coverage (Table S1).

For measuring polyphenolic metabolites, three to five fruits with uniform ripeness from at least two trees for one accession were collected as one biological replicate. Three biological replicates were used for each accession. In addition, young leaves of 283 wild species and 96 cultivated varieties, used for identifying In-868 through PCR, were gathered from the Apple Germplasm Resource Garden at the Horticultural Farm of the Northwest A&F University (NWAFU, Yangling, China).

Extraction and measurement of polyphenolic metabolites

Polyphenols were extracted as previously described by Zhang et al. [43]. Briefly as follows: first, a 0.05 g sample of apple powder was taken and placed in a 1.5 ml polypropylene (PP) tube. Then, 1 ml of extraction solution was added, which consisted of a mixture of methanol, deionized water (ddH2O), and formic acid in a ratio of 25:24:1. Next, the tubes were sonicated in an ultrasonic cleaner for 20 min, and then the tubes were placed on an oscillator for 20 min. After shaking was completed, the tubes were placed in a centrifuge and centrifuged at 10 000 g for 15 min to separate the solid residue from the extraction solution. Subsequently, the supernatant was aspirated using a syringe and filtered through a 0.22-μm nylon filter (Osmonics, Fisher Scientific, Pittsburgh, USA). The filtered extraction was collected and suitably diluted 5-folds and then stored in amber glass vials for LC–MS analysis.

The HPLC (AB SCIEX ExionLC™ AC System) combined with a triple-quadruple linear ion trap mass spectrometer (Triple Quad™ 5500) was used for the precise analysis of polyphenol compounds. The chromatographic conditions were as follows: (i) Chromatographic column: InertSustain AQ-C18 column (4.6 mm × 150 mm, 5 μm, GL Sciences, Tokyo, Japan), with the column temperature set at 30°C. (ii) Flow rate: 300 μl/min. (iii) Solvent system: the mobile phase solutions were water with 0.1% formic acid (A) and 100% methanol (B). (iv) Gradient elution process: 0–1 min, 75% A and 25% B; 1–5 min, linear gradient to 5% A and 95% B; 5–6.5 min, maintain 5% A and 95% B; 6.5–12 min, return to initial conditions, 75% A and 25% B. (v) Injections for samples and standards: 5 μl.

The main parameters of the mass spectrometry were as follows: (i) Ion source: electrospray ionization (ESI). (ii) Ion source temperature (TEM): 600°C. (iii) Ion-spray voltages (IS): 5500 V. (iv) Nebulizer gas (GS1): 60 psi. (v) Heater gas (GS2): 60 psi. (vi) Curtain gas (CUR): 35 psi. (vii) Scanning mode: multireaction monitoring (MRM) was adopted as the method for efficient fragmentation and selective detection of polyphenol metabolites.

Population structure analysis

Raw reads were trimmed using FASTP software (version 0.20.0) [44] and then mapped to the M. domestica ‘Golden Delicious’ reference genome (GDDH13 v1.1) by using BWA-MEM (version 0.7.17, https://github.com/lh3/bwa) with default parameters. Then, duplicated reads were detected and removed with the Picard tools (version 2.1.1) (http://broadinstitute.github.io/picard). SNP calling was performed using GATK (version 3.8) Best Practices workflow [45] as previously described [46].

An NJ tree was constructed using VCF2Dis software (version 1.47, https://github.com/BGI-shenzhen/VCF2Dis) on the basis of a distance matrix with 1000 bootstrap replicates. The web-based tool iTOL (https://itol.embl.de) was used to visualize the phylogenetic tree. PCA was performed using smartpca program embedded in EIGENSOFT (version 6.1.4) [47], and the first three eigenvectors were plotted. Population structure was assessed utilizing ADMIXTURE (version 1.3) [48] which is a model-based clustering method for inferring population structure. Genetic cluster (K) values ranging from 2 to 12 were predefined, and 20 replicates were executed for each K to estimate standard errors.

Linkage disequilibrium analysis

In this study, PopLDdecay software (version 3.41) [49] was used to calculate the LD decay of SNPs in four groups (Wild, M. sieversii, Cultivar, and Hybrid) and all sample set (All). The parameter settings used were as follows: MaxDist = 20 Kb, MAF = 0.05, Het = 0.8.

Genome-wide association study

In the study, we constructed a panel of association maps containing 134 Malus accessions. After genotype imputation, we identified 10 987 271 SNPs with a minimum allele frequency (MAF) >5%, which composed the genotype dataset for GWAS. In addition, we performed log transformation to the phenotypic data for the 15 traits to ensure that the phenotypes conformed to a normal distribution for use as phenotypes in GWAS.

We used the MLM of GEMMA software (version 0.98.5) [27] to explore the association between genotype and phenotype. To correct for the potential effects of population structure and kinship on the association analyses, we considered the population structure and kinship of the samples as cofactors in the model. The kinship correlation matrix among samples was calculated based on SNPs using a built-in program (−gk 2 parameter) in the GEMMA software. Besides, we used the first three principal components (PC1, PC2, and PC3) of PCA as covariates to further control for group stratification.

The Genetic Type I error calculator (GEC) software (version 0.2, https://pmglab.top/gec) was used to perform an accurate count of effective markers in the SNP dataset, which ultimately confirmed a total of 5 501 094 effective markers (ne). Then, we set a threshold of 1/ne to identify statistically significant association loci based on the false discovery rate (FDR) criterion [50]. We identified neighboring genes located within 5 kb upstream and downstream of the leading SNPs as significantly associated genes and obtained functional annotations of these genes through homology alignment to further identify candidate genes.

Haplotype block analysis

We analyzed haplotype blocks surrounding significantly trait-associated SNPs and candidate genes (TT4, C2H2, MYB94, and MdMYB7) using the LDBlockShow software (version 1.40) [51] and genotyped them. The association of haplotype genotypes with trait phenotypes was evaluated using one-way analysis of variance (ANOVA) statistics followed by Duncan’s multiple comparison test.

Vector construction and genetic transformation

To generate overexpressing MdPAT10 or MdMYB7 apple calli, the coding sequences (CDS) of MdPAT10 and MdMYB7 were cloned and introduced into the pK2GW7 vector, respectively. To obtain RNA-interfering MdPAT10 or MdMYB7 apple calli, fragments (200 bp) of MdPAT10 or MdMYB7, respectively, were cloned and introduced into the pK7GWIWG2D vector. The resulting recombinant vectors were transformed into Agrobacterium tumefaciens strain EHA105, respectively. Then, A. tumefaciens carrying the plasmid was transformed into apple calli (‘Orin’), which were used as the WT for genetic transformation.

Transient expression assay

To obtain transiently transformed apple fruits to overexpress the MdPAT10 and MdDof2.4 genes, we transformed recombinant plasmids pK2GW7-MdPAT10 and pK2GW7-MdDof2.4 into A. tumefaciens strain GV3101. The transformed strain was allowed to stand under light-avoidance conditions for 2–4 h and was subsequently injected into fruits of ‘Qin Guan’at 100 days after flowering (DAF). Quantitative real-time PCR (qRT-PCR) was used to detect the expression levels of MdPAT10 and MdDof2.4 after 7 days.

To generate MdPAT10 and MdDof2.4 gene-silenced transient transformation Apple materials, a VIGS assay was performed according to the previously described method [52]. Briefly, 200-bp fragments of MdPAT10 and MdDof2.4 were introduced into the pTRV2 vector, respectively. The pTRV1 plasmid, empty vector pTRV2, pTRV2-MdPAT10, and pTRV2-MdDof2.4 were transformed into A. tumefaciens strain GV3101, respectively. The four aforementioned strains (pTRV1, pTRV2-MdPAT10, pTRV2-MdDof2.4, and pTRV2) were cultured to OD600 = 0.5, and then pTRV1 was mixed with three other (pTRV2-MdPAT10, pTRV2-MdDof2.4, and pTRV2 empty vectors) strains, respectively, in equal volume. After standing for 2–4 h without light, the mixture was injected into ‘Gan Hong’ and ‘Qin Guan’ apple fruits, respectively, at 100 DAF. qRT-PCR was performed to analyze the expression levels of MdPAT10 and MdDof2.4 after 7 days.

Staining and determination of PC

Procyanidin extraction and assay were performed according to the instructions in the Plant Procyanidins Content Assay Kit from Solarbio (product no. BC1350). Staining of apple calli was referred to previously reported [53], and the apple calli was stained using DMACA reagent [1% (w/v) ethanol: 6 M HCl (1:1, v/v)] for 2 h and then observed and photographed for documentation.

qRT-PCR

RNA was extracted from apple calli or fruits using RNAprep Pure Plant PlusKit (DP441, TIANGEN, Beijing, China) in accordance with the manufacturer’s instructions. For qRT-PCR analysis, 1 μg of total RNA was reverse transcribed to first-strand cDNA with oligo-dT using the HiScript III RT SuperMix for qPCR kit (R323–01, Vazyme, Nanjing, China). ChamQ Universal SYBR qPCR Master Mix (C601, Vazyme, Nanjing, China) was used to perform qRT-PCR on a LightCycler® 480 II Real-Time System (Roche, Mannheim, Germany). Malate dehydrogenase (MdMDH) was used as a reference gene to calculate relative expression by the 2-ΔΔCt method, with three biological and three technical replicates for each condition. All primers used are listed in Table S6.

Selective sweep

We calculated the nucleotide diversity (π) [54] and Tajima’s D values [55] for the gene body, upstream, and downstream regions of the MdPAT10. SNP variation data from Chen et al. [24] for wild (n = 97) and cultivars (n = 106) was used, and the calculations were carried out using the VCFtools (version 0.1.17) [56] in the sliding-window approach, with the window size set to 1000 bp and a step size of 100 bp. R package (ggplot2) was used for the visualization of the calculated data.

Y1H assay

Sequences of two promoters (MdPAT10pro and MdPAT10proIn-868) were submitted to the Plant Transcriptional Regulatory Map (PlantRegMap) database [57] for analysis of transcription factor binding sites. The results showed a potential binding site for MdDof2.4 on In-868 of the promoter.

To validate the interactions of MdDof2.4 with MdPAT10pro or MdPAT10proIn-868, we used a Y1H. Briefly, sequences without and with In-868 of MdPAT10 promoter were constructed into the pAbAi vector and transformed into the Y1H yeast strain, which was inhibited from self-activation in SD/-Ura/AbA medium (AbA was used to inhibit self-activation). Subsequently, Y1H yeast strains containing pAbAi-MdPAT10pro and pAbAi-MdPAT10proIn-868 were made into receptor cells, and the pGADT7-MdDof2.4 plasmid was transferred into the receptor cells, and pGADT7-Empty was transferred as a negative control. The transformed yeast cells were screened on SD/−Leu/AbA medium.

Dual-luciferase reporter assay

The CDS sequence of MdDof2.4 was cloned from a cDNA library of ‘Golden Delicious’ apple leaves and constructed into the pGreenII62-SK vector. Meanwhile, two promoter variants of the MdPAT10 (MdPAT10pro and MdPAT10proIn-868) were also constructed into the pGreenII0800-LUC vector, respectively. Subsequently, these recombinant vectors were transferred into A. tumefaciens strain GV3101 containing pSoup, respectively. Suspensions of different combinations of pGreenII62-SK and pGreenII0800-LUC recombinant vector strains were mixed at a ratio of 1:1 and injected into 4-week-old tobacco leaves. Leaves were collected after 3 days of coinfiltration, and firefly luciferase (LUC) and renilla luciferase (REN) signals were quantified using the Dual Luciferase Reporter Gene Assay Kit (11402ES60, Yeasen, Shanghai, China).

Electrophoretic mobility shift assay

The DNA-binding domain sequence of MdDof2.4 was cloned into pMAL-c5X (MBP) expression vector and subsequently transformed into Transetta (DE3) Chemically Competent Cell (CD801–02, TransGen, Beijing, China). Subsequently, positive monoclones were selected and amplified in culture, and the recombinant proteins were induced to be expressed by 0.5 mM IPTG. The cells were crushed using an ultrasonic instrument, followed by recombinant protein purification using Amylose resin. Additionally, the probes used for the EMSA experiments were synthesized by Sangon Biotech (Shanghai, China). EMSA experiments were performed on the purified proteins according to the guidelines of the Lightshift Chemiluminescent EMSA kit (20 148, Thermo Scientific, Pittsburgh, USA).

Statistical analysis

Differences between two samples were analyzed with the use of a two-tailed Student’s t-test. For multiple comparisons, differences were analyzed using one-way ANOVA followed by Duncan’s multiple comparison test. Statistical analyses were performed using IBM SPSS Statistics 24.

Supplementary Material

Web_Material_uhae349
web_material_uhae349.zip (4.4MB, zip)

Acknowledgements

We would like to thank Dr. Chunxiang You from Shandong Agricultural University for providing wild-type ‘Orin’ calli, Dr. Jing Zhang from Horticulture Science Research Center, Northwest A&F University (NWAFU), for providing professional technical assistance, and Arij Khalil from Department of Horticulture, Ghazi University, for the language modifications. We also thank the High-Performance Computing (HPC) platform of NWAFU for providing computing resources. This work was supported by the National Natural Science Foundation of China (32402541), the National Science Fund for Distinguished Young Scholars (32325045), China Postdoctoral Science Foundation (2024M752635), Shaanxi Postdoctoral Research Project (2023BSHEDZZ123), the Key S&T Special Projects of Shaanxi Province, China (2020zdzx03-01-02), and Chinese Universities Scientific Fund (2452023067).

Contributor Information

Zhongxing Li, State Key Laboratory for Crop Stress Resistance and High-Efficiency Production/Shaanxi Key Laboratory of Apple, College of Horticulture, Northwest A&F University, Yangling, Shaanxi 712100, China; College of Grassland Agriculture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Cai Gao, College of Grassland Agriculture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Tianle Fan, State Key Laboratory for Crop Stress Resistance and High-Efficiency Production/Shaanxi Key Laboratory of Apple, College of Horticulture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Yilin Cui, College of Grassland Agriculture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Zeyuan Liu, State Key Laboratory for Crop Stress Resistance and High-Efficiency Production/Shaanxi Key Laboratory of Apple, College of Horticulture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Lei Li, State Key Laboratory for Crop Stress Resistance and High-Efficiency Production/Shaanxi Key Laboratory of Apple, College of Horticulture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Qian Qian, State Key Laboratory for Crop Stress Resistance and High-Efficiency Production/Shaanxi Key Laboratory of Apple, College of Horticulture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Mengjie Cheng, State Key Laboratory for Crop Stress Resistance and High-Efficiency Production/Shaanxi Key Laboratory of Apple, College of Horticulture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Xiangqiang Zhan, State Key Laboratory for Crop Stress Resistance and High-Efficiency Production/Shaanxi Key Laboratory of Apple, College of Horticulture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Chundong Niu, State Key Laboratory for Crop Stress Resistance and High-Efficiency Production/Shaanxi Key Laboratory of Apple, College of Horticulture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Fengwang Ma, State Key Laboratory for Crop Stress Resistance and High-Efficiency Production/Shaanxi Key Laboratory of Apple, College of Horticulture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Peizhi Yang, College of Grassland Agriculture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Qingmei Guan, State Key Laboratory for Crop Stress Resistance and High-Efficiency Production/Shaanxi Key Laboratory of Apple, College of Horticulture, Northwest A&F University, Yangling, Shaanxi 712100, China.

Author contributions

Q.G. and P.Y. conceived and supervised the project; F.M. provided the Malus samples. ZX.L., C.G., Y.C., and X.Z. performed bioinformatics and statistical analyses; ZX.L., C.G., T.F., Y.C., ZY.L., L.L., Q.Q., M.C., and C.N. conducted the experiments. ZX.L., C.G., and Q.G. wrote the manuscript. All authors read and approved of the content.

Data availability statement

The raw whole-genome resequencing data of 134 Malus accessions were retrieved from the NCBI Sequence Read Archive (SRA) with the accession number PRJNA728537, and details in Table S1. Sequence data used in this article can be found in GenBank under the following accession numbers: XM_008339732.3 (MdPAT10); PQ100214 (MdDof2.4); XM_008354691.3 (MdMYB7).

Conflict of Interests

The authors declare no competing interests.

Supplementary information

Supplementary data is available at Horticulture Research online.

References

  • 1. FAO FAOSTAT. 2023.  https://www.fao.org/faostat/en/#data/QCL
  • 2. Hisanaga  A, Mukai  R, Sakao  K. et al.  Anti-inflammatory effects and molecular mechanisms of 8-prenyl quercetin. Mol Nutr Food Res. 2016;60:1020–32 [DOI] [PubMed] [Google Scholar]
  • 3. Tijjani  H, Zangoma  MH, Mohammed  ZS. et al.  Polyphenols: classifications, biosynthesis and bioactivities. In: Egbuna  C, Dable Tupas  G, eds. Functional Foods and Nutraceuticals. Springer International Publishing: Cham, 2020,389–414 [Google Scholar]
  • 4. Lin  Q, Chen  J, Liu  X. et al.  A metabolic perspective of selection for fruit quality related to apple domestication and improvement. Genome Biol. 2023;24:95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Zhao  J, Liu  G, Chang  R. et al.  Diversity of flesh polyphenols and their progressive dilution during fruit expansion in Malus germplasm. Sci Hortic. 2015;197:461–9 [Google Scholar]
  • 6. Liu  Y, Qian  J, Li  J. et al.  Hydroxylation decoration patterns of flavonoids in horticultural crops: chemistry, bioactivity, and biosynthesis. Hortic Res. 2022;9:uhab068 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Marchiosi  R, dos  Santos  WD, Constantin  RP. et al.  Biosynthesis and metabolic actions of simple phenolic acids in plants. Phytochem Rev. 2020;19:865–906 [Google Scholar]
  • 8. Han  Y, Vimolmangkang  S, Soria-Guerra  RE. et al.  Introduction of apple ANR genes into tobacco inhibits expression of both CHI and DFR genes in flowers, leading to loss of anthocyanin. J Exp Bot. 2012;63:2437–47 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Liao  L, Vimolmangkang  S, Wei  G. et al.  Molecular characterization of genes encoding leucoanthocyanidin reductase involved in proanthocyanidin biosynthesis in apple. Front Plant Sci. 2015;6:243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Xu  W, Dubos  C, Lepiniec  L. Transcriptional control of flavonoid biosynthesis by MYB–bHLH–WDR complexes. Trends Plant Sci. 2015;20:176–85 [DOI] [PubMed] [Google Scholar]
  • 11. Wang  N, Liu  W, Zhang  T. et al.  Transcriptomic analysis of red-fleshed apples reveals the novel role of MdWRKY11 in flavonoid and anthocyanin biosynthesis. J Agric Food Chem. 2018;66:7076–86 [DOI] [PubMed] [Google Scholar]
  • 12. Sun  Q, Jiang  S, Zhang  T. et al.  Apple NAC transcription factor MdNAC52 regulates biosynthesis of anthocyanin and proanthocyanidin through MdMYB9 and MdMYB11. Plant Sci. 2019;289:110286 [DOI] [PubMed] [Google Scholar]
  • 13. Bai  S, Tuan  PA, Saito  T. et al.  Epigenetic regulation of MdMYB1 is associated with paper bagging-induced red pigmentation of apples. Planta. 2016;244:573–86 [DOI] [PubMed] [Google Scholar]
  • 14. An  J-P, Wang  X-F, Zhang  X-W. et al.  An apple MYB transcription factor regulates cold tolerance and anthocyanin accumulation and undergoes MIEL1-mediated degradation. Plant Biotechnol J. 2020;18:337–53 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Zhou  L-J, Li  Y-Y, Zhang  R-F. et al.  The small ubiquitin-like modifier E3 ligase MdSIZ1 promotes anthocyanin accumulation by sumoylating MdMYB1 under low-temperature conditions in apple. Plant Cell Environ. 2017;40:2068–80 [DOI] [PubMed] [Google Scholar]
  • 16. Fukata  Y, Fukata  M. Protein palmitoylation in neuronal development and synaptic plasticity. Nat Rev Neurosci. 2010;11:161–75 [DOI] [PubMed] [Google Scholar]
  • 17. Hemsley  PA, Kemp  AC, Grierson  CS. The TIP GROWTH DEFECTIVE1 S-acyl transferase regulates plant cell growth in Arabidopsis. Plant Cell. 2005;17:2554–63 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Chai  S, Ge  F-R, Zhang  Y. et al.  S-acylation of CBL10/SCaBP8 by PAT10 is crucial for its tonoplast association and function in salt tolerance. J Integr Plant Biol. 2020;62:718–22 [DOI] [PubMed] [Google Scholar]
  • 19. Zhou  L-Z, Li  S, Feng  Q-N. et al.  PROTEIN S-ACYL TRANSFERASE10 is critical for development and salt tolerance in Arabidopsis. Plant Cell. 2013;25:1093–107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Batistič  O. Genomics and localization of the Arabidopsis DHHC-cysteine-rich domain S-acyltransferase protein family. Plant Physiol. 2012;160:1597–612 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Li  Y, Scott  R, Doughty  J. et al.  Protein S-acyltransferase 14: a specific role for palmitoylation in leaf senescence in Arabidopsis. Plant Physiol. 2016;170:415–28 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Liu  F, Qu  P-Y, Li  J-P. et al.  Arabidopsis protein S-acyl transferases positively mediate BR signaling through S-acylation of BSK1. Proc Natl Acad Sci. 2024;121:e2322375121 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Jiang  H, Ma  Q, Zhong  M. et al.  The apple palmitoyltransferase MdPAT16 influences sugar content and salt tolerance via an MdCBL1–MdCIPK13–MdSUT2.2 pathway. Plant J. 2021a;106:689–705 [DOI] [PubMed] [Google Scholar]
  • 24. Chen  P, Li  Z, Zhang  D. et al.  Insights into the effect of human civilization on Malus evolution and domestication. Plant Biotechnol J. 2021;19:2206–20 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Duan  N, Bai  Y, Sun  H. et al.  Genome re-sequencing reveals the history of apple and supports a two-stage model for fruit enlargement. Nat Commun. 2017;8:249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Zhang  Z, Ersoz  E, Lai  C-Q. et al.  Mixed linear model approach adapted for genome-wide association studies. Nat Genet. 2010b;42:355–60 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Zhou  X, Stephens  M. Genome-wide efficient mixed-model analysis for association studies. Nat Genet. 2012;44:821–4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Mo  Y, Nagel  C, Taylor  LP. Biochemical complementation of chalcone synthase mutants defines a role for flavonols in functional pollen. Proc Natl Acad Sci. 1992;89:7213–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Su  Z, Sun  M, Cai  Z. et al.  Identification and expression analysis of chlorogenic acid biosynthesis key gene PpHCT in peach. Hortic Plant J. 2023;9:670–80 [Google Scholar]
  • 30. Chagné  D, Krieger  C, Rassam  M. et al.  QTL and candidate gene mapping for polyphenolic composition in apple fruit. BMC Plant Biol. 2012;12:12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Khan  SA, Chibon  P-Y, de  Vos  RCH. et al.  Genetic analysis of metabolites in apple fruits indicates an mQTL hotspot for phenolic compounds on linkage group 16. J Exp Bot. 2012;63:2895–908 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Fornalé  S, Lopez  E, Salazar-Henao  JE. et al.  AtMYB7, a new player in the regulation of UV-sunscreens in Arabidopsis thaliana. Plant Cell Physiol. 2014;55:507–16 [DOI] [PubMed] [Google Scholar]
  • 33. Kalinowska  M, Bielawska  A, Lewandowska-Siwkiewicz  H. et al.  Apples: content of phenolic compounds vs. variety, part of apple and cultivation model, extraction of phenolic compounds, biological properties. Plant Physiol Biochem. 2014;84:169–88 [DOI] [PubMed] [Google Scholar]
  • 34. Zheng  H-Z, Kim  Y-I, Chung  S-K. A profile of physicochemical and antioxidant changes during fruit growth for the utilisation of unripe apples. Food Chem. 2012;131:106–10 [Google Scholar]
  • 35. McClure  KA, Gong  Y, Song  J. et al.  Genome-wide association studies in apple reveal loci of large effect controlling apple polyphenols. Hortic Res. 2019;6:107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. McGhie  TK, Hunt  M, Barnett  LE. Cultivar and growing region determine the antioxidant polyphenolic concentration and composition of apples grown in New Zealand. J Agric Food Chem. 2005;53:3065–70 [DOI] [PubMed] [Google Scholar]
  • 37. Liao  L, Zhang  W, Zhang  B. et al.  Unraveling a genetic roadmap for improved taste in the domesticated apple. Mol Plant. 2021;14:1454–71 [DOI] [PubMed] [Google Scholar]
  • 38. Jiao  T, Huang  Y, Wu  Y-L. et al.  Functional diversity of subgroup 5 R2R3-MYBs promoting proanthocyanidin biosynthesis and their key residues and motifs in tea plant. Hortic Res. 2023;10:uhad135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Jiang  S, Sun  Q, Zhang  T. et al.  MdMYB114 regulates anthocyanin biosynthesis and functions downstream of MdbZIP4-like in apple fruit. J Plant Physiol. 2021b;257:153353 [DOI] [PubMed] [Google Scholar]
  • 40. Comino  C, Hehn  A, Moglia  A. et al.  The isolation and mapping of a novel hydroxycinnamoyltransferase in the globe artichoke chlorogenic acid pathway. BMC Plant Biol. 2009;9:30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Navarre  DA, Payyavula  RS, Shakya  R. et al.  Changes in potato phenylpropanoid metabolism during tuber development. Plant Physiol Biochem. 2013;65:89–101 [DOI] [PubMed] [Google Scholar]
  • 42. Sonnante  G, D’Amore  R, Blanco  E. et al.  Novel hydroxycinnamoyl-coenzyme a quinate transferase genes from artichoke are involved in the synthesis of chlorogenic acid. Plant Physiol. 2010;153:1224–38 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Zhang  Y, Li  P, Cheng  L. Developmental changes of carbohydrates, organic acids, amino acids, and phenolic compounds in ‘Honeycrisp’ apple flesh. Food Chem. 2010a;123:1013–8 [Google Scholar]
  • 44. Chen  S. Ultrafast one-pass FASTQ data preprocessing, quality control, and deduplication using fastp. iMeta. 2023;2:e107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. McKenna  A, Hanna  M, Banks  E. et al.  The genome analysis toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20:1297–303 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Li  Z, Wang  L, He  J. et al.  Chromosome-scale reference genome provides insights into the genetic origin and grafting-mediated stress tolerance of Malus prunifolia. Plant Biotechnol J. 2022;20:1015–7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Price  AL, Patterson  NJ, Plenge  RM. et al.  Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet. 2006;38:904–9 [DOI] [PubMed] [Google Scholar]
  • 48. Alexander  DH, Lange  K. Enhancements to the ADMIXTURE algorithm for individual ancestry estimation. BMC Bioinformatics. 2011;12:246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Zhang  C, Dong  S-S, Xu  J-Y. et al.  PopLDdecay: a fast and effective tool for linkage disequilibrium decay analysis based on variant call format files. Bioinformatics. 2019;35:1786–8 [DOI] [PubMed] [Google Scholar]
  • 50. Benjamini  Y, Hochberg  Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc. 1995;57:289–300 [Google Scholar]
  • 51. Dong  S-S, He  W-M, Ji  J-J. et al.  LDBlockShow: a fast and convenient tool for visualizing linkage disequilibrium and haplotype blocks based on variant call format files. Brief Bioinform. 2021;22:bbaa227 [DOI] [PubMed] [Google Scholar]
  • 52. Niu  C, Jiang  L, Cao  F. et al.  Methylation of a MITE insertion in the MdRFNR1-1 promoter is positively associated with its allelic expression in apple in response to drought stress. Plant Cell. 2022;34:3983–4006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53. An  X-H, Tian  Y, Chen  K-Q. et al.  MdMYB9 and MdMYB11 are involved in the regulation of the JA-induced biosynthesis of anthocyanin and proanthocyanidin in apples. Plant Cell Physiol. 2015;56:650–62 [DOI] [PubMed] [Google Scholar]
  • 54. Nei  M, Li  WH. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proc Natl Acad Sci. 1979;76:5269–73 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Tajima  F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics. 1989;123:585–95 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56. Danecek  P, Auton  A, Abecasis  G. et al.  The variant call format and VCFtools. Bioinformatics. 2011;27:2156–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57. Tian  F, Yang  D-C, Meng  Y-Q. et al.  PlantRegMap: charting functional regulatory maps in plants. Nucleic Acids Res. 2019;48:D1104–13 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Web_Material_uhae349
web_material_uhae349.zip (4.4MB, zip)

Data Availability Statement

The raw whole-genome resequencing data of 134 Malus accessions were retrieved from the NCBI Sequence Read Archive (SRA) with the accession number PRJNA728537, and details in Table S1. Sequence data used in this article can be found in GenBank under the following accession numbers: XM_008339732.3 (MdPAT10); PQ100214 (MdDof2.4); XM_008354691.3 (MdMYB7).

Articles from Horticulture Research are provided here courtesy of Oxford University Press

RESOURCES