Genome-Wide Identification of Histone Deacetylases and Their Roles Related with Light Response in Tartary Buckwheat (Fagopyrum tataricum)

Histone deacetylases (HDACs), known as histone acetylation erasers, function crucially in plant growth and development. Although there are abundant reports focusing on HDACs of Arabidopsis and illustrating their important roles, the knowledge of HDAC genes in Tartary buckwheat (Polygonales Polygonaceae Fagopyrum tataricum (L.) Gaertn) is still scarce. In the study, a total of 14 HDAC genes were identified and divided into three main groups: Reduced Potassium Dependency-3/His-52 tone Deacetylase 1 (RPD3/HDA1), Silent Information Regulator 2 (SIR2), and the plant-53 specific HD2. Domain and motif composition analysis showed there were conserved domains and motifs in members from the same subfamilies. The 14 FtHDACs were distributed asymmetrically on 7 chromosomes, with three segmental events and one tandem duplication event identified. The prediction of the cis-element in promoters suggested that FtHDACs probably acted in numerous biological processes including plant growth, development, and response to environmental signals. Furthermore, expression analysis based on RNA-seq data displayed that all FtHDAC genes were universally and distinctly expressed in diverse tissues and fruit development stages. In addition, we found divergent alterations in FtHDACs transcript abundance in response to different light conditions according to RNA-seq and RT-qPCR data, indicating that five FtHDACs might be involved in light response. Our findings could provide fundamental information for the HDAC gene family and supply several targets for future function analysis of FtHDACs related with light response of Tartary buckwheat.


Introduction
In Eukaryotes, gene expression can be modulated by epigenetic regulation, which involves the switches between condensed and relaxed states of chromatin structure [1]. Chromatin is formed by DNA and nuclear histone, modifications of which play essential roles in making chromatin into either a transcriptionally active or an inert state [1]. Histone acetylation, as a major form of epigenetic modification, is a reversible biological process mediated by histone acetyltransferases (HATs) and histone deacetylases (HDACs) functioning on the ε-amino group of conserved lysine residues (such as H3K9, H4K5, H2AK5, H2BK5 and so on) within the core histones [2]. In general, the positive charge of histone tails can be neutralized by the acetylation of lysine residues [1,3,4]. Therefore, histone In addition, the expression patterns of FtHDACs in various tissues, fruits of different development stages, and seedlings under different light conditions were analyzed. The objective of the current study was to shed light on potential roles of FtHDACs in F. tataricum growth, development, and light response.

Identification and Characterization of HDACs in F. tataricum
The HDACs protein sequences in Cruciformes Cruciferae Arabidopsis thaliana (L.) Heynh. and Cruciformes Cruciferae Oryza sativa L. were employed as queries to search against the Tartary buckwheat genome through the BLASTP program. In total, 14 FtH-DAC genes were identified from the F. tataricum genome after eliminating the redundant sequences. A phylogenetic analysis of HDAC families in Arabidopsis, rice, and Tartary buckwheat was performed to analyse the evolutionary relationships and classification of FtHDACs ( Figure 1). The results demonstrated that 14 FtHDACs were divided intoto three subfamilies as Arabidopsis thaliana and Oryza sativa, with 9 members in the RPD3/HDA1 subfamily, 2 members in the SIR2 subfamily, and 3 members in the HD2 subfamily. Furthermore, we renamed the FtHDAC genes according to their homologs in Arabidopsis thaliana. The protein properties analysis displayed that the protein lengths of the 14 FtHDACs were distributed a broad range, spanning from 183 AA (FtHDA5) to 502 AA (FtHDA19) ( Table 1). As a result, FtHDA5 exhibited the smallest predicted molecular weight (MW) (20.68 kDa), while FtHDA19 presented the biggest predicted MW (56.19 kDa). In addition, the predicted isoelectric point (pI) distributed from 4.11 (FtHDT1) to 9.39 (FtHDA5).
In this study, the genome-wide analysis of HDACs was conducted on the F. tataricum genome. We investigated chromosomal localization, gene structure, conserved motifs, phylogenetic relationships, and putative cis-elements in the promoters of FtHDAC genes. In addition, the expression patterns of FtHDACs in various tissues, fruits of different development stages, and seedlings under different light conditions were analyzed. The objective of the current study was to shed light on potential roles of FtHDACs in F. tataricum growth, development, and light response.

Identification and Characterization of HDACs in F. tataricum
The HDACs protein sequences in Cruciformes Cruciferae Arabidopsis thaliana (L.) Heynh. and Cruciformes Cruciferae Oryza sativa L. were employed as queries to search against the Tartary buckwheat genome through the BLASTP program. In total, 14 FtHDAC genes were identified from the F. tataricum genome after eliminating the redundant sequences. A phylogenetic analysis of HDAC families in Arabidopsis, rice, and Tartary buckwheat was performed to analyse the evolutionary relationships and classification of FtHDACs ( Figure 1). The results demonstrated that 14 FtHDACs were divided intoto three subfamilies as Arabidopsis thaliana and Oryza sativa, with 9 members in the RPD3/HDA1 subfamily, 2 members in the SIR2 subfamily, and 3 members in the HD2 subfamily. Furthermore, we renamed the FtHDAC genes according to their homologs in Arabidopsis thaliana. The protein properties analysis displayed that the protein lengths of the 14 FtHDACs were distributed a broad range, spanning from 183 AA (FtHDA5) to 502 AA (FtHDA19) ( Table 1). As a result, FtHDA5 exhibited the smallest predicted molecular weight (MW) (20.68 kDa), while FtHDA19 presented the biggest predicted MW (56.19 kDa). In addition, the predicted isoelectric point (pI) distributed from 4.11 (FtHDT1) to 9.39 (FtHDA5).

Motif and Domain Composition of the FtHDAC Genes
The putative motifs of the FtHDACs were detected via the program MEME ( Figure 2A). Thirty conserved motifs differing in length and amino acid component were identified and named: from motif 1 to motif 30. There were at least two motifs present in each subfamily, suggesting that they belonged to the same family. Motifs 1 and 3 were specific to the RPD3/HDA1 subfamily members. Motifs 11,12,14,17,19,20 and 21 were specific to SIR2 proteins, while motifs 9, 13 and 24 were specific to HD2 proteins. The conserved domains were further detected with SMART and HMMER, and visualized via IBS software (version 1.0.3) ( Figure 2B). There was a deacetylase catalytic domain (HDAC domain) in all RPD3/HDA1 subfamily members. The members of the SIR2 subfamily all contained a SIR2 domain. A Nucleoplasmin-like domain (NPL) was present in the three members of the HD2 subfamily. Furthermore, there was a conserved ZnF_C2H2 domain in FtHDT2 and FtHDT3 but not in FtHDT1.

Motif and Domain Composition of the FtHDAC Genes
The putative motifs of the FtHDACs were detected via the program MEME ( Figure  2A). Thirty conserved motifs differing in length and amino acid component were identified and named: from motif 1 to motif 30. There were at least two motifs present in each subfamily, suggesting that they belonged to the same family. Motifs 1 and 3 were specific to the RPD3/HDA1 subfamily members. Motifs 11,12,14,17,19,20 and 21 were specific to SIR2 proteins, while motifs 9, 13 and 24 were specific to HD2 proteins. The conserved domains were further detected with SMART and HMMER, and visualized via IBS software (version 1.0.3) ( Figure 2B). There was a deacetylase catalytic domain (HDAC domain) in all RPD3/HDA1 subfamily members. The members of the SIR2 subfamily all contained a SIR2 domain. A Nucleoplasmin-like domain (NPL) was present in the three members of the HD2 subfamily. Furthermore, there was a conserved ZnF_C2H2 domain in FtHDT2 and FtHDT3 but not in FtHDT1.

Phylogenetic Analysis and Gene Structure of the FtHDAC Gene Family
The protein sequences of FtHDACs were aligned and applied for the Neighbor-Joining (NJ) phylogenetic tree construction via MEGA6.06. As shown in Figure 3A, there were three subgroups in the tree, with member names in three different colors (RPD3/HDA1, black; SIR2, red; HD2, blue). There were 3, 1, and 1 major groups in the RPD3/HDA1, SIR2, and HD2 branch, respectively. As for the first group of the RPD3/HDA1 branch, the paralogous RPD3/HDA1 pair consisted of FtHDA8-1 and FtHDA8-2 on the outermost side and FtHDA14 and FtHDA2 located inside. Diversely, the paralogous RPD3/HDA1 pair consisted of FtHDA6-1 and FtHDA6-2, located on the innermost side, and FtHDA9 and FtHDA19 located outside in the second group. Moreover, there was only one member (FtHDA5) in the third group. Additionally, there was one paralogous pair formed by FtSRT1 and FtSRT2 in the SIR2 branch, and there was a paralogous HD2 pair consisting of FtHDT2 and FtHDT3 with FtHDT1 distributed outside in the HD2 branch. The gene structure was displayed by the TBtools software (version 1.108) to demonstrate the structures of the FtHDACs ( Figure 3B). As shown in Figure 3B, the gene structures of FtHDACs were diverse, and the number of introns in FtHDACs varied from two to thirteen. As for members of the SIR2 subfamily, the gene structures were conserved and there were thirteen introns present in each SIR2 gene. In addition, the data displayed that FtHDT1, 2, and 3 contained seven, nine, and seven introns, respectively. Taken together, the intron/exon distribution pattern was varying in members of the RPD3/HDA1 and HD2 subfamilies, while that of the SIR2 subfamily was conserved.

Phylogenetic Analysis and Gene Structure of the FtHDAC Gene Family
The protein sequences of FtHDACs were aligned and applied for the Neighbor-Joining (NJ) phylogenetic tree construction via MEGA6.06. As shown in Figure 3A, there were three subgroups in the tree, with member names in three different colors (RPD3/HDA1, black; SIR2, red; HD2, blue). There were 3, 1, and 1 major groups in the RPD3/HDA1, SIR2, and HD2 branch, respectively. As for the first group of the RPD3/HDA1 branch, the paralogous RPD3/HDA1 pair consisted of FtHDA8-1 and FtHDA8-2 on the outermost side and FtHDA14 and FtHDA2 located inside. Diversely, the paralogous RPD3/HDA1 pair consisted of FtHDA6-1 and FtHDA6-2, located on the innermost side, and FtHDA9 and FtHDA19 located outside in the second group. Moreover, there was only one member (FtHDA5) in the third group. Additionally, there was one paralogous pair formed by FtSRT1 and FtSRT2 in the SIR2 branch, and there was a paralogous HD2 pair consisting of FtHDT2 and FtHDT3 with FtHDT1 distributed outside in the HD2 branch. The gene structure was displayed by the TBtools software (version 1.108) to demonstrate the structures of the FtHDACs ( Figure 3B). As shown in Figure 3B, the gene structures of FtHDACs were diverse, and the number of introns in FtHDACs varied from two to thirteen. As for members of the SIR2 subfamily, the gene structures were conserved and there were thirteen introns present in each SIR2 gene. In addition, the data displayed that FtHDT1, 2, and 3 contained seven, nine, and seven introns, respectively. Taken together, the intron/exon distribution pattern was varying in members of the RPD3/HDA1 and HD2 subfamilies, while that of the SIR2 subfamily was conserved.

Chromosomal Location and Duplications of the FtHDACs in the F. tataricum Genome
Chromosomal location of FtHDACs in the F. tataricum genome was carried out via the TBtools program ( Figure 4). The results illustrated that FtHDACs were located asymmetrically on seven F. tataricum chromosomes and no FtHDAC was distributed on chromosome Ft2. Chromosome Ft6 contained the highest density of HDACs, with four members (FtHDA8-1, FtHDA8-2, FtHDA14 and FtHDA19), followed by chromosome Ft1 with three genes (FtHDA6-1, FtHDT1, and FtSRT2) and chromosome Ft4 with two genes (FtHDA6-2 and FtHDA5), while there was only one HDAC gene present on each of the chromosomes Ft3, Ft5, Ft7 and Ft8. Four paralogous HDAC gene pairs were detected in F. tataricum according to the phylogenetic tree of FtHDACs ( Figure 3A). The data illustrated three duplications consisting of genes from distinct chromosomes and one duplication consisting of genes (FtHDA8-1 and FtHDA8-2) distributed at the same chromosome,

Chromosomal Location and Duplications of the FtHDACs in the F. tataricum Genome
Chromosomal location of FtHDACs in the F. tataricum genome was carried out via the TBtools program ( Figure 4). The results illustrated that FtHDACs were located asymmetrically on seven F. tataricum chromosomes and no FtHDAC was distributed on chromosome Ft2. Chromosome Ft6 contained the highest density of HDACs, with four members (FtHDA8-1, FtHDA8-2, FtHDA14 and FtHDA19), followed by chromosome Ft1 with three genes (FtHDA6-1, FtHDT1, and FtSRT2) and chromosome Ft4 with two genes (FtHDA6-2 and FtHDA5), while there was only one HDAC gene present on each of the chromosomes Ft3, Ft5, Ft7 and Ft8. Four paralogous HDAC gene pairs were detected in F. tataricum according to the phylogenetic tree of FtHDACs ( Figure 3A). The data illustrated three duplications consisting of genes from distinct chromosomes and one duplication consisting of genes (FtHDA8-1 and FtHDA8-2) distributed at the same chromosome, which was characterized as a tandem duplication since the two homologous genes were located on the chromosome Ft6 within a 20 kb physical distance (7061 bp) ( Table 1) with 70.6% and 69.63% nucleotide and protein sequence similarity, respectively. To analyze the evolutionary selection process of the FtHDAC gene family, the ratio of Ka/Ks was computed via the TBtools program (Table 2). Notably, there was no result for Ks of the gene pair consisting of FtHDT2 and FtHDT3 since the sequence divergence value (0.83) was higher than 0.75. As for the other three gene duplication events, the Ka/Ks ratios were less than one, indicating that negative selection was the main driving force for the expansion of FtHDACs. To track the dates of duplication events, their approximate dates were computed. As shown in Table 2, the two segmental duplication events of the FtHDACs originated from 38.77 to 88.57 million years ago, while the tandem duplication of FtHDA8-1 and FtHDA8-2 occurred 2.47 million years ago. which was characterized as a tandem duplication since the two homologous genes were located on the chromosome Ft6 within a 20 kb physical distance (7061 bp) ( Table 1) with 70.6% and 69.63% nucleotide and protein sequence similarity, respectively. To analyze the evolutionary selection process of the FtHDAC gene family, the ratio of Ka/Ks was computed via the TBtools program (Table 2). Notably, there was no result for Ks of the gene pair consisting of FtHDT2 and FtHDT3 since the sequence divergence value (0.83) was higher than 0.75. As for the other three gene duplication events, the Ka/Ks ratios were less than one, indicating that negative selection was the main driving force for the expansion of FtHDACs. To track the dates of duplication events, their approximate dates were computed. As shown in Table 2, the two segmental duplication events of the FtHDACs originated from 38.77 to 88.57 million years ago, while the tandem duplication of FtHDA8-1 and FtHDA8-2 occurred 2.47 million years ago.

Prediction of Cis-Regulatory Elements in the Promoter Regions of the FtHDACs
Up-to-2-kb upstream regions of the FtHDACs were analyzed with the Plant Care server to detect probable cis-regulatory elements. There were in total 22 cis-regulatory elements found in the promoters of FtHDACs ( Figure 5). More than six types of cis-regulatory elements existed in the promoters of 13 FtHDACs, while there were only three types of cis-regulatory elements present in the promoter of FtHDA14. Among the identified cisregulatory elements, the element related to light response was the most abundant and

Prediction of Cis-Regulatory Elements in the Promoter Regions of the FtHDACs
Up-to-2-kb upstream regions of the FtHDACs were analyzed with the Plant Care server to detect probable cis-regulatory elements. There were in total 22 cis-regulatory elements found in the promoters of FtHDACs ( Figure 5). More than six types of cis-regulatory elements existed in the promoters of 13 FtHDACs, while there were only three types of cis-regulatory elements present in the promoter of FtHDA14. Among the identified cis-regulatory elements, the element related to light response was the most abundant and present in the promoters of all the FtHDACs but FtHDA14, followed by the element related to MeJA-responsiveness which was present in the promoters of 12 FtHDACs but not FtHDA2 and FtHDA14. As for the cis-regulatory elements involved in abscisic acid, salicylic acid, gibberellin, and auxin response: these were present in promoters of 9, 3, 5, and 5 FtHDACs, respectively. There was a cis-regulatory element related to anaerobic induction present in the promoters of 10 FtHDACs. In addition, the promoters of 3, 3, 3, 5, 6, 6, 7, and 7 FtHDACs contained the cis-regulatory element involved in zein metabolism regulation, MYBHv1 binding site, enhancer-like element involved in anoxic-specific inducibility, cisacting regulatory element related to meristem expression, MYB binding site involved in drought-inducibility, cis-acting regulatory element involved in circadian control, lowtemperature responsiveness, and defense and stress responsiveness, respectively. The cis-regulatory element involved in endosperm expression, wound-responsive element, and binding site of AT-rich DNA binding protein (ATBP-1) only existed in the promoters of 1or 2 FtHDACs. These results indicated that FtHDACs might be involved in not only plant-environment interaction, but also phytohormone and stress signaling. present in the promoters of all the FtHDACs but FtHDA14, followed by the element related to MeJA-responsiveness which was present in the promoters of 12 FtHDACs but not FtHDA2 and FtHDA14. As for the cis-regulatory elements involved in abscisic acid, salicylic acid, gibberellin, and auxin response: these were present in promoters of 9, 3, 5, and 5 FtHDACs, respectively. There was a cis-regulatory element related to anaerobic induction present in the promoters of 10 FtHDACs. In addition, the promoters of 3, 3, 3, 5, 6, 6, 7, and 7 FtHDACs contained the cis-regulatory element involved in zein metabolism regulation, MYBHv1 binding site, enhancer-like element involved in anoxic-specific inducibility, cis-acting regulatory element related to meristem expression, MYB binding site involved in drought-inducibility, cis-acting regulatory element involved in circadian control, low-temperature responsiveness, and defense and stress responsiveness, respectively. The cis-regulatory element involved in endosperm expression, wound-responsive element, and binding site of AT-rich DNA binding protein (ATBP-1) only existed in the promoters of 1or 2 FtHDACs. These results indicated that FtHDACs might be involved in not only plant-environment interaction, but also phytohormone and stress signaling.

Tissue-Specific Expression Profiles of the FtHDACs
To enrich the knowledge of potential roles, the transcript abundance of FtHDACs in root, stem, leaf, flower, and fruit of three diverse developmental phases was extracted from the Tartary Buckwheat Database with gene IDs and utilized to build a heat map ( Figure S1). The data displayed that 14 FtHDACs were expressed at different levels in distinct tissues, which suggested that FtHDACs might play various roles in plant growth and fruit development in F. tataricum. The transcript of 14 FtHDACs could be detected in all the abovementioned tissues. Among these FtHDACs, FtHDT1 showed the highest abundance of transcript, while FtHDA6-1 was expressed at extremely low levels in all the detected tissues. Five FtHDACs (FtHDA2, FtHDA5, FtHDA9, FtHDA19, and FtHDT1) and seven FtHDACs (FtHDA6-1, FtHDA6-2, FtHDA8-1, FtHDA8-2, FtSRT2, FtHDT2, and FtHDT3) illustrated the most abundant transcript levels in roots and fruits at different stages, respectively. In addition, FtHDA14 and FtSRT1 exhibited the highest transcript abundance in leaf and flower, respectively. On the other hand, five (FtHDA6-1, FtHDA8-2, FtHDA9, FtHDA14, and FtSRT1) and four (FtHDA6-2, FtHDA8-1, FtSRT2, and FtHDT2) FtHDACs were expressed at their lowest levels in stem and leaf, respectively. Four FtHDACs (FtHDA5, FtHDA19, FtHDT1, and FtHDT3) displayed the lowest transcript levels in fruits at different stages, while only FtHDA2 was expressed at the least level in flowers. Interestingly, the transcript levels of all the FtHDACs showed a decreased tendency from Fruit-13 to Fruit-19, followed by subsequent decrement or increment in 8 (FtHDA5 ,  FtHDA6-1, FtHDA8-1, FtHDA8-2, FtHDA14, FtSRT1, FtSRT2, and FtHDT2) and 6 (FtHDA2 , FtHDA6-2, FtHDA9, FtHDA19, FtHDT1, and FtHDT3) FtHDACs from Fruit-19

Tissue-Specific Expression Profiles of the FtHDACs
To enrich the knowledge of potential roles, the transcript abundance of FtHDACs in root, stem, leaf, flower, and fruit of three diverse developmental phases was extracted from the Tartary Buckwheat Database with gene IDs and utilized to build a heat map ( Figure S1). The data displayed that 14 FtHDACs were expressed at different levels in distinct tissues, which suggested that FtHDACs might play various roles in plant growth and fruit development in F. tataricum. The transcript of 14 FtHDACs could be detected in all the abovementioned tissues. Among these FtHDACs, FtHDT1 showed the highest abundance of transcript, while FtHDA6-1 was expressed at extremely low levels in all the detected tissues. Five FtHDACs (FtHDA2, FtHDA5, FtHDA9, FtHDA19, and FtHDT1) and seven FtHDACs (FtHDA6-1, FtHDA6-2, FtHDA8-1, FtHDA8-2, FtSRT2, FtHDT2, and FtHDT3) illustrated the most abundant transcript levels in roots and fruits at different stages, respectively. In addition, FtHDA14 and FtSRT1 exhibited the highest transcript abundance in leaf and flower, respectively. On the other hand, five (FtHDA6-1, FtHDA8-2, FtHDA9, FtHDA14, and FtSRT1) and four (FtHDA6-2, FtHDA8-1, FtSRT2, and FtHDT2) FtHDACs were expressed at their lowest levels in stem and leaf, respectively. Four FtH-DACs (FtHDA5, FtHDA19, FtHDT1, and FtHDT3) displayed the lowest transcript levels in fruits at different stages, while only FtHDA2 was expressed at the least level in flowers. Interestingly, the transcript levels of all the FtHDACs showed a decreased tendency from Fruit-13 to Fruit-19, followed by subsequent decrement or increment in 8 (FtHDA5, FtHDA6-1, FtHDA8-1, FtHDA8-2, FtHDA14, FtSRT1, FtSRT2, and FtHDT2) and 6 (FtHDA2, FtHDA6-2, FtHDA9, FtHDA19, FtHDT1, and FtHDT3) FtHDACs from Fruit-19 to Fruit-25, respectively. Altogether, these results demonstrated that the expression profiles of the FtHDACs in distinct tissues were overlapping but unique, indicating that FtHDACs might play essential roles in sundry physiological processes in F. tataricum.

Expression Profile of the FtHDACs in Response to Diverse Light Condition
Light is one of the most significant environmental factors and triggers diverse photomorphogenic events with a set of light-dependent alterations to regulate the whole plant growth and development process. The above cis-regulatory element prediction ( Figure 5) indicated that FtHDACs might be involved in light response, and so we investigated the probable functions of FtHDACs in response to diverse light wavelength and duration time with transcriptome or RT-qPCR data. The transcriptome data of F. tataricum seedlings treated by diverse light wavelengths (Darkness, Blue, Far-red, and Red) illustrated that the expression levels of the FtHDACs varied considerably when responding to diverse light wavelengths ( Figure S2) [37]. Compared to the transcript levels of FtHDACs under darkness, eight FtHDACs (FtHDA2, FtHDA5, FtHDA6-1, FtHDA8-1, FtHDA19, FtSRT1, FtSRT2, and FtHDT1) showed fluctuated expression patterns under different lightwavelengths, while four (FtHDA6-2, FtHDA9, FtHDT2, and FtHDT3) and two (FtHDA8-2 and FtHDA14) FtHDACs were inhibited or induced by diverse light wavelengths, respectively. Specifically, blue, far-red, and red light downregulated the expression of nine, six, and eight FtHDACs, while five, eight, and six FtHDACs were upregulated by blue, far-red and red light, respectively. FtHDA6-2, FtHDA9, FtHDT2 and FtHDT3 were expressed at their highest expression levels under darkness, while FtHDA14 and FtSRT1 were most intensely induced by blue light. FtHDA2, FtHDA5, FtHDA8-2 and FtHDA19 displayed the most abundant transcripts under far-red light treatment, while red light induced the transcripts of FtHDA6-1, FtHDA8-1, FtSRT2, and FtHDT1 to the most abundant extent.
Furthermore, the relative expression levels of FtHDACs under different light duration conditions were analyzed by RT-qPCR with two reference genes to investigate the role of FtH-DACs in the photomorphogenic process under white light treatment ( Figure 6 and Figure S3). Integratedly, the relative expression levels of FtHDACs fluctuated considerably along the light treatment. There was a slight difference between the relative expression levels of FtH-DACs derived from FtH3 ( Figure 6) and FtACTIN7 ( Figure S3), and the expressions similarly and significantly changed under both reference genes were called significant changes. After one hour of white light treatment, only the expression of FtHDA8-2 was significantly inhibited. With the duration of white light treatment, three FtHDACs (FtHDA6-1, FtHDA8-1, and FtHDA8-2) were obviously inhibited at 2 h, while three (FtHDA6-1, FtHDA9, and FtHDA19) and one (FtHDA6-2) FtHDACs were notably induced or inhibited at 6 h, respectively. After 12 h of white light treatment, four FtHDACs (FtHDA14, FtHDA19, FtHDT2, and FtSRT2) were markedly upregulated, while FtHDA6-1 was strongly downregulated. When it came to 24 h, FtHDA6-1, FtHDT1, and FtSRT2 were inhibited, while only FtHDA19 was induced. Similarly, only FtHDA19 was induced at 48 h, but five FtHDACs (FtHDA8-1, FtHDA9, FtHDA14, FtHDT1, and FtSRT2) were clearly inhibited. For specific FtHDACs: FtHDA19 was obviously induced at 6 h, 12 h and 48 h, and FtHDT1 was notably induced at 24 h and 48 h, while FtHDA6-1 was obviously inhibited at 2 h, 12 h, and 24 h, FtHDA8-1 was inhibited at 2 h and 48 h, and FtHDA8-2 was inhibited at 1 h and 2 h. In addition, FtSRT2 and FtHDA14 were induced at 12 h, but inhibited at 24 h and 48 h, respectively. Jointly, the distinct expression patterns of FtHDACs under diverse light condition suggested that FtHDACs were critical for the light response of F. tataricum. Figure 6. Expression profiles of 14 FtHDAC genes in response to light. FtH3 was used as reference gene. Expression levels of each gene were expressed as a ratio relative to that of untreated seedlings (Darkness 0 h), which was set as 1. Each data point represents a mean ± standard error (n = 3). Asterisks above the bars indicate significant differences (* p < 0.05, ** p < 0.01, ***p < 0.001).among the treatments.
Plant HDACs have been identified to play crucial roles in numerous biological processes, such as seed germination, hypocotyl elongation, organ development, flowering, leaf senescence, and stress response [44]. In our study, the potential cis-regulatory elements in promoters of FtHDACs illustrated that there were diverse cis-regulatory elements involved in light response, hormone response, and stress response existing in promoters of most FtHDACs, suggesting HDACs might act in plant growth, development, and plantenvironment interaction. In particular, the cis-element involved in light response was the only cis-element existing in the promoters of all the FtHDACs, indicating the essential role of FtHDACs in the light response of F. tataricum ( Figure 5). Furthermore, transcriptome data downloaded from the Tartary buckwheat database was used to investigate the tissue-specific expression patterns of all the FtHDACs ( Figure S1). FtHDA6-1 exhibited extremely low expression levels in all the detected tissues ( Figure S1), which seemed contradictory to its homologous gene AtHDA6. AtHDA6 has been identified to participate in numerous biological processes of plant growth and development [44]. Notably, there was a paralogous gene of FtHDA6-1 in F. tataricum, which was expressed at abundant levels in all the detected tissues ( Figure 1 and Figure S1). Basing on the close evolutionary relationship and similar expression pattern of FtHDA6-2 and AtHDA6, FtHDA6-2 might play similar roles in F. tataricum's growth and development as AtHDA6. Jointly, these results indicated that there might be function redundancy of paralogous genes (FtHDA6-1 and FtHDA6-2) in F. tataricum. The FtHDA5, the closest homolog of AtHDA5, was expressed at relatively abundant levels in root and flower (Figure 1 and Figure S1), indicating it might be involved in root and flower development. The FtHDA19 was homologous to AtHDA19 and OsHDA702 (Figure 1), which were characterized to play a part in embryonic and flower development, and root development, respectively [44,51]. The FtHDA19 was expressed at relatively abundant levels in root, stem, and flower ( Figure S1), suggesting its potential role in root and flower development. The FtHDT1, FtHDT2 and FtHDT3 were expressed at their highest levels at Fruit_13 among the three fruit development stages (Figure S1), similar to their homologous genes: AtHDT1, AtHDT2, and AtHDT4 ( Figure 1) [5,52]. AtHDT1, AtHDT2, and AtHDT4 were reported to be involved in young siliques development [52]; thus, FtHDT1, FtHDT2, and FtHDT3 might also act in the fruit development of F. tataricum. In addition, combining FtHDA2 was significantly induced from Fruit_19 to Fruit_25 and its homolog: AtHDA2 was expressed in developing embryos and dry seeds (Figure 1) [53], and so we could hypothesize that FtHDA2 might be involved in embryo development. AtHDA19 has been identified to interact with SCARECROW-LIKE15 or HSL1 to repress seed maturation [54,55]. The FtHDA19, the homolog of AtHDA19, was downregulated along with fruit development (Figure 1 and Figure S1), suggesting it might also play a negative role in the seed maturation of Tartary buckwheat. Moreover, unlike AtHDA8, which was induced in late-stage seeds [53], FtHDA8-1 and FtHDA8-2 were expressed with a decreased tendency in the fruit development of F. tataricum ( Figure S1). Taken together, the functions of FtHDACs might be partially conservative and diverse compared to those of other plants.
F. tataricum is planted mostly in mountainous alpine areas with altitudes of more than 2500 m, where the light condition is distinct from that in the plains. The light condition is crucial for the phenotype and flavonoid biosynthesis of F. tataricum [37]. In the current study, we found that the expression of all the FtHDACs was altered by different light wavelengths ( Figure S2). Furthermore, their expression was diverse under different light treatment time ( Figure 6). These data indicated FtHDACs might play an essential role in the light response of F. tataricum. A deacetylase complex formed by AtHDA19 and SNL1-SNL6 was recruited by HY5 to the chromatin regions of HY5 and BBX22, and thus reduced the accessibility and histone acetylation and repressed their expression, resulting in inhibited photomorphogenesis. Specifically, the hda19-2 mutant exhibited enhanced photomorphogenesis under far-red, red, and blue light treatments compared with the wild type [56]. FtHDA19 was downregulated by red and blue light, while far-red light slightly induced its expression ( Figure S2). Moreover, AtHDA19 were inhibited from dark to white light in wild type Arabidopsis seedlings [56], but FtHDA19 was significantly induced by 6 h, 12 h, and 48 h of white light treatment ( Figure 6). These results suggested that the function of FtHDA19 in light response is partly diverse from that of AtHDA19. AtHDA15 has been reported to be recruited by HY5 or NF-YCs to decrease the histone acetylation levels of hypocotyl cell elongation related genes, resulting in inhibited hypocotyl elongation. Phenotypic analysis displayed that AtHDA15 played a negative role in hypocotyl cell elongation under red and far-red light treatments in Arabidopsis seedlings [34,57]. FtHDA5, the closest homolog of AtHDA15, was markedly induced by far-red and blue light and slightly inhibited by red light (Figure S2), suggesting its potential role in light response. The AtHDA6 was characterized as a positive regulator of light-regulated chromatin compaction. The athda6 mutant displayed lower chromatin compaction and reduced methylation levels of DNA and histone H3K9Me2 at Nucleolar Organizing Regions [58]. FtHDA6-2 was repressed by far-red, red, and blue light, while FtHDA6-1 was repressed by far-red and blue light but induced by red light ( Figure S2). However, the transcript levels of FtHDA6-1 were also extremely low under different light conditions ( Figure S2), and we could speculate that there was function redundancy between FtHDA6-1 and FtHDA6-2 in light response of Tartary buckwheat. Jointly, FtHDA6-2 might act in a similar way to AtHDA6 to positively regulate the chromatin compaction of the light response process. These data were in accordance with the findings that HDACs play critical roles in plant growth and development related with light response [34,56,57].

Plant Materials and Treatments
F. tataricum seeds (Cultivar Xiqiao No. 1 of F. tataricum) were retrieved from the Coarse Cereal Processing Center of Chengdu University (Chengdu, China). The seeds were immersed for 18-22 h in 0.2% sodium chloride solution in darkness after being rinsed in deionized water (dH 2 O) until the liquid was transparent. Afterwards, the seeds were sown on filter paper moistened with liquid medium containing 0.083 g/L of nitrogen fertilizer, 0.083 g/L of potassium fertilizer, and 0.166 g/L of phosphate fertilizer, and were kept in darkness for 4 days at 22 • C. Thereafter, the seedlings were illuminated with light-emitting diodes (LEDs) with the following intensities: 0 and 50 µmol m −2 s −1 , respectively. The samples after 0, 1, 2, 6, 12, 24, and 48 h of light treatments were frozen with liquid nitrogen and kept at −80 • C for future analysis.

Identification and Classification of FtHDAC Genes
The protein sequences of the Arabidopsis thaliana and Oryza sativa HDACs were obtained from the Arabidopsis Information Resource [59] and the Rice Genome Annotation Project [60], respectively. FtHDACs were identified in Tartary buckwheat genome, obtained from the Tartary Buckwheat Genome Project [35], via the BLASTP method using the Arabidopsis thaliana and Oryza sativa HDACs as query sequences. The conserved domains of non-redundant FtHDACs were detected by the online NCBI conserved domain searches [61], and the online SMART (version 9) [62] and HMMER (version 3.3.2) [63] softwares. The online ExPASy tool [64] were applied to predict the protein sequence length, molecular weight (MW), and isoelectric point (pI) of the FtHDACs.

FtHDACs Protein Motif, Domain and Gene Structure Analysis
The MEME (Multiple Em for Motif Elicitation) program [65] was used to discover the conservative motifs of the FtHDACs. The domain architecture of FtHDACs was predicted with SMART (version 9) [62] (http://smart.embl-heidelberg.de/accessed on 10 December 2021) and HMMER (version 3.3.2) [63] and visualized via the IBS (Illustrators of Biological Sequences) software (version 1.0.3) [66]. The TBtools software (version 1.108) was applied to visualize the gene structure of the FtHDACs.

Phylogenetic Analysis, Genome Distribution and Gene Duplication Analysis of FtHDACs
All the HDAC protein sequences from F. tataricum, Arabidopsis thaliana, and Oryza sativa were aligned using ClustalW and used for phylogenetic analysis with the MEGA 6.06 software in the Neighbor-Joining model with 1000 replicated bootstrap values. The initial chromosomal positions of the FtHDACs and chromosome lengths of F. tataricum were retrieved from the F. tataricum genome, obtained from the Tartary Buckwheat Genome Project [35], and the TBtools software(version 1.108) was employed to visualize the chromosome localization of the FtHDACs [30]. The parameters (Ks and Ka: synonymous and nonsynonymous substitution rate, respectively) of the duplication events were analyzed by the TBtools software (version 1.108). The Ka/Ks ratio of homologous gene pairs was applied to define the type of selection they were under. When Ka/Ks is greater than one, genes are under positive selection. A Ka/Ks less than one indicates negative selection. When Ka/Ks is equal to one, genes are under neutral selection. The time (T) of the duplication events was computed with the formula T = Ks/(2 × 1.5 × 10 −8 ) Mya [31]).

Prediction and Analysis of Cis-Regulatory Elements in the Promoter Regions of FtHDAC Genes
The 2.0 kb upstream sequences of the transcription start site of FtHDACs were retrieved from the Tartary Buckwheat Genome Project [35] and subjected to the online PlantCARE database [67] to detect cis-regulatory elements in the putative promoter regions. The identified cis-regulatory elements were visualized with the TBtools software (version 1.108).

Transcriptome Data Analysis
The transcript profiles of FtHDAC genes in root, stem, leaf, flower, and fruits from diverse development stages of F. tataricum were retrieved from the Tartary Buckwheat Database [68]. The transcript profiles of FtHDAC genes in seedlings of F. tataricum under different light wavelengths (red light (670 nm), blue light (470 nm), and far-red light (735 nm)) were downloaded from Zhang et al. [37]. The RPKM (Reads Per Kilobase of exon model per Million mapped reads) values were transformed to log 2 (value + 1) and visualized by the TBtools software (version 1.108).

RNA Isolation and Quantitative Real-Time PCR Analyses (RT-qPCR)
The total RNA of F. tataricum seedlings after light treatments was extracted with the RNAprep Pure Plant Plus Kit (TIANGEN BIOTECH, Beijing, China) according to the manufacturer's command. TransScript ® All-in-One First-Strand cDNA Synthesis SuperMix for qPCR (One-Step gDNA Removal) (Trans, Beijing, China) was employed to synthesise the first strand of cDNA with the extracted RNA as a template. RT-qPCR primers were devised via the Primer Premier 5 software with the corresponding sequences of FtHDACs downloaded from the F. tataricum genome. The expression patterns of all the FtHDACs after light treatment were investigated by RT-qPCR with FtH3 and FtACTIN7 as the internal reference genes based on their similar expression levels in nearly all the tissues. RT-qPCR was performed with TransStart ® Top Green qPCR SuperMix (Trans, Beijing, China) via the qTOWER3G system (Analytik Jena GmbH, Jena, Germany) as described by Yan et al. [69]. The relative expression levels of FtHDAC genes were calculated with the formula 2 −∆∆CT . A total of three independent biological replicates were utilized in the investigation. The primers applied for RT-qPCR are listed in Table S1.

Statistical Analysis
Experiments were performed in a totally randomized plan with three biological replicates. Data were demonstrated as the mean ±standard error (SE). Data were tested by Student's t test using SPSS version 26.0 (SPSS Inc., Chicago, IL, USA).

Conclusions
In the current study, in total 14 histone deacetylase genes, consisting of 9, 2, and 3 members of the RPD3/HDA1, SIR2, and HD2 subfamilies, respectively, were identified and characterized. There were conservative domains and motifs in members of the same subfamilies. FtHDACs were distributed unevenly on all the chromosomes except chromosome Ft2, with overlapping and distinct genres and numbers of cis-regulatory elements in their promoter regions. Negative selection was the major driving force for HDAC gene expansion in F. tataricum. The 14 FtHDACs were expressed in all the detected tissues with diverse expression profiles, indicating that FtHDACs might be involved in the regulation of various aspects of F. tataricum growth and development. Furthermore, study of expression pattern under different light conditions demonstrated that two induced FtHDACs (FtHDA19 and FtHDT1) and three inhibited FtHDACs (FtHDA6-1, FtHDA8-1 and FtHDA8-2) might be crucial regulators of light response in F. tataricum. In summary, a potential working model for FtHDACs in the light response of F. tataricum is proposed (Figure 7). Our results will be helpful for expanding our understanding of the participation of epigenetic regulators-histone modifiers-in the regulation of light response and supplying some targets for future function and molecular mechanism investigation of HDACs in response to different light conditions.

Conflicts of Interest:
The authors declare no conflict of interest.