Identification of GT64 family
Following a comparative search using the local BLASTP program, candidate sequences were identified and their conservation domains confirmed through Pfam, SMART, and CDD analysis. Subsequently, a total of 39 GT64 genes were identified across eight cotton species, with three genes in each of G. herbaceum (A1), G. arboreum (A2), and G. raimondii (D5), and six genes in G. hirsutum (AD1), G. mustelinum (AD4), G. barbadense (AD2), G. tomentosum (AD3), and G. darwinii (AD5). Notably, the number of genes in the D subgenome G. raimondii (D5) and the A subgenomes G. herbaceum (A1) and G. arboreum (A2) are all three, while the count of GT64 family members in the five tetraploid cotton species is twice that of the three diploid cotton species, with each having six genes. These genes were subsequently renamed based on their chromosomal locations(Table S1).
The physicochemical properties of the GT64 gene family members in the eight cotton species were then assessed. The GT64 amino acid length ranged from 329 to 783 residues, with an average of 487 residues. The range of molecular weights for the proteins was 37.74 to 88.81 kDa, with an average of 55.22 kDa, while the isoelectric point (pI) varied from 8.31 to 9.4, averaging 8.92(Table S1).
Construction of the phylogenetic tree of GT64 gene family members
A systematic phylogenetic analysis was conducted to explore the evolutionary relationships within the GT64 gene family across eight cotton species(Fig. 1). The study included the creation of phylogenetic trees of 39 GT64 protein sequences.The GT64 proteins were categorized into five subfamilies labeled as Class1 through Class5. Class1 exhibited the highest membership with 13 members, followed by Class2 with nine members, and the fewest members belonged to Class5, totaling four. In Class1, each of the five tetraploid cotton species contained two members, while each of the three diploid cotton species had one member. Notably, G. herbaceum (A1) was absent in Class2. In Class3, all cotton species, except G. herbaceum (A1) and G. arboreum, were represented by one member each. G. raimondii was not present in Class4. Within Class5, G. herbaceum (A1), G. mustelinum (AD4), G. tomentosum (AD3), and G. darwinii (AD5) were the only species with one member each. Of particular interest is that in phylogenetic analysis, two diploid Gossypium species and two tetraploid Gossypium species often cluster together, indicating that upland cotton and island cotton have originated from two diploid Gossypium species[20].
Chromosomal location of GT64 genes in eight cotton species
We conducted a visual analysis of 39 GT64 genes in the cotton genome. For instance, in G. hirsutum(Fig. 2C), six genes were identified across chromosomes A05, A12, D04, D05, and D12, with an equal representation of genes in both the A and D subgenomes. Notably, chromosome A05 harbored two genes, while the remaining chromosomes each contained one gene. The distribution pattern of GT64 genes in G. barbadense mirrored that of G. hirsutum(Fig. 2D). In G. arboreum(Fig. 2A), three GT64 genes were located on chromosomes Chr05 and Chr12. Similarly, in G. raimondii, three GT64 genes were present on chromosomes Chr04, Chr05, and Chr12(Fig. 2B), with the distinct occurrence of one GT64 genes solely on chromosome Chr04, setting it apart from G. arboreum.
In G. herbaceum(Fig. 3A), similar to G. raimondii, the GT64 genes were distributed on chromosomes Chr04, Chr05, and Chr12. In the remaining cotton species, tetraploid species G. darwinii, G. mustelinum, G. tomentosum, and G. mustelinum exhibited a distribution pattern consistent with G. hirsutum and G. barbadense, with all six GT64 genes located on chromosomes A05, A12, D04, D05, and D12 (Fig. 3B-D).
Basic analysis of GHGT64s
We conducted phylogenetic tree construction and analysis of motifs and gene structures of six genes in upland cotton (Fig. 4). The results indicated that among the six members in G. hirsutum, a total of 10 motifs were identified. Specifically, GHGT64_1 and GHGT64_5 encompassed all 10 motifs, GHGT64_2 and GHGT64_4 contained six motifs each (motif1-motif4, motif6, and motif7), while GHGT64_3 and GHGT64_6 contained motifs 1 to 6.
Furthermore, an analysis of intron-exon structures was performed. As depicted in Fig. 4, members within the same group exhibited similar intron-exon arrangements. GHGT64_1 and GHGT64_5 featured four exons and three introns, GHGT64_2 and GHGT64_4 comprised five exons and four introns, and GHGT64_3 and GHGT64_6 possessed only one exon.
We further analyzed the cis-acting elements in the promoter regions of the six GT64 genes in upland cotton. In G. hirsutum (Fig. 4), the predicted elements included MYB binding sites related to drought response and flavonoid biosynthesis, light-responsive elements, and various phytohormone-related elements such as those responsive to abscisic acid, salicylic acid, methyl jasmonate, and auxin. Other identified cis-regulatory elements were involved in endosperm expression, low-temperature response, zein metabolism regulation, gibberellin response, seed-specific regulation, defense and stress response, palisade mesophyll cell differentiation, meristem expression, and cell cycle regulation. Through promoter analysis, these findings will support the validation of subsequent gene functions.
Analysis of gene duplication and synteny
In addition, we identified the repeat type of the GT64 genes in eight cotton genera (Table S2). Among the three diploid cotton species, namely Gossypium arboreum, Gossypium raimondii, and Gossypium herbaceum, all three genes are categorized as Dispersed type. Conversely, in tetraploid cotton species, all genes are classified under the WGD or Segmental duplication types.
We conducted multiple collinearity analyses of GT64 genes in eight cotton genus (Fig. 5). We observed five homologous gene pairs between G. barbadense and G. arboreum, five homologous gene pairs between G. hirsutum and G. arboreum, six homologous gene pairs between G. barbadense and G. hirsutum, six homologous gene pairs between G. barbadense and G. raimondii, six homologous gene pairs between G. hirsutum and G. raimondii, 12 homologous gene pairs between G. darwinii and G. hirsutum, six homologous gene pairs between G. darwinii and G. barbadense, five homologous gene pairs between G. darwinii and G. arboreum, six homologous gene pairs between G. darwinii and G. raimondii, six homologous gene pairs between G. darwinii and G. herbaceum, 12 homologous gene pairs between G. darwinii and G. mustelinum, 12 homologous gene pairs between G. darwinii and G. tomentosum, six homologous gene pairs between G. mustelinum and G. herbaceum, five homologous gene pairs between G. mustelinum and G. arboreum, six homologous gene pairs between G. mustelinum and G. raimondii, 12 homologous gene pairs between G. mustelinum and G. barbadense, 12 homologous gene pairs between G. mustelinum and G. tomentosum, 12 homologous gene pairs between G. mustelinum and G. hirsutum, five homologous gene pairs between G. tomentosum and G. arboreum, six homologous gene pairs between G. tomentosum and G. raimondii, 12 homologous gene pairs between G. tomentosum and G. hirsutum, 12 homologous gene pairs between G. tomentosum and G. barbadense, six homologous gene pairs between G. tomentosum and G. herbaceum, three homologous gene pairs between G. herbaceum and G. arboreum, three homologous gene pairs between G. herbaceum and G. raimondii, six homologous gene pairs between G. herbaceum and G. hirsutum, and six homologous gene pairs between G. herbaceum and G. barbadense, along with three homologous gene pairs between G. arboreum and G. raimondii. Our hypothesis, based on these observations, is that the GT64 gene family's evolution and gene amplification primarily stem from whole-genome duplication and segmental duplication events.
Subsequently, we conducted collinearity analysis among upland cotton and identified a total of three orthologous/paralogous pairs (Fig. 6B). Within the G. barbadense species (Fig. 6A), we identified three orthologous/paralogous pairs. In the other tetraploid cotton species, namely G. tomentosum (AD3), G. mustelinum (AD4), and G. darwinii (AD5), we observed three orthologous/paralogous pairs in each species (Fig. 6C-E). Additionally, no orthologous/paralogous pairs were found in the three diploid cotton species.
Selection pressure analysis of eight cotton genus
To explore the GT64 gene differentiation mechanism in cotton polyploid duplication events, we assessed the Ka/Ks ratio to discern selection pressure types on homologous gene pairs (Table S3). The Ka/Ks ratios were computed for 217 homologous gene pairs in eight cotton species individually (Fig. 6F-H). Notably, Ka/Ks ratios below 0.5 were observed between the diploid species G. raimondii and G. arboreum, as well as between G. raimondii and G. herbaceum. However, between G. arboreum and G. herbaceum, one homologous gene pair exhibited a Ka/Ks ratio exceeding 0.5, with another pair exceeding 1, indicating prevalent purifying selection among most homologous gene pairs in diploid cotton species, alongside a few instances of positive selection. Subsequent analyses extended to comparisons between diploid and tetraploid cotton species. Specifically, Ka/Ks ratios remained below 0.5 between G. herbaceum and both G. barbadense and G. hirsutum. Similarly, all homologous gene pairs between G. herbaceum and G. tomentosum displayed Ka/Ks ratios lower than 0.5. Notably, between G. herbaceum and G. darwinii, three pairs had ratios below 0.5, while between G. herbaceum and G. mustelinum, three pairs exhibited ratios below 0.5 and one pair had a ratio exceeding 1. These findings collectively suggest complex evolutionary dynamics involving diverse selection pressures in distinct cotton species and ploidy levels.
In tetraploid species, all homologous gene pairs within G. barbadense, G. hirsutum, G. tomentosum, G. mustelinum, and G. darwinii exhibited Ka/Ks ratios less than 0.5. Between G. barbadense and G. hirsutum, G. barbadense and G. mustelinum, and G. barbadense and G. darwinii, all homologous gene pairs had Ka/Ks ratios less than 0.5; however, between G. barbadense and G. tomentosum, two gene pairs had ratios greater than 0.5, while the rest were less than 0.5 between G. hirsutum and G. darwinii, one gene pair had a ratio greater than 0.5, and one pair had a ratio greater than 1, with the others less than 0.5 between G. hirsutum and G. mustelinum, one gene pair had a ratio greater than 0.5, while the rest were less than 0.5 between G. hirsutum and G. tomentosum, two gene pairs had ratios greater than 0.5, with the remaining pairs less than 0.5. Between G. darwinii and G. mustelinum, one gene pair had a ratio greater than 0.5, and the rest were less than 0.5 between G. darwinii and G. tomentosum, three gene pairs had ratios greater than 0.5, with the others less than 0.5 between G. mustelinum and G. tomentosum, two gene pairs had ratios greater than 0.5, while the remaining pairs were less than 0.5. These findings highlight the differential selection pressures and evolutionary dynamics among tetraploid cotton species.
In summary, among the eight cotton species, most GT64 genes have experienced intense purifying selection throughout evolution, with a few homologous gene pairs showing evidence of positive selection effects.
GT64 Genes Expression Profiles in G. hirsutum
In order to study the expression patterns of GT64 family genes, we utilized transcriptome data from various tissues of upland cotton[17]. The results indicated (Fig. 7A) that GHGT64_2 and GHGT64_4 exhibited higher expression levels compared to other genes in all tissues. Specifically, GHGT64_2 showed the highest expression in leaf, stem, receptacle, and stamen, followed by GHGT64_4, GHGT64_4 displayed the highest expression in pistil and root, followed by GHGT64_2(Fig. 7B). Furthermore, during cotyledon development (Fig. 7C), the expression level of GHGT64_4 increased gradually with time, reaching its peak at 96 h, followed by a decrease starting from 120 h. In the root development process, both GHGT64_2 and GHGT64_4 showed a gradual increase in expression level, reaching their peaks at 120 h. GHGT64_2 and GHGT64_4 exhibited a trend of initially increasing and then decreasing expression levels during seed development with time. In the process of fiber development (Fig. 7D), GHGT64_2 and GHGT64_4 had higher expression levels in ovules than fibers in the early stages of fiber development, but the opposite was observed in the later stages, where fiber expression levels were higher than ovule expression levels. GHGT64_2 displayed the highest expression at -1 DPA(days post-anthesis) during the entire fiber development process; GHGT64_4 exhibited the topmost expression at 20 DPA during development. GHGT64_5 showed the topmost expression at 35 DPA during ovule development compared to other genes, possibly related to the oil content of cotton seeds. Interestingly, through other studies[25], it was found that GHGT64_4 showed a gradual increase in expression levels at 2 DPA, 4 DPA, and 6 DPA in fuzz material, but a gradual decrease in fuzzless material, indicating that this gene may be involved in the development of fuzz fiber in upland cotton (Fig. 7E). Subsequently, transcriptome data from high lint percentage (LP) material LMY22 and low lint percentage (LP) material LY343 [26] revealed that GHGT64_2 and GHGT64_4 had higher expression levels in LMY22 compared to LY343 at the same stage of fiber development, suggesting that these two genes may be involved in regulating the changes in lint percentage (LP) during upland cotton fiber development (Fig. 7F).
Cottonseed oil has a wide range of applications and at the same time has a certain influence on the quality of cotton [27]. The results showed (Fig. 7G) that GHGT64_5 exhibited a rapid increase in expression levels at 20 DPA to 30 DPA in the low-oil material, suggesting that this gene may have a negative regulatory effect on the oil content in cotton materials. Cotton is highly susceptible to prolonged waterlogging stress. Then, we draw on previous studies [28]. The results indicated that the expression levels of GHGT64_2 and GHGT64_4 were higher compared to other genes, suggesting that these two genes play important roles in cotton's tolerance to flood stress and may be key genes for upland cotton's resistance to waterlogging (Fig. 7H).
The development of pigment glands plays a crucial role in cotton, based on previous research[29]. We found that the expression levels of GHGT64_2 and GHGT64_4 were higher in the four materials compared to other genes. GHGT64_2 showed significantly lower expression in Z17YW compared to Z17, while GHGT64_4 exhibited markedly higher expression in L7XW compared to L7, indicating that these two genes may regulate the development of pigment glands in upland cotton (Fig. 8A). Additionally, based on transcriptome data from defoliant-sensitive materials CIR12 and CCIR50 [30], the results showed that under different temperature treatments, the expression levels of GHGT64_2 and GHGT64_4 were higher in both the early and late stages of TDZ (Thidiazuron) treatment compared to the control, indicating that these two genes are involved in cotton's response to TDZ under different temperature conditions(Fig. 8B).
To study the response mechanism of the GT64 gene to abiotic stress, based on previous studies [17]. The results (Fig. 8C) showed that under salt and PEG stress, the expression level of GHGT64_4 at 12 h was higher compared to other genes, followed by GHGT64_2; while under hot and cold stress, the expression level of GHGT64_2 at 12 h was higher compared to other genes, followed by GHGT64_4. GHGT64_2 and GHGT64_4 exhibited an initial decrease followed by an increase in expression level under salt stress; under PEG stress, the expression level showed a continuous increase, reaching its peak at 12 h; under hot stress, the expression level displayed an initial increase followed by a decrease. Additionally, based on previous studies [17], it was observed that GHGT64_2 and GHGT64_4 exhibited a trend of initial increase, decrease, and subsequent increase in expression level with increasing time after Verticillium dahliae infection. Furthermore, compared to other genes, their expression levels increased after inoculation, indicating that these two genes may play a role in cotton's response to Verticillium dahliae (Fig. 8D).
GT64 Genes Expression Profiles in G. barbadense
Utilizing expression data of the GT64 genes in different tissues and fiber development stages of G. barbadense [17], we found that GBGT64_4 exhibited higher expression levels in calycle, pistil, petal, receptacle, and leaf compared to other genes; while GBGT64_2 demonstrated elevated expression levels in root, stamen, and stem. During ovule development, GBGT64_2 showed higher expression levels at 1 DPA, 10 DPA, and 20 DPA compared to other genes, while GBGT64_4 displayed increased expression levels in the beginning of fiber development. In the process of fiber development, GBGT64_4 showed the greatest expression levels at middle stages of fiber development, followed by GBGT64_2; GBGT64_2 had the topmost expression level at 25 DPA, followed by GBGT64_4 (Fig. 8E).
The expression levels data of materials with high and low fiber strength of island cotton were also utilized [31]. It was observed that GBGT64_2 and GBGT64_4 displayed significant differences in expression levels at 20 DPA, 25 DPA, 30 DPA, and 35 DPA in both materials, indicating the involvement of these two genes in the late-stage fibers development of G. barbadense, potentially regulating the quality of G. barbadense fiber strength (Fig. 8F). Subsequently, we utilized the expression levels data of disease-resistant and disease-susceptible materials of island cotton [32]. The results depicted (Fig. 8G) that the majority of GT64 genes showed no significant expression level changes before and after infection. However, The expression levels of GBGT64_2 and GBGT64_4 show significant differences in several extreme materials, which may indicate that these two genes play crucial roles in the resistance process of island cotton against FOV.
GHGT64_4 enhances the disease resistance of tobacco
Based on previous work, to identify the gene function of GH_D04G0699 (GHGT64_4), We transformed the gene into tobacco or transgenic lines to observe the disease resistance of tobacco.
We selected tobacco lines OE1, OE2, and OE3 with high expression levels for disease resistance assessment. The results showed that plants overexpressing GHGT64_4 were more resistant to Vd592 than the wild type. After 20 days of inoculation, the disease index of wild type was 76.94, and that of overexpressed plants was 26.02. Compared with wild type, the disease severity of overexpressed plants was significantly reduced (Fig. 9A). By day 20 post-inoculation, the wild-type plants displayed significant leaf necrosis, whereas the overexpressing plants exhibited leaf yellowing without necrosis. Over time, the wild type plants began to exhibit whole-plant necrosis (30 days), whereas the overexpressing plants showed signs of necrosis at 45 days post-inoculation, indicating that GHGT64_4 enhances tobacco resistance to Vd592 (Fig. 9B-D). qRT-PCR analysis of disease-related genes in tobacco (NbPR1a, NbPR2, NbPR9, NbPR10a, NbLOX, NbERF1) revealed that, except for NbERF1 and NbPR9, in the overexpressing tobacco, the expression levels of the other genes were generally elevated compared to those in the wild type. This indicates that GHGT64_4 can rapidly respond to Verticillium dahliae pathogen stress in the initial stage. Genes associated with the JA pathway, such as Lox6, reached their peak expression at 72 hours, while genes related to SA synthesis, such as PR1a, showed a rapid increase in expression after V. dahliae treatment, reaching a level over 10 times higher at 72 hours than the control. The PR9 protein, with peroxidase activity, thickens the cell wall by catalyzing lignin synthesis to resist pathogen invasion. This gene showed a rapid increase after V. dahliae treatment for 12 hours, followed by a decrease. These results demonstrate that GHGT64_4 can be heterologously expressed in tobacco, by activating disease-related protein genes, the tobacco's tolerance to Verticillium wilt was enhanced(Fig. 9E-J).
Validation of GHGT64_4 in cotton
In addition, VIGS experiments were conducted in cotton, we found that after injection for 15 days, the true leaves and stem veins of cotton exhibited a whitening phenotype, indicating successful gene silencing in cotton (Fig. 10A). qRT-PCR analysis of the target gene silencing efficiency in the experimental plants revealed a significant decrease in the expression levels of the target gene, indicating successful gene silencing in the plants (Fig. 10B-D). After silencing the GHGT64_4, the resistance to Verticillium wilt was weakened compared to the control group pTRV2:00. At 15 days post-inoculation with the V. dahliae pathogen, plants with silenced GHGT64_4 showed significantly reduced resistance, with a disease index of 42.92 (Fig. 10E), indicating a marked increase in disease severity compared to pTRV2:00. The results above indicate that GHGT64_4 is involved in the resistance of upland cotton to Verticillium wilt.
Furthermore, we used qRT-PCR technology to examine the expression levels of resistance-related genes in silenced plants. The results demonstrated (Fig. 10F) that, compared to control plants, pTRV2:GHGT64_4 plants exhibited significantly reduced expression levels of PAL, 4CL, PPO, PR1, and AOC, while CHI, SOD, CAT, and ACO showed significantly increased expression levels. These findings suggest that GHGT64_4 positively regulates the expression of PAL, 4CL, PPO, PR1, and AOC genes, while negatively regulating the expression of CHI, SOD, CAT, and ACO, indicating that GHGT64_4 mainly influences the synthesis of PAL, 4CL, and AOC, thereby affecting lignin and JA biosynthesis pathways, ultimately impacting cotton resistance. Following inoculation with Vd592 (Fig. 10G), 4CL, POD, EDS1, and ACO exhibited significant reductions compared to pTRV:00, while SOD and AOC showed significant increases. In addition, the pathogenic bacteria isolated on potato dextrose agar (PDA) showed that a large amount of V. dahliae grew in the silenced cotton stems with gene GHGT64_4, while no mycelium was observed in the control (Fig. 10H).
Lignin is thought to have a significant impact on shielding cotton plants from V. dahliae infection. To validate this finding, we conducted additional measurements of the overall lignin levels. Following inoculation with V. dahliae, the stems of TRV:GHGT64_4 plants exhibited reduced lignin content compared to TRV:00 plants, while exposure to V. dahliae resulted in an elevation of lignin levels in the plants(Fig. 10I).
This suggests that the expression levels of certain genes were elevated under the induction of other pathways. Additionally, the interplay or antagonistic effects among JA, SA, and ET pathways could lead to one pathway being enhanced while strongly inhibiting another[33], such as AOC and ACO. The signaling pathway is a vast and intricate network, where the suppression of one gene may be compensated by others, thus, following induction of Verticillium wilt in cotton, the expression of some disease-resistant genes showed an increase.
Transcription analysis GHGT64_4 between G. hirsutum and G. barbadense
Although a series of bioinformatics analyses have been conducted on the GT64 gene family and we have gained a basic understanding, their potential role in resistance to Verticillium wilt in upland and island cotton is still unclear. Based on the transcriptome data of 90 published samples [34] (derived from TM-1 and Hai7124 at time points 0 h, 12 h, 24 h, 48 h, 72 h, 96 h, 120 h, 144 h pre- and post-inoculation). TM-1 is a variety with low susceptibility to the disease, while Hai7124 is disease-resistant. We visualized the expression levels of six GHGT64 genes in G. hirsutum and G. barbadense materials (Fig. 11A, Fig. 11B) and found that genes GHGT64_2 and GHGT64_4 were highly expressed in both materials. Of note, GHGT64_4 exhibited consistently higher expression levels in G. barbadense compared to G. hirsutum from 12 h post-inoculation up to 144 h, indicating its significant role in disease resistance in G. barbadense (Fig. 11C).
Subsequently, we conducted Weighted Gene Co-Expression Network Analysis (WGCNA) on 9486 genes with FPKM > 10 in G. hirsutum and 9357 genes in G. barbadense. The result shows that, in the TM-1 material, a total of 19 modules were identified (Fig. 12A), with the MEturquoise module containing the highest number of genes at 3088, and the MElightgreen module having the fewest genes at only 44, averaging 499 genes per module. In the Hai7124 material, 10 modules were identified (Fig. 13A), with the MEturquoise module containing the most genes at 2558, and the MEgrey module containing the fewest genes at 114, averaging 935 genes per module. Core modules were chosen in both materials according to the criteria (|r|>0.50 and P < 0.001). Notably, the GHGT64_4 was found in the MEturquoise module in both G. hirsutum and G. barbadense. Comparison of genes in the MEturquoise module between these two materials revealed 1667 genes that were common to both (Fig. 11D), with 1422 genes and 892 genes unique to G. hirsutum and G. barbadense, respectively. We performed separate KEGG enrichment analysis on the gene sets of G. hirsutum and G. barbadense, showing common enrichments in pathways such as Spliceosome, Ribosome, mRNA surveillance pathway, Glutathione metabolism, Valine, leucine, and isoleucine biosynthesis in both materials. In G. hirsutum, enrichment was mainly observed in pathways like SNARE interactions in vesicular transport, Oxidative phosphorylation, Sphingolipid metabolism, Ubiquinone, and other terpenoid-quinone biosynthesis (Fig. 11E), whereas in G. barbadense, enrichment was primarily seen in pathways such as Endocytosis, Arachidonic acid metabolism, Lipoic acid metabolism, Ubiquitin-mediated proteolysis, and Sesquiterpenoid and triterpenoid biosynthesis (Fig. 11F).
Furthermore, we observed that in G. hirsutum, the MEturquoise module was significantly positively correlated with 144 hours post-inoculation (Fig. 12B), while in G. barbadense, the MEturquoise module was significantly positively correlated with 72 hours post-inoculation (Fig. 13B), indicating that GHGT64_4 may have initiated its immune response against Verticillium wilt in G. barbadense earlier than in G. hirsutum. We selected 150 genes with weight values greater than 0.02 from the MEturquoise module in each material as potential interacting partners with GH_D04G0699 (Table S4). To investigate the potential role of the GH_D04G0699 interaction network (Fig. 12C, Fig. 13C), we conducted KEGG pathway analysis on the set of 150 genes in each sample material (Fig. 12D). In both materials, the enriched pathways for these genes were primarily related to Plant-pathogen interaction, Endocytosis, alpha-Linolenic acid metabolism, and MAPK signaling pathway - plant. We hypothesize that during infection with Verticillium wilt, the two cotton species may employ metabolic pathways and signaling pathways to resist the disease. Interestingly, in G. barbadense (Table S5), processes such as RNA degradation and Ubiquitin-mediated proteolysis played significant roles in combating Verticillium wilt, unlike in G. hirsutum (Fig. 13D).
GHGT64_4 localization analysis
To examine the subcellular localization of GHGT64_4, a transient expression assay was performed by transfecting GFP-GHGT64_4 and GFP constructs into tobacco epidermal cells. GFP fluorescence in cells expressing GFP alone was observed in the nucleus and at the cell membrane, whereas GFP fluorescence from GHGT64_4 was specifically detected in the endoplasmic reticulum and Golgi apparatus. The results (Fig. 14) indicate that GHGT64_4 is primarily localized to the endoplasmic reticulum and Golgi apparatus.
Expression analysis of GT64s in different cotton varieties
To examine the level of GT64s expression under different stress conditions in various cotton varieties. Through the previous studies of expression and cis-acting elements, we propose that GHGT64_1, GHGT64_2, GHGT64_4, GHGT64_5, as well as GBGT64_2, GBGT64_4 may play roles in responding to various biotic and abiotic stress factors, fuzz fiber development, and modulation of oil content in cottonseeds.To further elucidate these roles, we conducted expression pattern analyses on selected G. hirsutum and G. barbadense cultivars to monitor the expression dynamics of these genes across different cotton species and temporal stages.
Initially, GHGT64_1 and GHGT64_5 were selected for fluorescence quantitative PCR analysis in cottonseeds with varying oil content, specifically low oil content from Emian22 and high oil content from 3–79, as depicted in Fig. 15A. Over the course of cottonseed maturation from 10 DPA to 30 DPA, a time-dependent increase in the expression levels of both genes was observed. Particularly noteworthy was the significant divergence in expression levels at 20 DPA and 30 DPA between the two varieties, suggesting a potential regulatory role of these genes in cottonseed oil production.
Subsequently, GHGT64_2 and GHGT64_4 were investigated via fluorescence quantitative PCR in fuzzless series material, as illustrated in Fig. 15B. While GHGT64_2 exhibited consistent expression levels, GHGT64_4 displayed significant differences in expression at 1 DPA and 3 DPA between the two variants. Notably, at 3 DPA, the expression of GHGT64_4 was higher in the fuzzless mutant, indicating a potential positive regulatory role in fuzz fiber development in G. hirsutum.
Following inoculation with V. dahliae, the transcription levels of these genes were markedly induced at specific time points in both resistant and susceptible variants, as shown in Fig. 15C. GHGT64_2 and GHGT64_4 displayed distinct expression patterns, with peak levels at different time points post-inoculation, suggesting their involvement in the response to V. dahliae invasion in cotton plants.
Upon exposure to PEG-induced drought stress, the transcription levels of GHGT64_2 and GHGT64_4 indicated a potential contribution to G. hirsutum's response to drought conditions, as depicted in Fig. 15D. The genes exhibited a dynamic expression pattern of initial downregulation, subsequent upregulation, and final downregulation, highlighting their potential role in drought stress response.
Subsequent examination of salt stress response in G. hirsutum variants Xinluzao26 (resistant to salt stress) and Xinluzhong30 (susceptible to salt stress) revealed that GHGT64_2 and GHGT64_4 may participate in the response to salt stress conditions, as shown in Fig. 15E. Both genes exhibited similar expression dynamics under salt stress, there are significant differences between the two extreme materials at specific points in time, suggesting their involvement in salt stress response.
The expression profiles of GBGT64_2 and GBGT64_4 were examined in G. barbadense, specifically in the FOV-resistant cultivar 06-146 and the FOV-susceptible cultivar Xinhai14, under FOV stress conditions, as presented in Fig. 15F. The results indicated contrasting roles of these genes in response to FOV stress, with significant differences in expression levels at various time points, underscoring their potential involvement in different stress conditions.