An InDel variant in the promoter of the NAC transcription factor MdNAC18.1 plays a major role in apple fruit ripening

Abstract

A complex regulatory network governs fruit ripening, but natural variations and functional differentiation of fruit ripening genes remain largely unknown. Utilizing a genome-wide association study (GWAS), we identified the NAC family transcription factor MdNAC18.1, whose expression is closely associated with fruit ripening in apple (Malus × domestica Borkh.). MdNAC18.1 activated the transcription of genes related to fruit softening (Polygalacturonase, PG) and ethylene biosynthesis (1-aminocyclopropane-1-carboxylic acid synthase, ACS), thereby promoting fruit ripening of apple and tomato (Solanum lycopersicum). There were two single-nucleotide polymorphisms (SNP-1,545 and SNP-2,002) and a 58-bp insertion-deletion (InDel-58) in the promoter region of MdNAC18.1. Among these, InDel-58 serves as the main effector in activating the expression of MdNAC18.1 and driving fruit ripening. InDel-58 determines the binding affinity of the class D MADS-box protein AGAMOUS-LIKE 11 (MdAGL11), a negative regulator of fruit ripening. The InDel-58 deletion in the early-ripening genotype reduces the inhibitory effect of MdAGL11 on MdNAC18.1. Moreover, MdNAC18.1 and its homologous genes originated from a common ancestor across 61 angiosperms, with functional diversification attributed to tandem replications that occurred in basal angiosperms. In summary, our study revealed how a set of natural variations influence fruit ripening and explored the functional diversification of MdNAC18.1 during evolution.

Figure 1.

MdNAC18.1 is associated with fruit ripening in apple. A) A Manhattan plot of fruit ripening. Each point represents a SNP. The chromosome number is plotted on the x axis, and the −log₁₀ value of the P value is plotted on the y axis. The gray horizontal dashed line indicates the genome-wide significance threshold. Chr, chromosome. B) A quantile–quantile (Q–Q) plot of fruit ripening. The x axis indicates the expected −log₁₀ value of the P value, and the y axis represents the observed −log₁₀ value of the P value. C) Local Manhattan plot of fruit ripening using GWAS with SNPs (top) and gene distribution (bottom). D) LD plot of the region surrounding the MdNAC18.1. E) FPKM of 4 candidate genes in “Geneva Early” and “Hanfu” fruits throughout development. Error bars indicate Sd (n = 3). The expression level of MdNAC18.1 in F) early-ripening and G) late-ripening cultivars throughout fruit development. DAP, days after pollination.

Figure 2.

Overexpressing MdNAC18.1 in tomatoes and apples accelerates fruit ripening. A) Morphology of wild type (AC) and MdNAC18.1 OE tomato fruits at the indicated stages. Scale bars, 2 cm. B) Statistics of days to color breaker of fruits in A). C–F) Chlorophyll and carotenoid levels of fruit ripening in wild type and MdNAC18.1 OE fruits. G and H) Determination of ethylene (ETH) production rate and fruit firmness of wild type and MdNAC18.1 OE tomato fruits at 4 indicated stages. I) Representative photographs showing the apple fruits of “Gala” at 120 DAP infiltrated with the indicated plasmids. pK7, empty overexpression vector; pK7-MdNAC18.1, overexpression construct for MdNAC18.1; pTRV, empty virus-induced gene silencing vector; pTRV-MdNAC18.1, knockdown construct for MdNAC18.1. Photograph was taken 6 days after infiltration. Scale bars, 2 cm. J–M) Quantification of the fruit firmness and ethylene production rate in apples shown in I). Error bars indicate Sd (n = 15 in B, n = 3 in C–H, L–M, n = 30 in J–K). Student’s two-tailed t-test is used for statistical analyses, and statistically significant differences are indicated by **P < 0.01 and ***P < 0.001. dpa, days post-anthesis; OE, overexpression.

Figure 3.

MdNAC18.1 regulates the expression of ripening-related genes. A) Volcano plots of fold changes of DEGs between WT and MdNAC18.1 OE at 37 and 42 dpa stages. Red and blue dots represent upregulated and downregulated genes, respectively (P < 0.05, absolute fold change >2.0). Three biological replicates were used for each sample at each time point. B) Venn diagram showing the overlap of ripening-related genes and MdNAC18.1-regulated genes. C) Distribution of ripening-related genes regulated by MdNAC18.1. Genes upregulated during ripening (272) are shown on the left, and those downregulated during ripening (1,050) are shown on the right. D) Heatmap of ripening-related genes regulated by MdNAC18.1. Each gene’s average FPKM (fragments per kilobase of exon per million fragments mapped) values at 37 and 42 dpa stages were analyzed, and expression levels were normalized relative to the WT. E) Relative expression levels of SlACS2, SlPSY1, SlPG2a, SlERF2, SlSGR1, and SlGgpps2 in WT and MdNAC18.1 OE fruits at 32, 37, and 42 dpa stages. F) ChIP-qPCR showing MdNAC18.1 enrichment in SlPG2a (top) and SlACS2 (bottom) promoters in MdNAC18.1-GFP fruits. WT and No-antibody serve as the negative controls. G) Dual-LUC assay for MdNAC18.1 activating the SlPG2a and SlACS2. Empty vector was included as a negative control. Error bars indicate Sd (n = 3 in E, n = 8 in G). Student’s two-tailed t-test is used for statistical analyses, and statistically significant differences are indicated by ***P < 0.001. WT, wild type; dpa, days post-anthesis; DEGs, differentially expressed genes; OE, overexpression.

Figure 4.

Analysis of variations in the promoter of MdNAC18.1. A) Distribution of SNPs and InDel in the Sanger-sequenced promoters of MdNAC18.1 in 153 out of 204 Malus accessions. The dotted lines represent the region of InDel-58. B) Sanger sequencing results of SNP-1,545 and SNP-2,002 in the promoters of MdNAC18.1. Red markers represent the SNP locations. C) Distribution of InDel-58 among Malus accessions detected by RT-PCR. D) Association of 3 natural variations and apple fruit ripening. Whiskers represent the range of data distribution. Circles indicate outlier data. MdNAC18.1^E represents the genotype T or A in SNP-2,002, T or A in SNP-1,545, and 58-bp deletion in InDel-58. MdNAC18.1^L represents the genotype C or G in SNP-2,002, A or T in SNP-1,545, and 58-bp insertion in InDel-58. Median values are indicated in bold lines. Statistical significance is estimated using two-tailed Wilcoxon tests (***P < 0.001, ****P < 0.0001).

Figure 5.

Natural variations in the MdNAC18.1 promoter affect fruit ripening. A) Diagram showing MdNAC18.1 promoters in early-ripening (MdNAC18.1^E) and late-ripening (MdNAC18.1^L) cultivars containing the 2 SNPs and InDel-58, and various mutant versions of the 2 promoters. MdNAC18.1^mE1, MdNAC18.1^mE2, and MdNAC18.1^mE3 represent substitution of T to C, T to A, and deletion to insertion at position SNP-2,002, SNP-1,545, and InDel-58, in the background of MdNAC18.1^E, respectively. MdNAC18.1^mL1, MdNAC18.1^mL2, and MdNAC18.1^mL3 represent substitution of C to T, A to T, and insertion to deletion at position SNP-2,002, SNP-1,545, and InDel-58, in the background of MdNAC18.1^L, respectively. Mutations were marked in red. B) Transcriptional activity of the promoters shown in A) as determined by promoter:LUC assay. C) Representative fruits of WT and various transgenic tomatoes expressing MdNAC18.1 driven by the native promoters shown in A) at different stages. D) Days to color breaker in WT and various transgenic tomatoes shown in C). Error bars indicate Sd (n = 6). Asterisks indicate the significant differences (one-way ANOVA with Dunnett’s multiple comparisons test; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns, not significant). dpa, days post-anthesis.

Figure 6.

Functional analysis of natural variations in the MdNAC18.1 promoter through transient transformation assays. A) Transient overexpression of 8 constructs combining distinct MdNAC18.1:CDS shown in Fig. 5A in apple fruits of “Gala” at 120 DAP. Photograph was taken 6 days after infiltration. Scale bars, 2 cm. B and C) Changes in fruits firmness and ethylene production rate in apples infiltrated with the constructs shown in Fig. 5A. Error bars indicate Sd (n = 30 in B, n = 3 in C). Asterisks indicate the significant differences (one-way ANOVA with Dunnett’s multiple comparisons test; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns, not significant).

Figure 7.

MdAGL11 represses MdNAC18.1 expression by binding to CArG motif in the InDel-58 region of MdNAC18.1 promoter. MdAGL11 represses MdNAC18.1 expression by binding to CArG motif in the InDel-58 region of MdNAC18.1 promoter. A) The promoter structure of MdNAC18.1^E and MdNAC18.1^L. Blue and yellow boxes represent CArG motif (CAATTAAATTATG) and SNP, respectively. The dotted line represents InDel. Red-marked bases represent the 58-bp insertion. B) Y1H assay demonstrates that MdAGL11 can bind to the promoters of MdNAC18.1 in early- and late-ripening cultivars. SD/-Leu, SD medium without Leu; SD/-Leu/Aba³⁰⁰, SD medium without Leu supplemented with 300 ng mL⁻¹ AbA. MdNAC18.1⁻⁵⁸ (−657∼−720 bp) and MdNAC18.1⁺⁵⁸ (−657∼−778 bp) shown in A) were used. C) EMSA shows the direct binding of MBP-MdAGL11 to the MdNAC18.1 promoter. The probe was biotin-labeled MdNAC18.1⁺⁵⁸ (−657∼−717 bp) containing CArG motif. Mut represents the mutant probe. Binding sites in mutant probe were replaced with AAAAAAAAAAAAAA. D) MdAGL11 represses MdNAC18.1 expression as determined by Dual-LUC assay in apple calli. MdNAC18.1^m−58. and MdNAC18.1^m+58 represents CArG motifs on MdNAC18.1 promoter were replaced with AAAAAAAAAAAAAA. Empty vector was included as a negative control. Error bars indicate Sd (n = 6). Different letters indicate differences (P < 0.05) determined by one-way ANOVA with Tukey’s multiple comparisons test.

Figure 8.

Phylogenetic tree and synteny network of MdNAC18.1 and its homologous in 61 species. Phylogenetic tree and synteny network of MdNAC18.1 and its homologous in 61 species. Maximum-likelihood gene tree for MdNAC18.1 and its homologous genes. Terminal branches of the outer circle represent 2 broad categories: clade A (light yellow) and clade B (blue). The genes of the collinearity network in the inner circle belongs to rosids (pink), asterids (yellow), monocots (green), magnoliids (sky blue), basal eudicots (orange), and basal angiosperm (deep blue).

Figure 9.

A proposed model of MdNAC18.1 in regulating apple fruit ripening. As a transcriptional activator, MdNAC18.1 promotes apple fruit ripening via activating the expression of ripening-related genes, including directly accelerating expression of fruit softening gene PG and ethylene biosynthesis gene ACS. Three variations within the promoter of MdNAC18.1 are associated with fruit ripening, with InDel-58 playing a major role. In the early-ripening cultivars, the InDel-58 is a deletion in MdNAC18.1 promoter, which leads to attenuated repression of MdNAC18.1 by MdAGL11, whose homolog is a negative regulator for fruit ripening. In the late-ripening cultivars, the InDel-58 is an insertion that results in 2 binding motifs of MdAGL11, thereby increasing the repression of MdNAC18.1.