Identification of BvUGT90 Family Members and Analysis of Drought Resistance Gene Screening in Sugar Beet

TL;DR

This study identifies and analyzes the BvUGT90 gene family in sugar beet to find genes that help the plant resist drought stress.

Contribution

The study systematically identifies and functionally characterizes the BvUGT90 gene family in sugar beet under drought stress.

Findings

121 BvUGT90 family members were identified, unevenly distributed across 9 chromosomes.

45 BvUGT90 genes showed significant responsiveness to drought stress, with 10 upregulated at the protein level.

Transgenic experiments confirmed Bv_005070_jjst.t1's role in drought resistance.

Abstract

The sugar beet (Beta vulgaris L.) industry in China occupies a pivotal position in the national sugar supply, yet drought in its major cultivation areas has become a key limiting factor for its high-quality development. Glycosyltransferases (GTs) play a pivotal role in plant responses to abiotic stress, particularly in the regulation of drought resistance. However, the systematic identification of the BvUGT90 gene family in sugar beet and the functional characterization of its members under drought stress remain largely unexplored. In this study, drought stress was simulated in the sugar beet cultivar ‘HI0466’ using the weighing method to regulate soil moisture. Samples were collected at different stress durations and after rewatering for subsequent experimental analyses. In this study, 121 members of the BvUGT90 family were identified in sugar beet, and a comprehensive analysis was conducted on their gene structures, phylogenetic relationships, promoter cis-acting elements and expression patterns under drought stress. The results showed that these 121 members were unevenly distributed across 9 chromosomes. The proteins they encode had an average amino acid length of 474, with molecular weights ranging from 10.78 to 99.10 kDa and theoretical isoelectric points (pI) from 4.68 to 8.69 (with an average of 5.76). Notably, 110 of these members (accounting for 90.91%) were identified as hydrophilic proteins. Synteny analysis indicated a high degree of homology between the BvUGT90 family members in sugar beet and their orthologous genes in Arabidopsis thaliana. Analysis of promoter cis-acting elements revealed the presence of six major categories of core elements in the promoter regions of BvUGT90 genes, including hormone-responsive elements, stress-responsive elements and pathway regulatory elements. Transcriptomic data showed that 45 BvUGT90 family members exhibited significant responsiveness to drought stress. Proteomic analysis demonstrated that 10 of these members were significantly upregulated at the protein level under drought stress, and these results were further validated by quantitative real-time polymerase chain reaction (qRT-PCR). Integrated transcriptomic and proteomic analyses identified Bv_005070_jjst.t1 and Bv6_140060_stjc.t1 as the family members with the most prominent responses to drought stress. Furthermore, transgenic transformation of sugar beet was performed, which confirmed that Bv_005070_jjst.t1 plays an important role in drought stress resistance. The findings of this study provide direct candidate genes from this family for drought-tolerant sugar beet breeding.

Figures

Full text

1. Introduction

Sugar beet (Beta vulgaris L.) is an important sugar-producing crop in China, mainly cultivated in regions such as North China, Northeast China, and Northwest China [1]. Currently, commercially cultivated sugar beet varieties generally exhibit low abiotic stress resistance, a problem that has been closely linked to the domestication process of cultivated beets [2]. Throughout the domestication of wild beets to cultivated accessions, breeding efforts have prioritized sugar content and biomass enhancement, which has indirectly diminished the stress resistance of cultivated beets. Studies have shown that compared with cultivated beets, the proline synthetase gene (P5CS) in wild beets is significantly upregulated by 3- to 4-fold under abiotic stress, whereas no significant change in P5CS expression is detected in cultivated beets, resulting in reduced proline accumulation and decreased drought tolerance [3]. Similarly, the betaine aldehyde dehydrogenase gene (BADH) in wild beets is markedly upregulated by 2- to 3-fold under salt stress, while cultivated beets show only a ~1-fold upregulation [3]. Furthermore, wild beets exhibit significantly stronger reactive oxygen species (ROS) scavenging capacity than cultivated beets [3]. GTs play a pivotal role in plant secondary metabolism, with their core function being the catalysis of glycosylation reactions—specifically, transferring sugar molecules from activated donors to specific acceptor molecules [4]. Glycosylation modification is a key step in the biosynthesis of plant secondary metabolites and often occurs in cascades with processes such as acylation and ubiquitination [5], thereby participating in the regulation of secondary metabolic pathways closely associated with plant stress resistance traits [6]. To date, extensive bioinformatics analyses have been performed on GT1 family members in plants. In maize, 107 GT1 family members have been identified, which are unevenly distributed across 10 chromosomes; among these, 16 GT1 genes exhibit highly similar expression patterns to anthocyanidin synthase (ANS) [7]. In flax, 137 UDP-glycosyltransferase genes have been characterized, all of which contain the PSPG box within their coding sequences (CDS) [8]. Further analysis of the identified family members revealed that the expression of some glycosyltransferase genes is highly responsive to abiotic stresses, and family members sharing conserved domains display analogous expression trends. For example, the UGT76E11 gene in Arabidopsis thaliana is significantly upregulated under salt and drought stresses; overexpression of this gene significantly enhances drought tolerance in transgenic lines at both the seedling and mature stages while also participating in flavonoid biosynthesis [9]. Similarly, UGT85E1 in rice is sensitive to drought stress; its overexpression improves drought resistance by inducing abscisic acid (ABA) accumulation, activating signal transduction pathways, promoting leaf stomatal closure, and increasing proline content and reactive oxygen species (ROS) scavenging capacity [10]. These studies confirm the conserved function of UGTs in plant drought resistance, but limited information is available for sugar beet—especially the BvUGT90 family, which hinders the exploration of drought-tolerant genetic resources in this crop.

This study aimed to: (1) systematically identify BvUGT90 family members in sugar beet; (2) analyze their sequence characteristics, phylogenetic relationships, and promoter cis-acting elements; (3) screen drought-responsive members via transcriptomic, proteomic, and qRT-PCR analyses; and (4) identify core candidate genes for drought-tolerant breeding. This study lays a reliable theoretical foundation for subsequent research on the drought-resistant molecular mechanisms of UGT90-related genes in sugar beet and provides novel candidate genes and insights for drought-resistant genetic breeding of sugar beet.

2. Results

2.1. Identification of BvUGT90 Family Members and Their Sequence Characteristics

A total of 121 UGT90 gene family members were identified from the sugar beet genome through BLAST-based (ncbi-blast-2.11.0+) homology screening using TBtools v2.129 software and validated using the Pfam database. The physicochemical properties of these members were systematically analyzed, covering amino acid composition, protein instability index, hydrophilicity/hydrophobicity, and subcellular localization. The BvUGT90 family members had an average length of 474 amino acid residues, with molecular weights ranging from 10.78 to 99.10 kDa. Their theoretical isoelectric points (pI) ranged from 4.68 to 8.69 (mean = 5.76), and 117 members (96.69%) had a pI below 7. The protein instability index varied between 73 and 57.24, with 83 members scoring above 40. Most members (110, 90.91%) were predicted to be hydrophilic proteins, and subcellular localization analysis indicated a predominant cytoplasmic localization (Table 1).

2.2. Phylogenetic Analysis of BvUGT90s

To systematically characterize the structural features and evolutionary patterns of the BvUGT90 family, this study first clarified the intrafamilial phylogenetic relationships of UGT90 in sugar beet via circular phylogenetic tree analysis (Figure 1a). The results showed that this family could be clustered into six distinct subfamilies (Groups I–VI), with Group IV containing the largest number of members (53) and Group VI the fewest (only 2); most Bootstrap values on the tree branches were ≥0.85, with some nodes reaching 1, which strongly supported the statistical reliability of the clustering relationships among the subfamilies, laying a core phylogenetic framework and classification basis for subsequent analyses of functional divergence, molecular evolutionary rules, and biological function specificity among different subfamilies. To further expand the understanding of the evolutionary characteristics of this family and explore its cross-species conservation and functional specialization rules, based on the single-species analysis, this study incorporated homologous UGT90 genes from the dicot model plant Arabidopsis thaliana and the monocot representative species Oryza sativa and constructed a cross-species phylogenetic tree together with members of the BvUGT90 family in sugar beet (Figure 1b). Each branch exhibited significant evolutionary conservation.

2.3. Analysis of the Gene Structure and Protein Conserved Motifs of BvUGT90s Members

Conserved motif analysis showed that the BvUGT90 family contains six highly conserved motifs (Figure 2a). Conserved domain analysis revealed that the vast majority of members harbor the typical Glycosyltransferase_GTB-type superfamily domain, while a few members contain the PLN02448, MGT superfamily, and GT1_Gtf-like domains (Figure 2b).

2.4. Chromosomal Localization of BvUGT90 Members

Based on the sugar beet genome annotation file and the identified BvUGT90 family members, this study performed a visualization analysis of their chromosomal localization (Figure 3). The results indicated that the BvUGT90 genes were unevenly distributed across nine chromosomes, with all members localized to the subtelomeric regions.

2.5. Collinearity Analysis of the BvUGT90s Gene

To further dissect the syntenic relationships and evolutionary connections between the UGT90 gene family in sugar beet and its homologous families in other species, we selected the dicot model plant Arabidopsis thaliana (A. thaliana) and the monocot representative plant Oryza sativa (O. sativa) and performed syntenic relationship analysis on UGT90 family members across these three species (Figure 4). The results showed that there are 35 pairs of UGT90 homologous gene pairs between sugar beet and A. thaliana; this relatively high number of pairs reflects that the gene structure and member composition of the UGT90 family have retained high conservation between the two species during evolution. Additionally, 9 homologous gene pairs were detected between A. thaliana and O. sativa, and 9 conserved homologous gene pairs are shared among the three species. These results indicate that compared with O. sativa (a monocot plant of the Poaceae family), sugar beet exhibits higher homology in the UGT90 family with A. thaliana (a core eudicot plant of the Brassicaceae family), and this difference is likely closely associated with the functional specialization process of the gene family following the divergence of monocot and dicot plants. Furthermore, the 9 homologous gene pairs shared by the three species are most likely conserved members of the UGT90 family that undertake basic physiological functions (e.g., core glycosylation modification), which can serve as target references for subsequent dissection of the core functions of this family.

2.6. Analysis of Cis-Acting Elements of BvUGT90s Promoter

To further elucidate the regulatory functions of the BvUGT90 family, a systematic analysis of cis-acting elements within the 2000 bp promoter region upstream of the translation start site of its members was performed using the online tool PlantCARE (Figure 5a). The results revealed that the promoters of the BvUGT90 family contain six major categories of core cis-acting elements, namely hormone-responsive elements, stress-responsive elements, light-responsive elements, plant growth and development regulatory elements, pathway regulation and binding site-related elements, and metabolism-related elements. Each category encompasses diverse functional subtypes with significant differences in abundance (Figure 5b,c).

Specifically, hormone-responsive elements include six functional subtypes, comprising various specific elements with distinct abundances, such as abscisic acid (ABA)-responsive element (ABRE), methyl jasmonate (MeJA)-responsive elements (CGTCA-motif, TGACG-motif), auxin-responsive elements (AuxRE, AuxRR-core, TGA-box, TGA-element), gibberellin (GA)-responsive elements (GARE-motif, P-box), and salicylic acid (SA)-responsive elements (SARE, TCA-element). Among these, MeJA-responsive elements are the most abundant, totaling 251, and represent the dominant elements in this category. Stress-responsive elements consist of six functional types, namely anaerobic-responsive element (GC-motif), drought-responsive element (MBS), low-temperature-responsive element (LTR), antioxidant stress-responsive element (ARE), defense and stress-responsive elements (TC-rich repeats), and wound-responsive element (WUN-motif). Anaerobic-responsive elements account for the highest proportion, with 189 identified.

Light-responsive elements include a variety of components, such as MRE, G-box, circadian, ACA-motif, C-box, CAG-motif, Box 4, ATC-motif, AE-box, ACE, 3-AF1 binding site, 4cl-CMA2b, AAAC-motif, Box II, chs-CMA1a, GA-motif, Gap-box, GATA-motif, GT1-motif, GTGGC-motif, I-box, LAMP-element, L-box, Sp1, TCCC-motif, TCT-motif, ATCT-motif, LS7, and chs-CMA2a. Their total abundance ranks first among the six categories, reaching 294. Plant growth and development regulatory elements involve eight types, including A-box, HD-Zip 1, CAT-box, NON-box, AACA_motif, GCN4_motif, RY-element, and motif I. Among these, regulatory elements associated with meristem expression are the most numerous, with 37 identified. Pathway regulation and binding site-related elements consist of seven functional subtypes, specifically ATBP-1 binding site, CCAAT-box, HD-Zip 3, Box III, 3-AF3 binding site, AT-rich sequence, and AT-rich element. MYBHv1 binding sites exhibit the highest occurrence frequency, totaling 21. Metabolism-related elements include three types: MBSI, A-box, and O2-site. Among these, zein metabolism-related regulatory elements are the most abundant, with 57 identified.

In summary, the promoter regions of the BvUGT90 family are enriched with diverse cis-acting elements involved in hormone signaling, abiotic stress responses, light signal transduction, growth and development regulation, and secondary metabolic pathways, with distinct specificity in the abundance distribution of each category.

2.7. Analysis of the Expression Pattern of BvUGT90s Under Drought Stress

In the cis-acting elements within the promoters of BvUGT90 genes, a large number of stress-responsive elements were identified in this study. To further investigate the potential functions of BvUGT90s under drought stress, RNA sequencing (RNA-seq) was performed on sugar beet seedlings exposed to 4 days (d), 6 days, or 10 days of drought stress and rehydration (RW) treatment (Figure 6a). The results demonstrated that the expression levels of BvUGT90 family members are highly sensitive to stress duration: 23 BvUGT90 members were significantly upregulated after 4 days of drought stress (DS4), 38 members after 6 days (DS6), 45 genes after 10 days (DS10), and 22 genes following RW treatment. At the protein level, 14 family members were significantly upregulated after DS4, 10 members after 8 days of drought stress (DS8), and 16 members post-RW treatment (Figure 6b). Specifically, the 10 BvUGT90 family members that were upregulated at DS8 are Bv6_155170_umck.t1, Bv_005070_jjst.t1, Bv6_140060_stjc.t1, Bv1_003390_zciq.t1, Bv4_077610_cdyk.t1, Bv9_224280_wrdn.t1, Bv7_168720_zftj.t1, Bv7_175150_yuax.t1, Bv6_138370_eghx.t1, and Bv6_155180_ytax.t1. These ten members were validated via quantitative real-time polymerase chain reaction (qRT-PCR) (Figure 6c). The results indicated that, except for Bv7_168720_zftj.t1 (RNA-seq data showed that the expression of this gene was significantly repressed at DS4, whereas qRT-PCR results revealed no significant difference compared with the control (CK)), the expression levels of the remaining 9 members were consistent with the RNA-seq data.

2.8. GO Enrichment Analysis of BvUGT90s

To further characterize the functions of BvUGT90 family members at the protein level, this study performed Gene Ontology (GO) enrichment analysis on the 10 members that were upregulated at the protein level under drought stress, using all BvUGT90 family members identified in the proteome as the background annotation file (Figure 7). The results demonstrated that these 10 BvUGT90 family members take glycosyltransferase activity as their core molecular function, participate in flavonoid metabolism-related biological processes, and are primarily localized to intracellular membrane-bound organelles.

2.9. Multi-Omics Venn Diagram Analysis

To clarify the expression overlap characteristics of BvUGT90 family members at the transcriptomic and proteomic levels, this study first performed Venn diagram analysis on the BvUGT90 family members commonly covered in the transcriptomic and proteomic data under different treatments (4 days (d), 8 days (d) of drought stress, and rehydration (RW) treatment). The results showed that a total of 26 BvUGT90 family members were commonly detected in the transcriptomic and proteomic data under these conditions (Figure 8a). To further screen the family members that synergistically respond to drought stress at both the transcriptional and proteomic levels, this study subsequently conducted cross-omics Venn diagram overlap analysis on the significantly upregulated BvUGT90 family members identified in the transcriptomic data (4 d, 6 d, 10 d of drought stress, and RW treatment) and the proteomic data (4 d, 8 d of drought stress, and RW treatment), respectively. The results indicated that among all treatment conditions (Figure 8b), only 2 BvUGT90 family members exhibited significant upregulation at both the transcriptional and proteomic levels, namely Bv_005070_jjst.t1 and Bv6_140060_stjc.t1.

3. Identification of Drought Resistance Function of Bv_005070_jjst.t1

3.1. Effects of Drought Stress on the Phenotype of Transgenic Lines

After drought stress treatment, significant phenotypic differences were observed among the various lines (Figure 9). Specifically, the overexpression lines exhibited robust overall growth, with leaves consistently maintaining a dark green hue and a flat and upright morphology; their mesophyll tissues were plump without obvious signs of water loss, thereby preserving a favorable photosynthetic physiological structure. In contrast, both the wild-type lines and RNAi-silenced lines displayed typical phenotypic symptoms of drought stress: the leaves showed marked wilting and drooping, accompanied by chlorosis and yellowing in partial leaves; the lower leaves of some individual plants exhibited curling, and their overall growth was severely inhibited.

3.2. Effects of Drought Stress on Physiological Indices of Transgenic Sugar Beets

Under drought stress, various physiological indices of wild-type lines and transgenic lines exhibited significant differential responses to the overexpression and silencing of the Bv_005070_jjst.t1 gene (Figure 10). Specifically, the activities of superoxide dismutase (SOD) and peroxidase (POD) in the overexpression lines were significantly higher than those in the wild-type lines (Figure 10a,b), whereas the activities of these two enzymes in the silenced lines were significantly lower than those in the wild-type lines. In detail, under drought stress, compared with the wild-type lines, the SOD activities in overexpression lines OE1, OE2 and OE3 were significantly increased by 17.25%, 20.35% and 25.93%, respectively, and their POD activities were significantly elevated by 16.08%, 18.02% and 17.77%, respectively. In contrast, the SOD activities in the three silenced lines—Ri1, Ri2 and Ri3—were significantly reduced by 8.40%, 11.22% and 7.31%, respectively, relative to the wild-type lines, and their POD activities were significantly decreased by 47.92%, 44.60% and 44.58%, respectively.

With regard to malondialdehyde (MDA) content (Figure 10c), that in the overexpression lines was significantly reduced by 32.60%, 42.42% and 44.42%, respectively, compared with the wild-type lines, whereas the MDA content in the silenced lines was significantly increased by 65.17%, 53.10% and 72.86%, respectively, relative to the wild-type lines. As for proline content (Figure 10d), it was significantly lower in the overexpression lines but significantly higher in the silenced lines in comparison with the wild-type lines. More precisely, relative to the wild-type lines, the proline content in the overexpression lines was significantly decreased by 63.18%, 61.42% and 61.85%, respectively, while that in the silenced lines was significantly increased by 153.39%, 151.16% and 161.63%, respectively.

4. Discussion

4.1. Bioinformatic Characterization and Evolutionary Analysis of the BvUGT90 Gene Family

Glycosyltransferases (GTs) represent a large, multi-gene enzyme superfamily that catalyzes the glycosylation of plant secondary metabolites. The glycosylation reactions they mediate are crucial for modulating metabolite stability, water solubility, and bioactivity [11]. To date, 139 GT families have been identified in plants, among which the GT1 family, due to its numerous members and diverse functions, is a central focus in plant stress resistance and secondary metabolism research [5]. In this study, we systematically identified 121 UGT90 gene family members from the sugar beet genome for the first time using BLAST-based homology screening with TBtools software combined with validation against the Pfam database. This family belongs to the GT1 family, with the majority of its members containing the Glycosyltransferase_GTB-type superfamily domain. The genes were unevenly distributed across all nine sugar beet chromosomes and were primarily localized to subtelomeric regions. This uneven distribution pattern may be associated with processes of gene duplication and functional divergence. Notably, these characteristics—the uneven chromosomal distribution and the widespread presence of the GTB-type domain—closely resemble those reported for the GT1 family in maize [7], collectively underscoring the significant structural conservation of the GT1 family across species.

A systematic analysis of the physicochemical properties of the BvUGT90 family, encompassing amino acid composition, protein instability index, hydrophilicity/hydrophobicity, and subcellular localization, revealed the following: The encoded proteins had an average length of 474 amino acids and molecular weights ranging from 10.78 to 99.10 kDa. The average theoretical isoelectric point (pI) was 5.76, with 96.69% of members being acidic proteins (pI < 7). This acidic nature confers a net negative charge under physiological conditions, potentially enabling functions such as binding intracellular calcium ions, modulating calcium signaling, influencing chromatin structure and gene expression, and affecting protein subcellular localization and trafficking. These attributes may represent important molecular features shaped during sugar beet’s long-term evolutionary adaptation to its environment [11,12]. Furthermore, 90.91% of the members were predicted to be hydrophilic proteins, and most were localized to the cytoplasm, providing a spatial foundation for substrate interaction and participation in secondary metabolic reactions within the cell. The finding that 83 members had an instability index greater than 40 suggests that a portion of the BvUGT90 family proteins may be stress-inducible, potentially participating in the plant’s stress response through rapid changes in expression.

Interspecies phylogenetic analysis demonstrated significant evolutionary conservation of the UGT90 gene family in land plants, along with evident functional specialization following the divergence of monocots and dicots. Sugar beet BvUGT90 genes were distributed across nine cross-species subfamilies. Among these, Group IV from the single-species analysis emerged as the dominant subfamily, comprising 41.32% of the total BvUGT90 family members, indicating its targeted expansion in sugar beet, likely accompanied by functional differentiation. The co-clustering relationship between sugar beet BvUGT90 genes and their Arabidopsis UGT90 homologs provides direct evolutionary clues for the functional prediction of BvUGT90 members. This analysis also confirmed a correlation between the differentiation processes of the UGT90 family and plant taxonomic status, offering a comprehensive evolutionary framework for functional dissection of this family and the screening of candidate targets for sugar beet drought-tolerant breeding. Gene structure analysis further indicated that members within the same evolutionary clade shared highly similar exon–intron patterns regarding number, length, and arrangement, whereas structurally significant differences existed between distantly related clades. This finding aligns with general principles of gene evolution and is consistent with conclusions from studies on GT families in rice [13]. It suggests that closely related BvUGT90 members may exhibit functional redundancy, while distantly related ones might have developed functional specificity through structural divergence, providing important clues for subsequent functional subdivision within the family.

Cis-acting elements in gene promoters are key targets for deciphering transcriptional regulatory networks [14]. Analysis of the promoters of the 121 BvUGT90 genes identified six major categories of core elements: hormone response, stress response, light response, growth and development regulation, pathway regulation-related elements and transcription factor binding sites, and metabolism-related elements. This compositional profile indicates that the BvUGT90 family can integrate multiple signals—hormonal, environmental stress-related, and developmental—thereby playing a pleiotropic regulatory role in sugar beet growth, development, signal transduction, and abiotic stress responses. Specifically, the abundant presence of drought-responsive elements (MBS), low-temperature-responsive elements (LTR), and MYB-binding sites associated with flavonoid biosynthesis suggests that this family can respond to various abiotic stresses like drought and cold, potentially participating in sugar beet stress resistance through the regulation of flavonoid secondary metabolism. The high abundance of MBS elements, which are binding sites for MYB transcription factors, further confirms that the BvUGT90 family is likely regulated by MYB TFs. This regulatory axis may activate downstream antioxidant enzyme genes and osmoprotectant biosynthesis genes to mediate the drought stress response [15], providing clear targets for future research into the regulatory mechanisms of this family.

4.2. Transcriptomic and Proteomic Analysis of the BvUGT90 Gene Family and Comparative Assessment of RNA-Seq and qRT-PCR Technologies

The pivotal regulatory roles of glycosyltransferases in plant growth, stress responses, and secondary metabolism are well-established [16,17]. Drought, as a major abiotic stress, induces extensive transcriptional reprogramming in plants, activating a series of stress resistance networks, including secondary metabolism and antioxidant defense [17,18]. Previous studies have shown significant enrichment of the phenylpropanoid biosynthesis pathway in pea under drought stress [19], an increase in the number of differentially expressed genes (DEGs) in Chinese cabbage with prolonged drought duration [19], and the involvement of Arabidopsis UGT78D2 and rice OsUGT74E2 in enhancing drought tolerance through flavonoid glycosylation [20,21]. These findings confirm the conserved mechanism by which UGT family members mediate plant drought resistance via flavonoid metabolism. In this study, using the drought-tolerant sugar beet cultivar HI0466 subjected to gradient drought treatments (4, 6, and 10 days), we observed a progressive increase in the number of significantly upregulated genes within the BvUGT90 family (23 → 38 → 45). This trend aligns with the expression pattern reported in Chinese cabbage [22], indicating a pronounced time-dependent response of this family to drought stress, suggesting the BvUGT90 family’s potential role in the hierarchical regulation of sugar beet’s progressive defense against drought. To further identify core members involved in the drought response, we performed an integrative transcriptomic and proteomic analysis. Twenty-six family members showed co-expression patterns at both transcriptional and translational levels across drought and re-watering treatments. Notably, only two members, Bv_005070_jjst.t1 and Bv6_140060_stjc.t1, exhibited coordinated upregulation at both omics levels. This transcription–translation coherence is a hallmark of functionally critical genes, ensuring efficient translation of stress-responsive transcripts into functional proteins, thereby establishing these two genes as prime candidates central to sugar beet drought tolerance. Gene Ontology (GO) enrichment analysis of drought-responsive BvUGT90 members revealed significant enrichment in glycosyltransferase catalytic activity and involvement in flavonoid metabolic processes. This result echoes findings from drought-stressed peas [23]. Given that flavonoid glycosylation enhances antioxidant activity and water solubility [24], we propose that the BvUGT90 family may strengthen oxidative stress defense in sugar beet by catalyzing the glycosylation of flavonoids, thereby promoting the accumulation of drought-resistance-related derivatives such as quercetin and kaempferol glycosides. This proposed regulatory mechanism shows evolutionary conservation with drought resistance pathways involving UGT families in Arabidopsis and rice [20]. Our study, for the first time, delineates the role of the sugar beet BvUGT90 family in mediating drought tolerance through the regulation of flavonoid secondary metabolism, filling a gap in the understanding of the molecular mechanisms underlying stress resistance in sugar beet UGT families.

To validate the reliability of our RNA sequencing (RNA-seq) data, we selected 10 BvUGT90 members upregulated at the protein level after 8 days of drought stress for quantitative real-time PCR (qRT-PCR) validation. The expression trends of nine genes were highly consistent with the RNA-seq data, while a minor discrepancy was observed for Bv7_168720_zftj.t1. This subtle difference reflects the inherent distinctions between the two technologies in gene expression profiling. Specifically, RNA-seq is a high-throughput, unbiased whole-transcriptome detection technology. It comprehensively sequences cDNA libraries from total RNA, enabling simultaneous detection of expression changes in all genes in the genome and facilitating high-throughput screening of DEGs. In this study, RNA-seq provided a complete expression profile for all 121 BvUGT90 family members. In contrast, qRT-PCR is a targeted amplification technology based on fluorescent signals. It amplifies pre-selected individual genes or a few genes using specific primers, limiting its detection scope to the target genes. In this study, it served primarily for targeted validation of core candidate genes. Together, they form a complementary “high-throughput screening → targeted verification” system. Regarding sensitivity and quantitative characteristics, RNA-seq offers higher sensitivity for detecting low-abundance transcripts, novel transcripts, and splice variants, capturing subtle expression changes, although its quantification is relative and semi-quantitative, influenced by sequencing depth and alignment efficiency. qRT-PCR also exhibits high sensitivity, particularly in reliably detecting expression changes in low-abundance transcripts, and achieves precise absolute or relative quantification through real-time fluorescence monitoring, making it the “gold standard” for gene expression detection. The observed discrepancy for Bv7_168720_zftj.t1, aside from minor experimental variances, may stem from its potentially low expression abundance, where the RNA-seq depth might have been insufficient for precise detection, and qRT-PCR also failed to show a significant change. Furthermore, complex post-transcriptional regulation in plants can lead to inconsistencies between transcript levels and protein expression, a factor that neither technique can distinguish, as both only measure transcript levels. Additionally, DEG identification in RNA-seq is influenced by the number of biological replicates and the setting of significance thresholds, whereas the higher technical reproducibility of qRT-PCR might also contribute to minor deviations.

4.3. Functional Characterization of Drought Resistance and Regulatory Mechanisms of the Bv_005070_jjst.t1

Glycosylation modification is a crucial step in the biosynthesis of plant secondary metabolites. Acting in concert with hydroxylation and acylation, it modifies a variety of small-molecule compounds—such as flavonoids, hormones, and alkaloids—thereby diversifying the structures and functions of secondary metabolites and enhancing plant tolerance to abiotic stress [25]. Previous studies have confirmed that overexpression of drought-responsive glycosyltransferase genes such as UGT76C2 and UGT85E1 in Arabidopsis significantly improves drought resistance [10,26]. However, glycosyltransferases such as UGT73B2, which exert a negative regulatory effect on oxidative stress responses, also exist [27,28,29], indicating the functional diversity of UGT family members in modulating plant stress adaptation.

Based on multi-omics screening results, we performed functional validation of the core candidate gene Bv_005070_jjst.t1 via transgenic overexpression and RNAi-mediated silencing. Under drought stress, overexpression lines exhibited markedly superior growth phenotypes compared to wild-type plants. Concurrently, the activities of antioxidant enzymes—superoxide dismutase (SOD) and peroxidase (POD)—were significantly elevated, while malondialdehyde (MDA) content was substantially reduced. These results demonstrate that Bv_005070_jjst.t1 enhances drought tolerance in sugar beet by boosting antioxidant enzyme activity and alleviating membrane lipid peroxidation, thereby corroborating the conserved role of glycosyltransferases in regulating plant abiotic stress tolerance.

Notably, we observed an unexpected pattern in proline content: it was significantly lower in overexpression lines and higher in silencing lines, which contrasts with conventional research findings. We propose two plausible explanations for this discrepancy. First, the severe drought stress simulated by 30% PEG6000 in this study may have disrupted cellular homeostasis in sugar beet, leading to metabolic dysregulation in osmolyte biosynthesis and consequently altering the typical proline accumulation pattern. Second, overexpression of Bv_005070_jjst.t1 may have activated complex regulatory networks in sugar beet. Beyond its role in flavonoid metabolism, this gene might indirectly inhibit proline synthesis through interactions with other osmoregulatory pathways. This observation highlights the complexity of the molecular network underlying drought resistance in sugar beet, which likely involves multi-gene and multi-pathway crosstalk, warranting further mechanistic investigation.

4.4. Summary of Research Findings and Application Prospects

4.4.1. Research Findings

This study conducted a systematic investigation of the BvUGT90 gene family in sugar beet, encompassing bioinformatic analysis, expression profiling, and functional validation of core genes. The key findings are as follows: First, we systematically identified 121 BvUGT90 family members from the sugar beet genome for the first time. We characterized their physicochemical properties, chromosomal localization, gene structures, evolutionary relationships, and cis-regulatory elements in their promoters. This research fills a significant gap in the study of the UGT90 gene family in sugar beet. Second, through integrated transcriptomic and proteomic analyses under gradient drought stress, we elucidated the drought-responsive expression patterns of the BvUGT90 family. This approach enabled the identification of two core candidate drought tolerance genes, Bv_005070_jjst.t1 and Bv6_140060_stjc.t1, which exhibited significantly coordinated upregulation at both transcriptional and translational levels. Third, functional validation via transgenic approaches confirmed that Bv_005070_jjst.t1 enhances drought tolerance in sugar beet by increasing antioxidant enzyme activities and alleviating membrane lipid peroxidation. This study is the first to clarify the drought resistance function of this gene, thereby enriching our understanding of the molecular regulatory mechanisms underlying drought tolerance in sugar beet.

4.4.2. Application Prospects

The findings of this study hold considerable practical value for sugar beet molecular breeding and stress biology research:

• (1)The identified core drought tolerance gene, Bv_005070_jjst.t1, serves as a valuable genetic resource for molecular breeding. It can be applied to the development of drought-resistant sugar beet varieties through transgenic technology, gene editing, or marker-assisted selection. This application offers a potential solution to the significant yield and quality losses caused by drought stress in major sugar beet cultivation areas.
• (2)The research framework established for studying the sugar beet UGT90 gene family provides an important reference for mining stress-related genes and investigating their molecular mechanisms in other sugar crops, such as sugarcane and sweet sorghum.
• (3)The clarified regulatory relationship between the BvUGT90 family and MYB transcription factors, along with its association with flavonoid secondary metabolism, offers a new theoretical foundation for deciphering the stress-responsive molecular network in sugar beet. It also suggests novel technical strategies for improving stress resistance by modulating secondary metabolism.

4.5. Overall Scientific Significance of This Study

This study presents the first systematic characterization of the traits and drought stress response mechanisms of the BvUGT90 gene family in sugar beet while also identifying and functionally validating a core drought tolerance gene, Bv_005070_jjst.t1. This research not only enriches our understanding of the evolution and functional diversity of plant glycosyltransferase families but also fills a critical gap in our understanding of the molecular mechanisms underlying stress resistance in sugar beet. The results confirm the conserved role of the UGT family in mediating plant drought tolerance through the regulation of flavonoid secondary metabolism, thus providing new insights into the molecular regulatory networks underlying plant drought resistance. Furthermore, the identified core drought tolerance gene offers a direct genetic resource for the molecular breeding of drought-tolerant sugar beet varieties. These findings hold significant theoretical and practical importance for enhancing the drought stress resilience of sugar beet and ensuring the production security of sugar crops.

5. Materials and Methods

5.1. Plant Materials and Treatments

The sugar beet cultivar ‘HI0466’ (Swiss breed) was used in this study. Seeds were sown in plastic pots filled with a 3:1 (v/v) mixture of nutrient soil and vermiculite. All experiments were conducted in an artificial climate chamber located at the Sugar Beet Physiology Research Institute, West Campus of Inner Mongolia Agricultural University (Latitude: 40.807560, Longitude: 111.707220). The plants were cultivated under controlled conditions at 22 °C with a 16 h light/8 h dark photoperiod. Upon reaching the six-true-leaf stage (30–35 days after sowing), uniformly grown seedlings were selected for drought stress treatment, with three biological replicates per treatment. Drought stress was imposed using the weighing method to regulate soil moisture content. Based on preliminary experiments, seedlings were sampled at 0 (control, CK), 4 (DS4), 6 (DS6), 8 (DS8), and 10 (DS10) days after stress initiation, with an additional group rewatered for 3 days after DS10 treatment (RW). The DS8 time point was specifically set to capture key mid-stage drought stress responses for subsequent proteomic analysis. All collected samples were immediately frozen in liquid nitrogen and stored at −80 °C (Haier Twin-core Ultra-low Temperature Freezer DW-86L578ST, made in Shandong, China) for further use.

5.2. Identification of BvUGT90 Gene Family Members

To systematically identify the UGT90 gene family in sugar beet, this study used the protein sequence of BvUGT90A1 as the query sequence for BLASTP homologous search. BvUGT90A1 is a previously reported UGT family gene, and its encoded protein contains the signature conserved domain of the UGT family—the PSPG (Plant Secondary Product Glycosyltransferase) box. The sugar beet reference genome (RefBeet-1.2 version) used for the search was retrieved from the Boku Sugar Beet Genome Database (https://bvseq.boku.ac.at/Genome/Download/RefBeet-1.2/, Austria, accessed on 14 October 2025).

The hidden Markov model (HMM) profile corresponding to the PSPG box (Pfam accession number: PF00201) (EMBL-EBI) was downloaded from the Pfam database (https://pfam.xfam.org/, accessed on 14 October 2025) [30]. To ensure the reliability of candidate sequences, two complementary screening strategies were adopted: (1) HMM search using TBtools v2.129 software [31] with a threshold of E-value < 1 × 10^−10^; (2) conserved domain validation via the InterPro (EMBL-EBI) database (https://www.ebi.ac.uk/interpro/, accessed on 14 October 2025) [32] to confirm the presence of the PSPG box (PF00201) in candidate sequences. Candidate sequences identified by both strategies (intersection of the results) were retained for further analysis. Redundant sequences and those with incomplete open reading frames (ORFs) were removed using the ORFfinder tool (https://www.ncbi.nlm.nih.gov/orffinder/, accessed on 14 October 2025) and CD-HIT v4.8.1 software (sequence identity cutoff = 95%) [33]. Finally, 121 non-redundant members of the UGT90 gene family in sugar beet were confirmed.

For comparative analysis, UGT90 family members in Arabidopsis thaliana and Oryza sativa were identified using the same workflow. The A. thaliana reference genome (TAIR10 version) was retrieved from The Arabidopsis Information Resource (TAIR, Phoenix Bioinformatics, Stanford, CA, USA, https://www.arabidopsis.org/, accessed on 14 October 2025,), and the O. sativa reference genome (RGAP 7 version) was obtained from the Rice Genome Annotation Project (RGAP, Michigan State University, East Lansing, MI, USA, https://rice.plantbiology.msu.edu/, accessed on 14 October 2025).

The physicochemical properties of UGT90 family proteins, including molecular weight, theoretical isoelectric point (pI), instability index, and grand average of hydropathicity (GRAVY), were predicted using the ExPASy ProtParam tool (Swiss Institute of Bioinformatics, Geneva, Switzerland, https://web.expasy.org/protparam/, accessed on 14 October 2025) with default parameters. Subcellular localization was predicted using TargetP 2.0 (DTU Health Tech, Technical University of Denmark, Kongens Lyngby, Denmark, https://services.healthtech.dtu.dk/service.php?TargetP-2.0, accessed on 14 October 2025), a web-based tool from DTU Health Tech, with the “plant” model selected for analysis.

5.3. Construction of the Phylogenetic Evolutionary Tree of the BvUGT90 Gene Family

The sugar beet genome and its annotation files were obtained from the Beta vulgaris Resource (https://bvseq.boku.ac.at/Genome/Download/RefBeet-1.2/, accessed on 14 October 2025). The Arabidopsis thaliana genome and its annotation files were sourced from the TAIR database (https://www.Arabidopsis.org/, accessed on 14 October 2025). The rice genome and its annotation files were obtained from the Ensembl database (https://plants.ensembl.org/index.html, accessed on 14 October 2025). A phylogenetic evolutionary tree was constructed using the maximum likelihood method in MEGA 12 software, and the tree was visually refined using the online tool iTOL (https://itol.embl.de/, accessed on 14 October 2025).

5.4. Analysis of the Structure and Conserved Motifs of BvUGT90 Gene Family Members

The structural information of the BvUGT90 gene family members was obtained from the sugar beet database (https://bvseq.boku.ac.at/Genome/Download/RefBeet-1.2/, accessed on 14 October 2025). The conserved motifs were analyzed using the online tool MEME (https://meme-suite.org/meme/, accessed on 14 October 2025), and the results were visualized using TBtools.

5.5. Analysis of Promoter Cis-Acting Elements in BvUGT90 Gene Family Members

The 2000 bp DNA sequences upstream of the 5′UTR of the BvUGT90 gene family members were extracted using TBtools, and the promoter cis-acting elements were analyzed using the online website PlantCARE. The results were visualized with TBtools.

5.6. Collinearity Analysis of BvUGT90 Gene Family Members

Collinearity analysis of UGT90 gene family members in sugar beet, Arabidopsis thaliana, and rice was performed using TBtools, and the results were visualized.

5.7. Analysis of Expression Patterns of BvUGT90 Gene Family Members Under Drought Stress

The transcriptomic (RNA-seq) and proteomic data used in this study were derived from previous research conducted by our group [34]. The details are as follows: Transcriptomic Sequencing and Analysis RNA sequencing (RNA-seq) was performed by BGI Genomics on the Illumina HiSeq 2000 platform, generating approximately 50 million raw reads per sample (Table 2). Raw data were subjected to quality control using SOAPnuke software, with specific criteria including the removal of adapter-containing reads, reads with unknown bases (N) exceeding 10%, and low-quality reads in which bases with a quality score Q ≤ 10 constituted over 50% of the entire read. This process yielded high-quality clean reads, and quality control results indicated that the number of clean reads for each sample ranged from 50.33 to 50.41 million, with balanced base composition and Q20/Q30 ratios meeting experimental standards. Clean reads were aligned to the sugar beet reference genome (RefBeet-1.2) using BWA software, achieving alignment rates between 85.31% and 88.93%. Gene expression levels were normalized using the FPKM (Fragments Per Kilobase of transcript per Million mapped reads) method. To enhance the robustness of differentially expressed gene (DEG) identification, a combined approach that utilized three methods was employed. For pairwise sample comparisons, a Poisson distribution-based method was used with thresholds of FDR ≤ 0.001 and |log_2_FC| ≥ 1. For comparisons between groups with biological replicates, the Noiseq (thresholds: |log_2_FC| ≥ 1 and probability ≥ 0.8) and EBSeq (thresholds: |log_2_FC| ≥ 1 and PPEE < 0.05) methods were applied. Genes commonly identified by all three methods were considered DEGs for subsequent analysis.

Proteomics [34]: Total proteins were extracted from beet leaves subjected to different drought stress treatments. After enzymatic digestion, the resulting peptides were labeled using iTRAQ reagents and then desalted. The labeled peptides were subsequently separated by high-pH reversed-phase liquid chromatography and analyzed by mass spectrometry using a Q Exactive™ mass spectrometer coupled with nanoLC (Thermo Fisher Scientific, Waltham, MA, USA). The acquired spectra were processed using Mascot software (version 2.6.1), and protein identification and quantification were performed using MaxQuant (version 1.5.5.1). Differentially expressed proteins were screened based on the criteria of a fold change ≥ 1.2 and a p-value < 0.05. Functional annotation and pathway analysis of the differentially expressed proteins were conducted by aligning them against databases such as Gene Ontology (GO) and KEGG using the BLAST2go software.

5.8. Real-Time Quantitative PCR (RT-qPCR) Gene Expression Analysis

All RT-qPCR experiments in this study, including experimental design, execution, data analysis and result reporting, were strictly performed in accordance with the MIQE [35] guidelines. The raw Cq values and statistical results of all replicate analyses are compiled in Supplementary Table S1.

5.8.1. Experimental Design

RT-qPCR analysis was performed on samples subjected to drought stress for 0, 4, 6, and 10 days, as well as rewatered (RW) samples, and on transgenic plant lines, including wild-type (WT), overexpression (OE1–3), and RNA interference silencing (Ri1–3) lines. For each RNA sample, three independent biological replicates and two technical replicates were used to ensure the reproducibility and reliability of the results. All RT-qPCR experiments were conducted independently in our laboratory.

5.8.2. Sample Collection

Plant samples were collected and immediately frozen in liquid nitrogen. They were then transferred to a −80 °C ultra-low-temperature freezer and stored in sealed conditions until RNA extraction to prevent RNA degradation.

5.8.3. Total RNA Extraction and Quality Control

RNA extraction and genomic DNA removal: Total RNA was extracted from plant samples using the Universal Plant RNA Extraction Kit (DNase I) (suitable for polysaccharide- and polyphenol-rich plant samples, Cat. No. CW2598S) manufactured by CoWin Biosciences (Beijing, China). The extraction procedure was performed strictly according to the manufacturer’s instructions. A rigorous genomic DNA removal step was included during the extraction process to completely eliminate genomic DNA contamination. This treatment system was validated through preliminary experiments and by the manufacturer’s official protocols, demonstrating its effectiveness in completely removing genomic DNA residues from sugar beet leaf samples. Following extraction, RNA concentration and purity were assessed using a NanoDrop 2000 spectrophotometer (Gene Company Limited, Hong Kong, China). For each sample, 2 µL of extracted RNA was applied to the detection pedestal for measurement. The results are presented in Table 3.

5.8.4. Reverse Transcription Reaction

Reverse transcription was performed using the HiFiScript gDNA Removal and cDNA Synthesis Kit (Cat. No. CW2582M) manufactured by CoWin Biosciences (Beijing, China). According to the manufacturer’s validation, the reverse transcription efficiency of this kit exceeds 95%. Based on the statistical analysis of the first dataset in Supplementary Table S1, the coefficient of variation (CV) of Cq values for the reference gene ACT was 2.17%, demonstrating excellent reproducibility and stability of the reverse transcription reaction. Template amounts were added according to the manufacturer’s instructions and the measured RNA concentrations. The reaction systems and procedures are presented in Table 4 and Table 5.

5.8.5. RT-qPCR Primer Design and Validation

Primers were designed using the online primer design platform of Sangon Biotech (Shanghai, China) Co., Ltd. (https://store.sangon.com/primerDesign, accessed on 24 September 2025). The parameters were set specifically for SYBR Green-based quantitative real-time PCR, with primer lengths of 18–22 bp, amplicon lengths of 80–200 bp, and Tm values of 58–60 °C. All primers were designed to span exon-exon junctions to effectively prevent non-specific amplification from genomic DNA (Table 6). Additionally, BLAST analysis against the sugar beet reference genome sequence was performed to verify that all primers specifically matched only the target gene sequences, with no homologous non-specific binding sites and no risk of primer dimer formation. All primers produced consistent amplification results across all samples. The standard deviations (SD) of Cq values for technical replicates were <0.4 for all samples, and the coefficients of variation (CV) for biological replicates were <5% for all samples. Detailed data are provided in Supplementary Material Table S1.

5.8.6. RT-qPCR Reaction System and Protocol

Real-time quantitative PCR (RT-qPCR) was performed using the YALEPIC Universal SYBR Green qPCR MasterMix (Cat. No. YQ51003, Yeasen Biotechnology (Shanghai, China) Co., Ltd.) on a CFX Connect™ Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Hercules, CA, USA). cDNA synthesized from reverse transcription was used as the template. The reaction systems and thermal cycling protocols are detailed in Table 7 and Table 8.

5.8.7. Data Analysis and Statistical Processing

Raw qPCR data were analyzed using the instrument’s proprietary software, Bio-Rad CFX Manager 3.1, with automatic baseline and threshold settings to determine the quantification cycle (Cq) values for each sample. Relative expression levels of target genes were calculated using the 2^−ΔΔCq^ method [35], provided that the amplification efficiency difference between the target and reference genes was less than 5%. The specific calculation procedure was as follows: (i) The ΔCq value for each sample was calculated as ΔCq = Cq (target gene) − geometric mean Cq (reference gene); (ii) The ΔΔCq value was calculated as ΔΔCq = ΔCq (test sample) − mean ΔCq (control group); (iii) The relative gene expression level of the test sample compared to the control group was calculated using the formula 2^−ΔΔCq^.

5.9. Construction, Transformation, and Validation of Overexpression and RNAi Silencing Vectors for Bv_005070_jjst.t1

To further explore the biological function of this gene in sugar beet, we constructed overexpression and RNAi silencing vectors targeting Bv_005070_jjst.t1 via homologous recombination combined with Golden Gate cloning. Subsequently, Agrobacterium-mediated genetic transformation was performed on the high-sugar sugar beet cultivar ‘BS02’, employing a laboratory-optimized transformation system adapted to our specific experimental conditions. The sugar beet genetic transformation protocol, based on Agrobacterium-mediated methods and optimized for this study, has been partially described in our previous publications [36].

The detailed procedure was as follows:

First, sugar beet seeds were rinsed under tap water for 30 min, surface-sterilized with 0.1% (w/v) HgCl_2_ solution for 10 min, and rinsed thoroughly with sterile distilled water. The sterilized seeds were then sown on germination medium. Seedlings germinated from these seeds were used as explants. The cotyledonary node region was wounded, pre-cultured for 1 day, and then immersed for 10 min in an Agrobacterium suspension (containing the overexpression or silencing vector) adjusted to an OD_600_ of 0.5. The bacterial culture had been grown overnight to an OD_600_ of 0.6–0.8 before resuspension. Excess bacterial suspension was blotted dry, and the explants were co-cultivated in the dark for 2–3 days. Following co-cultivation, explants were rinsed with sterile distilled water and transferred to a bacteriostatic medium for 5–7 days. Subsequently, the sterile explants were placed on a selection medium containing antibiotics for 15–20 days to select for transformed tissues. Resistant shoots were then transferred to a differentiation medium for shoot proliferation. When the plantlets reached 3–5 cm in height, they were transferred to a rooting medium to induce root formation. After approximately one month, robustly rooted plantlets were selected, acclimatized by gradually opening the culture vessel lids for 3–5 days, and then transplanted into pots filled with a 3:1 (v/v) mixture of nutrient soil and vermiculite. The transplanted plantlets were kept in the dark for 2–3 days before being moved to normal growth conditions.

The media compositions were as follows: Germination medium consisted of tap water supplemented with 1 mg/L 6-BA and 8.0 g/L agar (pH 5.8). Both pre-culture and co-cultivation media were based on MS medium, supplemented with 30 g/L sucrose, 1 mg/L 6-BA, 0.2 g/L casein hydrolysate, and 7 g/L agar (pH 5.8). The co-cultivation medium was additionally supplemented with 100 μM acetosyringone (AS). The resuspension solution was ½-strength MS medium containing 50 g/L sucrose, 5 g/L MES, and 0.02% (v/v) Silwet-77 (pH 5.8). The bacteriostatic medium comprised MS medium with 30 g/L sucrose, 1 mg/L 6-BA, 0.25 mg/L NAA, 0.5 mL/L PPM™, 8.0 g/L agar, and 400 mg/L timentin (pH 5.8). The selection medium was identical to the bacteriostatic medium, except that timentin was replaced with 6 mg/L of the appropriate selective antibiotic (G418) (pH 5.8). The differentiation medium had the same composition as the selection medium but without the antibiotic (pH 5.8). The rooting medium consisted of MS medium supplemented with 30 g/L sucrose, 2 mg/L NAA, 0.5 mL/L PPM™, and 8.0 g/L agar (pH 5.8). The pH of all media was adjusted to 5.8 prior to autoclaving.

The drought stress treatment conditions were determined based on our previous research [37] and preliminary experiments, considering seasonal temperature, ambient humidity, and other relevant factors. Transgenic plants were subjected to drought stress simulated by treatment with 30% PEG6000 solution.

A total of 30–35 independent overexpression and RNAi silencing lines were generated. For this study, lines exhibiting consistent phenotypes under stress were selected for subsequent measurements and analyses. Transgenic lines were initially identified by PCR amplification of the resistance gene (G418) present on the vector backbone. Further validation was performed by qRT-PCR (reaction system and program as described in Section 5.8) to confirm the expression levels of the target gene. The primers and reaction conditions used for identification are listed in Table 9, Table 10 and Table 11. Representative electrophoresis results of the PCR identification and the relative expression levels of the target gene in the selected lines are shown in Figure 11 and Figure 12, respectively.

6. Conclusions

This study systematically identified, for the first time, 121 members of the BvUGT90 gene family in sugar beet. The encoded proteins have an average length of 474 amino acids and molecular weights ranging from 10.78 to 99.10 kDa. Among them, 96.69% are acidic proteins (pI < 7), 90.91% are hydrophilic, and most members are localized to the cytoplasm. Synteny analysis indicated evolutionary conservation between this family and its homologs in Arabidopsis thaliana and Oryza sativa. The cis-acting elements identified in their promoters, which are involved in hormone response, stress defense, and metabolic regulation, suggest that this family plays a pleiotropic regulatory role in the growth, development, and stress adaptation of sugar beet. Further expression analysis revealed that under drought stress, 45 and 10 members were significantly upregulated at the transcriptomic and proteomic levels, respectively. Cross-omics Venn analysis identified 26 co-expressed members under both drought and re-watering conditions, among which only Bv_005070_jjst.t1 and Bv6_140060_stjc.t1 showed coordinated upregulation at both transcriptional and protein levels, establishing them as core candidate genes for regulating the drought stress response. Genetic transformation experiments confirmed that Bv_005070_jjst.t1 significantly enhances antioxidant enzyme activity, mitigates membrane damage, and thereby effectively improves drought tolerance in sugar beet.