Skip to main content
NCBI home page
Plant Physiology logoLink to Plant Physiology
. 2022 Feb 16;189(1):315–328. doi: 10.1093/plphys/kiac063

High mobility group A3 enhances transcription of the DNA demethylase gene SlDML2 to promote tomato fruit ripening

Zhifei Li 1, Ying Pi 2, Junmiao Fan 3, Xinxin Yang 4, Changsheng Zhai 5, Hong Chen 6, Feng Wang 7, Jing Ding 8, Tingting Gu 9, Yi Li 10,†, Han Wu 11,✉,†
PMCID: PMC9070846  PMID: 35171288

Abstract

DNA methylation plays an important role in regulating tomato (Solanum lycopersicum) fruit ripening. Although SlDML2, a DNA demethylase (DML) gene, is critically involved in tomato fruit ripening, little is known about genes that regulate its expression. Using yeast one-hybrid screening, we identified a High Mobility Group A protein, named SlHMGA3, and demonstrated its binding activity to the AT-rich region of the SlDML2 promoter. We produced slhmga3 tomato mutants using CRISPR/Cas9 and observed that slhmga3 fruit reached the breaker stage much later than fruit from the wild-type. We further demonstrated that at the initiation stage of fruit ripening, the increased expression of SlDML2 and ethylene biosynthetic and signaling genes was significantly delayed in slhmga3 fruit, along with delays in ethylene production and demethylation and activation of ripening-associated transcription factor genes. Our results demonstrate that SlHMGA3 plays a role in enhancing SlDML2 expression, and its effects on tomato fruit ripening are largely through DNA demethylation of ripening-associated transcription factor genes.

A chromatin architectural protein binds to the promoter of a DNA demethylase gene to influence initiation of tomato fruit ripening through demethylation of ripening-related genes.

Introduction

Genomic DNA methylation is a type of epigenetic modification that involves the addition of a methyl group to carbon 5 of cytosine, which can play a role in transcriptional regulation and chromosome interactions. As a result, DNA methylation can affect many aspects of plant growth and development, as well as plant responses to biotic and abiotic stresses (Chan et al., 2005; Mirouze et al., 2012; Saze et al., 2012; Yang et al., 2019).

DNA methylation is also associated with fleshy fruit ripening (Tang et al., 2020; Nie, 2021). In nonclimacteric fruits, strawberry (Fragaria × ananassa) fruit undergo an overall loss of DNA methylation during ripening which is caused by a downregulation of the RNA-directed DNA methylation pathway rather than the upregulation of DNA demethylase (DML) genes. Inhibition of DNA methylation causes early ripening of strawberry fruit (Cheng et al., 2018). In sweet orange (Citrus sinensis) fruit, although the whole-genome DNA methylation is increased during fruit ripening, inhibition of DNA methylation delays fruit ripening (Huang et al., 2019). Tomato (Solanum lycopersicum) is a model plant to study climacteric fruit ripening. In tomato, the overall level of DNA methylation is reduced in pericarp tissues during fruit ripening, and inhibition of DNA methylation in immature fruit induces premature fruit ripening (Zhong et al., 2013). Liu et al. (2015a, 2015b) and Lang et al. (2017) have shown that the upregulation of a DML gene SlDML2 is critical for the reduction of genome-wide DNA methylation levels during tomato fruit ripening. The knockdown or knockout of SlDML2 dramatically delays ripening. The active DNA demethylation induced by SlDML2 is required for both the activation of ripening-induced genes and the inhibition of ripening-repressed genes involved in carotenoid, ethylene biosynthesis and ethylene signaling, and known ripening-associated transcription factors (Liu et al., 2015a, 2015b; Lang et al., 2017). Zhou et al. (2019) report that a tomato RNA demethylase SlALKBH2 mediates the ripening-associated dynamics of mRNA m6A methylation, and regulates the stability of SlDML2 mRNA to affect fruit ripening. Li et al. (2020a, 2020b) demonstrate that histone demethylase SlJMJ6 promotes tomato fruit ripening by removing H3K27 methylation of ripening-related genes. SlJMJ6 can activate the expression of SlDML2 gene by demethylating its histone of chromatin rather than carbon 5 of cytosine.

High Mobility Group (HMG) proteins are nonhistone chromatin architectural proteins composed of three families: HMGA, HMGB, and HMGN (Bustin, 2001; Catez and Hock, 2010). The HMGA family in mammals includes three proteins: HMGA1a, HMGA1b, and HMGA2. HMGA is a transcriptional regulatory factor necessary for embryonic development, and its expression is downregulated during postembryonic development (Lanahan et al., 1992; Zhou et al., 1995). However, HMGA levels are often re-expressed during oncogenesis (Reeves, 2001; Takaha et al., 2004). As architectural transcription factors, each HMGA protein in mammals has three or four conserved AT-hook DNA-binding motifs with a characteristic stretch of nine residues containing the invariant core peptide sequence RGRP (Aravind and Landsman, 1998; Cui et al., 2005), which can bind to AT-rich DNA via the minor groove and initiate four-way functioning of DNA to regulate gene expression (Aravind and Landsman, 1998; Zhang et al., 2003). In addition, HMGA proteins can specifically interact with other proteins, most of which are transcription factors, to induce the formation of stereo-specific complexes called the “enhanceosome” on gene promoter or enhancer regions, and in turn activate or inhibit gene transcription (Reeves, 2001; Sgarra et al., 2018). HMGA proteins can be modified through many biochemical processes, which greatly affect their ability to interact with chromatin, DNA, and other proteins (Sgarra et al, 2006; Zhang and Wang, 2010).

In angiosperms, HMGA proteins contain AT-hook motifs at the C-terminus. An additional highly conserved GH1 domain usually found in histone H1 occurs at the N-terminus of HMGAs, which is not present in their mammalian homologs (Jerzmanowski et al., 2000; Klosterman and Hadwiger, 2002; Kotlinski et al., 2017). Some plant HMGAs have been shown to bind to an AT-rich DNA sequence in vitro. For example, pea (Pisum sativum) HMGA can bind to a positive regulatory region of the plastocyanin gene promoter (Pwee et al., 1994; Webster et al., 1997). Rice (Oryza sativa) HMGA protein PF1 can stimulate binding of the transcriptional activator GT-2 to the PHYA gene promoter (Nietosotelo et al., 1994; Martinez-Garcia and Quail, 1999). The maize (Zea mays) HMGA protein has a high capacity to bind AT-rich DNA, and a SUC1-associated kinase can strongly phosphorylate HMGA to reduce its capacity to bind to AT-rich DNA (Zhao et al., 2009). Arabidopsis (Arabidopsis thaliana) has four HMGA proteins: GH1-HMGA1 (HON4), HON5, GH1-HMGA3, and GH1-HMGA4. GH1-HMGA1 and GH1-HMGA2 have 50% identity and contain six AT-hook motifs, GH1-HMGA3 contains four AT-hook motifs, whereas GH1-HMGA4 has an N-terminal GH1 domain, but contains C-terminus without recognizable AT-hook (Kotlinski et al., 2017; Zhao et al., 2021). In angiosperms, only Arabidopsis HMGA genes have been functionally analyzed. GH1-HMGA1 is involved in DNA repair and telomere stability, and the gh1-hmga1 mutant exhibits developmental and growth defects (Charbonnel et al., 2018). Zhao et al. (2021) report that members of the GH1-HMGA family in Arabidopsis act as chromatin architecture modulators, redundantly promoting the floral transition through repression of FLC by precluding FLC gene looping.

DNA methylation plays an important role in tomato fruit ripening. The upregulation of SLDML2 is a key factor that decreases the global DNA methylation level to induce tomato fruit ripening. To clarify the protein that directly influences SlDML2 expression, we used the yeast one-hybrid (Y1H) system to identify proteins that may bind to the SlDML2 gene promoter. The HMGA family protein SlHMGA3 was identified. Electrophoretic mobility shift assay (EMSA) and Dual-Luciferase (LUC) Reporter Assay were performed to test the binding and activated capacity of SlHMGA3 in modulating SlDML2 expression. Then, we obtained slhmag3 mutants using the CRISPR/Cas9 system to analyze SlHMGA3 function during fruit ripening. We further analyzed the expression levels of ethylene biosynthetic and signaling genes, as well as the methylation levels of ripening-associated transcription factor gene promoters in slhmga3 mutants. These results provide important insights into the molecular basis of SlDML2-mediated tomato fruit ripening.

Results

SlHMGA3 was identified to bind to the SlDML2 promoter through Y1H screening

We performed a Y1H screening assay to identify proteins capable of binding to the SlDML2 promoter, in which 2-kb promoter sequences of SlDML2 were used as a bait to screen target proteins in a cDNA library constructed from tomato fruit pericarp tissues at the breaker stage. After sequencing of thirty positive clones, among which 14 clones encoded the same protein, HMG-Y-related protein A (XP_004236944), which belongs to an uncharacterized HMGA family in tomato. Four HMGA proteins were identified in both tomato and Arabidopsis, and phylogenetic analysis showed that XP_004236944 was clustered together with AtGH1-HMGA3 in Arabidopsis (Figure 1A). Therefore, we designated XP_004236944 as SlHMGA3. HMGA proteins usually have a GH1 domain at the N-terminus and varying numbers of AT-hook motifs at the C-terminus. Unlike other HMGA proteins with three, six or eight AT-hook motifs, we found that SlHMGA3 and AtGH1-HMGA3 had four AT-hook motifs (Figure 1B). Multi-sequence alignment showed that SlHMGA3 has high sequence similarity to other canonical plant HMGA proteins that can bind to AT-rich DNA nucleotides in vitro (Figure 1C), suggesting that SlHMGA3 may be capable of binding to DNA sequences.

Figure 1.

Figure 1

Open in a new tab

SlHMGA3 in tomato is a canonical HMGA protein. A, SlHMGA3 (XP_004236944) in tomato is clustered together with AtGH1-HMGA3 in Arabidopsis. The GenBank accession numbers of the HMGA-like family in tomato (S. lycopersicum) are XP_004235775, XP_004236944, XP_004250181, and XP_004240031. The TAIR accession numbers of the HMGA-like family in Arabidopsis are AT3G18035 (GH1-HMGA1), AT1G48620 (GH1-HMGA2), AT1G14900 (GH1-HMGA3), and AT5G08780 (GH1-HMGA-related 4). B, SlHMGA3 has one H15 domain and four AT-hook motifs. The H15 domain is represented by green, and the AT-hook motifs are represented by blue. C, SlHMGA3 has high sequence similarity to other plant canonical HMGA proteins. The H15 domain is marked with a blue box, and the four AT-hook motifs are marked with red boxes. GenBank accession numbers of proteins are: S. lycopersicum, SlHMGA3 (XP_004236944.1); A. thaliana, AtGH1-HMGA3 (CAA71797.1); O. sativa, OsHMGA (AAA33914.1); Z. mays, ZmHMGA1 (CAB40848.2), and ZmHMGA2 (AAG00601.1); C. gladiata (Japanese jack bean), CgHMGA1 (BAA13133.1), and CgHMGA2 (BAA11767.1); A. sativa, AsHMGA (AAA32718.1); P. sativum, PsHMGA (CAA67752.1); B. napus, BnHMGA (AAF22135.1); G. max, GmHMGA1 (Q00423.1), and GmHMGA2 (NP_001347278.1).

SlHMGA3 is a nuclear protein that is ubiquitously expressed in different organs

To study the subcellular localization of SlHMGA3, the full-length open reading frame (ORF) of SlHMGA3 was fused to the N-terminus of the green fluorescent protein (GFP) reporter under the control of CaMV35S promoter, and the recombinant vector (SlHMGA3-GFP) was introduced into the Nicotiana benthamiana leaf epidermis via Agrobacterium tumefaciens-mediated transient transformation. Microscopic visualization showed that green fluorescence of the SlHMGA3-GFP fusion protein was present only in the nucleus of N. benthamiana leaves (Figure 2A), demonstrating that SlHMGA3 is a nuclear protein. Then, the yeast two-hybrid (Y2H) result showed that yeast strains containing a DNA-binding domain fused to SlHMGA3 (SlHMGA3-BD) and an activation domain (AD) could not grow on synthetic defined (SD) Leu–Trp–His medium or SD–Ade–Leu–Trp–His medium (Supplemental Figure S1), demonstrating that SlHMGA3 is unable to produce a transcriptional output in yeast.

Figure 2.

Figure 2

Open in a new tab

SlHMGA3 is a nuclear protein and is ubiquitously expressed in different organs. A, SlHMGA3 is localized in cell nucleus. SlHMGA3-GFP fusion construct was transiently transformed into N. benthamiana epidermal cells. Green fluorescence images (first), DAPI fluorescent (second), bright-field image (third), and the merged image (fourth) are shown. Scale bars, 20 μm. B, SlHMGA3 is ubiquitously expressed in different organs and has increased transcript levels during tomato fruit ripening. The fruit pericarp tissues of tomato variety “Micro-Tom” were collected for RT-qPCR analysis at 5, 15, and 30 dpa and Breaker (BR, 39 dpa), Orange (O), and Red Ripe (RR) stages. Error bars represent the ±sd (standard deviation) for three biological replications.

We used reverse transcription-quantitative PCR (RT-qPCR) to analyze expression of SlHMGA3 in various tomato organs and fruits at different developmental stages. The results showed that the SlHMGA3 gene was ubiquitously expressed in vegetative and reproductive organs. The transcript levels of SlHMGA3 were similar in roots, stems, and leaves, but much higher in flowers and fruits. In particular, SlHMGA3 expression was elevated during fruit ripening, and its increase in expression started at the breaker stage and reached its peak at the red-ripening stage (Figure 2B), consistent with SlHMGA3 expression patterns derived from the online RNA-seq data (Supplemental Figure S2).

SlHMGA3 protein enhances transcription of SlDML2 by binding to its promoter

The Y1H assay was performed to verify the interaction between SlHMGA3 and SlDML2 promoter, and identify their binding sites. First, four promoter fragments (P1, P2, P3, and P) upstream of the ATG site of SlDML2 were tested for binding to SlHMGA3. The results showed that SlHMGA3 bound to fragments containing the AT-rich nucleotide sequence of the SlDML2 promoter (P3 and P) (Figure 3A). Second, we explored which conserved domain (H15 domain or AT-hook motifs) of SlHMGA3 bound to the SlDML2 promoter, and found that AT-hook motifs, not the H15 domain, could interact with the SlDML2 promoter (Figure 3B).

Figure 3.

Figure 3

Open in a new tab

SlHMGA3 protein enhances the transcription of SlDML2 by binding to its promoter. A, SlHMGA3 binds to the SlDML2 promoter fragment containing the AT-rich region using Y1H system. The promoter fragment (P) of SlDML2 was divided into three fragments (P1, P2, and P3). B, The AT-hook motifs of SlHMGA3 protein have the ability of binding to the SLDML2 promoter using Y1H system. C, SlHMGA3-HIS fusion protein binds to the AT-rich region of the SlDML2 promoter using EMSA method. SlDML2-labeled probe is a biotin-labeled fragment of the SlDML2 promoter containing the AT-rich base. Unlabeled probe is used as a competitor. The mutant probe is the labeled probe sequence with mutated nucleotides. D, SlHMGA3 promotes the transcription of SlDML2 in tomato pericarp protoplasts using Dual-LUC Reporter Assay. Error bars represent the ±sd for three biological replications.

Bioinformatic analysis showed that the promoter fragment of P3 contained two AT-rich DNA sites. To confirm which AT-rich DNA site is specifically bound to SlHMGA3, we purified the SlHMGA3-His fusion protein through prokaryotic expression and performed EMSA using two biotin-labeled fragments of the SlDML2 promoter containing AT-rich DNA sequences as the probe. The results showed that the AT-rich DNA site containing ATTATAAAAAATTA (−1,485 to −1,471) could not bind to SlHMGA3-His protein. On the other hand, when the AT-rich DNA site containing ATAAATAAATA (−1,528 to −1,517) oligonucleotides were used as the labeled probe, binding of SlHMGA3-His was detected (Figure 3C), and the addition of varying amounts of unlabeled ATAAATAAATA oligonucleotides as a competitor decreased the binding signal intensity. In addition, when the AT-rich site was mutated from ATAAATAAATA to GCGGGCGGGCG, no SlHMGA3-DNA binding signal was detected (Figure 3C). These results demonstrate that SlHMGA3 can specifically bind to AT-rich DNA sequences of the SlDML2 promoter.

We further studied whether SlHMGA3 could promote expression of the SlDML2 gene with the Dual-LUC Reporter Assay using tomato protoplast prepared from fruit pericarp tissues at the breaker stage. The results showed that when Pro35S:SlHMGA3 was co-introduced into the protoplasts with the ProSlDML2:LUC vector, the ratio of firefly LUC to renilla LUC (REN) signals was much higher than that in the control (Figure 3D), strongly suggesting that SlHMGA3 can promote the transcription of the SlDML2 gene in tomato fruit.

SlHMGA3 mutations delay the initiation of tomato fruit ripening

To investigate the function of SlHMGA3 in tomato fruit ripening, slhmga3 mutants were generated using the CRISPR/Cas9 gene-editing system. We designed two single-guide RNA (sgRNA) sequences to specifically target the second exon of SlHMGA3 (Figure 4A). Two independent homozygous mutant lines with no CRISPR/Cas9 vector, designated slhmga3-1 and slhmga3-2, were used for further analysis. The slhmga3-1 mutant had a 6-bp deletion at target 1 and a 4-bp deletion at target 2, and thus encoded a truncated SlHMGA3 protein of 116 amino acids (aa). The slhmga3-2 mutant had 1-bp insertion at target 1 and 2-bp deletion at target 2, encoding a truncated SlHMGA3 protein of 86 aa (Figure 4, B and C). No unanticipated editing events were identified in six potential off-target sites (Supplemental Table S1). In the slhmga3-1 mutant, the H15 domain had a deletion of 2 aa; the last three AT-hook motifs of SlHMGA3 were absent, but the first AT-hook motif remained present. slhmga3-2 contained a truncated H15 domain of SlHMGA3, and all four AT-hook motif were absent (Figure 4C). Therefore, slhmga3-1 and slhmga3-2 are partial and complete loss-of-function mutants, respectively, and are suitable for further analysis of the functions of the SlHMGA3 gene. In addition, we analyzed expression of the SlHMGA3 gene at various stages of fruit ripening in wild-type (WT) and slhmga3 mutants. The results showed that expression of mutated slhmga3 was significantly reduced in slhmga3 mutant fruit compared to SlHMGA3 gene in WT fruit, and slhmga3-2 had much lower slhmga3 gene expression than that of slhmga3-1 (Supplemental Figure S3). These results demonstrate that SlHMGA3 mutations also reduce the gene’s own mRNA levels.

Figure 4.

Figure 4

Open in a new tab

SlHMGA3 mutations delay the initiation of tomato fruit ripening. A, The two sgRNA target sites in SlHMGA3 locus (T1: 331–350 bp, T2: 502–521 bp) used for CRISPR/Cas9 gene-editing system. B, The mutation types of SlHMGA3 in slhmga3-1 line (six bases of ACCAGC located at 342–347 bp are deleted and four bases of ATGG located at 503–506 bp were deleted, respectively) and slhmga3-2 (one base of C is inserted after 347 bp and two bases of GG located at 506–507 bp are deleted, respectively). Purple letters indicate the target sites, green letters represent edited sites, blue letters represent the PAM. C, The structure of SlHMGA3 protein in WT and two slhmga3 mutants. The full-length of intact SlHMGA3 protein is 179 amino acid in WT, but a truncated HMGA3 protein with 116 amino acids is in slhmga3-1, and a truncated HMGA3 protein with 86 amino acids is in slhmga3-2. D and E, SlHMGA3 mutations delay the initiation of tomato fruit ripening. The ripening process was photographed (D) and the number of days from anthesis to the breaker stage was counted (E) in WT and slhmga3 mutants according to the methods of previous studies (Gao et al., 2018; Li et al., 2020a; Liang et al., 2020; Liu et al., 2021). Values are means ± sd of 16 individual fruits. ***P < 0.001 (Student’s t test). F, SlHMGA3 mutations delay fruit coloring during fruit ripening. The fruit color saturation was assessed by Hue angle. Values are means ± sd of 16 individual fruits. *P < 0.05; **P < 0.01; ***P < 0.001 (Student’s t test). G and H, SlHMGA3 mutations delay carotenoid accumulation and chlorophyll degradation during fruit ripening. Values are means ± sd of three biological replications. *P < 0.05; **P < 0.01; ***P < 0.001 (Student’s t test).

We first investigated the effect of SlHMGA3 mutation on plant architecture and fruit size. The results showed that the plant height of slhmga3-1 showed no significant difference from WT plants, while slhmga3-2 exhibited reduced plant height (Supplemental Figure S4, A and B). In both mutants, fruit size was not significantly altered at any point of the fruit development and ripening periods compared to the WT (Supplemental Figure S4, C and D).

Next, we investigated the effect of SlHMGA3 mutation on fruit ripening. We recorded the number of days from anthesis to the breaker stage. The results showed that the WT fruit reached the breaker stage at 36 d postanthesis (dpa), while slhmga3-1 and slhmga3-2 fruits reached the breaker stage at 39 and 42 dpa, 3- and 6-d delays observed in mutant fruit relative to the WT (Figure 4, D and E; Supplemental Figure S5). Consistently, our results also showed that fruit coloring of slhmga3 mutants was delayed relative to WT plants (Figure 4F), as well as chlorophyll degradation and carotenoid accumulation (Figure 4, G and H). Together, these results demonstrate that complete or partial loss of the SlHMGA3 function leads to delays in tomato fruit ripening.

SlHMGA3 mutations delay the initiation of SlDML2 expression and ethylene biosynthesis and signaling in tomato fruit

As SlHMGA3 protein enhances the transcription of SlDML2 by binding to its promoter, we analyzed the effects of SlHMGA3 mutation on the expression level of SlDML2. The RT-qPCR results showed that the transcript level of SlDML2 began to increase at 32 dpa, and reached a peak at 38 dpa in WT fruit. However, this initiation and peak expression of SlDML2 was substantially delayed in slhmga3-1 and slhmga3-2 fruits (Figure 5A), suggesting that SlHMGA3 mutations delay the upregulation of SlDML2 expression during tomato fruit ripening.

Figure 5.

Figure 5

Open in a new tab

SlHMGA3 mutations delay SlDML2 expression and expression of ethylene biosynthesis and signaling genes in tomato fruit. A, SlHMGA3 mutations delay the upregulation of expression of SlDML2 during fruit ripening. B–D, SlHMGA3 mutations delay the upregulation of expression of ethylene biosynthetic genes during fruit ripening. E and F, SlHMGA3 mutations delay the upregulation of expression of ethylene-responsive genes during fruit ripening. G–K, SlHMGA3 mutations delay the upregulation of expression of ethylene signaling genes during fruit ripening. Error bars represent the ±sd for three biological replications.

Ethylene plays critical role in tomato fruit ripening, so we investigated the effect of SlHMGA3 mutation on ethylene biosynthesis in tomato fruit. RT-qPCR results showed that the dramatic increases in the expression of three key ethylene biosynthetic genes, ACS2, ACS4, and ACO1, were delayed in slhmga3 compared to WT fruit (Figure 5, B–D). This delayed increase in ethylene biosynthetic gene expression was accompanied by a later burst of ethylene production and ethylene-regulated E4 and E8 expression in slhmga3 mutant fruits (Figures 5, E and F and 6), demonstrating that SlHMGA3 mutations can lead to delayed increase in production of ethylene and expression of ethylene regulated ripening-related genes.

Figure 6.

Figure 6

Open in a new tab

SlHMGA3 mutations delay the peak of ethylene biosynthesis during fruit ripening. Error bars represent the ±sd for three biological replications.

We also analyzed the expression of ethylene signaling genes using RT-qPCR, and showed that transcript levels of ethylene receptor genes ETR3 and ETR4 increased sharply from 36 to 38 dpa in WT fruit, while this increase occurred from 38 to 42 dpa in slhmga3-1 fruit and 42 to 46 dpa in slhmga3-2 fruit, respectively, thus showing delays of about 3–6 d (Figure 5, G and H). Additionally, the expression of three ethylene signaling genes, CTR1, EIL3, and EIN2, in WT fruit began to increase at 36 dpa, but this increase was delayed to 38 and 42 dpa in slhmga3-1 and slhmga3-2 fruits (Figure 5, I–K), respectively. Taken together, these results show that SlHMGA3 mutations affect the ethylene signaling pathway during the tomato fruit ripening process.

SlHMGA3 mutations delay the upregulation of the expression of ripening-associated transcription factors in tomato

Some ripening-associated transcription factor genes that are directly associated with ethylene production and signaling play important roles in the regulation of tomato fruit ripening (Liu et al., 2015a, 2015b; Ito et al., 2017; Gao et al., 2019, 2020). Therefore, we analyzed expression changes in slhmga3 mutants of six ripening-associated transcription factor genes, namely NON RIPENING (NOR), COLORLESS NON RIPENING (CNR), RIPENING INHIBITOR (RIN), TOMATO AGAMOUS-LIKE 1 (TAGL1), APETALA2a (AP2a), and FRUITFULL1 (FUL1). RT-qPCR analysis showed that the expression levels of NOR and CNR in WT fruit began to increase at 32 dpa, peaked at 38 dpa, and subsequently decreased. Meanwhile, in slhmga3-1 and slhmga3-2 fruits, the dramatic increase and peak of NOR and CNR expression was substantially delayed (Figure 7, A and B). In addition, RIN, TAGL1, AP2a, and FUL1 expression increased dramatically after 36 dpa in WT, while this increase was delayed to 38 dpa in slhmga3 mutants (Figure 7, C–F). These results demonstrate that SlHMGA3 mutations delay the upregulation of the expression of ripening-associated transcription factor genes in tomato fruit.

Figure 7.

Figure 7

Open in a new tab

SlHMGA3 mutations delay the upregulated expression of ripening-associated transcription factor genes in tomato. Error bars represent the ±sd for three biological replications.

SlHMGA3 mutations delay the demethylation process of the NOR and CNR gene promoters during tomato fruit ripening

To assess whether the delayed induction of ripening-associated transcription factor gene expressions in slhmga3 fruits results from maintenance of a high cytosine methylation level in their promoters, we selected CNR and NOR as example to analyze changes in the cytosine methylation levels of representative promoter regions using bisulfite sequencing (Liu et al., 2015a, 2015b). The total cytosine methylation level of the CNR promoter in WT fruit showed a substantial reduction from 32 to 38 dpa, and then increased from 38 to 42 dpa, whereas in slhmga3 mutants, the CNR promoter region maintained higher cytosine methylation levels from 32 to 42 dpa (Figure 8A). Specifically, CG, CHG and CHH sites on the CNR promoter all had higher methylation levels in slhmga3 compared to WT fruit (Figure 8, B–D). For the NOR promoter, although the methylation levels of CG, CHG, and CHH sites, and the total cytosine methylation level in WT and slhmga3 mutants all decreased from 32 to 42 dpa, slhmga3 mutants showed a delay in these demethylation processes compared to WT fruit (Figure 8, E–H). In conclusion, our results suggest that the delayed increase of CNR and NOR expressions in slhmga3 mutants results from the delayed demethylation process of their promoters.

Figure 8.

Figure 8

Open in a new tab

SlHMGA3 mutations cause a delayed demethylation process of the CNR and NOR promoters during fruit ripening. A, slhaga3 mutants have delayed demethylation of the CNR promoter compared to the WT. B–D, The CG, CHG, CHH sites of the CNR promoter have delayed demethylation in slhmga3 fruits compared to WT fruit. Fifteen individual clones were sequenced and analyzed. E, slhaga3 mutants have delayed demethylation of the NOR promoter compared to the WT. F–H, The CG, CHG, CHH sites of the NOR promoter have delayed demethylation in slhmga3 fruits compared to WT fruit. Fifteen individual clones were sequenced and analyzed.

Discussion

A function of canonical HMGA protein SlHMGA3 in tomato fruit ripening

DNA methylation modification has been shown to be closely associated with fruit ripening. In tomato, the decrease of global DNA methylation levels during fruit ripening is due to upregulation of the DML gene SlDML2 (Lang et al., 2017). SlDML2 is involved in demethylation of promoter regions of the ripening-associated transcription factor genes, NOR, CNR, and RIN, and subsequent their upregulation during fruit ripening (Liu et al., 2015a, 2015b). In addition, SlDML2 expression is reduced in rin, nor, and Cnr mutants, which suggests that SlDML2 may also be regulated by RIN, NOR, or CNR (Liu et al., 2015a, 2015b). Further, a tomato RNA demethylase SlALKBH2 and a histone demethylase SlJMJ6 can directly interact with SlDML2 gene and affect tomato fruit ripening (Zhou et al., 2019; Li et al., 2020a, 2020b). In this study, we identified the SlHMGA3 gene and showed that its protein most likely binds to a promoter sequence of the SlDML2 gene and subsequently promotes SlDML2 expression (Figure 3). Further, our slhmga3 mutants exhibited delayed fruit ripening (Figure 4). Thus we conclude that SlHMGA3 protein can influence SlDML2 expression and affect tomato fruit ripening.

Four HMGA proteins have been identified in Arabidopsis, two of which (AtGH1-HMGA1 and AtGH1-HMGA2) cluster together and have six AT-hook motifs. AtGH1-HMGA4 contains C-terminus without recognizable AT-hook motifs. AtGH1-HMGA3 that has four AT-hook motifs is considered the canonical HMGA protein (Charbonnel et al., 2018). AtGH1-HMGA1 protein plays a role in DNA repair and telomere stability, with the atgh1-hmga1 mutant exhibiting developmental and growth defects (Charbonnel et al., 2018). The canonical HMGA protein AtGH1-HMGA3 has been found to localize to the nucleus in Arabidopsis, and to be excluded from the nucleolus (Launholt et al., 2006). A recent study has shown that all the four GH1-HMGA genes in Arabidopsis redundantly control the floral transition (Zhao et al., 2021). In tomato, the SlHMGA3 protein has high sequence similarity to AtGH1-HMGA3 and has four AT-hook motifs (Figure 1), so it is considered as a canonical HMGA protein in tomato. SlHMGA3 gene is expressed in vegetative organs, including roots, stems, and leaves (Figure 2B). Consistently, when all four AT-hook motifs of the HMGA3 protein are deleted in the slhmga3-2 mutant, the plant height is reduced (Supplemental Figure S4, A and B), suggesting that SlHMGA3 plays a role in tomato vegetative organ growth. We have not observed remarkable changes in growth and development of flowers or immature fruits (Supplemental Figure S4, C and D), although SlHMGA3 is highly expressed in both flowers and immature fruits.

Our data showed that SlHMGA3 expression is enhanced during tomato fruit ripening and slhmga3 mutants showed delayed fruit ripening compared to the WT fruit. It is interesting to note that mutations in the SlHMGA3 gene lead to reduction in its own expression levels, and we also observed much more reduction in expression of mutated slhmga3 gene in the slhmga3-2 mutant (Supplemental Figure S3). One possibility is premature reading frame stops that induce nonsense-mediated decay/RNA stability. In short, in addition to their roles in plant growth and flowering in Arabidopsis (Charbonnel et al., 2018; Zhao et al., 2021), we have now demonstrated a function of HMGA protein in fleshy fruit ripening. Further, we also show that SlHMGA3 protein is most likely involved in promoting SlDML2 expression, and subsequently influences the initiation of tomato fruit ripening through SlDML2-mediated DNA demethylation.

SlHMGA3 may act as a cofactor to promote SlDML2 expression

HMGA proteins are architectural transcription factors with no transcriptional activation ability of their own, but can interact with the transcriptional machinery to modify chromatin structure and thereby positively or negatively regulate gene transcription, so HMGAs are considered as cofactors (Parisi et al., 2020). In Arabidopsis, GH1-HMGA3 is localized to the nucleus, and HMGA has been reported to bind to DNA sequences in plants. Pea HMGAs are the main proteins interacting with multiple sites on the plastocyanin gene promoter (Pwee et al., 1994; Webster et al., 1997) and binding to the AT-rich distal promoter regions of zein genes (Guillen et al., 1998). Our study demonstrates that tomato SlHMGA3 is located in the nucleus, and has the ability to bind to AT-rich DNA sequences of the SlDML2 promoter (Figures 2 and 3). These findings are in accordance with those HMGA proteins in other plants, and suggest that SlHMGA3 is not a typical transcription factor, but instead may act as a cofactor to promote SlDML2 expression.

In this study, we found that SlHMGA3 can enhance the transcription of SlDML2 by binding to its promoter, and that the expression pattern of SlHMGA3 is similar to that of SlDML2 (Liu et al., 2015a, 2015b). However, the transgenic plants of SlDML2, constructed using RNA interference (RNAi) and the sldml2 mutant, show much more severe defects than slhmga3 mutants, including abnormally shaped leaves and extreme dwarfing; these severe defects are not observed in slhmga3 mutants (Liu et al., 2015a, 2015b; Supplemental Figure S6). The ethylene production of SlDML2-RNAi plants is dramatically reduced throughout the ripening process, so the fruit do not turn red (Liu et al., 2015a, 2015b). However, the slhmga3 mutant fruit only exhibits a delay in starting ethylene production with the fruit eventually turning red (Figures 4–6). Together, these results suggest that SlHMGA3 may be only one of several factors affecting the expression of SlDML2.

Proposed function of SlHMGA3 in promoting SlDML2 expression and tomato fruit ripening

As chromatin architectural factors, HMGA proteins do not have transcriptional activation ability per se (Parisi et al., 2020); however, they are widely involved in the assembly of multi-subunit protein–DNA complexes, and in protein–protein interactions that induce chromatin structural changes in an ATP-independent fashion due to their unusual intrinsic plasticity (Tessari et al., 2003; Reeves, 2010; Sgarra et al., 2010). Structural modification of chromatin caused by HMGA proteins always affects the ability of transcription factors to bind to the promoter, repressor, or enhancer, and thereby negatively or positively regulates the transcription of multiple genes (Dragan et al., 2003; Thomae et al., 2008; Mansoori et al., 2021). For example, HMGA family member PF1 can stimulate binding of the transcription activator GT-2 to the promoter of the PHYA gene, in both oat (Avena sativa) and rice plants (Pwee et al., 1994; Martinez-Garcia and Quail, 1999). Pea HMGB and HMGA proteins interact physically, and HMGB enhances HMGA binding to an AT-rich enhancer element of the plastocyanin gene (Pwee et al., 1994; Webster et al., 1997, 2001). The four GH1-HMGA proteins in Arabidopsis redundantly preclude FLC gene looping to repress FLC expression (Zhao et al., 2021). In this study, we believe that the interaction between SlHMGA3 and the AT-hook motifs of SlDML2 promoter can lead to decompactness of the chromatin. Then the relaxed chromatin structure becomes an anchoring site for the recruitment of other transcription factors or enhancers, which may facilitate the formation of transcription factor complexes to form and to bind to the SlDML2 gene promoter region and thereby promotes the SlDML2 expression and subsequent fruit ripening. Alternatively, SlHMGA3 may act a passive player. The binding of other transcription factors to SlDML2 promoter regions may recruit SlHMGA3 binding to the AT-hook motifs of SlDML2 promoter, which further alters chromatin architecture to promote SlDML2 expression. We are currently further dissecting the possible mechanism of SlHMGA3 action on tomato fruit ripening.

Conclusions

Our studies show that SlHMGA3 is involved in the initiation of tomato fruit ripening. Molecular studies demonstrate that SlHMGA3 is capable of binding to the AT-hook motifs of the SlDML2 promoter sequence and likely promotes SlDML2 expression, thereby affecting the DNA methylation and expression levels of ripening-associated transcription factor genes, and subsequent ethylene biosynthesis and signaling, and fruit ripening in tomato.

Materials and methods

Plant materials and growth conditions

The tomato variety “Micro-Tom” (S. lycopersicum) was grown in a growth chamber at 23°C for 16 h during the day and at 18°C for 8 h during the night cycles with 60% humidity. Roots, stems, leaves, flowers, and fruit pericarp tissues at different fruit developing and ripening stages of WT plants were collected to study the expression pattern of SlHMGA3. Fruit pericarp samples of WT and slhmga3 mutants harvested at 32, 36, 38, 42, 48, and 50 dpa were immediately frozen in liquid nitrogen and stored at −80°C for quantifying gene transcript levels.

CRISPR/Cas9 plasmid construction of SlHMGA3 and tomato transformation

Two specific sgRNAs at the second exon of tomato HMGA3 gene were designed using CRISPR-P (http://cbi.hzau.edu.cn/crispr/). The sgRNAs along with the corresponding promoter were synthesized and cloned into the 2300GN-Ubi-Cas9 binary vector that was modified from pYLCRISPR/Cas9Pubi-N binary vector (Ma et al., 2015). The recombinant vector was transformed into the WT “Micro-Tom” using A. tumefaciens infection of leaf explants. Genomic DNA was extracted from true leaves of transgenic seedlings, and then PCR amplification and sequencing were used to identify suitable mutations. The synthesized sgRNA along with the corresponding promoters and the 2300GN-Ubi-Cas9 map are listed in Supplemental Figure S7. In addition, six potential off-target sites were tested for off-target editing events according to the method from Gao et al. (2020). The oligonucleotide primers used for off-target analysis are listed in Supplemental Table S2.

Sequence analysis of the SlHMGA3 protein

ClustalW program in BioEdit software was used to perform the sequence alignments of HMGA protein family from tomato (S. lycopersicum), Arabidopsis (A. thaliana), rice (O. sativa), maize (Z. mays), Japanese jack bean (Canavalia gladiata), oat (A. sativa), pea (P. sativum), Brassica napus, and soybean (Glycine max). The exon and intron structures of SlHMGA3 gene was analyzed by Exon-Intron Graphic Maker (version 4, http://www.wormweb.org/exonintron). The phylogenetic tree of HMGA family was constructed using MEGA version 6.0 software with 1,000 bootstrapping replicates (Tamura et al., 2013).

Subcellular localization of SlHMGA3

The ORF without terminator codon of the SlHMGA3 was polymerase chain reaction (PCR) amplified using special primers (Supplemental Table S3). The amplification product was inserted into the N-terminal of GFP protein of pCAMBIA2300GN-GFP binary vector to generate a fusion protein (SlHMGA3-GFP). The recombinant vector was transferred into A. tumefaciens strain GV3101. The leaves of 4-week-old N. benthamiana were infiltrated with the Agrobacterium suspension (OD600 = 0.3). After 48 h, the live cell imaging of leaves was observed by a confocal laser scanning microscope (Zeiss, Jena, Germany). 49,6-diamidino-2-phenylindole (DAPI) staining was used as a nuclear localization control. The experimental setups are as follows: GFP laser 488 nm, intensity 10.0%, collection bandwidth 493–598 nm, gains 790 V; DAPI laser 405 nm, intensity 2.0%, collection bandwidth 410–585 nm, gains 635 V.

Y1H assay

The Y1H assay was performed as previously described (Li et al., 2016). The full-length CDS of SlHMGA3 was ligated into the pGADT7 vector (Clontech, Mountain View, CA, USA), creating a prey vector (pGADT7-SlHMGA3). SlDML2 promoter fragments (P, P1, P2, and P3) were ligated into the pAbAi vector (Clontech), creating bait plasmids (pABAi-SlDML2-P, pABAi-SlDML2-P1, pABAi-SlDML2-P2, and pABAi-SlDML2-P3). Bait plasmids were digested with restriction enzyme BbsI and then transferred into the Y1H-Gold yeast strain on SD-Ura medium. Different concentrations of yeast cell growth inhibitor Aureobasidin A (AbA) were used to test the self-activation ability of bait plasmids. Then the pGADT7-SlHMGA3 or empty pGADT7 vectors were transformed into Y1H-Gold yeast strains containing the bait plasmids on SD-Ura-Leu medium added with 200 ng/mL AbA, and incubated for 3 d at 30°C. All primers used are listed in Supplemental Table S3.

EMSA

The coding sequence (CDS) regions of SlHMGA3 were amplified with special primers (Supplemental Table S3) and cloned into a pET-32a (+) expression vector. The recombinant vector was then transformed into Escherichia coli BL21 (DE3) (TransGen, Beijing, China). The induction and purification of SlHMGA3-His fusion proteins were performed with 100-mM IPTG and a column containing Ni-NTA-agarose resin as previously described (Feng et al., 2018). The concentration of the purified protein was quantified using a Bradford assay kit.

The recombined protein of SlHMGA3-His was used for EMSA assays with a LightShift Chemiluminescent EMSA Kit (Thermo Fisher Scientific Inc., Waltham, MA, USA) according to the previously described (Hu et al., 2016). Oligonucleotide probe containing the ATAAATAAATA sequence of SlDML2 gene promoter was synthesized with a biotin label (General Biol, Chuzhou, China). Unlabeled DNA fragment of the same sequence was used as a competitor probe. The ATAAATAAATA DNA fragment was modified into GCGGGCGGGCG and then used as a mutant probe. The DNA–protein complexes were separated with 6.5% (v/v) nondenaturing polyacrylamide gels and detected according to the manufacturer’s instructions (Thermo Fisher Scientific Inc.).

Dual LUC reporter assay of transient expression

The CDS of SlHMGA3 was cloned into the pGreenII 62-SK vector to generate the effector construct. The promoters of SlDML2 (1,619-bp upstream of the putative start codon) were inserted into the pGreenII 0800-LUC vector to generate the reporter construct. Protoplasts of tomato breaker fruit were extracted as described previously (Sakamoto et al., 2013), and the transient transformation assays of protoplasts were performed as described (Agarwal et al., 2006). A Dual-LUC Reporter Assay System (Promega, Madison, WI, USA; Cat. #E1910) was used to measure the ratio of LUC/REN in a Tanon-5500Multi spectrophotometer. Protoplasts hosting recombinant reporter gene construct and empty pGreenII 62-SK vector were used as control.

Chlorophyll and carotenoid analyses

Fresh weighed tomato pericarps at 32, 36, 38, 42, 46, and 50 dpa were collected and ground into a homogenate in a mortar adding quartz sand, calcium carbonate, and 95% ethanol, and then the homogenate was filtered into a 25-mL brown volumetric flask and diluted to 25 mL. The OD value of the supernatant was measured at 665, 649, and 470 nm by SpectraMax ID5 spectrophotometer to analyze the content of chlorophyll a, chlorophyll b, and carotenoids.

RNA extraction and RT-qPCR analysis

Total RNA was extracted using with a CTAB method from fruit pericarp materials stored at –80°C. The 1-μg total RNA was further used as a template to generate cDNA using HiScript II QRT SuperMix for RT-qPCR (+gDNA wiper) (Vazyme Biotech Co., Ltd., Nanjing, China). RT-qPCR was performed in a 96-well iCycler on a CFX96 Real-time system (Bio-Rad, Hercules, CA, USA) with the ChamQ qPCR SYBR Green Master Mix (Vazyme Biotech Co., Ltd.). The tomato Actin gene (Solyc11g005330) was used as the internal reference gene and each sample was run in triplicate. Relative gene expression values were on the basis of the 2−ΔCt method (Livak and Schmittgen, 2001). All primers are listed in Supplemental Table S3.

Measurement of ethylene and fruit color

Fresh tomato fruits of WT and slhmga3 mutants were harvested at 32, 36, 38, 42, 46, and 50 dpa and placed on an open petri dish for 3 h at room temperature to remove the stress-induced ethylene release. Then the fruit was transferred to a 50 mL jar and sealed with thin film for 8 h. Then 1-mL headspace gas in the jar was collected using a syringe and analyzed with gas chromatography. Ethylene content was determined by comparison against an ethylene standard curve and normalized with processing times and fruit weight.

Fruit color was evaluated as previously described (Li et al., 2019). Tomato fruit of WT and slhmga3 mutants at 32, 36, 38, 42, 46, and 50 dpa was used to measure the fruit color using a Konica Minolta Chromameter CR-400 at four symmetrical locations around the equator.

Bisulfite sequencing

Genomic DNA was isolated from tomato fruit pericarp of WT and slhmga3 mutants at 32, 36, 38, and 42 dpa with the CTAB method. Briefly, 1.5 μg of genomic DNA was converted with Bisulfite Mix using the EpiMark Bisulfite Conversion Kit (NEB, Ipswich, MA, USA) according to the manufacturer instructions. The DNA with bisulfite conversion was washed and dissolved in 30-μL Elution buffer, and then 5 μL of this solution was used as the template for PCR amplification with EpiMark Hot Start Taq DNA Polymerase (NEB, USA). The PCR product was purified and cloned into pEASY-Blunt Cloning vectors (TransGen biotech Inc., Beijing, China). Fifteen individual clones were sequenced, and the sequencing result was analyzed with Kismeth software (Gruntman et al., 2008). The primer sequences for bisulfite sequencing are listed in Supplemental Table S3.

Accession numbers

Sequence data from this article can be found in GenBank under the following accession numbers: SlHMGA3 (XP_004236944), HMGA-like family in tomato (XP_004235775, XP_004236944, XP_004250181), OsHMGA (AAA33914.1), ZmHMGA1 (CAB40848.2) and ZmHMGA2 (AAG00601.1), CgHMGA1 (BAA13133.1), and CgHMGA2 (BAA11767.1), AsHMGA (AAA32718.1), PsHMGA (CAA67752.1), BnHMGA (AAF22135.1), GmHMGA1 (Q00423.1), and GmHMGA2 (NP_001347278.1), promoter sequences of SlDML2 (101263652). The TAIR accession numbers of the HMGA-like family in Arabidopsis are AT3G18035 (GH1-HMGA1), AT1G48620 (GH1-HMGA2), AT1G14900 (GH1-HMGA3), and AT5G08780 (GH1-HMGA-related 4).

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. SlHMGA3 is unable to produce a transcriptional output in yeast.

Supplemental Figure S2. SlHMGA3 is ubiquitously expressed in different organs and has increased transcript levels during fruit ripening in tomato varieties “Ailsa Craig” and “Micro-Tom.”

Supplemental Figure S3. Expression levels of the SlHMGA3 gene during fruit ripening in WT and slhmga3 mutants.

Supplemental Figure S4. SlHMGA3 mutations affect plant height but not fruit size.

Supplemental Figure S5. SlHMGA3 mutations delay tomato fruit ripening.

Supplemental Figure S6. sldml2 mutant plants show more severe alterations in plant growth and development than slhmga3 mutant plants.

Supplemental Figure S7. gRNA sequence and map of the Cas9/sgRNA vector used in this study.

Supplemental Table S1. Detection of mutations on putative off-target sites.

Supplemental Table S2. Primers used for off-target site mutation analysis.

Supplemental Table S3. Primers used in this research.

Supplementary Material

kiac063_Supplementary_Data
Click here for additional data file. (936.6KB, pdf)

Acknowledgments

We thank Dr Lijie Zhou in Nanjing Agricultural University for technical support of the EMSA experiment.

Funding

This work was supported by the National Key Research and Development Program of China (2020YFE0202900), the Jiangsu Agricultural Science and Technology Innovation Fund [CX(20)3105], the Jiangsu Seed Industry Revitalization Project [JBGS(2021)015], and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

Conflict of interest statement. The authors declare that they have no competing interests.

Contributor Information

Zhifei Li, State Key Laboratory of Crop Genetics and Germplasm Enhancement, College of Horticulture, Nanjing Agricultural University, Nanjing 210095, China.

Ying Pi, State Key Laboratory of Crop Genetics and Germplasm Enhancement, College of Horticulture, Nanjing Agricultural University, Nanjing 210095, China.

Junmiao Fan, State Key Laboratory of Crop Genetics and Germplasm Enhancement, College of Horticulture, Nanjing Agricultural University, Nanjing 210095, China.

Xinxin Yang, State Key Laboratory of Crop Genetics and Germplasm Enhancement, College of Horticulture, Nanjing Agricultural University, Nanjing 210095, China.

Changsheng Zhai, State Key Laboratory of Crop Genetics and Germplasm Enhancement, College of Horticulture, Nanjing Agricultural University, Nanjing 210095, China.

Hong Chen, Institute of Botany, Jiangsu Province and Chinese Academy of Sciences, Nanjing 210014, China.

Feng Wang, State Key Laboratory of Crop Genetics and Germplasm Enhancement, College of Horticulture, Nanjing Agricultural University, Nanjing 210095, China.

Jing Ding, State Key Laboratory of Crop Genetics and Germplasm Enhancement, College of Horticulture, Nanjing Agricultural University, Nanjing 210095, China.

Tingting Gu, State Key Laboratory of Crop Genetics and Germplasm Enhancement, College of Horticulture, Nanjing Agricultural University, Nanjing 210095, China.

Yi Li, Department of Plant Science and Landscape Architecture, University of Connecticut, Storrs, Connecticut 06269, USA.

Han Wu, State Key Laboratory of Crop Genetics and Germplasm Enhancement, College of Horticulture, Nanjing Agricultural University, Nanjing 210095, China.

 

H.W. and Y.L. designed research. Z.L., Y.P., J.F., X.Y., C.Z., H.C., F.W., J.D., and T.G. performed the research. Z.L. and Y.P. analyzed data. Z.L., H.W., and Y.L. wrote the paper.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/general-instructions) is: Han Wu (wuhan@njau.edu.cn).

References

  1. Agarwal M, Hao Y, Kapoor A, Dong CH, Fujii H, Zheng X, Zhu JK (2006) A R2R3 type MYB transcription factor is involved in the cold regulation of CBF genes and in acquired freezing tolerance. J Biol Chem  281:  37636–37645 [DOI] [PubMed] [Google Scholar]
  2. Aravind L, Landsman D (1998) AT-hook motifs identified in a wide variety of DNA-binding proteins. Nucleic Acids Res  26:  4413–4421 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bustin M (2001) Revised nomenclature for high mobility group (HMG) chromosomal proteins. Trends Biochem Sci  26:  152–153 [DOI] [PubMed] [Google Scholar]
  4. Catez F, Hock R (2010) Binding and interplay of HMG proteins on chromatin: lessons from live cell imaging. BBA-Gene Regul Mech  1799:  15–27 [DOI] [PubMed] [Google Scholar]
  5. Chan SWL, Henderson IR, Jacobsen SE (2005) Gardening the genome: DNA methylation in Arabidopsis thaliana. Nat Rev Genet  6:  351–360 [DOI] [PubMed] [Google Scholar]
  6. Charbonnel C, Rymarenko O, Da Ines O, Benyahya F, White CI, Butter F, Amiard S (2018) The linker histone GH1-HMGA1 is involved in telomere stability and DNA damage repair. Plant Physiol  177:  311–327 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cheng J, Niu Q, Zhang B, Chen K, Yang R, Zhu JK, Zhang Y, Lang Z (2018) Downregulation of RdDM during strawberry fruit ripening. Genome Biol  19:  212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cui TJ, Wei S, Brew K, Leng FF (2005) Energetics of binding the mammalian high mobility group protein HMGA2 to poly(dA-dT)(2) and poly(dA)-poly(dT). J Mol Biol  352:  629–645 [DOI] [PubMed] [Google Scholar]
  9. Dragan AI, Liggins JR, Crane-Robinson C, Privalov PL (2003) The energetics of specific binding of AT-hooks from HMGA1 to target DNA. J Mol Biol  327:  393–411 [DOI] [PubMed] [Google Scholar]
  10. Feng K, Xu ZS, Liu JX, Li JW, Wang F, Xiong AS (2018) Isolation, purification, and characterization of AgUCGalT1, a galactosyltransferase involved in anthocyanin galactosylation in purple celery (Apium graveolens L.). Planta  247:  1363–1375 [DOI] [PubMed] [Google Scholar]
  11. Gao Y, Wei W, Fan Z, Zhao X, Zhang Y, Jing Y, Zhu B, Zhu H, Shan W, Chen J, et al. (2020) Re-evaluation of the nor mutation and the role of the NAC-NOR transcription factor in tomato fruit ripening. J Exp Bot  71: 3560–3574. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gao Y, Wei W, Zhao X, Tan X, Fan Z, Zhang Y, Jing Y, Meng L, Zhu B, Zhu H, et al. (2018) A NAC transcription factor, NOR-like1, is a new positive regulator of tomato fruit ripening. Hortic Res  5:  75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gao Y, Zhu N, Zhu XF, Wu M, Jiang CZ, Grierson D, Luo YB, Shen W, Zhong SL, Fu DQ, et al. (2019) Diversity and redundancy of the ripening regulatory networks revealed by the fruitENCODE and the new CRISPR/Cas9 CNR and NOR mutants. Hortic Res  6:  39. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gruntman E, Qi Y, Slotkin RK, Roeder T, Martienssen RA, Sachidanandam R (2008) Kismeth: analyzer of plant methylation states through bisulfite sequencing. BMC Bioinformatics  9:  371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Guillen P, Debon RM, Grasser KD, Harrsch P, Grimm R, Ponte I, Palau J (1998) Isolation and characterization of a 28-kDa HMG-like protein that binds to A/T-rich distal promoter regions of zein genes. Plant Sci  135:  31–38 [Google Scholar]
  16. Hu DG, Sun CH, Ma QJ, You CX, Cheng LL, Hao YJ (2016) MdMYB1 regulates anthocyanin and malate accumulation by directly facilitating their transport into vacuoles in apples. Plant Physiol  170:  1315–1330 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Huang H, Liu R, Niu Q, Tang K, Zhang B, Zhang H, Chen K, Zhu JK, Lang Z (2019) Global increase in DNA methylation during orange fruit development and ripening. Proc Natl Acad Sci USA  116:  1430–1436 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Ito Y, Nishizawa-Yokoi A, Endo M, Mikami M, Shima Y, Nakamura N, Kotake-Nara E, Kawasaki S, Toki S (2017) Re-evaluation of the rin mutation and the role of RIN in the induction of tomato ripening. Nat Plants  3:  866–874 [DOI] [PubMed] [Google Scholar]
  19. Jerzmanowski A, Przewloka M, Grasser KD (2000) Linker histones and HMG1 proteins of higher plants. Plant Biol  2:  586–597 [Google Scholar]
  20. Klosterman SJ, Hadwiger LA (2002) Plant HMG proteins bearing the AT-hook motif. Plant Sci  162:  855–866 [Google Scholar]
  21. Kotlinski M, Knizewski L, Muszewska A, Rutowicz K, Lirski M, Schmidt A, Baroux C, Ginalski K, Jerzmanowski A (2017) Phylogeny-based systematization of Arabidopsis proteins with histone H1 globular domain. Plant Physiol  174:  27–34 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lanahan A, Williams JB, Sanders LK, Nathans D (1992) Growth factor-induced delayed early response genes. Mol Cell Biol  12:  3919–3929 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lang Z, Wang Y, Tang K, Tang D, Datsenka T, Cheng J, Zhang Y, Handa AK, Zhu JK (2017) Critical roles of DNA demethylation in the activation of ripening-induced genes and inhibition of ripening-repressed genes in tomato fruit. Proc Natl Acad Sci USA  114:  E4511–E4519 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Launholt D, Merkle T, Houben A, Schulz A, Grasser KD (2006) Arabidopsis chromatin-associated HMGA and HMGB use different nuclear targeting signals and display highly dynamic localization within the nucleus. Plant Cell  18:  2904–2918 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Li H, Wu H, Qi Q, Li H, Li Z, Chen S, Ding Q, Wang Q, Yan Z, Gai Y, et al. (2019) Gibberellins play a role in regulating tomato fruit ripening. Plant Cell Physiol  60:  1619–1629 [DOI] [PubMed] [Google Scholar]
  26. Li S, Zhu BZ, Pirrello J, Xu CJ, Zhang B, Bouzayen M, Chen KS, Grierson D (2020b) Roles of RIN and ethylene in tomato fruit ripening and ripening-associated traits. New Phytol  226:  460–475 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Li T, Jiang Z, Zhang L, Tan D, Wei Y, Yuan H, Li T, Wang A (2016) Apple (Malus domestica) MdERF2 negatively affects ethylene biosynthesis during fruit ripening by suppressing MdACS1 transcription. Plant J  88:  735–748 [DOI] [PubMed] [Google Scholar]
  28. Li Z, Jiang G, Liu X, Ding X, Zhang D, Wang X, Zhou Y, Yan H, Li T, Wu K, et al. (2020a) Histone demethylase SlJMJ6 promotes fruit ripening by removing H3K27 methylation of ripening-related genes in tomato. New Phytol  227:  1138–1156 [DOI] [PubMed] [Google Scholar]
  29. Liang Q, Deng H, Li YX, Liu ZY, Shu P, Fu R, Zhang YX, Pirrello J, Zhang Y, Grierson D, et al. (2020) Like Heterochromatin Protein 1b represses fruit ripening via regulating the H3K27me3 levels in ripening-related genes in tomato. New Phytol  227:  485–497 [DOI] [PubMed] [Google Scholar]
  30. Liu M, Pirrello J, Chervin C, Roustan JP, Bouzayen M (2015a) Ethylene control of fruit ripening: revisiting the complex network of transcriptional regulation. Plant Physiol  169:  2380–2390 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Liu R, How-Kit A, Stammitti L, Teyssier E, Rolin D, Mortain-Bertrand A, Halle S, Liu M, Kong J, Wu C, et al. (2015b) A DEMETER-like DNA demethylase governs tomato fruit ripening. Proc Natl Acad Sci USA  112:  10804–10809 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Liu Y, Shi Y, Su D, Lu W, Li Z (2021) SlGRAS4 accelerates fruit ripening by regulating ethylene biosynthesis genes and SlMADS1 in tomato. Hortic Res  8:  3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCT method. Methods  25:  402–408 [DOI] [PubMed] [Google Scholar]
  34. Ma X, Zhang Q, Zhu Q, Liu W, Chen Y, Qiu R, Wang B, Yang Z, Li H, Lin Y, et al. (2015) A robust CRISPR/Cas9 system for convenient, high-efficiency multiplex genome editing in monocot and dicot plants. Mol Plant  8:  1274–1284 [DOI] [PubMed] [Google Scholar]
  35. Mansoori B, Mohammadi A, Ditzel HJ, Duijf PHG, Khaze V, Gjerstorff MF, Baradaran B (2021) HMGA2 as a critical regulator in cancer development. Genes  12:  269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Martinez-Garcia JF, Quail PH (1999) The HMG-I/Y protein PF1 stimulates binding of the transcriptional activator GT-2 to the PHYA gene promoter. Plant J  18:  173–183 [DOI] [PubMed] [Google Scholar]
  37. Mirouze M, Lieberman-Lazarovich M, Aversano R, Bucher E, Nicolet J, Reinders J, Paszkowski J (2012) Loss of DNA methylation affects the recombination landscape in Arabidopsis.  Proc Natl Acad Sci USA  109:  5880–5885 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Nie WF (2021) DNA methylation: from model plants to vegetable crops. Biochem Soc Trans  49:  1479–1487 [DOI] [PubMed] [Google Scholar]
  39. Nietosotelo J, Ichida A, Quail PH (1994) PF1: an A-T hook-containing DNA-binding protein from rice that interacts with a functionally defined d(AT)-rich element in the Oat Phytochrome A3 gene promoter. Plant Cell  6:  287–301 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Parisi S, Piscitelli S, Passaro F, Russo T (2020) HMGA proteins in stemness and differentiation of embryonic and adult stem cells. Int J Mol Sci  21:  362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Pwee KH, Webster CI, Gray JC (1994) Hmg protein-binding to an A/T-rich positive regulatory region of the pea plastocyanin gene promoter. Plant Mol Biol  26:  1907–1920 [DOI] [PubMed] [Google Scholar]
  42. Reeves R (2001) Molecular biology of HMGA proteins: hubs of nuclear function. Gene  277:  63–81 [DOI] [PubMed] [Google Scholar]
  43. Reeves R (2010) Nuclear functions of the HMG proteins. BBA-Gene Regul Mech  1799:  3–14 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Sakamoto S, Fujikawa Y, Esaka M (2013) Analysis of ascorbic acid biosynthesis using a simple transient gene expression system in tomato fruit protoplasts. Biosc Biotech Bioch  77:  673–675 [DOI] [PubMed] [Google Scholar]
  45. Saze H, Tsugane K, Kanno T, Nishimura T (2012) DNA methylation in plants: relationship to small RNAs and histone modifications, and functions in transposon inactivation. Plant Cell Physiol  53:  766–784 [DOI] [PubMed] [Google Scholar]
  46. Sgarra R, Lee J, Tessari MA, Altamura S, Spolaore B, Giancotti V, Bedford MT, Manfioletti G (2006) The AT-hook of the chromatin architectural transcription factor high mobility group A1a is arginine-methylated by protein arginine methyltransferase 6. J Biol Chem  281:  3764–3772 [DOI] [PubMed] [Google Scholar]
  47. Sgarra R, Pegoraro S, Ros G, Penzo C, Chiefari E, Foti D, Brunetti A, Manfioletti G (2018) High Mobility Group A (HMGA) proteins: molecular instigators of breast cancer onset and progression. BBA-Rev Cancer  1869:  216–229 [DOI] [PubMed] [Google Scholar]
  48. Sgarra R, Zammitti S, Lo Sardo A, Maurizio E, Arnoldo L, Pegoraro S, Giancotti V, Manfioletti G (2010) HMGA molecular network: from transcriptional regulation to chromatin remodeling. BBA-Gene Regul Mech  1799:  37–47 [DOI] [PubMed] [Google Scholar]
  49. Takaha N, Resar LM, Vindivich D, Coffey DS (2004) High mobility group protein HMGI(Y) enhances tumor cell growth, invasion, and matrix metalloproteinase-2 expression in prostate cancer cells. Prostate  60:  160–167 [DOI] [PubMed] [Google Scholar]
  50. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S (2013) MEGA6: Molecular evolutionary genetics analysis version 6.0. Mol Biol Evol  30:  2725–2729 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Tang DG, Gallusci P, Lang ZB (2020) Fruit development and epigenetic modifications. New Phytol  228:  839–844 [DOI] [PubMed] [Google Scholar]
  52. Tessari MA, Gostissa M, Altamura S, Sgarra R, Rustighi A, Salvagno C, Caretti G, Imbriano C, Mantovani R, Del Sal G, et al. (2003) Transcriptional activation of the cyclin A gene by the architectural transcription factor HMGA2. Mol Cell Biol  23:  9104–9116 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Thomae AW, Pich D, Brocher J, Spindler MP, Berens C, Hock R, Hammerschmidt W, Schepers A (2008) Interaction between HMGA1a and the origin recognition complex creates site-specific replication origins. Proc Natl Acad Sci USA  105:  1692–1697 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Webster CI, Packman LC, Gray JC (2001) HMG-1 enhances HMG-I/Y binding to an A/T-rich enhancer element from the pea plastocyanin gene. Eur J Biochem  268:  3154–3162 [DOI] [PubMed] [Google Scholar]
  55. Webster CI, Packman LC, Pwee KH, Gray JC (1997) High mobility group proteins HMG-1 and HMG-I/Y bind to a positive regulatory region of the pea plastocyanin gene promoter. Plant J  11:  703–715 [DOI] [PubMed] [Google Scholar]
  56. Yang Y, Tang K, Datsenka TU, Liu W, Lv S, Lang Z, Wang X, Gao J, Wang W, Nie W, et al. (2019) Critical function of DNA methyltransferase 1 in tomato development and regulation of the DNA methylome and transcriptome. J Integr Plant Biol  61:  1224–1242 [DOI] [PubMed] [Google Scholar]
  57. Zhang Q, Wang Y (2010) HMG modifications and nuclear function. Bba-Gene Regul Mech  1799:  28–36 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Zhang W, Wu Q, Pwee KH, Manjunatha Kini R (2003) Interaction of wheat high-mobility-group proteins with four-way-junction DNA and characterization of the structure and expression of HMGA gene. Arch Biochem Biophys  409: 357–366 [DOI] [PubMed] [Google Scholar]
  59. Zhao B, Xi Y, Kim J, Sung S (2021) Chromatin architectural proteins regulate flowering time by precluding gene looping. Sci Adv  7: eabg3097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Zhao J, Paul LK, Grafi G (2009) The maize HMGA protein is localized to the nucleolus and can be acetylated in vitro at its globular domain, and phosphorylation by CDK reduces its binding activity to AT-rich DNA. BBA-Gene Regul Mech  1789: 751–757 [DOI] [PubMed] [Google Scholar]
  61. Zhong SL, Fei ZJ, Chen YR, Zheng Y, Huang MY, Vrebalov J, McQuinn R, Gapper N, Liu B, Xiang J, et al. (2013) Single-base resolution methylomes of tomato fruit development reveal epigenome modifications associated with ripening. Nat Biotechnol  31: 154–159 [DOI] [PubMed] [Google Scholar]
  62. Zhou L, Tian S, Qin G (2019) RNA methylomes reveal the m(6)A-mediated regulation of DNA demethylase gene SlDML2 in tomato fruit ripening. Genome Biol  20: 156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Zhou X, Benson KF, Ashar HR, Chada KJN (1995) Mutation responsible for the mouse pygmy phenotype in the developmentally regulated factor HMGI-C. Nature  376: 771–774. [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

kiac063_Supplementary_Data
Click here for additional data file. (936.6KB, pdf)

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES