Biodiverse Responses of O-methyltransferase Genes to Salt Stress and Fiber Development of Gossypium Species

O-methyltransferases (OMTs) are an important group of enzymes that catalyze the transfer of a methyl group from S-adenosyl-L-methionine to their acceptor substrates. OMTs are divided into several groups according to their structural features. In Gossypium species, they are involved in phenolics and avonoid pathways. Phenolics defend the cellulose ber from dreadful external conditions of biotic and abiotic stresses, promoting strength and growth of plant cell wall. In this study, an OMT gene family, containing a total of 192 members, has been identied and characterized in three main Gossypium species, G. hirsutum, G. arboreum and G. raimondii. Cis-regulatory elements analysis suggested important roles of OMT genes in growth, development, and defense against stresses. Transcriptome data of different ber developmental stages in Chromosome Substitution Segment Lines (CSSLs), Recombination Inbred Lines (RILs) with excellent ber quality, and standard genetic cotton cultivar TM-1 demonstrate that up-regulation of OMT genes at different ber developmental stages, and abiotic stress treatments have some signicant correlations with ber quality formation, and with salt stress response. Quantitative RT-PCR results revealed that GhOMT43 and GhOMT27 genes had a specic expression in response to salt stress while GhOMT16, GhOMT55, and GhOMT33 in ber elongation and secondary cell wall stages. Our results indicated that these genes might contribute to salt tolerance or ber quality traits respectively in Gossypium.


Introduction
Cotton (Gossypium Species) has the importance for natural ber all over the globe. The primary goals of upland cotton (G. hirsutum) perspectives have been always to achieve better quality with higher yield (Lee et al. 2007). Mostly G. hirsutum bears staple bers 25-40 mm in length and 15 µm in thickness at their full maturity. Fiber cells must undergo four distinct but partially overlapped developmental stages, including initiation, elongation, secondary cell wall deposition, and maturation. The secondary cell wall of ber, which is mainly composed of cellulose, is important especially for ber quality perspectives.
However, some studies have shown that secondary cell wall of bers of ax (Linum usitatissimum L.), ramie (Boehmeria nivea L.), and Spanish broom (Spartium junceum L.) also contain phenolics along with cellulose. Their bers are known for their physical properties such as length and strength and have been used for textile purposes. A thicker secondary cell wall was estimated to contain no less than 70% cellulose content while the cotton ber contains almost 90% cellulose ( (Zhong 1998). The involvement of OMTs mediate normal plant growth in the presence of lignin (Ye and Varner 1995). The initial OMT cDNA was described in 1991 (Bugos et al. 1991), then a series of OMT cDNAs have been cloned from diverse plants species, including Zea mays, Arabidopsis thaliana, Iris hollandica, and Nicotiana tabacum (Vincent et al. 2005).
According to substrate classi cation, plant methyltransferases have three major categories, I. Omethyltransferases (OMTs), II. N-methyltransferases (NMTs), and III. C-methyltransferases (CMTs). The category I OMTs are further classi ed into ve sub-categories. Sub-category I-a comprises caffeoyl coenzyme A 3-O-methyltransferase (CCoAOMT) and caffeic acid 3-O-methyltransferases (COMTs), which are involved in methylation in phenylpropanoids. Sub-categories I-b, I-c, and I-d act in methylation of hydroxyl in avonoid, alkaloids, and myoinositol, respectively. The fth sub-category I-e takes part in methylation of carboxyl of diverse acids. The results of a study (Zubieta et al. 2002) discovered the crystal structure of OMTs from Medicago sativa. In the light of the explanations, the OMT gene that was cloned and characterized from a medicinal plant Ligusticum chuanxiong and contained higher ferulic acid was named as LcCOMT. The LcCOMT gene was differentially expressed under cold stress treatments, which suggested that it can assemble the ferulic acid under chilling stress. BLAST analysis identi ed 23.9-40.2% similarity of LcCOMP with the OMTs of alkaloid, avonoid, iso avonoid, and phenylpropanoids ).
In the whole life cycle of cotton plant, it undergoes various environmental conditions from the early spring in April when it is sowed to mid-summer when it grows rapidly in vegetation and reproduction and to late autumn when it gets mature and is harvested. During the whole growth procedure, the cotton plant maintains an exquisite molecular controls and regulations. But little is known what roles the OMT family genes have played in cotton plant especially in early or late growth stage when season transition occurs, or in various stress conditions. Therefore, in this study, we identi ed the OMT family genes in the genome-wide scale and made detailed bioinformatics analysis of gene structure, chromosomal distribution, selection pressure during their evolution, sub-cellular localization, cis-regulatory elements etc, together with their expression pro ling in different developmental stages and in responses to various stresses. Their expression pro ling in developing ber cells was veri ed using RNA sequencing data from RILs, CSSLs, and TM-1 at different ber development stages. This study could open the way to comprehend the functions of OMTs in ber quality advancement and in cotton plant responses to abiotic stresses, and thus could assume a noteworthy part for further investigation in the molecular mechanism of ber improvement and stress tolerance.

Identi cation of OMT protein family members, sequences alignment, and phylogenetic tree construction
The hidden Markov model (HMM) (PF00891 and PF01596) was downloaded from Pfam (https://pfam.xfam.org/). The HMMER 3.0 software was used to acquire the OMT genes of Pfam (PF00891 and PF01596) with default parameters. Then the evaluated genes were con rmed by using Pfam (https://pfam.xfam.org/) and SMART (Simple Modular Architecture Research Tool) (Khan et al.

2018
). After con rmation of evaluated results, Manual check for the presence of methyltransferase domains performed. Members with the absence of required domains were manually removed, while some potential OMT genes were retrieved according to some other features including chromosomal positions, protein length (aa), and molecular weight (kDa) by using cotton functional genomic database (http://www.cottonfgd.org/). The full length amino acid sequences of G. hirsutum, G. arboreum, G. raimondii, A. thaliana, and T. cacao encoded by OMT genes were aligned with clustalx2 software (http://www.clustal.org/)(Arai et al. 2019) with default parameters for the neighbor-joining phylogenetic tree as 1000 bootstraps. Subsequently, two neighbor-joining phylogenetic trees were generated by using Mega 7 (Khan et al. 2018). The topology of both phylogenetic trees was con rmed to understand the phylogenic relationship within the ve plant species.
Nomenclature of these members was based on their chromosomal locations and numbers in each Gossypium species.

Chromosomal mapping and collinearity analysis
Gene IDs were used for blast within cotton genome les to estimate the positions of OMT genes. The physical positions of OMT genes in three cotton species were visualized by using TBtools software . Circle gene viewer model of TBtools software was used to visualize collinearity between homologous sequences.

Gene structure and conserved motifs
The structure of the OMT genes was analyzed using the online server of Gene  MBI9749, and Hai1 have high ber quality traits, while 693062, MBI7285, sGK9708, CCRI36, and CCRI45 have low ber quality traits. Detailed information of these referenced materials is presented in S1 Table. Transcriptome data of G. arboreum (PRJNA179447) , and G. raimondii (PRJNA79005)  were also included to compare the comparative expression of these OMT genes.
2.7 Plant material, RNA isolation, cDNA synthesis, and qRT-PCR Cultivars sGK9708 and 0-153 (S1 Table) were planted in April 2018 in the experimental elds of the Institute of Cotton Research, Chinese Academy of Agricultural Sciences, Anyang, Henan. Flowers were tagged on the day of anthesis for ber sampling in July 2018. Bolls of tagged owers were sampled in the morning between 9:00 and 10:00 AM at 10, 15, 20, and 25 days' post anthesis (DPA). The bers were dissected from the developing seeds right after boll picking and immediately stored at -80 o C for RNA extraction.
To examine the expression pro ling of OMT genes under salt stress, seeds of sGK9708 cultivar were germinated in wet lter papers for 72 hours and then were transferred to hydroponic conditions. The seedlings were treated with 200 mM NaCl at three leaves stage. The true leaves, stems, and roots were sampled at 0hr, 2hrs, and 6hrs of the treatment. The 0 h of treatment was considered as control sample to compare the expression pro ling with treated samples.
Total RNA isolation was performed with the RNAprep Pure Plant Kit by (Tiangen, Beijing, China). To eliminate the genomic DNA contamination, the RNA samples were treated with DNase1. RNA concentration and integrity was observed on Nano Drop 2000 spectrophotometer (Thermo scienti c, USA) and 1% agarose gel electrophoresis. cDNAs of the RNA samples that the A260/280 ratio reached 2.00 were synthesized using PrimeScript® RT Reagent Kit (Perfect Real Time, Takara Biotechnology Co., Ltd., Dalian, China). qRT-PCR was performed with ABI 7500 fast Real-Time PCR system (Applied Biosystems, USA), with Gh-Histone3 gene was used as reference to normalize the relative expression level. Primers pairs of ve OMT genes were designed by using Oligo 7 (S2 Table)

Characterization of putative OMT genes
The genome wide analyses revealed that the OMT family genes have varying sizes and diverse annotations in Gossypium species (S3 Table. Sheet A). There various features and characterizations of OMTs in Gossypium might result in differentiation of functions of this gene family. A total of 192 OMT members were identi ed in three Gossypium species, including 82 in G. hirsutum, 55 in G. arboreum, and 55 in G. raimondii (S3 Table. Sheet A). For phylogenetic analysis (Ma et al. 2016a), 33 OMT members in A. thaliana, and 26 members in T. cacao species were also identi ed (https://phytozome.jgi.doe.gov/pz/portal.html) (S3 Table. Sheet B). Retrieving information of OMT genes in G. hirsutum revealed that GhOMT81, which was detected in scaffold, coded the smallest peptide of 62 amino acids (aa) with a molecular weight of 6.642 kDa. While GhOMT40, which was identi ed on chromosome D t 02, coded the largest peptide of 969 aa with a molecular weight of 108.296 kDa among all OMT members in three Gossypium species.
As domain wise characterizing OMT family genes in Gossypium species, the results revealed that almost 3/4 (G. hirsutum) to more than 4/5 (G. arboretum and G. raimondii) of the OMT family genes harbored methyltransf_2 domain, while only from 1/4 to less than 1/5 harbored methyltransf_3 domain. As comparison, a little less than 2/3 of the OMT family genes in A. thaliana harbored methyltransf_2 domain and in T. Cacao, this number was almost 2/2.

Chromosomal distribution, collinearity, duplication, and loss of OMT genes
We con rmed chromosomal locations of the OMT family genes in Gossypium as previously described . A total of 161 OMT genes were positioned on their respective chromosomes, while seven of G. raimondii, one of G. arboreum, and 23 of G. hirsutum were positioned in scaffolds (S4 Figure). In G. raimondii (D genome), chr11 was mapped with 13 genes followed by chr08 with nine genes. The minimum number of genes in a chromosome was one in chr2, chr6, and chr10 respectively. There was no OMT family members identi ed in chr01 and chr07 (S4 Figure.a). In G. arboreum (A-genome) (S4 Figure.b), 54 OMT genes were mapped in all chromosomes except chr1. Chr10 harbored 13 OMT genes which were the highest per chromosome, followed by chr12 and chr04 with 10 and 9 genes respectively.
The minimum number of genes located in a chromosome was one in chr02 and chr11 respectively. In G. hirsutum (A t D t genome) (S4 Figure.c), unexpectedly, there were no OMT genes in A t 02, A t 05, A t 07, D t 03, D t 09, and D t 11 chromosomes. The distribution of genes in D t sub-genome (33 genes) was higher than in A t sub-genome (26 genes). The maximum number of genes in a chromosome was seven in D t 04 and A t 12, followed by four in D t 10 and A t 10 chromosomes, respectively. D t 01, D t 05, A t 01, A t 06, and A t 11 only had one OMT gene, and D t 06, D t 07, A t 03, A t 08, A t 09, A t 13 two OMT genes and D t 02, D t 08, and D t 13 three OMT genes respectively (S4 Figure.c). A collinearity analysis of the OMT family genes in Gossypium species chromosomes was shown in Fig. 1. The results demonstrated a pair wise collinearity of OMT genes between the chromosomes on which OMT family genes were mapped. Noticeably, a number of available genes in A t and D t scaffolds were collinear with their homologues in A and D genomes suggesting the collinearity of the DNA fragments between the scaffolds and chromosome where these OMT genes locate (Fig. 1). Taken the OMT gene numbers identi ed in each A/D genome or A t /D t subgenome, collinearity analysis also revealed that there were totally 21 and 19 OMT genes exclusively detected in A and D genomes respectively. Their homologous counterparts in A t D t genomes of G.
hirsutum are lost. There are also a few OMT genes that are exclusively detected in A t D t genome of G.
hirsutum without homologous counterparts in A and D genomes (S4 Figure).
According to studies there are ve types of duplications including singleton, dispersed, proximal, tandem, and segmental or whole-genome duplication (Qiao et al. 2019). A total of 73 members of the OMT genes of three Gossypium species were identi ed to have dispersed duplications, fty-one genes (including 15 in A genome, 12 in D genome, and 24 in A t D t genome) to have segmental duplications, while thirty-four to have singleton duplications (S5 Table Sheet

Analysis of selection pressure
The ratio of the number of non-synonymous substitutions per non-synonymous site (Ka) to the number of synonymous substitutions per synonymous site (Ks) represents selection pressure of the gene (Ma et al. 2016b). Ka/Ks < 1 demonstrates high purifying selection pressure, while Ka/Ks > 1 shows positive selection pressure. Analysis of Ka/Ks ratio of homologous OMTs in three Gossypium species revealed that they are under purifying selection pressure. The Ka/Ks ratio of homologous OMTs in G. raimondii and G. arboreum ranged from 0.09 to 0.8, in G. raimondii and G. hirsutum ranged 0 to 0.7, and in A t and D t of G. hirsutum ranged 0.4 to 0.7 (S5 Table Sheet

Phylogenetic Analyses, Sequences alignment, conserved motifs and gene structure
The phylogenetic analysis included all of the 251 OMT genes identi ed in this study, including 192 genes from three Gossypium species ( Fig. 2.a), 33 from A. thaliana, and 26 from T. cacao species (Fig. 2.b). The evolutionary relationship of OMT genes in three Gossypium species was closer and more similar with each other (Fig. 2.a) than with A. thaliana and T. cacao ( Fig. 2.b). According to the topology of constructed tree, the OMT gene family is divided into ve clades (I, II, III, IV, and V) in Gossypium, A. thaliana, and T. cacao species. Previous study has also identi ed ve clades of OMT genes in Catalpa bungei (Lu et al. 2019). The results showed that each clade of OMT genes were symmetrically distributed within Gossypium species ( Fig. 2.a), while in A. thaliana and T. cacao, OMT genes were identi ed in cluster forms ( Fig. 2.b). The results demonstrated that these Gossypium OMT members might be evolutionary close within respective species and their identi ed clades.
To examine the conserved motifs of each clade, the analysis of representative motif logo and conserved motifs prediction were conducted (S6 Figure). The results revealed that motif 1, enriched with leucine, valine, and glycine, motif 2, enriched with leucine and valine, and motif3, motif 4, motif 5, and motif 6 were common in clades I, II, III, IV and clade V. While motif 7 was found missing in some members of clade V, which was then replaced with motif 8 at same positions (S6 Figure). The enriched amino acid residues of conserved motif1 (L/VDVGGG/TG) was previously identi ed in S-adenosyl-l-methionine (SAM)-dependant OMTs that shared 95% similarity with G. hirsutum OMT which further evidenced the biological function of identi ed genes in response to diverse abiotic stresses (Kim et al. 2013).
Methyltransferases serve different functions although they may have high sequence similarity. Investigation of gene structure has uncovered the different number of exons and introns of OMT genes. Exon and intron number of OMT genes varied from the least one exon and no intron to the most 7 to 9 exons and 6 to 8 introns (S7 Table). Two members in G. hirsutum including GhOMT67 and GhOMT30 contain nine exons as the highest (S7 Table). Same as, two members including GaOMT47 in G. arboreum, and GrOMT21 in G. raimondii contain 9 exons (S7 Table). Gene structure analysis revealed that the OMT genes with higher number of exons had a shorter exons and introns, and vise versa. These results demonstrated that OMT members possess different structural patterns in accordance with their features.

Identi cation of cis-regulatory elements in OMT family
The promoter regions of the OMT family contain precisely a large number of cis-regulatory elements. The analysis of cis-regulatory elements revealed the enrichment of MYB cis-regulatory elements, which was detected more than 350 times in OMT genes (S8 Figure). The MYC was another important element that was found 183 times in enlisted OMT genes. Box 4 (part of a conserved DNA module involved in light responsiveness) was found 152 times in 43/82 genes in G. hirsutum. ABRE elements was detected 119 times in 29/82 OMT genes in G. hirsutum. The ERE element was detected 113 times in 37/82and G-Box 97 times in 37/82 G. hirsutum OMT genes. An auxin RR-core and cis-acting regulatory element involved in the MeJA-responsiveness (TGACG-motif) were also observed in Gossypium OMT genes where this element was identi ed 48 times in 25/82 genes. Some other important cis-regulatory elements including wun-motif 44 times in 26/82, W-box 39 times in 31/82, GATA-motif 32 times in 27/82, O2-site 30 times in 22/82 OMT genes respectively, in G. hirsutum (S8 Figure). These cis-regulatory elements might function collectively in accordance with their speci c roles and with speci c conditions as well as growth and development stages (S8 Figure) 3.6 Sub-cellular localization prediction of OMT genes Understanding and determining the sub-cellular localization of proteins is an important strategy to identify the function of protein at cellular level (Binder et al. 2014). This approach includes proteomicbased experiments and microscopic high throughputs (Andersen et al. 2002;Herold et al. 2009). Several sequence-based approaches have been developed to predict the sub-cellular localization by providing amino acid sequences including PSORT (Horton and Nakai 1997), Yloc (Briesemeister et al. 2010), BaCelLO (Pierleoni et al. 2006), LOCtree (Goldberg et al. 2012). According to Cello prediction, most of OMT genes were located in the cytoplasm (Table 1), while seven genes were predicted in periplasm, including, GhOMT31, GhOMT33, GhOMT16, GhOMT55, GhOMT68, GhOMT69, and GhOMT71. Five OMTs were predicted to be localized in both periplasm and cytoplasm, including GhOMT32, GhOMT5, GhOMT50, GhOMT20, and GhOMT70. Two genes GhOMT67 and GhOMT30 were predicted in the outer membrane. Only GhOMT29 was predicted in inner membrane and cytoplasm (Table. 1). The function of the OMT genes might be related to their predicted localizations, though the experimental approach is still needed for further con rmation.   (Fig. 3.a). In KEGG Pathway analysis, the OMT family genes are categorized into two groups according to their domain functions. Twenty-nine OMT family genes were involved in monolignol biosynthesis, phenylpropanoid, secondary metabolism, and metabolic pathways and eleven genes were involved in phenylalanine and avonoid biosynthesis pathways (Fig. 3.b). In InterPro analysis (http://www.ebi.ac.uk/interpro/), 82 OMT genes were categorized as common as the functions of Sadenosyl-L-methionine-dependent methyltransferase (Fig. 3.c). Followed by the second most enriched categories of methyltransferase_2 and O-methyltransferase COMT-type, with Sixty-two genes predicted in each of them (Fig. 3.c). Fifty-seven OMT genes were enriched in winged helix-turn-helix DNA-binding domain, among which 53 genes were also predicted in plant methyltransferase dimerization category ( Fig. 3.c).

Expression pro ling of OMT genes and their homologues
In can be assorted into three basic groups (Fig. 4A): Those that have a broad responses to different developmental stages from germination to ber maturation, typical examples of which included GhOMT33, GhOMT71, GhOMT16 and GhOMT55; those that have speci c responses to root development, including GhOMT28, GhOMT4 and GhOMT34; and those that have responses to early germination in seed, cotyledon, root and stem, including GhOMT58, GhOMT32 and GhOMT46. When ber speci c transcriptome data sets of G. arboreum, G. raimondii were applied to observe the expression pro ling diploid OMT family genes, the result also supported speci c expression pro ling of some OMT genes in diploid species of G. arboreum (Fig. 4.B) and G. raimondii (Fig. 4.C).
The gene expression pro ling was further veri ed with trancriptome datasets of RILs (Fig. 5.A) two CSSLs Based on the expression pro ling of the OMT gene family in responses to cold, hot, osmotic, and salt stress treatments (Fig. 6.A), two genes speci c in salt stress responses, GhOMT27 and GhOMT43, and three genes speci c in ber development, GhOMT16, GhOMT55, and GhOMT33 were veri ed by qRT-PCR with RNA samples extracted from salt treatment. The results indicated that both GhOMT27 and GhOMT43 showed an elevated expression in salt treatments in salt-tolerant cultivar as compare to the control treatments (Fig. 6.B and 6.C). These two genes had different expression pro les from 2 h to 6 h after salt treatment. GhOMT27 had the highest expression at 2 h and then its expression went down at 6 h; whereas GhOMT43 had an increasing expression pattern from 2 h to 6 h. Both genes had much higher expression in roots than in stem or leaf. hirsutum, which clued that these genes might have experienced abovementioned events during evolution processes. Collectively, a higher number of genes were also identi ed in dispersed duplication event. Analyses of selection pressures and phylogenesis revealed that most of OMT genes in Gossypium species were under a purifying selection pressure and that high similarity of OMTs within Gossypium species supported the conservative evolution mode of OMT genes. The purifying selection pressure might suggest the importance of OMT genes in Gossypium species. But noticeable exceptions were also observed in some interspeci c homologous pairs, in which their Ka/Ks values were above one, indicating these homologous pairs were under a positive selection pressure. These homologous pair exceptions included GrOMT16-GhOMT78 and GrOMT29-GhOMT25 in G. raimondii and G. hirsutum, GaOMT21-GhOMT15 in G. arboreum and G. hirsutum, GrOMT30-GaOMT39 and GrOMT29-GaOMT40 in G. raimondii and G. arboreum. These results suggested that the OMT genes might had experience positive selection pressures during the evolution from diploids to tetraploids. Previous studies have evidenced that the positive selection pressure might be associated with the onsets of new functions in genes (Conant and Wolfe 2008; Van Zee et al. 2016). Considering the fact that quite a proportion of OMT genes were lost during the formation and evolution of allotetraploid cotton (see afore discussion and S4 Figure). In the current study, two OMT family members, GhOMT40 and GaOMT5, were characterized as reticuline 7-Omethyltransferase. Since reticuline is unknown in higher plants and have been only reported in legumes (Akashi 2003; He 1998). Therefore, how these genes function is still open to discussion. Taken all ndings together, the results might suggest that the OMTs that experienced positive selective pressure be lost or take on some novel functions in G. hirsutum during the processes of its evolution and ancestor formation.

A genome-wide survey of OMTs
Previous ndings have reported that the G. raimondii (D-genome) and G. arboreum (A-genome) are the closest relatives to the D t and A t sub-genomes of allotetraploids, respectively (Hu et al. 2019). Each gene in A or D genome will always have a homolog in the correspondent A t or D t sub-genomes of G. hirsutum (Ge et al. 2020). However, in both A and D genomes we detected quite a large number of OMT genes that do not have homologs in their relative A t and D t sub-genomes (S4 Figure). Previous studies evidenced that such homolog loss could result from two possible reasons: one is that the homologs were lost during the procedure of polyploidization from diploids to tetraploid; the other is that after the tetraploid formation, the OMT members in each genome started their separate evolution procedure. This separate evolution procedure makes the newly evolved members have no homologs in its relative genomes (Hu et al. 2019). Previous studies revealed that in A, D, A t D t genomes do not maintain same speed of evolution.
A faster evolution rate was observed in allotetraploid cottons than in diploid cottons (Hu et al. 2019). Taken the fact that OMT genes undergo a purifying selection procedures (S5 Table. Sheet B), the rst reason is possibly endorsed as the main cause for the current evolution status of OMT gene family and the second reason may also played a role.

OMTs are involved in diverse cis-regulatory elements
Plants encounter various biotic and abiotic stresses during their entire life cycles that negatively affect growth, development, and productivity (Lamaoui et al. 2018). Under exposure of these stresses, plants require some potential mechanism, which can be activated in critical circumstances, to support whole plant life cycle (Rao et al. 2006). Excessive salinity is also a major factor that affects the cotton production all around the world (Xu et al. 2013). Identi cation of cis-regulatory elements revealed that the OMT genes are enriched with important cis-regulatory elements that are essential against negative environmental stresses. Some important regulatory elements, including W-box, During the inoculation of pathogens, changes in the expression patterns of phenylpropanoid related OMT genes were identi ed. These identi ed OMT genes included GhOMT49, GhOMT58, GhOMT20, and GhOMT61 that were found signi cantly expressed in 12 and 48 hours post inoculation Verticillium dahliae (Li et al. 2019a). In the current study, these genes were down-regulated under abiotic stresses and in ber development stages (Fig. 4). Previous reports have evidenced that desoxyhemigossypol-6-Omethyltransferase (dHG-6-OMT) catalyzed the biosynthesis of terpenoid and provided an effective defense mechanism to cotton plant against biotic stresses including insects and pathogens (Liu et al. 1999). In response to V. dahliae (V991) in CSSLs lines CCRI36 and MBI8255, diverse genes were found differentially expressed in lignin biosynthesis including CCoAOMT, which can adequately utilize lignin and has been characterized in several previous studies ( . In wheat, TaCOMT-3D contributes to stem mechanical support ). Another TaCOMT gene was also observed with constitutive expression in stem along with leaf and root (Ma 2009). The OMT gene (BdCOMT1) was strongly expressed in stem node and internode but poorly expressed in other tissues in Brachypodium distachyon plant (Wu et al. 2013). The expression pro les of OMT gene family in the transcriptome data of TM-1 (Hu et al. 2019) and veri cation results through qRT-PCR also suggested that two OMT members GhOMT43 and GhOMT27 might contribute to salt stress tolerance in G. hirsutum. In the qRT-PCR veri cations, GhOMT43 and GhOMT27 showed different expression pro ling from 2 h and 6 h after 200 mM NaCl treatment (Fig. 5). Probably they act differently in response to salt stress in G. hirsutum. Four genes including GhOMT28, GhOMT38, GhOMT32, GhOMT62, and GhOMT26 had signi cant expressions in stem (Fig. 4.A) where they might be the potential candidates to provide structural support and survival to plant in environmental stresses.
Cotton ber quality of is an important attribute to develop elite cultivars in the presence of negative environmental factors. Studies demonstrated that GhOMT16 and GhOMT33 were expressed at elongation stages of a CSSL (CS-B25) and TM-1 respectively (Hsu et al. 2018; Hu et al. 2019). In the current study, the ber speci c OMT genes were consistently identi ed across various populations and species including TM-1 (Fig. 4.A) G. arboreum (Fig. 4.B) G. raimondii (Fig. 4.C), RILs (Fig. 5.A), CSSLs (Fig. 5.B, 5.C), and. They also showed highly similar expression patterns in different ber development stages. The expression speci cities of GhOMT16, GhOMT33, and GhOMT55 in developing bers were further veri ed through qRT-PCR studies (Fig. 5

Conclusions
Methyltransferases are versatile class of enzymes. OMT contributes to diverse phenolics that are essential for plant growth and serves as protective shield against several kinds of stresses. Various bioinformatics analyses revealed that OMT gene family is a strong growth regulator, which not only provide protection to the plant, but also are involved in ber elongation and secondary cell wall synthesis stages. Furthermore, expression pro ling analysis based on several transcriptome data and qRT-PCR validation inferred that GhOMT43 and GhOMT27 might be the potential candidates for salt stress tolerance and that GhOMT16, GhOMT55, and GhOMT33 might have signi cant in uence in ber development at elongation and secondary cell wall stages of G. hirsutum. This proposed study concludes the important roles of OMT family genes in cotton ber development and in salt stress tolerance.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.
Availability of data and material Not applicable.

Competing interests
Authors declare that they have no competing interests for the publication of the manuscript.     Collinearity analysis of OMT genes between AtDt (G. hirsutum), A (G. arboreum), and D (G. raimondii) genomes.