DREB1 and DREB2 Genes in Garlic (Allium sativum L.): Genome-Wide Identification, Characterization, and Stress Response

Dehydration-responsive element-binding (DREB) transcription factors (TFs) of the A1 and A2 subfamilies involved in plant stress responses have not yet been reported in Allium species. In this study, we used bioinformatics and comparative transcriptomics to identify and characterize DREB A1 and A2 genes redundant in garlic (Allium sativum L.) and analyze their expression in A. sativum cultivars differing in the sensitivity to cold and Fusarium infection. Eight A1 (AsaDREB1.1–1.8) and eight A2 (AsaDREB2.1–2.8) genes were identified. AsaDREB1.1–1.8 genes located in tandem on chromosome 1 had similar expression patterns, suggesting functional redundancy. AsaDREB2.1–2.8 were scattered on different chromosomes and had organ- and genotype-specific expressions. AsaDREB1 and AsaDREB2 promoters contained 7 and 9 hormone- and stress-responsive cis-regulatory elements, respectively, and 13 sites associated with TF binding and plant development. In both Fusarium-resistant and -sensitive cultivars, fungal infection upregulated the AsaDREB1.1–1.5, 1.8, 2.2, 2.6, and 2.8 genes and downregulated AsaDREB2.5, but the magnitude of response depended on the infection susceptibility of the cultivar. Cold exposure strongly upregulated the AsaDREB1 genes, but downregulated most AsaDREB2 genes. Our results provide the foundation for further functional analysis of the DREB TFs in Allium crops and could contribute to the breeding of stress-tolerant varieties.


Introduction
Garlic (Allium sativum L.) is an economically important vegetable crop belonging to the most numerous (over 920 species) Allium genus of monocotyledons [1][2][3]. The A. sativum bulbs and green parts are used as spices and as traditional remedies for various diseases because of their anticarcinogenic, antioxidant, antidiabetic, renoprotective, antiatherosclerotic, antibacterial, antifungal, and antihypertensive properties [4]. In addition, garlic and other Allium plants are in demand for intercropping and crop rotation as they improve soil quality and reduce soil fungal communities, including Fusarium [5][6][7][8][9]. The versatile use of Allium spp. suggests their high adaptability to various environmental conditions.
A. sativum originated about 10,000-5000 years ago in the Mediterranean region and Western and Central Asia; since then, this species has undergone long-time artificial selection and is currently cultivated all over the world [1,3], and some researchers suggest that most of the available A. sativum genotypes have emerged as a result of mutations in single clones [2]. However, despite the lack of sexual reproduction, garlic is characterized by a high phenotypic diversity [3,10,11]. As a result, A. sativum demonstrates good adaptability to various stresses, among which ambient temperature changes and fungal infections are considered to be important factors affecting garlic development and yield [3,[12][13][14][15].
AsaDREB2 proteins contained a highly conserved GKGGPxN motif, which was a part of the NLS sequence upstream of the AP2 domain (Supplementary Figure S1b) and which is common for DREB2 TFs of other plant species [38].
As A. thaliana is a dicot plant, we included DREB1 and DREB2 proteins of a monocot Zea mays L. (Zm) in the phylogenetic analysis. This addition did not affect the separate clustering of AsaDREB1 proteins, but allowed for the determination of putative maize orthologs (ZmDREB2.4-2.6) for AsaDREB2.3 (high bootstrap values), whereas AsaDREB2.5, 2.7, and the AsaDREB2.1/2.2/2.4/2.6/2.8 group formed distinct branches in which any homology was supported by only low bootstrap values (Supplementary Figure S2).
The phylogenetic division was supported by MEME-based analysis, which revealed 10 and 11 conserved motifs in the AsaDREB1/ATDREB1 and AsaDREB2/ATDREB2 sets, respectively ( Figure 3). AsaDREB1 proteins had a highly conserved motif arrangement, whereas AsaDREB2 proteins showed at least two distinct motif patterns; in both groups, motifs 1 and 2 constituted NLS together with the AP2 domain.

Analysis of AsaDREB1 and AsaDREB2 Promoter Regions
Considering the role of DREB genes in plant development and stress response, we searched for cis-acting elements in the 5 untranslated regions (UTRs) and promoters (~1 kb upstream of the start codon). As a result, 7 hormone-and 9 stress-responsive elements and 13 sites associated with TF binding, meristem development, cell cycle regulation, and plant reproduction were identified ( Table 2). Table 2. cis-elements found in the AsaDREB1 and AsaDREB2 regulatory regions (~1000 bp).  -binding site  2  2  5  1  7  1  4  3  3  3  2  5  2  5  2  3 Re2f-1 E2F-binding sites; seed-specific, cell-cycle regulated gene expression

Motif Name Annotation
bZIP-binding site 1 HD-Zip IV HD-Zip IV-binding site 1 Color intensity (pale to dark) corresponds to the number of cis-elements (low to high).
The most common elements were ABA-responsive ABRE motifs (lacking only in AsaDREB1.7 and AsaDREB1.8) and MYB TF binding sites (present in all of the identified genes), which were closely related to the ABA-mediated plant stress response [39]. Promoters of the AsaDREB1.1, 1.3, 1.5, 2.5, and 2.7 genes had the largest number of ABRE motifs, whereas AsaDREB1.3, 1.5, 1.7, 2.4, and 2.6 had the largest number of MYB-binding sites ( Table 2).
Wounding-and pathogen-responsive elements (three different motifs) were identified in most genes (except AsaDREB1.1, 1.3, 1.5, and 2.7). Among the abiotic stress-related elements, many were associated with anaerobic conditions, such as ARE (11 genes), and low temperature, such as LTR (7 genes). The STRE element associated with response to other abiotic stresses (heat, osmotic stress, low pH, and nutrient starvation) was more characteristic for the AsaDREB1 genes (the largest number in AsaDREB1.2), whereas among the AsaDREB2 genes, only AsaDREB2.7 contained STRE. The least represented motif found only in AsaDREB1.6 was related to drought response. TC-rich repeats and W-box involved in plant defense were found in the promoters of three (AsaDREB1.3, 1.6, and 2.3) and four (AsaDREB1.3, 1.5, 2.1, and 2.2) genes, respectively ( Table 2).
The promoters of two AsaDREB1 and four AsaDREB2 genes contained nitrogenresponsive motifs associated with endosperm-specific gene expression. AsaDREB1.7, 1.8, and 2.5 had sites responsible for meristem-specific gene expression. Promoters of AsaDREB2.5 and 2.8 contained bZIP and HD-Zip IV TF-binding sites, respectively ( Table 2).
The other AsaDREB2 genes (AsaDREB2.1, 2.2, 2.5, and 2.8) were expressed in all of the analyzed tissues; among these genes, the strongest expression was observed for AsaDREB2.5 (maximum in the roots, flowers, and stage 4-8 bulbs, minimum in stage 1 and 3 bulbs). During bulb development, the mRNA levels of AsaDREB2.1 and AsaDREB2.2 fluctuated from stage 3 to 8, whereas those of AsaDREB2.5 gradually increased throughout stages 1-5, sharply decreased at stage 6, and again increased at stages 7 and 8, and those of AsaDREB2.8 gradually decreased from stage 1 to 8 ( Figure 4).
The other AsaDREB2 genes (AsaDREB2.1, 2.2, 2.5, and 2.8) were expressed in all of the analyzed tissues; among these genes, the strongest expression was observed for AsaDREB2.5 (maximum in the roots, flowers, and stage 4-8 bulbs, minimum in stage 1 and 3 bulbs). During bulb development, the mRNA levels of AsaDREB2.1 and AsaDREB2.2 fluctuated from stage 3 to 8, whereas those of AsaDREB2.5 gradually increased throughout stages 1-5, sharply decreased at stage 6, and again increased at stages 7 and 8, and those of AsaDREB2.8 gradually decreased from stage 1 to 8 ( Figure 4). The results indicated that the expression patterns of most AsaDREB1 genes (AsaDREB 1.1-1.5 and 1.8) in cv. Sarmat were similar to those in cv. Ershuizao. The two cultivars differed by the absence of AsaDREB1.6 and 1.7 transcripts in all cv. Sarmat tissues and by the ratio of expression levels in specific organs. The strongest expression was observed in the roots (AsaDREB1.1, 1.3, and 1.5), leaves (AsaDREB1.4, and 1.8), or both (AsaDREB1.2) and the weakest-in the peduncles and receptacles. AsaDREB1.1-1.5 and 1.8 had relatively low expression levels in the air bulbs, stems, and bulbs ( Figure 5a).
Compared with the AsaDREB1 genes, the AsaDREB2 genes showed more significant inter-group transcriptional differences. Thus, in cv. Sarmat, transcripts of AsaDREB2.1 were found only in the peduncles, receptacles, and air bulbs, and those of its closest structural homolog AsaDREB2.2 were observed in all organs, with the maximum in the leaves and pseudostems and minimum in the roots (Figure 5b), whereas in cv. Ershuizao, the two genes had a similar expression pattern with the maximum in the roots (Figure 4). In cv. Sarmat, AsaDREB2.3 transcription was strong in the pseudostems but weak in the bulbs, peduncles, and air bulbs (Figure 5b), whereas in cv. Ershuizao, it was not expressed in the pseudostems and stage 8 bulbs (Figure 4). AsaDREB2.6 was transcribed in all analyzed organs (maximum in the peduncles and air bulbs and minimum in the roots; Figure 5b), but was practically not expressed in cv. Ershuizao (Figure 4). In cv. Sarmat, the AsaDREB2.5 mRNA level in stage 8 bulbs was significantly lower than in the pseudostems and leaves (Figure 5b), whereas in cv. Ershuizao, the opposite pattern was observed (Figure 4). At the same time, in both cultivars, AsaDREB2.3 transcription was absent in the roots and leaves, AsaDREB2.4 was not transcribed in any organ, and AsaDREB2.5 and AsaDREB2.8 were expressed in all of the analyzed tissues (Figures 4 and 5b).
Next, we analyzed the transcription of the AsaDREB genes in the roots, stems, stage 8 bulbs, pseudostems, leaves, peduncles, receptacles, and air bulbs of A. sativum cv. Sarmat ( Figure 5) using gene-specific primers (Supplementary Table S1). The results indicated that the expression patterns of most AsaDREB1 genes (AsaDREB1.1-1.5 and 1.8) in cv. Sarmat were similar to those in cv. Ershuizao. The two cultivars differed by the absence of AsaDREB1.6 and 1.7 transcripts in all cv. Sarmat tis- Figure 5. Transcription of the AsaDREB1 and AsaDREB2 genes in A. sativum cv. Sarmat; ps.stems, pseudostems. The data were normalized to GAPDH and UBQ mRNA levels. The significance of differences in the gene expression between organ types was analyzed using one-way ANOVA; obtained p-values are given in the Supplementary Table S2 (p < 0.05 indicates significant gene expression difference between organ types).

AsaDREB1 and AsaDREB2 Expression in Garlic Seedlings in Response to Fusarium Infection
Considering the distinct patterns of hormone-and pathogen-responsive cis-regulatory elements in the promoters of the AsaDREB1 and AsaDREB2 genes (Table 2), we analyzed AsaDREB expression in the roots of F. proliferatum-infected garlic cultivars Sarmat and Strelets resistant and susceptible to FBR, respectively.
Among the AsaDREB1 genes, AsaDREB1.6 and 1.7 were not expressed, whereas all other genes were upregulated at 96 h post inoculation (hpi) in both cultivars, but the response in FBR-resistant cv. Sarmat was much stronger than in FBR-susceptible cv. Strelets. Thus, at 24 hpi, the mRNA levels of AsaDREB1.1-1.5 and 1.8 were similar in the two cultivars, whereas at 96 hpi, the upregulation was 3.5-10 times greater in cv. Sarmat (35-2500 times vs. control) than in cv. Strelets (2.5-300 times) (Figure 6a). . The data were normalized to GAPDH and UBQ mRNA levels and presented as fold change (mean ± SE) of control (expression in cv. Sarmat at 24 hpi taken as 1); * p < 0.05 compared with the uninfected control. The significance of differences in gene expression between the control and experiment, as well as between the time-dependent control and experiment data, was analyzed using one-way ANOVA; obtained p-values are given in Supplementary Table  S3.
Among the AsaDREB2 genes, four (AsaDREB2.2, 2.5, 2.6, and 2.8) were differentially regulated in response to F. proliferatum infection. AsaDREB2.2 and 2.6 were upregulated and AsaDREB2.5 was downregulated in both cultivars, whereas the AsaDREB2.8 gene was upregulated at 24 and 96 hpi in cv. Sarmat but downregulated at 24 hpi and upregulated at 96 hpi in cv. Strelets (Figure 6b). Overall, the activation of the AsaDREB2 genes was significantly stronger in the FBR-resistant than in the FBR-susceptible cultivar, whereas the inhibition (AsaDREB2.5) was weaker (Figure 6b).
Thus, in garlic, the magnitude of the AsaDREB transcriptional response to F. proliferatum infection appeared to depend on FBR susceptibility, although the pattern was the same for all of the genes (except AsaDREB2.8). Very similar infection response patterns of AsaDREB1 genes (except AsaDREB1.6 and 1.7) (Figure 6a) corresponded to the clustering of these genes on chromosome 1 (Figure 1a). The close structural homology of . The data were normalized to GAPDH and UBQ mRNA levels and presented as fold change (mean ± SE) of control (expression in cv. Sarmat at 24 hpi taken as 1); * p < 0.05 compared with the uninfected control. The significance of differences in gene expression between the control and experiment, as well as between the time-dependent control and experiment data, was analyzed using one-way ANOVA; obtained p-values are given in Supplementary Table S3. Among the AsaDREB2 genes, four (AsaDREB2.2, 2.5, 2.6, and 2.8) were differentially regulated in response to F. proliferatum infection. AsaDREB2.2 and 2.6 were upregulated and AsaDREB2.5 was downregulated in both cultivars, whereas the AsaDREB2.8 gene was upregulated at 24 and 96 hpi in cv. Sarmat but downregulated at 24 hpi and upregulated at 96 hpi in cv. Strelets (Figure 6b). Overall, the activation of the AsaDREB2 genes was significantly stronger in the FBR-resistant than in the FBR-susceptible cultivar, whereas the inhibition (AsaDREB2.5) was weaker (Figure 6b).
Thus, in garlic, the magnitude of the AsaDREB transcriptional response to F. proliferatum infection appeared to depend on FBR susceptibility, although the pattern was the same for all of the genes (except AsaDREB2.8). Very similar infection response patterns of AsaDREB1 genes (except AsaDREB1.6 and 1.7) (Figure 6a) corresponded to the clustering of these genes on chromosome 1 (Figure 1a). The close structural homology of AsaDREB2.2 and AsaDREB2.6 ( Figure 2) was consistent with the same infection-induced transcriptional profiles, whereas the phylogenetically distant AsaDREB2.5 (Figure 2) showed the opposite activation pattern (Figure 6b).

AsaDREB1 and AsaDREB2 Expression in Garlic Seedlings in Response to Cold Stress
Given that cv. Strelets and Sarmat also differed in sensitivity to low temperatures (winter-hardy and medium winter-hardy, respectively), we analyzed the expression of the AsaDREB genes in the leaves of seedlings subjected to cold stress (+4 • C) for different times. Short-term exposure (2-6 h) had no effect on the leaf status, but after 24 h, slight wilting was observed in both cultivars. However, the effects on the AsaDREB transcription could be detected already at 2 h post treatment when all AsaDREB1 and half of AsaDREB2 genes showed a differential expression in response to cold (Figures 7 and S3). transcriptional profiles, whereas the phylogenetically distant AsaDREB2.5 (Figure 2) showed the opposite activation pattern (Figure 6b).

AsaDREB1 and AsaDREB2 Expression in Garlic Seedlings in Response to Cold Stress
Given that cv. Strelets and Sarmat also differed in sensitivity to low temperatures (winter-hardy and medium winter-hardy, respectively), we analyzed the expression of the AsaDREB genes in the leaves of seedlings subjected to cold stress (+4 °C) for different times. Short-term exposure (2-6 h) had no effect on the leaf status, but after 24 h, slight wilting was observed in both cultivars. However, the effects on the AsaDREB transcription could be detected already at 2 h post treatment when all AsaDREB1 and half of AsaDREB2 genes showed a differential expression in response to cold (Figures 7 and S3). Strelets and medium winter-hardy cv. Sarmat after cold stress. Plants were exposed to cold (+4 °C) for 2, 4, 6, and 24 h and analyzed for the expression of the indicated AsaDREB genes in the leaves. The data were normalized to GAPDH and UBQ mRNA levels; transcriptions at 0 h (normal conditions before treatment) were taken as 1. The color gradient indicated expression decrease (blue) and increase (red) relative to the control. * p < 0.01 compared with the control (0 h).
Thus, in both cultivars, significant activation of AsaDREB1.1-1.5 and 1.8 genes (~6-430 times) was observed already at 2 h, which continued to increase after 4 h (~1.7-4.5 times) and remained strong thereafter, despite some decrease. The strongest transcriptional response was observed for AsaDREB1.4 and the weakest for AsaDREB1.2 ( Figure  7).
In contrast, the AsaDREB2 genes were mostly downregulated by cold (although slight activation (< 2.5 times) was observed for some genes at certain time points), and the overall transcriptional response was much weaker than that of AsaDREB1 (Figure 7). The AsaDREB2 expression patterns were somewhat different in cv. Strelets and Sarmat; only AsaDREB2.8 showed similar bell-shaped expression dynamics (Figure 7).
Among the 10 cold-responsive AsaDREB genes, only four (AsaDREB1.5, 1.8, 2.2, and 2.6) contained low temperature-sensitive (LTR) cis-regulatory elements in the promoters (Table 2), and the transcriptional response of these genes was cultivar-dependent ( Figure  7). Consistent with this pattern, AsaDREB1. 1-1.4, and AsaDREB2.8, which did not contain LTR motifs (Table 2), had similar expression changes in both cultivars (Figure 7). At the same time, the AsaDREB2.5 gene, which also did not have LTR motifs (Table 2), showed a cultivar-specific response to cold (Figure 7). As the analysis of the cis-element composition was based on cv. Ershuizao genes, it is quite possible that the AsaDREB genes of cv. Sarmat and Strelets could differ in their regulatory regions.
Overall, these results indicate that the AsaDREB1 genes played an important role in the reaction of garlic to cold stress, whereas the involvement of the AsaDREB2 genes was not obvious. The presence of LTR elements in the promoters of the AsaDREB genes was Figure 7. Heatmap of the time-dependent AsaDREB gene expression in A. sativum winter-hardy cv. Strelets and medium winter-hardy cv. Sarmat after cold stress. Plants were exposed to cold (+4 • C) for 2, 4, 6, and 24 h and analyzed for the expression of the indicated AsaDREB genes in the leaves. The data were normalized to GAPDH and UBQ mRNA levels; transcriptions at 0 h (normal conditions before treatment) were taken as 1. The color gradient indicated expression decrease (blue) and increase (red) relative to the control. * p < 0.01 compared with the control (0 h).
Thus, in both cultivars, significant activation of AsaDREB1.1-1.5 and 1.8 genes (~6-430 times) was observed already at 2 h, which continued to increase after 4 h (~1.7-4.5 times) and remained strong thereafter, despite some decrease. The strongest transcriptional response was observed for AsaDREB1.4 and the weakest for AsaDREB1.2 (Figure 7).
In contrast, the AsaDREB2 genes were mostly downregulated by cold (although slight activation (<2.5 times) was observed for some genes at certain time points), and the overall transcriptional response was much weaker than that of AsaDREB1 (Figure 7). The AsaDREB2 expression patterns were somewhat different in cv. Strelets and Sarmat; only AsaDREB2.8 showed similar bell-shaped expression dynamics (Figure 7).
Among the 10 cold-responsive AsaDREB genes, only four (AsaDREB1.5, 1.8, 2.2, and 2.6) contained low temperature-sensitive (LTR) cis-regulatory elements in the promoters (Table 2), and the transcriptional response of these genes was cultivar-dependent ( Figure 7). Consistent with this pattern, AsaDREB1.1-1.4, and AsaDREB2.8, which did not contain LTR motifs (Table 2), had similar expression changes in both cultivars (Figure 7). At the same time, the AsaDREB2.5 gene, which also did not have LTR motifs (Table 2), showed a cultivar-specific response to cold (Figure 7). As the analysis of the cis-element composition was based on cv. Ershuizao genes, it is quite possible that the AsaDREB genes of cv. Sarmat and Strelets could differ in their regulatory regions.
Overall, these results indicate that the AsaDREB1 genes played an important role in the reaction of garlic to cold stress, whereas the involvement of the AsaDREB2 genes was not obvious. The presence of LTR elements in the promoters of the AsaDREB genes was likely associated with the observed genotype-specific response of garlic to low temperatures.

Discussion
In this study, we aimed to identify and characterize DREB genes of the A1 and A2 subfamilies in the garlic genome. Overall, we found eight DREB1 (A1) and eight DREB2 (A2) genes in cv. Ershuizao through genome-wide analysis ( Table 1). The number of AsaDREB genes was close to that in other plant species such as dicot A. thaliana (six A1 and eight A2 genes [38]) and monocots Saccharum spontaneum L. (Commelinids; twelve A1 and six A2 genes [37]) and D. catenatum (Asparagales; 26 DREB genes [36]). Considering that the high phenotypic variability of modern garlic was primarily due to the initial crosspollination between fertile wild relatives in the center of origin [3,10,11], it can be assumed that the DREB gene family emerged in ancestral garlic accessions before the species lost fertility. Structural division of the AsaDREB proteins into A1 and A2 subfamilies ( Figure 2) suggests their functional conservation, including the role in the plant defense response to stress.
The AsaDREB1 genes form a tandem on chromosome 1 (Figure 1a), which may indicate their origin through tandem duplication of the precursor gene. This fact together with separate phylogenetic clustering of AsaDREB1.1-1.8 proteins (Figure 2) suggested the functional redundancy of AsaDREB1 in the transcriptional regulation in A. sativum. Consistent with this notion, we observed similar tissue expression patterns of the AsaDREB1 genes (Figures 4 and 5a) despite significant differences in the cis-regulatory element profiles ( Table 2).
The AsaDREB1.2 gene was expressed in cv. Ershuizao much stronger than the other AsaDREB1 (Figure 4). AsaDREB1.2 occupies a base position on the phylogenetic dendrogram in relation to the other AsaDREB1 ( Figure 2) and is a possible precursor of the AsaDREB1 cluster, carrying new valuable functions (compared with the earlier originated AsaDREB2). A common signature inherited from AsaDREB1.2 and present in the other AsaDREB1 genes includes a strong expression in the garlic roots and buds ( Figure 4). Variations in the AsaDREB1 expression levels in the other plant organs (Figure 4) may indicate functional changes in the A1 group. Thus, the increased expression of AsaDREB1.4 and 1.8 in the pseudostem suggests their predominant regulatory activity in this tissue, whereas very low transcriptional levels of AsaDREB1.5 in the seedlings, leaves, and bulbs ( Figure 4) could be associated with the loss of the AsaDREB1.5 regulatory function in these organs.
The differences in the AsaDREB1 tissue expression profiles observed between cv. Ershuizao and Sarmat suggest garlic genotype-dependent functions of the AsaDREB1 genes. Thus, in contrast with cv. Ershuizao, the expression of AsaDREB1.2 was similar to that of AsaDREB1.4, 1.5, and 1.8 (Figure 5a). At the same time, the activation of AsaDREB1.4 and 1.8 in the pseudostems was preserved in cv. Sarmat (Figure 5a), suggesting that the neofunctionalization of the two genes is conserved in different garlic cultivars. The absence of AsaDREB1.6 and 1.7 transcription in cv. Sarmat may indicate their genotype-specific pseudogenic nature.
Similar to DREB1 genes of other plants [23][24][25][26][27]29], AsaDREB1 genes (except for unexpressed AsaDREB1.6 and 1.7) were strongly activated by cold (Figure 7). Among them, the weakest induction was observed for AsaDREB1.2, which may indicate that the role in cold response acquired by this ancestral gene was later strengthened in the descendant AsaDREB1 genes. Thus, the highest induction level of AsaDREB1.4 in cv. Sarmat and Strelets (Figure 7) may indicate the predominant activity of this AsaDREB1 gene in the cold response of garlic; the strong expression of the other two genes, AsaDREB1.3 and 1.8, in cold-treated plants also suggests their significant contribution to garlic protection against low temperatures. The cultivar-dependent expression of AsaDREB1.3, 1.4, and 1.8 in response to cold treatment (Figure 7) could be associated with the difference in winter hardiness.
Our results indicated that the AsaDREB1 genes could also be involved in the garlic defense against F. proliferatum infection, as evidenced by the much stronger activation of AsaDREB1.1-1.5 and 1.8 in the FBR-resistant cv. Sarmat than in the FBR-susceptible cv. Strelets (Figure 6a). It should be noted that AsaDREB1.2 had the highest expression level (Figure 6a), suggesting that the responsiveness to infection was acquired by the AsaDREB1 genes relatively early in the evolution, but gradually weakened in the descendant genes. A still high activation level of AsaDREB1.4 and 1.8 (Figure 6a) suggests that these genes, along with AsaDREB1.2, are closely involved in the reaction of garlic to fungal infection.
Cumulatively, our findings indicate that AsaDREB1.2 is a precursor gene for the AsaDREB1 subfamily in garlic, which carried cold and infection sensitivity traits to the AsaDREB1 genes emerged later. The AsaDREB1 genes are most active in the roots, which represent one of the stress-sensitive plant tissues; among these genes, AsaDREB1.1-1.5 and 1.8 participate in plant response to cold stress and F. proliferatum infection. Three genes, AsaDREB1.2, 1.4, and 1.8, may perform most of the DREB1 function in garlic, whereas the other AsaDREB1 genes could play redundant roles.
The genes of the A2 subfamily are mainly scattered over different chromosomes (Figure 1a), which may indicate their origin through segmental gene duplications and suggest functional diversification. The latter notion is consistent with the distinct expression patterns of the AsaDREB2 genes. AsaDREB2.4 is not expressed in garlic, being probably a pseudogene, whereas the absence of AsaDREB2.6 and 2.7 transcripts in one cultivar and the presence in the other (Figures 4, 5b, 6b and 7) suggest a genotype-dependent pseudogene status of these AsaDREB2 genes.
The AsaDREB2 genes were transcribed in a tissue-specific manner both in cv. Ershuizao ( Figure 4) and cv. Sarmat (Figure 5b), which may indicate their differential functional activity in garlic organs. The highest expression levels in all tissues of both cultivars were observed for AsaDREB2.5 (Figures 4 and 5b), suggesting its primary role in performing the DREB2 function in garlic, while the other AsaDREB2 genes may complement AsaDREB2.5 activity in specific organs and/or genotypes.
The expression profiles of some AsaDREB2 genes (Figures 4 and 5b) were consistent with their phylogeny (Figure 2). Thus, AsaDREB2.1/2.2 and AsaDREB2.4/2.6 had similar expression levels and were clustered together in the dendrogram; the two clusters were grouped, but with low bootstrap support. At the same time, AsaDREB2.3, 2.5, and 2.7 with individual expression signatures were located distantly from the other AsaDREB2 genes ( Figure 2).
The closest homolog of the dominant AsaDREB2.5 gene is Arabidopsis AT2G40340 (DREB2C) (Figure 2), known to be involved in plant response to abiotic stresses and hormones [44][45][46][47]. It is likely that the AsaDREB2.5 gene plays similar roles in garlic, which would supplement its involvement in the response to cold and Fusarium infection (Figures 6b and 7).
AsaDREB2.2, 2.6, and 2.8 were significantly activated by Fusarium infection, whereas AsaDREB2.1, 2.3, 2.4, and 2.7 were not (Figure 6b), indicating the role of the former in the defense against fungal pathogens and the lack of it in the latter. At the same time, the expression of AsaDREB2.2, 2.6, and 2.8 was affected by cold exposure, and although the response was much weaker than that of the AsaDREB1 genes, it may indicate the contribution of AsaDREB2.2, 2.6, and 2.8 to the cold resistance of garlic. However, the involvement of the DREB2 genes in response to cold is still questionable [28,29,36] and requires further investigation.
Thus, AsaDREB2.2, 2.5, 2.6, and 2.8 are likely the main AsaDREB2 genes that perform the DREB2 function in garlic, among which AsaDREB2.5 plays a dominant role in the response to biotic and abiotic stresses.

AsaDREB1 and AsaDREB2 Gene Subfamilies: Identification and Analysis
In this study, the genome (PRJNA606385) and transcriptomes (PRJNA607255) of A. sativum (cv. Ershuizao) [1] were used to search for full-length sequences of the DREB1 and DREB2 genes. Structural and phylogenetic analyses of the AsaDREB1 and AsaDREB2 genes were carried out in MEGA 7.0.26 [48]: the dendrogram was constructed using the Neighbor-Joining method; the evolutionary distances were computed by the JTT matrix-based method and were in the units of the number of amino acid substitutions per site; bootstrap values based on 1000 replicates were applied to evaluate the confidence for the tree topologies. The localization of AsaDREB genes on A, sativum chromosomes was visualized in MG2C v2.1 (http://mg2c.iask.in/mg2c_v2.1/; accessed on 3 February 2023). The AsaDREB exon-intron structures were predicted with GSDS v2.0 [49]. The domain structure of the proteins was characterized using the Conserved Domain Database (https://www.ncbi.nlm.nih.gov/cdd; accessed on 3 February 2023). The conserved motifs of the proteins were identified by MEME 5.5.1 (http://meme-suite.org/tools/meme; accessed on 3 February 2023), as well as considering the available published data. Protein MW and pI values were predicted with ExPASy ProtParam (https://www.expasy.org/resources/protparam; accessed on 3 February 2023). The PANNZER2 server (Protein ANNotation with Z-scoRE) was used for protein functional annotation in terms of Gene Ontology (GO) (http://ekhidna2.biocenter. helsinki.fi/sanspanz/; accessed on 3 February 2023).

AsaDREB1 and AsaDREB2 Gene Expression: In Silico Analysis
Transcriptomic data for the bulbs (8 developmental stages corresponding to 192, 197, 202, 207, 212, 217, 222, and 227 days after sowing (DAS), respectively), roots, leaves, pseudostems, buds, sprouts (217 DAS), and flowers (217 DAS) of A. sativum cv. Ershuizao (PRJNA607255) normalized as FPKM (fragments per kilobase of transcripts per million reads mapped) [1] were used for AsaDREB1 and AsaDREB2 gene expression profiling. The criterion for the reliability of the presence of gene transcripts was FPKM ≥ 10 in at least one of the analyzed types of organs. A heatmap of the gene expression was constructed and visualized with Heatmapper [50].

cis-Regulatory Elements in the Promoters of the AsaDREB1 and AsaDREB2 Genes: In Silico Search
cis-acting regulatory elements associated with developmental processes and responses to hormones and stress were searched for in the AsaDREB1 and AsaDREB2~1.0 kb region upstream of the start codon. This analysis was performed using a regularly updated Plant-CARE database, which combines information on cis-elements, enhancers, and repressors in plant genes (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/; accessed on 3 February 2023).

Garlic Cultivars and F. proliferatum Strain
In this study, we used accessions of winter garlic common in Russia: the mediumhardy cv. Sarmat and the winter-hardy cv. Strelets, resistant and susceptible to Fusarium infection, respectively. In October 2021, garlic was sown in the fields of the Federal Scientific Vegetable Center (Moscow region, Russia), and in May 2022, the plants were moved to greenhouse conditions (16 h photoperiod; 23 • C). After 2.5 months, individual samples (roots, bulbs, stems (basal plates), leaves, pseudostems, peduncles, receptacles, and air bulbs (bulbils)) were collected and used to analyze the tissue-specific gene expression.
The F. proliferatum strain used in the study was isolated in 2021 from infected cv. Strelets bulbs and stored in the collection of microorganisms (Research Center of Biotechnology of the RAS, Moscow, Russia). The test of the isolate for pathogenicity showed the appearance of rot on the garlic cloves 5 days after infection [19,20].

Stress Assays
The response to short-term cold stress was analyzed by placing garlic seedlings (3-4 leaf stage) into a climatic chamber (+4 • C) for 2, 4, 6, and 24 h; untreated plants were used as the control. The leaves (two biological replicates) were harvested at each time point and stored at −80 • C.
The garlic cloves were infected with F. proliferatum as previously described [20]. Peeled and sterilized (3 min in 70% ethanol, then washed with water) cloves of a uniform size were kept on wet filter paper at room temperature in the dark until active root growth (~72 h). The formed roots and germinated cloves were immersed in a suspension of F. proliferatum conidia (~10 6 mL −1 ), blotted after 5 min, and transferred to fresh filter paper. Then, after 24 and 96 h (+25 • C, darkness), samples of roots, pseudostems, and cloves (n = 3 biological replicates) were collected; tissues of uninfected cloves were used as the controls. All of the samples were stored at −80 • C and were used for analysis of the gene expression.

AsaDREB Gene Expression Pattern in Garlic Cultivars: Quantitative (q) Real-Time (RT)-PCR Analysis
The collected garlic tissue samples (~0.1 g of each) were used for the total RNA isolation (RNeasy Plant Mini Kit and RNase free DNase set; QIAGEN, Hilden, Germany) followed by cDNA synthesis primed by oligo(dT) (GoScript Reverse Transcription System; Promega, Madison, WI, USA). RNA was qualified by gel electrophoresis. The concentration of nucleic acids was measured using a Qubit Fluorometer (Thermo Fisher Scientific, Waltham, MA, USA). Then, 3.0 ng of cDNA was taken for the gene expression analysis (qRT-PCR). The reaction also included SYBR Green RT-PCR mixture (Syntol, Moscow, Russia) and gene-specific primers. The reaction (95 • C 5 min and then 40 cycles (95 • C 15 s and 60 • C 40 s)) was run on a CFX96 Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, CA, USA).
Primers were designed by comparative analysis of AsaDREB mRNAs based on the cv. Ershuizao transcriptome [1] (Supplementary Table S1). Primer sequences were selected in the variable regions of the cDNA (forward and reverse primers were separated by at least one intron) and additionally validated using BLASTn (https://blast.ncbi.nlm.nih.gov/; accessed on 3 February 2023) and Primer3 (http://frodo.wi.mit.edu/primer3/, accessed on 3 February 2023). A mixture of cDNA preparations from different tissue samples at concentrations of 10.0, 1.0, and 0.1 ng per reaction was used to determine the efficiency and gene specificity of primers using qRT-PCR (three technical replicates). The criterion for the efficiency of a primer pair was a melting curve containing a single peak, which confirms the gain of only one amplicon during qRT-PCR. Testing showed 95-100% (R 2 = 0.943-0.999) efficiency of the selected primers.
The AsaDREB gene expression was normalized to the combination of two reference genes encoding ubiquitin (UBQ; NCBI ID MZ171222.1) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH; NCBI ID MZ171220.1) that were previously used separately in garlic [51,52] and onion (Allium cepa L.), respectively. The combination of two stably expressed references (GAPDH and UBQ) has been used to improve the accuracy of qRT-PCR analysis relative to single reference analysis [53], and has already shown its effectiveness in garlic studies [17,19,20].

Statistical Analysis of the AsaDREB Gene Expression Data
Statistical processing of gene expression analysis data (qRT-PCR) was carried out in GraphPad Prism version 9 (https://www.graphpad.com/scientific-software/prism/; accessed on 30 June 2023). On the graphs, the normalized expression data (considering three technical replicates of two biological replicates in each case) were presented as mean ± standard deviation (SD) (Figures 5, 6 and S3). In the case of Figures 7 and S3, the statistical significance of the differences in the detected level of gene transcripts was estimated using an unpaired (independent) t-test, assuming that expression differs significantly at p < 0.01. In the case of Figures 5 and 6, the statistical significance of differences in the detected level of gene transcripts was estimated using one-way ANOVA (multiple comparisons, corrected with Bonferroni test); obtained p-values are given in Supplementary  Tables S2 and S3, respectively. For Figure 5, the mean of each column was compared to the mean of every other column. For Figure 6, the means of the preselected pairs of columns (control (C) 24 h vs. experiment (E) 24 hpi; C 96 h vs. E 96 hpi; E 24 hpi vs. E 96 hpi; C 24 h vs. C 96 h; for each of two analyzed garlic cultivars) were compared. All p-values are given in Supplementary Table S2 (for Figure 5), Table S3 (for Figure 6) and Table S4 (for Figure 7).

Conclusions
We identified and characterized eight DREB1 and eight DREB2 subfamily genes in the A. sativum cv. Ershuizao genome and compared their organ-specific expression patterns with those in cv. Sarmat. Our results suggest that the tandemly clustered AsaDREB1 genes could have redundant functions in transcriptional regulation in garlic; among them, AsaDREB1.2, 1.4, and 1.8 may be the main genes performing the DREB1 function in garlic, whereas AsaDREB1.6 and 1.7 could be pseudogenes depending on the genotype. The AsaDREB1 genes are most active in the roots and are stimulated by cold stress and Fusarium infection, suggesting their role in garlic protection against abiotic and biotic stresses. The DREB2 subfamily genes AsaDREB2.2, 2.5, 2.6, and 2.8 are likely the functionally dominant DREB2 genes in garlic, among which AsaDREB2.5 has the highest expression level. The transcription of these four AsaDREB2 genes is affected by low temperatures and Fusarium infection, suggesting their contribution to garlic adaptive responses. AsaDREB2.4 could be a pseudogene, whereas the pseudogenic status of AsaDREB2.6 and 2.7 may be genotypespecific. The identification and characterization of the DREB genes in garlic performed in this study provides the foundation for further functional analysis of the DREB TFs in A. sativum and the other Allium spp., and could contribute to the breeding of stress-tolerant cultivars.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/plants12132538/s1, Table S1: List of primers for AsaDREB gene expression analysis; Table S2: Differences (p-values) in the expression level of the AsaDREB genes between different garlic organs; Table S3: Differences (p-values) in the expression level of the AsaDREB genes in the roots of A. sativum FBR-resistant cv. Sarmat and FBR-susceptible cv. Strelets in response to F. proliferatum infection compared to uninfected control; Table S4: Differences (p-values) in the expression level of the AsaDREB genes in the roots of A. sativum cv. Sarmat and cv. Strelets in response to cold stress compared to unstressed control; Figure S1: Alignment of the AsaDREB1.1-1.8 (a) and AtDREB2.1-2.8 (b) proteins with their homologs from Arabidopsis thaliana (NCBI IDs are indicated). Gray-shaded regions are 70-100% identical, red frame indicates AP2 domain. Consensuses typical for DREB proteins are underlined; Figure S2: Evolutionary relationships among DREB proteins from Allium sativum (blue), Arabidopsis thaliana (red), and Zea mays (green). The unrooted dendrogram was constructed using the Neighbor-Joining method (bootstrap test: 1000 replicates) in MEGA 7.0.26; Figure S3: Expression of the selected AsaDREB genes in the leaves of A. sativum cv. Strelets and cv. Sarmat after cold stress (+4 • C for 2, 4, 6, and 24 h). The data were normalized to GAPDH and UBQ mRNA levels; * p < 0.01 compared to control (0 h).  Table 1).