Introduction

Cotton is an important cash crop, and a pioneer crop in saline-alkali land. It has certain saline-alkali resistance and is an important model crop for us to study saline-alkali stress1. However, it can also be affected by various abiotic stresses during growth and development, such as salt, alkali, drought, low temperature, heat stress and so on. Plants can produce many secondary metabolites to eliminate harmful reactive oxygen species (ROS), thus promoting growth and development and resisting various biotic and abiotic stresses2. As an important secondary metabolite, isoflavones play a vital role in plant growth and development. Isoflavones belong to estrogen-like active substances, which can be classified as phytoestrogens3. Isoflavones are a subclass of flavonoids, and they are the major phytoestrogens naturally found in plants, such as peanuts, chickpeas, green peas, and alfalfa, and they are mainly produced in the leguminous plants. Previous studies had focused on the ability of flavonoids to relieve stress, mainly by increasing a wide range of biological activities including antioxidant and antifungal properties4. There were also investigations of flavonoids improving cotton's development5. In addition, flavonoid biosynthesis pathways regulate axillary bud growth by promoting the transport of auxin in upland cotton6. More recently, it was found that flavonoids were involved in response to low phosphorus stress7. However, the alleviation of abiotic stress in cotton by isoflavones has not been reported.

Isoflavone reductase (IFR) is a key enzyme in isoflavone synthesis pathway, which can control the synthesis of isoflavone in plants. At the same time, IFR is also involved in lignin synthesis to resist the stresses8. It was found that IFR gene was also involved in response to biotic and abiotic stresses. In soybeans, ABA treatment induces the expression of isoflavone reductase to resist ABA stress through the synthesis of secondary metabolites9. Under waterlogging stress, the expression of isoflavone reductase related genes was significantly up-regulated, which may play an important role in waterlogging stress10.

In addition, some studies have elucidated the function of IFR gene. In rice, overexpression of isoflavone reductase-like gene (OsIRL) can enhance tolerance to ROS stress11. Overexpression of GmIFR in soybeans can increase soybean resistance to Phytophthora sojae12. IRL gene in rice plays an important role in resisting heat stress13. IFR gene is also involved in the biosynthesis of tobacco alkaloids to resist various stresses14. However, it has not been known whether IFR gene is involved in abiotic stress in cotton.

In recent years, the genome sequencing of two diploid cotton Gossypium arboreum (G. arboreum)15 and Gossypium raimondii (G. raimondii)16 and two allotetraploid cotton Gossypium hirsutum (G. hirsutum)17,18 and Gossypium barbadense (G. barbadense)19 has been completed, which provides strong support for our research on the evolution of IFR gene family and the potential function of IFR genes.

To investigate the evolution and potential functions of the IFR gene family, we analyzed the IFR phylogenetic tree, gene structure, chromosomal position, collinearity relationship, and promoter cis-acting elements. In addition, the heat map of GhIFR gene under different stresses was analyzed. A key gene that is highly expressed in salt stress, GhIFR9A, was confirmed to positively regulate salt stress by VIGs assay. This study laid a foundation for further research on the function of IFR gene and its response to abiotic stresses.

Results

Identification of IFR family genes

The HiddenMarkov model of NmrA was used as a query file, and genes with conserved domain of NmrA were selected as candidate genes of IFR gene family, and Pfam database was used for further analysis based on the number PF05368. After further screening IFR genes by using CD search, we finally identified 28, 28, 14 and 15 IFR genes in G. hirsutum, G. barbadense, G. arboreum and G. raimondii, respectively. Similarly, we characterized IFR genes in other 7 species including A. thaliana, G. max, P. trichocarpa, O. sativa, T. cacao, V. vinifera and Z. mays, and 7, 36, 10, 7, 6, 7 and 4 IFR genes were identified. To facilitate subsequent identification of IFR genes, we renamed them according to the location of chromosomes (Supplementary Table S1). Among them, IFR genes in G. hirsutum were named GhIFR1A-GhIFR14A and GhIFR1D-GhIFR14D. The IFR genes in G. barbadense were named as GbIFR1A-GbIFR13A and GbIFR1D-GbIFR15D. The IFR gene in G. arboreum was renamed as GaIFR1-GaIFR14. The IFR gene in G. raimondii was renamed as GrIFR1-GrIFR15. The combined number of IFR genes of G. arboreum and G. raimondii was 29, almost equal to that of G. hirsutum or G. barbadense. We also analyzed the physical properties of IFR gene in G. hirsutum (Table 1). The protein length ranged from 270 amino acids (GhIFR8A) to 747 amino acids (GhIFR2A), and the molecular weight ranged from 30.343 kDa (GhIFR8A) to 81.643 kDa (GhIFR2A). The minimum isoelectric point (pI) was 5.005 (GhIFR14D), the maximum was 10.148 (GhIFR2A), and the average pI was 7.105. The grand average of hydropathy was negative, indicating that most GhIFR proteins were hydrophilic proteins.

Table 1 Physical properties of the GhIFR genes.
Full size table

Phylogenetic analysis of IFR gene family

To understand the evolutionary relationship of IFR gene families, MEGA7 software was used to construct phylogenetic trees of four cotton species and other 7 species (Fig. 1). According to the branch of evolutionary tree and gene structures, IFR genes were divided into 5 clades, among which clade I had the most IFR genes, 16 in G. hirsutum and 57 in four cotton species. Clade III contained the least IFR gene, including 1 in G. hirsutum and 3 in four cotton species. In addition, we also found the same trend in the evolutionary tree of four cotton species and seven species. Clade I contained the most IFR genes, accounting for 54.94% of the total, with 89 genes. The same clade contained IFR genes of four cotton species, and each clade contains T. cacao, indicating a close evolutionary relationship between cacao and cotton. In phylogenetic trees, the GhIFR gene pairs and GbIFR gene pairs were always clustered together, which could be the results of gene duplication. In addition, IFR genes in A. thaliana, G. max, P. trichocarpa and four cotton species were all distributed in clade I–V, indicating that IFR genes in these plants were evolutionarily linked.

Figure 1
figure 1

Phylogenetic trees of four Gossypium species and other 7 species. (A) Phylogenetic trees of the IFR genes from four Gossypium species, (B) Phylogenetic relationship of the IFR genes from four Gossypium species and other 7 species. The 5 clades, designated as I to V, are marked with different colored backgrounds.

Full size image

Chromosomal location of IFR gene family

To further study the distribution and evolutionary relationship of IFR genes on chromosomes, we mapped all genes onto corresponding chromosomes (Fig. 2). Only two genes were not located on the chromosome (GhIFR11A and GhIFR14D), and all the other genes were unevenly distributed on their chromosomes. In G. arboreum, Chr3 had relatively most GaIFR genes, with 3 genes, Chr1, Chr4, Chr5, Chr10, and Chr11 had the smallest number of IFR genes, with 1. No genes were distributed on Chr2, Chr7, and Chr9, and no tandem duplication events occurred. In G. raimondii, Chr5 and Chr8 had relatively most GrIFR genes, including 3 and 3 genes, respectively. No genes were found on Chr1, Chr3, and Chr6. In the At subgroup of G. hirsutum and G. barbadense, the number of IFR genes on Chr3 and Chr12 was the most, with 3 and 3, respectively. Although the number of IFR genes in A/D subgenome was similar, the distribution of these genes in chromosomes was not corresponding. However, the situation was different in the Dt subgroup, the most IFR genes on Chr2 in G. hirsutum, which was 3, and the most IFR genes on Chr2 and Chr12 of G. barbadense, with 3 (Table 2).

Figure 2
figure 2

Chromosome distribution map of four Gossypium species. (A) Chromosomal location of IFR genes on chromosomes in G. arboreum (Ga), (B) Chromosomal location of IFR genes on chromosomes in G. raimondii (Gr), (C) Chromosomal location of IFR genes on chromosomes in G. hirsutum At sub-genome (GhAt), (D) Chromosomal location of IFR genes on chromosomes in G. hirsutum Dt sub-genome (GhDt), (E) Chromosomal location of IFR genes on chromosomes in G. barbadense At sub-genome (GbAt), (F) Chromosomal location of IFR genes on chromosomes in G. barbadense Dt sub-genome (GbDt). The gene ID on the right side of each chromosome correspond to the approximate locations of each IFR gene. The scale of the genome size was given on the left.

Full size image
Table 2 Chromosome number distribution map of IFR genes in four Gossypium species.
Full size table

Gene duplication and collinearity analysis

Gene duplication events are one of the main contributors to evolutionary dynamics, and they play a significant role in the rearrangement and expansion of gene families20. Whole genome duplication, segmental duplication, and tandem duplication were the main causes of expansion in plant gene family during the evolution21. We identified 267 duplicated gene pairs in 10 combinations (Ga-Ga, Ga-Gb, Ga-Gr, Gb-Gb, Gb-Gr, Gh-Gh, Gh-Ga, Gh-Gb, Gh-Gr and Gr-Gr) (Supplementary Table S2). Among them, 1 gene pair (GhIFR8D/GhIFR9D) was identified as the tandem duplication, 40 were the segmental duplication and 226 were whole genome duplication (Fig. 3). The number of Lineal/Parallel homologous duplicated gene pairs of IFR gene in Gh-Ga, Gh-Gr, Gh-Gb, Gb-Gr, and Gb-Ga was 28, 30, 33, 32, and 93, respectively. The colinear gene pairs of Ga-Ga, Gb-Gb, Gh-Gh and Gr-Gr were 4, 18, 20 and 1, respectively. Therefore, segmental duplication and whole genome duplication were the main driving forces of IFR gene family evolution, leading to gene amplification.

Figure 3
figure 3

Collinearity analysis of IFR duplicated gene pairs in G. hirsutum, G. barbadense, G. arboretum, and G. raimondii. Chromosomal lines represented by various colors indicate the syntenic regions around the IFR genes.

Full size image

Selection pressure analysis

To study the mechanism of IFR gene differentiation after polyploid duplication events in cotton, Ka, Ks and Ka/Ks of 10 combinations were calculated (Fig. 4, Supplementary Table S3). In general, Ka substitution can cause amino acid changes that may alter the conformation and function of proteins, thus causing adaptive changes. Ka/Ks ratio can determine whether there is selective pressure on the protein-coding gene. The Ka/Ks ratio was used to infer the selection pressure of duplicated gene pairs. The results showed that 17 duplicated gene pairs with Ka/Ks > 1, 146 duplicated gene pairs with Ka/Ks < 1, indicating that most IFR genes underwent intense pure selection.

Figure 4
figure 4

Ka/Ks ratio analysis of 10 combinations. Ka and Ks divergence values for (Gh-Gh), (Gb-Gb), (Ga-Ga), (Gr–Gr), (Gh-Gb), (Ga-Gr), (Ga-Gh), (Ga-Gb), (Gr-Gh) and (Gr-Gb) are shown in radar chart. Different colors represent Ka/Ks gene pairs of 10 groups.

Full size image

Analysis of the conserved motifs and gene structures of GhIFR genes

Diversity of gene structure and differentiation of conserved motifs can promote the evolution of gene families22. We analyzed the conserved motifs and gene structures of GhIFR genes (Fig. 5). In general, most IFR genes belonging to the same clades had similar motif types, arrangements and quantities, but they were not strictly unified in the same clade. For example, each GhIFR gene contained 3–10 conserved motifs, especially in subgroups I, II and IV. Except for GhIFR8A, all the other GhIFR genes contained motif 2, suggesting that motif 2 may be conserved in evolution. In G. hirsutum, 2 (7.1%) IFR genes had 1 single exon, 5 (17.9%) IFR genes had 4 exons, 12 (42.9%) IFR genes owned 5 exons and 9 (32.1%) IFR gene had more than 5 exons (Fig. 5C).

Figure 5
figure 5

Phylogenetic tree, conserved motifs and gene structure analysis of GhIFR genes in G. hirsutum. (A) Phylogenetic tree of GhIFR genes, (B) Conserved motifs of GhIFR genes, (C) Gene structures of GhIFR genes. Green boxes indicated exons, and black lines indicated introns.

Full size image

Analysis of promoters and heatmap of GhIFR genes

By analyzing the upstream 2000 bp promoter region of the GhIFR genes, we found that most GhIFRs contained cis-acting elements related to plant hormones and abiotic stresses (Fig. 6B). All GhIFR genes contained light responsive elements, suggesting that all GhIFR genes may be involved in photosynthesis. 46.4% of the genes contained low-temperature responsive elements, 39.3% of the genes contained cis-acting elements responding to salicylic acid. 60.7% of the genes contained cis-acting elements related to MeJA. 46.4% of the genes contained gibberellin-responsive cis-acting elements. There are more cis-acting elements associated with plant hormones than with abiotic stresses. These results suggested that GhIFR family genes played an important role in hormone signal transduction and stress response in plants.

Figure 6
figure 6

Phylogenetic tree, cis-acting elements and heat map analysis of GhIFR genes in G. hirsutum. (A) Phylogenetic tree of GhIFR genes, (B) Cis-acting elements in promoters of GhIFR genes, (C) Heatmap of GhIFR genes under different abiotic stresses.

Full size image

To further investigate the potential role of GhIFR genes in G. hirsutum, we constructed phylogenetic tree and heat maps of GhIFR genes under different stresses (cold, heat, salt and drought stress) (Fig. 6A,C). Under cold stress, most GhIFR genes tended to be down-regulated in response to cold stress, while most GhIFR genes the expression of most GhIFR genes was up-regulated under heat, salt and drought stress. The results showed that most GhIFR genes responded positively to abiotic stresses.

qRT-PCR of GhIFR genes in response to abiotic stresses

To further confirm the GhIFR genes in response to various abiotic stresses, we randomly selected 10 GhIFR genes to detect the expression patterns in leaves under cold, heat, drought and salt stress by using qRT-PCR (Fig. 7). The results showed that GhIFR9A could be induced by all four different abiotic stresses. Most of GhIFR genes responded positively to heat stress, and their expressions were significantly up-regulated, such as GhIFR7A, GhIFR8A, GhIFR9A, GhIFR6D, GhIFR7D, GhIFR9D and GhIFR14D. In addition, 6 GhIFR genes (GhIFR8A, GhIFR9A, GhIFR13A, GhIFR6D, GhIFR7D and GhIFR14D) were significantly induced under cold stress, showing an upregulation of expression. Two genes, GhIFR7A and GhIFR9D, were only induced by heat stress, while GhIFR1D was only induced by salt stress. These genes that were actively responded to cold, heat, drought, and salt stress could be selected to further verify their functions, such as GhIFR9A, GhIFR8A and GhIFR13A, especially the GhIFR9A gene, which had high expression under four abiotic stresses. Subsequently, the function of the GhIFR9A gene will be verified because it is significantly upregulated in various stresses.

Figure 7
figure 7

Analysis of the expression patterns of GhIFR genes under cold, heat, drought and salt stress by qRT-PCR. The mean values were from three independent biological replicates. Statistical analyses were performed by Student’s t-test (*P < 0.05 and **P < 0.01).

Full size image

Cotton plants with the GhIFR9A silenced by VIGS were sensitive to salt stress

According to the results of qRT-PCR, GhIFR9A showed a significantly up-regulated trend to the four abiotic stresses, and its expression level was the highest under salt stress. Therefore, we decided to study the function of GhIFE9A under salt stress. First, we used VIGS technology to silence this gene, and found that the gene expression was significantly down-regulated after silencing, and under salt stress treatment, cotton seedling exhibited the wilting phenomenon, and pYL156: GhIFE9A cotton seedings was more severe than that of pYL156, indicating that GhIFE9A was positively regulating the tolerance to salt stress (Fig. 8).

Figure 8
figure 8

Silencing GhIFE9A via VIGS increased sensitivity to salt stress. (A) Phenotype of cotton leaves after VIGS, (B) qRT-PCR for GhIFE9A gene under salt stress. WT: No infection, salt stress: 200 mM NaCl.

Full size image

Discussion

IFR gene plays a crucial role in plant growth and development and tolerance to abiotic stress14. However, there still lack of systematic understanding of IFR gene family in cotton. With the completion of genome sequencing of G. hirsutum, G. barbadense, G. arboreum and G. raimondii, it is convenient for us to study the IFR gene family. In this study, we identified 28, 28, 14 and 15 IFR genes in G. hirsutum, G. barbadense, G. arboreum and G. raimondii, respectively. The number of IFR genes in two tetraploid cotton is almost the sum of the number of genes in two diploid cottons, which is consistent with the previous study that tetraploid may be formed through hybridization between two subgenomes A and D23. In G. hirsutum, IFR genes in At and Dt subgenome was identical, so we could assume that the translocations and reverse transcript insertion rarely occurred. In addition, 7, 36, 10, 7, 6, 7 and 4 IFR genes were identified in A. thaliana, G. max, P. trichocarpa, O. sativa, T. cacao, V. vinifera and Z. mays, respectively.

Phylogenetic analysis showed that all species had gene pairs from the same node, indicating that the IFR genes in all species had experienced gene duplication that making the expansion of the IFR gene family in the process of evolution. Notably, each clade had plants such as A. thaliana, G. max and P. trichocarpa, proving that they were evolutionarily close to cotton. In addition, we found that in every clade, cotton and cacao were on the same evolutionary branch, which was consistent with previous studies that cotton and cocoa came from the same ancestor18. The phylogenetic analysis indicated that the IFR genes in cotton might have similar biological functions as TcIFRs. At the same time, the uneven and random distribution of IFR genes suggested that the events of gene loss may occur during the process of evolution, and it might also be due to incomplete genome assembly. Motif is a short sequence of relatively conserved features shared among a group of genes. It may be a recognition sequence or it may encode a functional protein24. GhIFR genes in the same subfamily had similar gene structures and conserved motifs, which provided support for their clustering in the phylogenetic trees, and the highly conserved sequences in the same subfamily indicated that the GhIFR genes may have been duplicated in evolution. All genes contained motif 2, showing that motif 2 might be used to identify the of IFR gene family. Almost all genes contain motif 9, except for GhIFR11A and GhIFR11D, suggesting that these two genes may have lost this motif in evolution.

Gene duplication will lead to functional differentiation of genes, which is necessary for environmental adaptation and speciation25. Chromosome distributions and collinearity relationship analysis showed that the amplification of IFR genes in cotton was mainly derived from segmental duplication and whole genome duplication. By comparing the number of duplicated gene pairs of Gh-Gr, Gb-Gr, Gh-Ga and Gb-Ga, we found that the number of duplicated gene pairs in Gh-Gr and Gb-Gr was less than that in Gh-Ga and Gb-Ga. These results were consistent with previous results that the A-derived subgenome was more active than the D-derived subgenome during the evolution. There was only one tandem duplication on D08 of G. hirsutum, while no tandem duplication in G. barbadense, G. arboreum and G. raimondii, suggesting that there was a special evolutionary pattern in the evolution of different cotton species. In general, the ratio of Ka/Ks can reflect the evolutionary background of genes. Ka/Ks = 1 represents neutral selection, and natural selection will not lead to gene mutation. Ka/Ks > 1 indicates that the gene undergo positive selection, which accelerates the evolution of the gene. Ka/Ks < 1 indicates that the gene has been purified and selected to eliminate harmful mutations and retain important protein structures26. The Ka/Ks ratio for almost all duplicated gene pairs is less than 1, indicating that the cotton IFR gene family has undergone strong pure selection during evolution, with limited functional differentiation after segmental duplication and whole genome duplication.

Isoflavone reductase encoded by IFR gene is a class of key enzymes for the synthesis of secondary metabolites such as lignin and isoflavone, and actively responds to biological and abiotic stresses11,12,27. In maize, IFR gene was activated in response to sulfur starvation28. IFR protein was induced by ABA treatment in soybean9. Plant hormones play an essential role in plant resistance to abiotic stress. The promoter region of GhIFR genes in G. hirsutum contained cis-acting elements such as light responsive, low temperature responsive and plant hormone responsive. Salicylic acid (SA) could improve abiotic stress tolerance by regulating major metabolic processes in plants29. By analyzing cis-acting elements of the GhIFR genes, we found that 11 of 28 GhIFR genes contained SA responsive elements. MeJA is also a plant hormone that may be used against pathogens, salt stress, drought stress, low temperature stress and heavy metal stress30, and in this study, 18 (64.3%) GhIFR genes contained MeJA-related cis-acting elements, especially GhIFR1D and GhIFR7D, they had 3 MeJA responsive elements, indicating they may respond to adversity stress by regulating the synthesis of MeJA. In conclusion, GhIFR genes contained cis-acting elements related to abiotic stresses and plant hormone, which suggesting that they may play an important role in cotton growth and development and stress.

Under abiotic stress, plants will produce stress responses and related genes are induced to adapt to various developmental and physiological changes31. At the same time, GhIFR gene expression pattern analysis under different stresses showed that GhIFR gene actively responded to various stresses. It was found that the expression of IRL gene in wheat was increased, and the increased expression level was closely related to the synthesis of antioxidants under heat stress13. In cotton, we found that most GhIFR genes actively responded to heat stress, so we speculated that IFR genes may also resist heat stress by increasing the synthesis of antioxidants in cotton, but this needs to be verified in the future.qRT-PCR results showed a key gene, GhIFR9A, was significantly up-regulated under cold, heat, salt, and drought stress, which could be selected to further verify its function. There were still some GhIFR genes that were not significantly differentially expressed under cold, heat, salt, and drought stress, possibly because they mainly function in other aspects or had lost some functions in evolution. The silence of GhIFR9A gene by VIGS experiment showed that the plants that silenced GhIFR9A gene were more sensitive to salt stress, which proved that GhIFR9A gene could indeed positively regulate the tolerance of salt stress. Previous studies had found that salt stress produced a large amount of reactive oxygen species (ROS)32, so we speculated that the GhIFR9A gene will eliminate ROS against salt stress through a complex regulatory mechanism.

Conclusions

In this study, we analyzed the phylogenetic relationship, gene structure, chromosome distribution and cis-acting elements of the IFR gene family, which greatly enriched our understanding of the cotton IFR gene family. In addition, the gene expression profiles under different stress indicated that GhIFR genes actively participated in cold, heat, drought and salt stress, which laid a foundation for further analysis of the function of IFR genes. GhIFR9A gene was induced by salt stress, and silencing GhIFR9A gene would cause more severe phenotype in cotton.

Materials and methods

Identification of IFR gene family members in cotton

To identify the members of the IFR gene family in four cotton species (G. hirsutum, G. barbadense, G. arboreum and G. raimondii), we used PF05368 (NmrA-like family) in Cotton Functional Genomic Database (CottonFGD) (http://www.cottonfgd.org/) to preliminarily retrieve IFR gene family members33. The IFR genes of other 7 species including Arabidopsis thaliana (A. thaliana), Glycine max (G. max), Populus trichocarpa (P. trichocarpa), Oryza sativa (O. sativa), Theobroma cacao (T. cacao), Vitisvinifera Genoscope (V. vinifera) and Zea mays (Z. mays) were obtained from the online website Phytozome 13 (https://phytozome-next.jgi.doe.gov/). Then the NCBI CD search website (https://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi) was used to delete C and N terminal to further screen IFR family genes34. Finally, family genes of four cotton species and 7 species were obtained.

Phylogenetic tree analysis of IFR gene family members

CottonFGD was used to download the protein sequences of four cotton species, and Phytozome 13 was used to obtain the protein sequences of 7 species. MEGA7.0 software was used for multiple sequence alignment. Then, MEGA7.0 software was used to construct the phylogenetic tree of four cotton species using neighbor-joining (NJ) algorithm with 1000 bootstrap repetitions, and the parameter was set as default, and phylogenetic tree of four cotton species and other 7 species was constructed by using Maximum likelihood method with 1000 bootstrap replicates35.

Chromosomal locations of IFR genes in cotton

In order to better display the distribution of genes on each chromosome, we used the gff3 file of the genome and gene ID to carry out a visual analysis of the distribution of genes on chromosomes of four cotton species by using TBtools software36.

Duplicated gene pairs and collinearity relationship analysis

The duplicated gene pairs of four Gossypium species and the collinearity relationship between different gene pairs was performed using MCScanX37. The Advance Circos tool in TBtools was used to make visual analysis of the collinearity and homologous chromosome regions of the four cotton species.

Calculation of selection pressure

To understand the selection pressure experienced by the IFR duplicated gene pairs of four cotton species, the non-synonymous substitution rate (Ka), synonymous substitution rate (Ks) and Ka/Ks were calculated to investigate the selection pressure by using TB Tools software.

Analysis of the conserved motifs and gene structures of GhIFR genes in G. hirsutum

Multiple Em for Motif Elicitation (MEME, http://meme-suite.org/)38 was used to identify the conserved motifs of GhIFRs, and the maximum motif number was set to 10, other parameters were set as default. The newick file and gff3 file were then used to visualize the phylogenetic tree, conserved motifs and gene structure using TBtools.

Cis-acting element analysis of GhIFR genes in G. hirsutum

The upstream 2000 bp sequence was obtained from CottonFGD as the promoter. Then they were submitted to online website PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/)39 to get the cis-acting elements. TBtools software was used to construct the diagram of phylogenetic tree and cis-acting elements of GhIFR genes.

Expression pattern analysis of GhIFR genes

RNA-Seq data (PRJNA490626)40 including cold, heat, salt and drought stresses was downloaded from NCBI (https://www.ncbi.nlm.nih.gov/). TBtools software was used to visualize the heatmap with FPKM values of GhIFR genes.

qRT-PCR analysis of GhIFR genes under different abiotic stresses

To explore the expression patterns of GhIFR genes in different abiotic stresses, leaves of cotton exposed to cold (4 °C), heat (40 °C), drought (12% PEG6000) and salt (200 mM NaCl) stress at three leaf stage were collected for RNA extraction at 0 h and 12 h, respectively. Cotton plants treated with ddH2O were considered as control, and three biological replicates were taken for each treatment. The total RNA was isolated by using EASYspin Plus plant RNA quick isolation kit (Aidlab Co., LTD, Beijing, China). The pure RNA was reverse-transcribed using TransScript® II one-step gDNA removal and cDNA synthesis supermix (TransGen Biotech Co., LTD, Beijing, China) according to the manufacturer's instructions. Ten GhIFR genes were randomly selected for the qRT-PCR experiment, and the primer sequences were shown in Supplementary Table S4. 2−ΔΔCt method was used to measure relative expression levels of GhIFR genes41.

Virus‑induced gene silencing (VIGS) experiment

pYL156: GhIFR9A vector was constructed with the restriction enzyme cutting site BamHI and SacI. Primer sequences are shown in Supplementary Table S4. The GV3101 strains carrying pYL156, pYL156: GhIFR9A, pYL156: PDS, and pYL192 were cultured and injected into the underside of cotyledons of upland cotton material TM-1. When the plants injected with pYL156: PDS appeared an albino phenotype, it proved that the VIGs experiment was successful. Then the leaves were taken for qRT-PCR experiment. After that, the plants with pYL156, pYL156: GhIFR9A were treated with salt stress (200 mM NaCl), and their phenotypes were observed.