Phylogenetic and Transcriptional Analyses of the HSP20 Gene Family in Peach Revealed That PpHSP20-32 Is Involved in Plant Height and Heat Tolerance

The heat shock protein 20 (HSP20) proteins comprise an ancient, diverse, and crucial family of proteins that exists in all organisms. As a family, the HSP20s play an obvious role in thermotolerance, but little is known about their molecular functions in addition to heat acclimation. In this study, 42 PpHSP20 genes were detected in the peach genome and were randomly distributed onto the eight chromosomes. The primary modes of gene duplication of the PpHSP20s were dispersed gene duplication (DSD) and tandem duplication (TD). PpHSP20s in the same class shared similar motifs. Based on phylogenetic analysis of HSP20s in peach, Arabidopsis thaliana, Glycine max, and Oryza sativa, the PpHSP20s were classified into 11 subclasses, except for two unclassified PpHSP20s. cis-elements related to stress and hormone responses were detected in the promoter regions of most PpHSP20s. Gene expression analysis of 42 PpHSP20 genes revealed that the expression pattern of PpHSP20-32 was highly consistent with shoot length changes in the cultivar ‘Zhongyoutao 14’, which is a temperature-sensitive semi-dwarf. PpHSP20-32 was selected for further functional analysis. The plant heights of three transgenic Arabidopsis lines overexpressing PpHSP20-32 were significantly higher than WT, although there was no significant difference in the number of nodes. In addition, the seeds of three over-expressing lines of PpHSP20-32 treated with high temperature showed enhanced thermotolerance. These results provide a foundation for the functional characterization of PpHSP20 genes and their potential use in the growth and development of peach.


Introduction
During these times of global environmental change, temperature is arguably the most important factor affecting plant growth and geographical distribution [1,2]. Plants experience fluctuations in the thermal environment characterized by average, maximum, and minimum daily temperatures, which change over the course of the seasons. Plants are very sensitive to temperature, showing responses to slight changes of just 1 • C [3]. However, it remains unknown how temperature signals are perceived. Extremely high or low temperatures lead to temperature stress, which is one of the most severe abiotic stressors and severely impacts plant growth [4][5][6]. Plants have evolved multiple pathways to adapt to temperature stress [7][8][9][10]. Nevertheless, the precise mechanism of temperature sensing, particularly ambient temperature, is still unclear.
Tremendous amounts of research have shown that multiple heat shock proteins (HSPs) emerge as central players in the temperature response, including the growth/development and stress responses [11][12][13][14]. According to their molecular weight, HSPs are divided into five major categories, including HSP100s, HSP90s, HSP70s, HSP60s, and HSP20s [4].

Analysis of Cis-Acting Elements of PpHSP20s Promoters
The promoters in the upstream 2000 bp region of 42 PpHSP20 genes were analyzed to identify the cis-elements. Eleven types of cis-elements were detected. Most of the PpHSP20 genes possessed abscisic acid-responsive (ABRE), light-responsive, MeJA-responsive, and anaerobic-induction elements ( Figure 4A,B). The elements were grouped into three categories, including phytohormone-responsive, abiotic, and biotic stress-responsive and plant development-related cis-elements. The phytohormone-responsive classification accounted for the highest proportion (49.2%, 186 of 378 elements), including abscisic acid-responsive), MeJA-responsive, salicylic acid-responsive (SA), auxin-responsive and gibberellin-responsive (GA). The abiotic and biotic stress-responsive elements included anaerobic induction and low temperature-responsive elements. In the plant developmentrelated category, seed specific regulation, cell cycle regulation, and circadian control were identified. These results suggested that PpHSP20s were not only related to stress response, but also related to other physiological responses.

Analysis of cis-Acting Elements of PpHSP20s Promoters
The promoters in the upstream 2000 bp region of 42 PpHSP20 genes were analyzed to identify the cis-elements. Eleven types of cis-elements were detected. Most of the PpHSP20 genes possessed abscisic acid-responsive (ABRE), light-responsive, MeJAresponsive, and anaerobic-induction elements ( Figure 4A,B). The elements were grouped into three categories, including phytohormone-responsive, abiotic, and biotic stress-responsive and plant development-related cis-elements. The phytohormoneresponsive classification accounted for the highest proportion (49.2%, 186 of 378 elements), including abscisic acid-responsive), MeJA-responsive, salicylic acid-responsive (SA), auxin-responsive and gibberellin-responsive (GA). The abiotic and biotic stress-responsive elements included anaerobic induction and low temperature-responsive elements. In the plant development-related category, seed specific regulation, cell cycle regulation, and circadian control were identified. These results suggested that PpHSP20s were not only related to stress response, but also related to other physiological responses.

Expression of PpHSP20s during the Shoot Elongation of 'Zhongyoutao 14'
The expression patterns of the PpHSP20s were compared at four critical stages (initial period, IP; initial elongation period, IEP; rapid growth period, RGP; stable growth period, SGP) of shoot elongation in the temperature-sensitive semi-dwarf peach cultivar 'Zhongyoutao 14', grown in the field under regular management with natural ambient temperature. According to their expression patterns, the 42 PpHSP20s could be classified into four groups ( Figure 5A). Group I contained 19 PpHSP20s that showed the highest expression level during SGP. Group II contained 7 PpHSP20s that showed the lowest expression level in IP, including PpHSP20-6, PpHSP20-17, PpHSP20-18, PpHSP20-21, PpHSP20-33, PpHSP20-37, and PpHSP20-38. The transcript abundance of PpHSP20s in Group III was higher in the IP and the IEP (PpHSP20-32, PpHSP20-13, PpHSP20-14, and PpHSP20-35), compared to RGP and the SGP. Group IV, the second largest group containing 12 PpHSP20s, showed higher expression levels in IP or RGP, compared to IEP and RGP. The internodes length of IP (1.21 mm) and IEP (2.57 mm) were significantly less than that of RGP (11.27 mm) and SGP (12.54 mm) [10]. It showed a negative trend between the internode length and the expression levels of PpHSP20s in Group III (marked in red). Among the Group III genes, the expression level of PpHSP20-32 was most consistent with the terminal internode length, as shown in Figure 5B. We speculated that PpHSP20-32 might participate in temperature-induced shoot growth in this temperature-sensitive peach cultivar.

Expression of PpHSP20s during the Shoot Elongation of 'Zhongyoutao 14'
The expression patterns of the PpHSP20s were compared at four critical stages (initial period, IP; initial elongation period, IEP; rapid growth period, RGP; stable growth period, SGP) of shoot elongation in the temperature-sensitive semi-dwarf peach cultivar 'Zhongyoutao 14', grown in the field under regular management with natural ambient temperature. According to their expression patterns, the 42 PpHSP20s could be classified into four groups ( Figure 5A). Group I contained 19 PpHSP20s that showed the highest expression level during SGP. Group II contained 7 PpHSP20s that showed the lowest expression level in IP, including PpHSP20-6, PpHSP20-17, PpHSP20-18, PpHSP20-21, PpHSP20-33, PpHSP20-37, and PpHSP20-38. The transcript abundance of PpHSP20s in Group III was higher in the IP and the IEP (PpHSP20-32, PpHSP20-13, PpHSP20-14, and PpHSP20-35), compared to RGP and the SGP. Group IV, the second largest group containing 12 PpHSP20s, showed higher expression levels in IP or RGP, compared to IEP and RGP. The internodes length of IP (1.21 mm) and IEP (2.57 mm) were significantly less than that of RGP (11.27 mm) and SGP (12.54 mm) [10]. It showed a negative trend between the internode length and the expression levels of PpHSP20s in Group III (marked in red). Among the Group III genes, the expression level of PpHSP20-32 was most consistent with the terminal internode length, as shown in Figure 5B. We speculated that PpHSP20-32 might participate in temperature-induced shoot growth in this tem-

Overexpression of PpHSP20-32 in Arabidopsis Leads to an Increase in Plant Height
In order to study the function of PpHSP20-32, we constructed a PpHSP20-32 overexpression vector and transformed it into Arabidopsis thaliana using an Agrobacterium-mediated method. The phenotypes of three transgenic lines (L1, L2, and L3) and WT were recorded ( Figure 6). Two weeks after being transplanted into a substrate, their rosette leaves were longer and wider than WT (average length and width), but there was no significant difference in the number of rosette leaves ( Figure 6A-C). Four weeks after transplanting, the plant morphology was observed, and the height of the flowering bolt in the three transgenic lines was higher than that of WT ( Figure 6D

Overexpression of PpHSP20-32 in Arabidopsis Leads to an Increase in Plant Height
In order to study the function of PpHSP20-32, we constructed a PpHSP20-32 overexpression vector and transformed it into Arabidopsis thaliana using an Agrobacteriummediated method. The phenotypes of three transgenic lines (L1, L2, and L3) and WT were recorded ( Figure 6). Two weeks after being transplanted into a substrate, their rosette leaves were longer and wider than WT (average length and width), but there was no significant difference in the number of rosette leaves ( Figure 6A-C). Four weeks after transplanting, the plant morphology was observed, and the height of the flowering bolt in the three transgenic lines was higher than that of WT ( Figure 6D Figure 6G). There was also no significant difference in the number of branches among all lines ( Figure 6H).

PpHSP20-32-OE Seeds Exhibit Enhanced Thermotolerance
The seeds of three PpHSP20-32-OE lines and WT were treated at 46 • C for 30 min and transferred to 25 • C to assay thermotolerance (Figure 7). By 48 h after heat stress (HS), there was no seed germination in any of the four lines ( Figure 7A,B). After 60 h at high temperature, the germination rate of the three PpHSP20-32 transgenic lines was 100%, which was significantly higher than that of WT seeds ( Figure 7C,D,F). For the WT, the germination was less than 10% after 60 h of HS ( Figure 7C,F), but reached about 100% germination at 96 h ( Figure 7E,F). These results suggested that the overexpression of PpHSP20-32 improves the resistance of Arabidopsis seeds to high temperatures.

PpHSP20-32-OE Seeds Exhibit Enhanced Thermotolerance
The seeds of three PpHSP20-32-OE lines and WT were treated at 46 °C for 30 min and transferred to 25 °C to assay thermotolerance (Figure 7). By 48 h after heat stress (HS), there was no seed germination in any of the four lines ( Figure 7A,B). After 60 h at high temperature, the germination rate of the three PpHSP20-32 transgenic lines was 100%, which was significantly higher than that of WT seeds ( Figure 7C,D,F). For the WT, the germination was less than 10% after 60 h of HS ( Figure 7C,F), but reached about 100% germination at 96 h ( Figure 7E,F). These results suggested that the overexpression of PpHSP20-32 improves the resistance of Arabidopsis seeds to high temperatures.

PpHSP20-32-OE Seeds Exhibit Enhanced Thermotolerance
The seeds of three PpHSP20-32-OE lines and WT were treated at 46 °C for 30 min and transferred to 25 °C to assay thermotolerance (Figure 7). By 48 h after heat stress (HS), there was no seed germination in any of the four lines ( Figure 7A,B). After 60 h at high temperature, the germination rate of the three PpHSP20-32 transgenic lines was 100%, which was significantly higher than that of WT seeds ( Figure 7C,D,F). For the WT, the germination was less than 10% after 60 h of HS ( Figure 7C,F), but reached about 100% germination at 96 h ( Figure 7E,F). These results suggested that the overexpression of PpHSP20-32 improves the resistance of Arabidopsis seeds to high temperatures.

Discussion
As plants sense high temperatures or heat stress, gene expression patterns will vary, especially the up-regulation of the heat shock genes [11,32]. HSPs include the HSP100s, HSP90s, HSP70s, HSP60s, and HSP20s. HSP20s are a diverse, ancient, and important family among the HSPs [16]. The number of HSP20s has been determined in numerous plants, such as 31 in Arabidopsis thaliana [17,33], 51 in Glycine max [28], 35 in Capsicum annuum [19], 42 in Solanum lycopersicum [20], 63 in Panicum virgatum [34], 48 in Solanum tuberosum [35], 48 in Vitis vinifera [22], 47 in Sorghum bicolor [23], 41 in Malus pumila [13], 47 in Cucumis sativus [14], 45 in Cucumis melo [14], and 47 in Citrullus lanatus [14]. In peach, we identified 42 PpHSP20s, a number greater than in Arabidopsis thaliana and pepper, but lower than in switchgrass, potato, and grape. The varied numbers in different plants might be due to the difference in gene duplications during the evolution of the plants. The 42 PpHSP20s are unevenly mapped on the eight chromosomes, with Chr6, the second longest chromosome, containing the most HSP20s. The members of other gene families, such as E3 genes, were also mainly mapped on the longer chromosomes in peach [36], while the F-box genes showed a similar phenomenon in pear [37]. The E3 and F-box genes were mapped on the longer chromosome, similar to the distribution of PpHSP20s on the chromosomes. However, the biggest cluster of HSP20s was on the shortest chromosome, chromosome 8, in apple [13]. So, any rules of distribution of gene family members may be different among different families or different plants, and need to be further validated.
The PpHSP20 duplication in peach showed inconsistent patterns with those of other plants [13,28,35]. In apple, WGD and TD were the main duplication patterns [13]. In this study, the PpHSP20 family expanded more by DSD and TD. In peach, DSD was the major expansion route for other gene families, such as the F-box, U-box, RING, BTB, SKP [36], and HSF genes [38]. This phenomenon in peach might be explained by the fact that the peach genome has not undergone a recent whole-genome duplication [39].
The 42 PpHSP20s were divided into 11 classes (CI, CII, CIII, CIV, CV, CVI, MI, MII, P, ER, and Px), except for 2 unclassified PpHSP20s, based on the phylogenetic tree which was constructed using the amino acid sequences of peach, rice, Arabidopsis thaliana and soybean. In an earlier study, the AtHSP20s were divided into seven classes (CI, CII, CIII, M, P, ER, and Px), except for five genes that did not fall into any class [17]. Afterwards, the five unclassified AtHSP20s were categorized into five new classes (CIV, CV, CVI, and CVII) and MII [40]. Most of PpHSP20s were classified into nucleocytoplasmic subfamilies (CI-CVI), which indicated that the cytoplasm may be the primary site of action for the HSP20 proteins. This was also observed in other plants, for example, apple and soybean [13,28]. In this study, the HSP20s of peach lacked any proteins in the CVII subgroup, similar to soybean [28], rice [18], switchgrass [34], apple [13], and three cucurbit species (cucumber, melon, and watermelon) [14]. In Arabidopsis thaliana, the CVII subgroup gene AtHsp14.7 was constitutively expressed, and its transcript level did not change under different stresses [40]. It was speculated that AtHsp14.7-CVII was involved in specific housekeeping functions [40].
Plant HSPs are molecular chaperones that protect the functions of their target proteins under various stress conditions to help maintain growth and development [4,16]. In this study, PpHSP20s showed different expression patterns at non-stressful but elevated temperature. The expression patterns of four PpHSP20s, namely PpHSP20-13, PpHSP20-14, PpHSP20-35, and especially PpHSP20-32, showed a correlation with the length of the terminal internodes in the shoots of temperature-sensitive semi-dwarf peach. Populus trichocarpa, a transgenic line overexpressing PtHSP17.8, showed enhanced tolerance to heat and salt stresses [24]. In pepper, CaHSP16.4 participated in heat and drought stress by enhancing the scavenging of reactive oxygen species [27]. These results mainly focused on the function of HSP20s under stress conditions. Based on this study, PpHSP20s might play important role in the regulation of shoot elongation at non-stressful temperatures. In addition, HSP20 responded to the phytohormone ABA and modulated polar auxin transport [26].
In the ambient temperature-sensing pathway, AtHSP70 is expressed at a level proportionate to the ambient temperature [41]. AtHSP90 integrates environmental temperature and auxin signaling to regulate temperature-dependent plant growth by stabilizing the auxin co-receptor F-box protein TIR1 [42,43]. A recent study showed that the heat shock protein AtHSP22 promoted hypocotyl elongation at high temperatures by regulating polar auxin transport, which required the ABI1 protein phosphatase [26]. In this study, PpHSP20-32-overexpressing transgenic lines produced larger rosette leaves and taller plants than WT. The plant height of the transgenic lines was higher than that of WT. There was no significant difference in the length of the internodes between the transgenic lines and WT, indicating that the increase in plant height of the transgenic lines may be caused by the increase in the number of internodes. The PpHSP20-32-overexpressing lines also showed enhanced heat tolerance. Similar results were observed in rice, pepper, and poplar, which together demonstrate that HSP20 genes enhance thermotolerance [24,25,27].
It remains unknown how PpHSP20-32 regulates rosette leaf size and plant height. The promoter of PpHSP20-32 contained four types of phytohormone-responsive elements, namely ABRE, MeJA-responsive, salicylic acid-responsive, and gibberellin-responsive elements ( Figure 4A,B). This indicated that PpHSP20-32 might serve as a phytohormone responsive factor. In Arabidopsis thaliana, AtHSP22 is regulated by ABA and auxin, while AtHSP22 potentiates the auxin efflux PIN proteins, which promotes hypocotyl elongation [26]. These results suggested that PpHSP20-32 might serve to modulate rosette leaf and flower bolt growth by orchestrating phytohormone signaling.

Plant Materials
The peach cultivar 'Zhongyoutao14', a temperature-sensitive semi-dwarf, is planted in the experimental station of the Horticulture College, Henan Agricultural University (Zhengzhou, China). The shoot internode length was temperature-dependent. Shoot tips were collected at four critical growth stages, namely the initial period (IP), initial elongation period (IEP), rapid growth period (RGP), and stable growth period (SGP) [10]. The internodes' lengths were less than 3 mm at IP and IEP with lower environmental temperature (below 30 • C). While the internodes' lengths at RGP and SGP with higher temperatures (above 30 • C) were more than 10 mm [10]. All samples were quickly frozen in liquid nitrogen after collection and stored in the laboratory at −80 • C. Arabidopsis thaliana (L.) Heynh Columbia 0 (Col-0) was used for transformation with PpHSP20-32.

Chromosome Location and Gene Structure Analysis of the PpHSP20 Genes
According to the genome location annotation given by Phytozome V12.1, the chromosome location of each PpHSP20 was mapped using TBtools [44]. According to the DNA and CDS sequences data for the peach HSP20 gene, the gene structure map was drawn using the online tool GSDS (http://gsds.cbi.pku.edu.cn/, accessed on 10 May 2022).

Phylogeny and Motif Analysis of PpHSP20s
The amino acid sequences of the HSP20 genes of Arabidopsis thaliana, Oryza sativa, Glycine max, and peach were saved as FASTA format files. The phylogenetic tree was constructed by the maximum likelihood method using MEGA 7.0 software (v7.0, Sudhir kumar, Hachioji, Tokyo, Japan) [45]. The online software MEME5.0.4 (http://alternate. meme-suite.org/tools/meme, accessed on 12 May 2022) was used to analyze the motifs in each protein sequence.

Analysis of Cis-Acting Elements of PpHSP20s
The promoter sequence of each PpHSP20 gene (2000 bps upstream of the start codons) was downloaded from the peach genome. The cis-acting elements of the HSP20 promoters were detected using PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/ html/, accessed on 10 May 2022).

PpHSP20 Gene Expression in Different Growth Stages of Peach
The expression levels of the PpHSP20 genes were obtained from our previous transcriptome data of the four critical stages of shoot growth in the cultivar 'Zhongyoutao 14' (Table S3) [10]. The heatmap was generated by TBtools (v1.09876, Chengjie Chen, Guangzhou, Guangdong, China) [44]. The FPKM (fragments per kilobase of exon per million fragments mapped) values of the HSP20s and the terminal internode lengths of the stems were used for the correlation analysis.

Generation and Phenotypic Observation of PpHSP20-32-Overexpression in Arabidopsis Lines
The CDS of PpHSF20-32 was amplified using PpHSP20-32-F and PpHSP20-32-R primers (Table S4). The resulting PpHSP20-32 fragment was inserted into the pMD19T vector (Takara, Dalian, China). After sequence confirmation, the full coding sequence of PpHSP20-32 was amplified with primers including restriction sites Hind III and Xba I (Table S4), and the amplified fragment was directionally inserted into the vector pSAK277. Transgenic Arabidopsis plants were generated through the floral dip method using the Agrobacterium tumefaciens strain GV3101 [46].
After screening for kanamycin resistance and PCR verification (an initial denaturing step at 98 • C for 5 min, followed by 30 cycles of 98 • C for 10 s, 55 • C for 15 s, and 72 • C for 40 s, then 72 • C for 3 min), the transgenic plants were allowed to flower. Seeds from T 2 transgenic Arabidopsis lines were used for subsequent experiments. Three seedlings from each line with five rosette leaves per seedling of each line was considered one biological replicate and used for the observation of leaf phenotype (length and width). Three biological replicates were taken, for a total of nine seedlings observed. The height, the length of internodes, and the number of internodes and branches (five plants per line) in the different transgenic lines and WT were determined.

Heat Stress Treatment
To detect the function of PpHSP20-32 in heat tolerance, heat stress treatment (46 • C for 30 min) was performed. Before heat stress treatment, seeds of WT and transgenic Arabidopsis lines were sown on MS medium and kept dark at 4 • C for 2 d, and then at 22 • C for 2 d. Then, the seeds were exposed to 46 • C for 30 min, followed by being transferred into a climate chamber (22 • C, 16 h light/8 h dark cycles). The germination of seeds was counted daily and photographed. More than 60 seeds of each line were used in each plate, with three replicates. Differences in heat stress tolerance were confirmed using Student's t-test.

Statistical Analysis
Data were analyzed by ANOVA, Tukey HSDa, and Duncan's multiple range tests (at p < 0.05) using IBM SPSS Statistics 20 (SPSS, Armonk, New York, NY, USA).

Conclusions
In this study, 42 PpHSP20 genes, distributed on eight chromosomes randomly, were identified in the peach genome. Dispersed gene duplication (DSD) and tandem duplication (TD) were the primary modes of gene duplication of PpHSP20s. Except for two unclassified PpHSP20s, the other 40 PpHSP20s were classified into 11 subclasses. The gene structures, basic classification, conserved motifs, and cis-elements were also analyzed. The expression pattern of PpHSP20-32 was highly consistent with shoot length changes during four critical growth stages of temperature-sensitive semi-dwarf peach 'Zhongyoutao 14' in response to increasing temperature. Transgenic Arabidopsis lines overexpressing PpHSP20-32 demonstrated that the gene can increase plant height and enhance thermotolerance. The results in this study supplied general information on the PpHSP20 gene family and further revealed the potential roles of PpHSP20-32 in plant height, in addition to the response to heat stress.