Genome-Wide Identiﬁcation and Characterization of Aldo-Keto Reductase ( AKR ) Gene Family in Response to Abiotic Stresses in Solanum lycopersicum

: Tomato is one of the most popular and nutritious vegetables worldwide, but their production and quality are threatened by various stresses in the environment in which they are grown. Thus, the resistance and tolerance of tomatoes to various biotic and abiotic stresses should be improved. Aldo-keto reductases (AKR) are a superfamily of NAD(P)(H)-dependent oxidoreductases that play multiple roles in abiotic and biotic stress defenses by detoxiﬁcation and reactive oxygen species (ROS) clearance pathways. Here, 28 identiﬁed AKR family genes of tomatoes were identiﬁed genome-wide, and their characteristics, including chromosomal location, gene structures, protein motifs, and system evolution, were analyzed. Furthermore, the phylogenetic and syntenic relationships in Arabidopsis thaliana , rice, and tomatoes were compared. Expression patterns at different tissues and in response to abiotic stresses, such as drought and salt, were monitored to further explore the function of SlAKRs . Finally, three SlAKRs candidate genes were silenced by Virus induced gene silencing (VIGS) systems in Solanum lycopersicum , showing sensitivity to drought and salt stresses with low contents of proline (Pro) and peroxidase (POD) and high content of malonaldehyde (MDA). This study provides the characteristics and potential functions of SlAKRs in response to abiotic stresses that will be helpful for further studies in S. lycopersicum .


Introduction
Aldo-keto reductase (AKR) is a superfamily of NAD(P)(H)-dependent oxidoreductases, which is composed of many monomer types of protein members [1,2]. At present, more than 190 AKR family members have been identified because of the high conservation of the AKR family during evolution, and AKR members are classified into 16 families on the basis of amino acid sequence similarity, with more than 60% high homology in the same subfamily [2,3]. Each AKR member contains a conserved structural domain of AKR with classical (α/β) 8-barrel fold structures. Moreover, three loops (Loops A, B, and C) are observed at the carboxyl terminal of these structures with size and number differences according to their specificity. However, AKR family members have similar three-dimensional structures [4]. AKRs have been reported to be widely found in animals and plants involved in the modification of bioactive compounds and detoxification and have a wide range of substrates (such as reduction of sugars, aldehydes, and ketones; hormones; chemical carcinogens; and various carbonyl compounds) [5]. In animals, AKRs play multiple roles and are responsible for preventing mammalian cells from the toxic and carcinogenic effects of different genotoxic and nongenotoxic chemicals by reduction [6]. In plants, AKRs play roles in metabolic pathways, such as activated aldehyde detoxification, osmotic pressure synthesis, secondary metabolism, and membrane transport, and are reported to have functions in biotic and abiotic stress pathways by regulating secondary metabolism and maintaining cellular osmotic pressure [7,8].

Identification and Characteristics of SlAKR in S. lycopersicum
A total of 28 predicted AKR genes were exhibited on the SGN database and further confirmed by the NCBI database. The conserved domains of AKR candidate proteins were searched on the Pfam database with the No. of pfam00248. Conserved protein domains were further confirmed by InterPro, which showed that all 28 members contained at least one AKR domain. As a result, these members were considered the candidate genes of the AKR family. The total messages of the SlAKR family, including genomic location, coding sequence length (CDS), exon number, protein length, protein molecular weight (MW), and isoelectric point (PI), are listed in Table 1 and labeled using their gene loc no. on SGN.
Solyc09g074310.3.1 had the most exons, up to 12, with a CDS length of 1080 bp encoding 359 amino acids but was not the longest gene in the SlAKR family. Solyc11g067160.2.1 was the longest gene with 1197 bp CDS encoding 398 amino acids with nine exons, and the other members of the SlARK family had exons ranging from 2 to 10. The shortest gene was Solyc09g097990.1.1, with 273 nucleotides encoding only 90 amino acids. The MW of SlARK family proteins ranged from 10,277.02 D to 44,437.18 D, and PI was bridged from 5.12 to 9.28.

Locations of SlAKR on the S. lycopersicum Genome
Two chromosomes were found in S. lycopersicum, but 28 SlAKR family genes were unevenly distributed on eight chromosomes, i.e., Chr.01, Chr.03, Chr.04, Chr.06, Chr.07, Chr.09, Chr.11, and Chr.12 ( Figure 1). Most genes were located on Chr.09 with 12 members, and five genes each were found on Chr.01 and Chr.05 ( Figure 1). However, one gene each was observed on Chr.04, Chr.06, and Chr.07. Four members of the AtAKR family were found in the Tair database, with three genes located on Chr.2 and only one gene located on Chr.3 (Table S1). However, 22 AKR family genes were found in rice and unevenly distributed on 10 of 12 chromosomes in rice (Table S1). These results showed that the AKR genes of S. lycopersicum were 7-and 1.5-fold more than those of A. thaliana and rice, respectively. Moreover, the AKR genes of rice were distributed more asymmetrically than those of tomato.

Phylogenic Relationship and Gene Structures of AKR Genes in S. lycopersicum
The CDS sequences of SlAKRs were downloaded from the SNG database as fasta forms and aligned by MEGA11 to build the phylogenic tree by the neighbor-joining method. All of the 28 SlAKR genes were clustered into three subfamilies, i.e., I, II, and III. Subfamily I was the largest subfamily with 13 members, but the smallest subgroup, i.e., subfamily II, only had three members. Furthermore, the gene structures of SlAKR genes were analyzed in online tools (Figure 2). The Solyc09g074310.3.1 of subfamily II had the most exons, up to 12, but the two other members had 5 and 8 exons. However, the SlAKR genes of subfamily I had fewer exons ranging from 2 to 7 than subfamily II with 5 to 10 exons.

Phylogenic Relationship and Gene Structures of AKR Genes in S. lycopersicum
The CDS sequences of SlAKRs were downloaded from the SNG database as fasta forms and aligned by MEGA11 to build the phylogenic tree by the neighbor-joining method. All of the 28 SlAKR genes were clustered into three subfamilies, i.e., I, II, and III. Subfamily I was the largest subfamily with 13 members, but the smallest subgroup, i.e., subfamily II, only had three members. Furthermore, the gene structures of SlAKR genes were analyzed in online tools ( Figure 2). The Solyc09g074310.3.1 of subfamily II had the most exons, up to 12, but the two other members had 5 and 8 exons. However, the SlAKR genes of subfamily I had fewer exons ranging from 2 to 7 than subfamily II with 5 to 10 exons.

Alignment and Phylogenic Analysis of the AKR Protein in S. lycopersicum
First, all SlAKR protein sequences were downloaded from the SGN database and aligned by the DNAMAN8 software ( Figure 3A). On the basis of the multiple alignments of the SlAKR protein, four critical conserved sites were found in the front of these sequences labeled as g, g, d, and y in Figure 3A. These sequences were further subjected to cluster analysis by WEBLOGO [22], as shown by the increase in other conserved sites in Figure 3B. Furthermore, the conserved domains of AKR existed in SlAKRs. The positions of each conserved domain is listed in Table S2.
The phylogenetic relationships of 28 SlAKR proteins were calculated by the neighborjoining method in MEGA11 with a bootstrap of 1000, Figure 4A. The protein motifs of SlAKRs were searched by the online tool MEME and distributed on each SlAKR with 8 motifs in Figure 4B. Among these motifs, the most conservative motif 5 existed in 27 members of SlAKR, and motif 7 existed in 26 proteins. Following Motif 1 existed in 25 members except for Solyc03g082560.3.1, Solyc09g097950.3.1 and Solyc09g097990.1.1. The WEBlogo analysis results of each motif in SlAKR proteins are shown in Figure 4C and Table S3. Three loops, i.e., Loops A, B, and C, in the AKR domain were reported. Our results showed that motif 1 comprised the most conserved structures of AKR, such as Loop A.

Alignment and Phylogenic Analysis of the AKR Protein in S. lycopersicum
First, all SlAKR protein sequences were downloaded from the SGN database and aligned by the DNAMAN8 software ( Figure 3A). On the basis of the multiple alignments of the SlAKR protein, four critical conserved sites were found in the front of these sequences labeled as g, g, d, and y in Figure 3A. These sequences were further subjected to cluster analysis by WEBLOGO [22], as shown by the increase in other conserved sites in Figure 3B. Furthermore, the conserved domains of AKR existed in SlAKRs. The positions of each conserved domain is listed in Table S2.

Phylogenic Relationships of AKR Protein in S. lycopersicum, A. thaliana, and Oryza sativa
The protein sequences of 28 SlAKRs, 22 OsAKRs, and 4 AtAKRs (Table S4) were obtained from the database to construct the circular evolutionary tree by the neighbor-joining method to explore the phylogenic relationships of the AKR protein among S. lycopersicum, A. thaliana, and O. sativa. All AKR proteins of these three species were divided into six subgroups, i.e., I to VI, labeled with different colors ( Figure 5). All four AtAKRs were clustered into a sub-branch in subgroup IV. However, the AKRs of rice were distributed in many more subfamilies with similarities to S. lycopersicum in each subgroup. Subgroup II only had three AKR members.

Collinearity Relationship of AKR Genes in S. lycopersicum, A. thaliana, and O. sativa
The collinearity analysis of AKR genes was carried out ( Figure 6) to study the repetition events of the AKR gene family in S. lycopersicum, and the distribution and arrangement relationships of AKR genes between S. lycopersicum and other species were explored. First, the synteny of SlAKR genes was analyzed ( Figure 6A). A total of 34 075 genes were found in the genome of S. lycopersicum with the Version SL4.0, among which 7217 genes were collinearity genes accounting for 21.18%. However, no collinearity gene was observed among the SlAKR gene family. Second, the collinearity relationships of AKRs among the dicot plant A. thaliana, monocot plant O. sativa, and S. lycopersicum were determined ( Figure 6B). Nine pairs of collinearity genes, i.e., Solyc06g053600    Figure 4C and Table S3. Three loops, i.e., Loops A, B, and C, in the AKR domain were reported. Our results showed that motif 1 comprised the most conserved structures of AKR, such as Loop A.

Expression Levels of SlAKR Family Genes in Various Tissues
The transcriptional levels of SlAKR family genes were detected by qPCR in different organs and tissues, including roots, stems, leaves, flowers, and fruits with a diameter of 1 cm, to investigate the potential functions of SlAKRs in tomatoes (Figure 7). SlAKR genes widely enhanced their expression levels in stems, leaves, and flowers and leaves. Compared with roots, the expression levels of most SlAKRs (85.7%) were significantly increased in leaves, such as Solyc09g097980.4.1, with the highest expression. The transcriptional levels of the majority of SlAKRs genes were downregulated in fruits, but four genes, such as Solyc01g065490. 4

Phylogenic Relationships of AKR Protein in S. lycopersicum, A. thaliana, and Oryza sativa
The protein sequences of 28 SlAKRs, 22 OsAKRs, and 4 AtAKRs (Table S4) were obtained from the databas construct the circular evolutionary tree by the neighbor-joining method to explore the phylogenic relationships the AKR protein among S. lycopersicum, A. thaliana, and O. sativa. All AKR proteins of these three species w divided into six subgroups, i.e., I to VI, labeled with different colors (Figure 5). All four AtAKRs were cluster into a sub-branch in subgroup IV. However, the AKRs of rice were distributed in many more subfamilies with similarities to S. lycopersicum in each subgroup. Subgroup II only had three AKR members.

Expression Profiles of SlAKR Family Genes under Drought and Salt Stresses in AC Lines
The tomato AC inbred lines were treated with 10% PEG6000 as drought conditions and 200 mM NaCl as salt stresses for 0, 3, 6, 12, 24, 36, and 48 h to further explore the functions of SlAKRs in response to abiotic stresses. All plants were harvested for the expression profiles of the SlAKRs family by real-time fluorescent quantitative PCR (Figures 8 and 9). In response to drought stresses, the expression levels of most SlAKR genes (75%) were significantly induced at 24 h treatment of 10% PEG6000 treatments and then decreased at 48 h of drought treatments.

Expression Levels of SlAKR Family Genes in Various Tissues
The transcriptional levels of SlAKR family genes were detected by qPCR in different organs and tissues, including roots, stems, leaves, flowers, and fruits with a diameter of 1 cm, to investigate the potential functions of SlAKRs in tomatoes (Figure 7). SlAKR genes widely enhanced their expression levels in stems, leaves, and flowers and leaves. Compared with roots, the expression levels of most SlAKRs (85.7%) were significantly increased in leaves, such as Solyc09g097980. 4    were significantly changed in response to drought and salt stresses, which indicated that they might play roles in abiotic stresses. About 400-600 bp of these three genes were cloned and constructed to the pTRV2 vector derived by the 35S-mediating virus-induced gene silencing (VIGS) system. The recombinant plasmids of 35S:pTRV2-Solyc09g011240.3.1, 35S:pTRV2-Solyc07g043570.3.1, and 35S:pTRV2-Solyc01g106450.3.1 were transformed into tomato seedlings via Agrobacterium-mediated transformation. The pTRV2 vector and pTRV2-phytoene desaturase (PDS) were also transformed into the tomato, which seemed indicators. Recombinant plasmids were transformed into 50 independent tomato seedlings, and the expression levels of each target gene were detected by qPCR (Table S5) to calculate the silencing efficiency when the silenced transgenic tomatoes of PDS turned white. The silencing efficiency of these three target genes was up to 40-70% ( Figure S1), and 15 transgenic tomato lines with similar silencing efficiency were selected for the next detection of physiological indicators.
Malonaldehyde (MDA) content, proline (Pro) content, and peroxidase (POD) enzyme activity in plants are important physiological indices to evaluate the degree of plant resistance to abiotic stresses. MDA is one of the most important products after the lipid peroxidation of the membrane, aggravating membrane damage. POD enzyme is one of the key enzyme defense systems in plants and can eliminate excessive free radicals under adverse conditions to improve the stress resistance of plants. Proline can maintain osmotic balance at the cellular level and is an important osmotic regulator [23]. The si-

Discussion
AKRs encoding NADP(H)-dependent oxidoreductases exist in nearly all phyla and function in the phase 1 metabolism of endogenous substrates and xenobiotics [2,4]. More than 190 annotated proteins of AKRs were searched by the alignments of genome data classified into 16 families [13,24]. The AKRs of humans play multiple roles in a variety of disease processes, and selective enzyme inhibitors have been sought as chemical probes and as possible therapeutics [2]. Plant AKRs significantly participate in multiple stresses to confer tolerance, including abiotic responses (such as salt and drought stresses) and biotic stresses (such as pathogen defense) [25]. Moreover, AKRs function in primary and secondary metabolic pathways during the growth and development of the plant. With the continuous study of functions, increasing plant AKR families, such as A. thaliana [10], rice [11], H. vulgare [12], and M. truncatula [13], have been identified and excavated. However, the genome-wide analysis of the AKR gene family has not been performed in tomatoes yet. In our study, 28 SlAKR genes were identified from the tomato genome, and the bioinformation and expression levels under abiotic stresses were further analyzed to provide comprehensive information about the AKR family in tomatoes.

Analysis of SlAKRs Characterization and Phylogenetic Relationships
All gene structures of 28 SlAKR genes were analyzed, and results showed that similar gene structures were clustered in the same subfamily in accordance with the phylogenetic relationships of SlAKR genes (Figure 2). Multiple protein sequences, motifs, and conserved domains of SlAKRs were compared. The locations of each AKR conserved domain were similar in the whole protein, thus contributing to the classical (α/β) 8-barrel fold structures [4]. Moreover, the cluster analysis of AKR conserved domains of tomatoes found six highly conserved amino acid sites, such as glycine (G-78), glycine (G-112), aspartic acid (D-117), tyrosine (Y-122), lysine (K-160), and glycine (G-405) in Figure 3B, which were the conserved sites of most AKRs in 40 different plant species [25]. The 28 SlAKRs proteins encoded an average of 318 amino acids with MW of 35.6 KD, which was similar to the AKRs reported previously about 320 amino acids with 33-37 kD [2]. However, MtAKRs contained 336 amino acids on average, with 37 kDa in M. truncatula [13], and this value was slightly larger than average AKRs and even SlAKRs. Only one conservative motif was found in all 28 SlAKRs encoding the AKR domain.
SlAKRs were not evenly distributed on 12 chromosomes, as shown in Figure 1. No SlAKR gene was located on chromosomes 2, 5, 8, and 10. Most SlAKR genes(12) on chromosome 9 were not even either. The uneven distribution of genes may be related to species evolution and genetic variation [26]. No collinearity gene was observed among the SlAKR gene family ( Figure 6A), and no duplication event of SlAKR was observed on the genome. However, 9 and 8 pairs of collinearity genes were found among tomato, A. thaliana, and rice, indicating that SlAKR genes shared similar syntenic relationships with AtAKRs and OsAKRs. These collinearity genes shared close phylogenetic relationships in different species, but Solyc09g011240.3.1 had collinearity genes in A. thaliana and rice.

Expression Profiles and Candidate Genes of SlAKRs
The expression profiles of 28 SlAKRs in different tissues, such as root, stem, leaves, flower, and fruit, were detected by qPCR, which showed that most SlAKRs were induced in stem, leaves, and flower. SlAKRs had low expression profiles in root and fruit. Solyc03g093270.3.1, Solyc09g082720.3.1, and Solyc09g097960.3.1 had low expression levels in all tissues. Solyc09g074310.3.1 and Solyc09g097980.4.1 were specifically expressed in stems and leaves, respectively, which indicated that the tissue-specific expression of the SlAKR family members might be related to biological functions. SlAKRs consist of MtAKRs [13]. Plant AKRs play multiple roles in abiotic and biotic stress defenses, including ozone, drought, salinity, hypoxia, and P. syringae inoculation [18,[27][28][29]. MsALR, AKR4C8, and AKR4C9 are reported to be highly induced under abiotic stress [10,30] because AKR mediates stress tolerance by detoxification and ROS clearance in cell-damaging reactions [25]. Moreover, the overexpression of barley AKR4C9 enhances their tolerance in response to freezing, oxidative, and cadmium stresses [8,31]. Thus, the expression profiles of SlAKR genes in response to drought and salt stresses were detected in each treatment. Most SlAKR genes were induced by drought treatments at various times. Among these genes, Solyc01g065490.4.1, Solyc07g043570.3.1, and Solyc09g011240.3.1 showed significant induction. However, results due to salt stresses were less dramatically changed compared with those due to drought. Solyc07g043570.3.1, Solyc09g011240.3.1, Solyc09g082720.3.1, and Solyc09g097960.3.1 showed rapid increases in their expression levels in response to salt stresses. On the basis of expression characteristics, Solyc09g011240.3.1, Solyc07g043570.3.1, and Solyc01g106450.3.1 were silenced in tomatoes by the VIGS system, which showed sensitivity to drought and salt stresses with low contents of Pro and POD and high content of MDA at each stress stage ( Figure 10). The overexpression of OsAKR4C15 in A. thaliana lines resulted in lower contents of MDA and MG under both stresses and control conditions than wild types [32]. These results suggested that AKR members of tomatoes were in response to drought and salt stresses and that Solyc09g011240.3.1, Solyc07g043570.3.1, and Solyc01g106450.3.1 might play positive roles in regulating the tolerance of tomatoes in drought and salt stresses. stresses with low contents of Pro and POD and high content of MDA at each stress stage (Figure 10). The overexpression of OsAKR4C15 in A. thaliana lines resulted in lower contents of MDA and MG under both stresses and control conditions than wild types [32]. These results suggested that AKR members of tomatoes were in response to drought and salt stresses and that Solyc09g011240.3.1, Solyc07g043570.3.1, and Solyc01g106450.3.1 might play positive roles in regulating the tolerance of tomatoes in drought and salt stresses.

Plant Materials
Tomato Ailsa Craig (AC) were sown in the soil and transferred to a greenhouse with 16 h/8 h for day/night and temperatures of 25 • C/20 • C until four leaves. A total of 21 seedlings of tomatoes were treated with 200 mM NaCl for 0, 3, 6, 12, 24, 36, and 48 h with water as control, and 21 other seedlings were treated with 10% PEG 6000 for the same treated time as NaCl treatment with three biological replicates. A total of 126 tomato seedlings were harvested, frozen quickly in liquid nitrogen, and stored at −80 • C.
When AC tomato seedlings grew to the flowering stage, different tissues of three independent seedlings, including roots, stems, leaves, flowers, and 1 cm green fruits, were collected. All parts of the tissue were frozen with liquid nitrogen and stored at −80 • C.

RNA Extraction and Real-Time PCR
The plant samples of three biological replicates were ground into powder in liquid nitrogen, and the total RNA was isolated by the Trizol reagent (Invitrogen, Thermo Fisher Technology Co., LTD., Waltham, MA, USA) in accordance with the manufacturer's manual and reversely transcribed into cDNA by the Vazyme reverse transcription kit (Vazyme Code: R223-01, Vazyme Biotech Co., Ltd., Nanjing, China).
The total volume of qPCR was 15 µL:7.5 µL qPCR-Mix (Vazyme Biotech Co., Ltd.). The cDNA was diluted 20 times, and 3 µL was collected as a template. About 0.5 µL forward and reverse primers were added with dd H 2 O to obtain a volume of 3.5 µL. The PCR system was subjected to abi-q3 fluorescence quantitative PCR with a three-step procedure with three biological replicates [33]. The PCR primers of SlAKRs were designed by Premier Express software Ver in the nonconservative areas of each SlAKR (Table S5), and all genes were normalized by the reference gene SlActin.

Isolation of AKR Family Members
AKR family members were searched on the SGN (https://solgenomics.net/ (accessed on 12 July 2021)) and NCBI (https://www.ncbi.nlm.nih.gov/ (accessed on 20 July 2021)) databases, and candidate genes with the conserved domains of AKR, such as pfam00248, were screened out. The conserved domains of the AKR family were searched on the Pfam (http://pfam.xfam.org/ (accessed on 10 August 2021)). All CDS, genome, and protein sequences of AKR were downloaded in FASTA format from the SGN database. All proteins of AKR were confirmed by InterPro (http://www.ebi.ac.uk/interpro/ (accessed on 10 August 2021)). The MW and PI values of tomato AKR members were analyzed by the online ExPASy website (https://web.expasy.org/compute_pi/ (accessed on 10 August 2021)).

Location and Structures of ARK Members
The location of SlAKR members and gene distribution of tomato chromosomes were obtained from the annotations on SGN (https://solgenomics.net (accessed on 20 July 2021)), which were redrawn by TBtools [34]. The gene structures of SlAKR were analyzed by an online tool GSDS2.0 (http://gsds.gao-lab.org/ (accessed on 15 August 2021)) [35] in accordance with genome and CDS sequences.

Phylogenetic Analysis
The AKR protein sequences of A. thaliana and rice were downloaded from the Tair (https://www.arabidopsis.org (accessed on 25 July 2021)) and Gramene (https://www. gramene.org/ (accessed on 25 July 2021)) databases. The protein sequences of A. thaliana, rice, and tomatoes and the CDS sequences of tomatoes were blasted by MEGA 11, and phylogenetic trees were built using the online tool iTOL: Interactive Tree of Life (https: //itol.embl.de/upload.cgi (accessed on 15 January 2022)) through the neighbor-joining method. The motifs of the tomato AKR family were searched on the MEME database (http://meme-suite.org/tools/meme (accessed on 21 March 2022)).

Synteny
The distributions and interchromosomal relationships of SlAKRs were obtained by MCScanX associated with TBtools. The synteny relationships of AKRs family members between A. thaliana and rice were also determined by MCScanX [36] and TBtools [34]. The genome and coding sequences of AKR in A. thaliana and rice were obtained from Tair and Gramene databases.

Construction of Recombinant Plasmids and Transformation of Tomato
Three candidate genes of the AKR family were amplified for 300-500 bp fragments to be constructed to the VIGS vectors pTRV2 with specific primers (Table S6), such as Solyc01g106450.3.1, Solyc07g043570.3.1, and Solyc09g011240.3.1. Recombinant plasmids were screened, sequenced, and then transformed into tomato AC by Agrobacterium tumefa-ciens (EHA105) injection with pTRV2 as the negative control. The albino gene of PDS was also transferred in AS tomatoes as the display agent.

Conflicts of Interest:
All the authors in this research declare no conflict of interest.