GbEXPATR, a species‐specific expansin, enhances cotton fibre elongation through cell wall restructuring

Yang Li; Lili Tu; Filomena A Pettolino; Shengmei Ji; Juan Hao; Daojun Yuan; Fenglin Deng; Jiafu Tan; Haiyan Hu; Qing Wang; Danny J Llewellyn; Xianlong Zhang

doi:10.1111/pbi.12450

. 2015 Aug 13;14(3):951–963. doi: 10.1111/pbi.12450

GbEXPATR, a species‐specific expansin, enhances cotton fibre elongation through cell wall restructuring

Yang Li ¹, Lili Tu ^1,^✉, Filomena A Pettolino ², Shengmei Ji ¹, Juan Hao ¹, Daojun Yuan ¹, Fenglin Deng ¹, Jiafu Tan ¹, Haiyan Hu ¹, Qing Wang ¹, Danny J Llewellyn ², Xianlong Zhang ¹

PMCID: PMC11388876 PMID: 26269378

Summary

Cotton provides us the most important natural fibre. High fibre quality is the major goal of cotton breeding, and introducing genes conferring longer, finer and stronger fibre from Gossypium barbadense to Gossypium hirsutum is an important breeding strategy. We previously analysed the G. barbadense fibre development mechanism by gene expression profiling and found two homoeologous fibre‐specific α‐expansins from G. barbadense, GbEXPA2 and GbEXPATR . GbEXPA2 (from the D_T genome) is a classical α‐expansin, while its homoeolog, GbEXPATR (A_T genome), encodes a truncated protein lacking the normal C‐terminal polysaccharide‐binding domain of other α‐expansins and is specifically expressed in G. barbadense. Silencing EXPA in G. hirsutum induced shorter fibres with thicker cell walls. GbEXPA2 overexpression in G. hirsutum had no effect on mature fibre length, but produced fibres with a slightly thicker wall and increased crystalline cellulose content. Interestingly, GbEXPATR overexpression resulted in longer, finer and stronger fibres coupled with significantly thinner cell walls. The longer and thinner fibre was associated with lower expression of a number of secondary wall‐associated genes, especially chitinase‐like genes, and walls with lower cellulose levels but higher noncellulosic polysaccharides which advocated that a delay in the transition to secondary wall synthesis might be responsible for better fibre. In conclusion, we propose that α‐expansins play a critical role in fibre development by loosening the cell wall; furthermore, a truncated form, GbEXPATR , has a more dramatic effect through reorganizing secondary wall synthesis and metabolism and should be a candidate gene for developing G. hirsutum cultivars with superior fibre quality.

Keywords: cotton fibre, Gossypium barbadense (cotton), truncated α‐expansin, cell elongation, secondary cell wall

Introduction

Expansin is the first identified cell wall‐loosening protein that functions in a pH‐dependent manner but has no detectable hydrolytic or other enzymatic activity (McQueen‐Mason et al., 1992). A nonenzymatic model by Sampedro and Cosgrove (2005) postulated that expansins weaken noncovalent bonds between cell wall matrix polymers to promote slippage of cellulose microfibrils, stress relaxation and cell extension. Expansin constitutes a large multigene family of four groups: α‐expansin, β‐expansin, expansin‐like A and expansin‐like B (Sampedro and Cosgrove, 2005). Sequence analysis indicates that true expansins contain two domains and share several conserved motifs. The N‐terminal domain 1 has a structure similar to that of family‐45 glycoside hydrolases, while the C‐terminal domain 2 has similarity to the type‐A cellulose‐binding module that binds to microcrystalline cellulose (Georgelis et al., 2012). Over the past two decades, research on expansin has shown tremendous progress, and these outcomes showed that expansins were involved in many plant growth and development processes. Expansin down‐regulation inhibits cell wall extension, leading to phenotypic changes in the petal limb, leaf and storage root (Goh et al., 2012; Noh et al., 2013; Zenoni et al., 2004). Expansin overexpression in tomato, rice and petunia affects fruit softening, seedling growth and organ size (Brummell et al., 1999; Choi et al., 2003; Zenoni et al., 2011). Some expansins are also regulated by auxin and ethylene (Hutchison et al., 1999; Son et al., 2012) and also involved in abiotic stresses (Guo et al., 2011; Lü et al., 2013).

As one of the longest plant cells, the cotton fibre is approximately 1000–3000 times longer than its width (Kim and Triplett, 2001); however, the role of α‐expansin in fibre development has not yet been closely investigated. Two α‐expansin cDNAs specific to the developing cotton fibre have been isolated from G. hirsutum (Gh, (AD)₁ genome), which suggested that α‐expansin may be involved in fibre development (Harmer et al., 2002). However, Xu et al. (2013a) showed that simultaneous overexpression of GhEXPA1 and GhRDL1 that encoded a BURP domain‐containing protein, thought to interact with α‐expansin, did not affect fibre length or strength, although it had a significant effect on fruit production and hence fibre yield.

Cotton is the most important crop in providing a natural raw textile fibre for humans. There are two cultivated allotetraploid species of cotton, Gh and G. barbadense (Gb, (AD)₂ genome). Gh accounts for 90% of the world fibre production while Gb for 5%–8% (Qin and Zhu, 2011). However, the fibre quality of Gb is superior to that of Gh, with longer, finer and stronger fibres. Higher fibre quality equates to a more comfortable textile and better productivity in the spinning mill so manipulating fibre developmental processes to improve quality is a common target for breeding and biotechnology. At the cellular level, cotton fibre differentiates from the single outer epidermal cells on the seed, and the process can be divided into five developmental stages: initiation, elongation, transition, secondary wall synthesis and maturation (Haigler et al., 2012). Cotton fibre cell wall undergoes three phases: primary cell wall synthesis during expansion and elongation, transitional cell wall restructuring and then extensive secondary cell wall deposition. Primary cell wall synthesis commences from 0 day post anthesis (DPA) to approximately 20 DPA. Transition (approximately 16 to 20 DPA) is a key period that includes the primary cell wall ceasing elongation and the new secondary cell wall beginning to remodel cell wall polymer composition in preparation for the production of thick secondary cell wall (Haigler et al., 2012). During this phase, specific cellulose synthases are highly expressed, and the successive cellulose layers deposited during this stage put an end to any further elongation. The proportion of cellulose in mature fibre is approximately 95% of its dry weight. The research showed that the fibre cell wall properties play an important role in the fibre development process. For example, overexpression of GhXTH1, suppression of GhPEL and down‐regulation of GhPRP5 could effect on the fibre development and final quality (Lee et al., 2010; Wang et al., 2010; Xu et al., 2013b).

In addition to the intracellular osmotic pressure that drives cell expansion, the biomechanical properties of the fibre primary cell wall will play an important role in regulating fibre elongation. Cell wall‐loosening proteins, such as expansin, are likely to be key players in determining the rate and temporal period of fibre elongation and should have a role in determining fibre quality. In our study, a novel truncated α‐expansin, GbEXPATR, was found to specifically expressed at the fibre elongation stage in Gb. To compare the functions of GbEXPATR and the normal full‐length form of expansin, GbEXPA2, in fibre elongation, transgenic cotton lines with RNAi and overexpression of these two genes were produced. The cell wall composition and the fibre quality of the transgenic lines were altered with the GbEXPA2 and GbEXPATR expression level changes, thus an obvious role for α‐expansin in cell wall remodelling was demonstrated. In particular, GbEXPATR, that lacks the carbohydrate‐binding domain 2, had a strong effect on cell elongation through delaying secondary cell wall synthesis and, as a result, enhanced fibre length, fineness and strength.

Results

GbEXPATR is specifically expressed in G. barbadense fibre and is a functional protein

To increase our molecular understanding of fibre development in Gb, we constructed a normalized cDNA library of G. barbadense 3‐79 fibre and analysed the expression profiles of highly expressed genes during fibre growth (Tu et al., 2007). GbEXPA2 (DQ912952) and GbEXPATR (DQ912951), two very closely related α‐expansin genes with high expression levels, were identified amongst those ESTs. GbEXPA2 encoded a typical α‐expansin protein including N‐terminal signal peptide, domain 1 and domain 2, while GbEXPATR encoded a shorter protein containing only signal peptide and domain 1 (Figure 1a). The ORF sequences and deduced protein sequences of GbEXPA2 and GbEXPATR were highly similar in their overlapping regions (Figures 1a and S1a,b). However, the 3′ UTRs of the two ESTs were significantly different (Figure S1c). Genomic DNA sequences of GbEXPA2 and GbEXPATR genes were obtained from G. barbadense 3‐79. Comparing gDNA and cDNA sequences, we identified two introns located at the same positions in both genes that were highly similar in size and sequence (Figures 1a and S1d). These findings suggested that both GbEXPA2 and GbEXPATR were the members of α‐expansin family despite the absence of a domain 2 in GbEXPATR.

*GbEXPA2* and *GbEXPATR* protein domain structures and their gene expression patterns. (a) The domain structures of *GbEXPA2* and *GbEXPATR* . *GbEXPA2* has a signal peptide for entry into the secretory pathway, domain 1 and domain 2, found in other α‐expansins, but *GbEXPATR* only has a signal peptide and domain 1. There are amino acids differences at positions 12 (*GbEXPA2* is Ser, and *GbEXPATR* is Pro) and 37 (*GbEXPA2* is Gln, and *GbEXPATR* is Arg), indicated by black arrows. The intron positions are indicated by triangles. (b) *GbEXPA2* and *GbEXPATR* were expressed specifically during fibre cell development. qRT‐PCR analysis of *GbEXPA2* and *GbEXPATR* expression in root (R), hypocotyl (Hy), leaf (L), petal (Pe), anther (An), stigma (St), ovules with fibres attached (0 DPA) and the isolated fibres from 5 to 25 DPA. TM‐1 and 3‐79 are varieties of Gh and Gb, respectively. The data were calculated relative to the cotton ubiquitin gene ( *UBQ7*). Error bars represent the standard deviation (SD) of three biological replicates. (c) *GbEXPA2* transcript accumulated in 5–15 DPA fibre cells. Northern blot analysis of *GbEXPA2* expression in R, Hy, L, Pe, An, ‐3 DPA ovules, 0 DPA ovules with fibres attached, and the fibres from 5 to 25 DPA. The full‐length *GbEXPA2* cDNA was used as the probe, and *18S RNA* was used as the endogenous standard. (d) *GbEXPATR* is specifically expressed in *G. barbadense* fibre. Relative expression levels of *GbEXPA2* and *GbEXPATR* in the 15 DPA fibres of different cotton species (■Gh varieties Xu142, YZ1, *morrilli* and *marie‐galante*; □Gb varieties Hai7124, Giza47 and *peruvianum*). *UBQ7* was set to ‘1′. Error bars represent the SD of three biological replicates. MG: *marie‐galante*.

The expression patterns of GbEXPA2 and GbEXPATR were investigated by qRT‐PCR, which showed that GbEXPA2 transcripts accumulated to high levels during rapid fibre cell elongation stage (5 to 15 DPA) and low levels during the fibre secondary cell wall deposition in both G. hirsutum TM‐1 and G. barbadense 3‐79 (Figure 1b). No expression was detected in the root, hypocotyl, leaf, petal, anther or stigma confirming its fibre specificity. Northern blotting analysis (Figure 1c) further verified the qRT‐PCR data. GbEXPATR was also expressed at high level in developing fibre cells and showed a similar expression pattern to GbEXPA2 (Figure 1b). We designed the qRT‐PCR primers specific to each gene according to the 3′ UTRs to differentiate expression levels of the two similar genes. To further confirm the results of qRT‐PCR, we checked the digital expression data of GbEXPA2 and GbEXPATR in the G. barbadense 3‐79 whose genome was sequenced in our laboratory (http://cotton.cropdb.org/cotton/index.php). In fibre elongating stage, the RPKMs (specific to 3′ UTR) of GbEXPA2 and GbEXPATR were 80,124 and 28,103, respectively. However, unlike GbEXPA2, GbEXPATR expression was only detected in Gb varieties such as 3‐79, Hai7124, Giza47 and peruvianum, but was not detected in Gh varieties such as TM‐1, Xu142, YZ1, morrilli and marie‐galante (Figure 1b,d), thus indicated its specificity to Gb species.

Consistent with the previous studies (Guo et al., 2011; Zenoni et al., 2011), green fluorescent protein (GFP)‐tagged GbEXPA2 and GbEXPATR were shown to be distributed in both protoplasts and cell wall, when transformed in Arabidopsis (Figure S2). The sequence and subcellular localization suggested that GbEXPATR may be a variant form of α‐expansin with at least some functional attributes in common with the full‐length form. To test this hypothesis, GbEXPATR and GbEXPA2 were overexpressed in Arabidopsis. Previous work has shown that ectopic expression of functional α‐expansins such as soybean GmEXP1 and GmEXPB2 in tobacco and Arabidopsis increased root length (Guo et al., 2011; Lee et al., 2003). In addition, mutation of OsEXPA17 and overexpression of OsEXPA8 in rice can decrease and improve root hair elongation, respectively (Ma et al., 2013; Yu et al., 2011). If GbEXPATR is an authentic α‐expansin, it should increase Arabidopsis root and root hair length similar to GbEXPA2. Seven and nine independent homozygous transgenic lines with 35S::GbEXPA2 and 35S::GbEXPATR, respectively, were obtained (Figure S3a). Two overexpression lines with higher expression levels and one line with lower expression for each vector were used for further analysis. The seedlings of transgenic lines A2‐2, A2‐9, ATR‐8 and ATR‐10 with high expression levels exhibited increased primary root length (T₄, T₅ and T₆ generations) and root hair length (T₆ generation) compared with wild‐type seedlings or lower expressing lines A2‐13 and ATR‐12 (Figures S3b–d and S4). These results indicated that GbEXPATR had a function similar to GbEXPA2 at least in enhancing Arabidopsis primary root and root hair elongation.

GbEXPATR and GbEXPA2 are homoeologous genes

To better understand the evolution of GbEXPATR and GbEXPA2, the full‐length GhEX1 (AF043284) sequence was used as a query to blast the EST database in TIGR (http://plantta.jcvi.org/index.shtml). A total of 1187 ESTs from four cotton species (G. arboreum (Ga, A₂ genome), G. raimondii (Gr, D₅ genome), Gh and Gb) were obtained and analysed (Figure S5 and Tables S1–S3). Four contigs with close ORF sequence identity to GbEXPA2 (≥98%) were identified from different cotton species, including Gb‐contig1 (identical to GbEXPA2), Gh‐contig29, Ga‐contig4 and Gr‐contig12 (Figure S6a), suggesting that the four α‐expansins may be orthologs. Interestingly, Gh‐contig29 that was specific to the developing fibre of 15 Gh contigs (Figure S5), consisting of 482 ESTs, was analysed in detail, and the ESTs could be divided into two roughly equal groups according to SNP loci (Figures 2a and S6b). These findings indicated that Gh‐contig29 included transcripts from the A_T and D_T genomes, respectively, that were equally expressed. Distinctive SNP loci, similar to those found in the Gh‐contig29, were presented in Ga‐contig4 (A₂) and Gr‐contig12 (D₅) (Figure 2b) and could be used to infer the subgenome of origin of the two Gh‐contig29 subgroups. The SNP loci of Gh‐contig29‐1 were highly similar with those in Ga‐contig4, while those in Gh‐contig29‐2 were highly similar to Gr‐contig12 (Figure 2b), indicating that they are encoded from the A_T and D_T genomes, respectively. Gh‐contig29‐1 and Gh‐contig29‐2 had additional SNP loci compared with the diploid genes (Figure 2a), which allowed them to be assigned to the previously reported homoeologous Gh α‐expansin genes GhExp1 and GhExp2, respectively (Harmer et al., 2002). The SNP loci of GbEXPA2 were closest to those in GhExp2 (Gh‐contig29‐2; Figure 2b), justifying the naming of this full‐length α‐expansin as GbEXPA2. GbEXPATR was closest to GhExp1 (Gh‐contig29‐1; Figure 2b), but appeared to be a truncated form of that gene. Comparisons with the diploid and Gh genes indicated that GbEXPATR and GhExp1 were homoeologs with GbEXPA2 encoded by the D_T genome, and GbEXPATR encoded by the A_T genome of tetraploid Gb. The recent release of the full Gr and Ga genome sequence indicated that the ortholog of GbEXPA2 corresponded to the gene model Gorai011G128500.1 and Cotton_A_38175, respectively (Li et al., 2014; Paterson et al., 2012). Interestingly, the 3′ UTR of GbEXPATR related to the flanking sequences immediately downstream of Gorai011G128500 or Cotton_A_38175, supporting an origin for GbEXPATR as a deletion of about 450 bp from within exon 3 to just before end of the 3′ UTR of the homolog in Gb of GhExp1. As the Gh homologs did not have the same deletion, this must have occurred at or soon after the polyploidization event leading to Gb evolution.

A schematic diagram of the SNP loci differences in fibre‐specific *EXPAs* from different cotton species. (a) A schematic diagram of four EST sequences from Gh‐contig29 showing the different SNP loci, as marked with red boxes. According to SNP loci, the ESTs from Gh‐contig29 could be divided into two roughly equal groups. Yellow stars indicate the positions of two additional SNP loci relative to Ga‐contig4 and Gr‐contig12. (b) SNP loci alignment of fibre‐specific *EXPAs* in four cotton species showed that the SNP loci of Gh‐contig29‐1 and Gb‐contig2 (*GbEXPATR* ) were most similar with those in Ga‐contig4, while those in Gh‐contig29‐2 and Gb‐contig1 (*GbEXPA2*) were most similar to Gr‐contig12. Numbering is from the first base of the coding region. The red letters represent the variant SNP loci.

Manipulation of α‐expansin expression in transgenic cotton

To further characterize functions of GbEXPA2 and GbEXPATR, two overexpression (OE) vectors containing the full‐length ORF of GbEXPA2 or GbEXPATR and two RNA interference (RNAi) vectors, one targeting the 196‐bp common coding sequence (CDS) of the fibre‐specific EXPAs (Figure S1a) and a second the 158‐bp 3′ UTR of GbEXPA2 (Figure S1c), were constructed. Because Gb is very difficult to regenerate via somatic embryogenesis, these four constructs were transformed into G. hirsutum YZ1, a standard cultivar of tetraploid cotton for genetic transformation, by Agrobacterium tumefaciens‐mediated protocols. Transgenic lines with single‐ or two‐copy inserts were identified by Southern blotting and following lines were selected for further analysis: IE6, IE9 and IE11 for 3′ UTR suppressed; IE5 and IE27 for CDS suppressed; OE25, OE29 and OE30 for GbEXPA2 overexpressed; and OE2, OE3 and OE4 for GbEXPATR overexpressed (Figure S7).

The EXPA (both subgenome copies) expression levels from RNAi transgenic lines were significantly decreased in 10 DPA fibres as detected by Northern blotting and qRT‐PCR (Figure S8a). The EXPA expression levels in 10 DPA fibre of 35S::GbEXPA2 overexpression lines did not significantly change, because of the high endogenous EXPA. However, higher levels were found at 15 and 20 DPA compared with wild type and line 138 (an independently transformed 35S::GUS transgenic control line; Figure S8b). The qRT‐PCR analysis of 35S::GbEXPATR lines using GbEXPATR‐specific primers showed very high levels of GbEXPATR transcripts in the 10, 15 and 20 DPA fibres (Figure S8c). Endogenous EXPA expression levels in the 35S::GbEXPATR lines did not show any substantial changes in response to ectopic GbEXPATR expression (Figure S8c). Furthermore, qRT‐PCR and RT‐PCR analysis showed that 35S::GbEXPA2 and 35S::GbEXPATR expression levels were strongly increased compared with controls at 0 DPA in whole ovules and leaves of all the relative overexpression lines examined, respectively (Figure S9a,b).

GbEXPATR overexpression enhances fibre length, fineness and strength, but GbEXPA2 overexpression had little positive effect

The mature fibre phenotypes in transgenic lines were analysed by high volume instrumentation, an instrument used for commercial quality assessment of cotton fibre. For RNAi lines (T₆ generation), the mature fibre lengths were 5.3%–14.8% decreased compared with wild type (Table 1). GbEXPA2 overexpression lines (T₃ generation) did not obviously affect fibre length, although changes in overall EXPA expression relative to controls during fibre elongation were modest. For the GbEXPATR overexpression lines (T₅ generation), the fibre length was 5.9%–7.7% longer than wild‐type fibres. The length improvement of transgenic lines showed similar trends for the other generations tested (Table S4). To further characterize the fibre development and elongation rate in RNAi, 35S::GbEXPA2, 35S::GbEXPATR and control lines, different fibre development stages were investigated (Figures 3 and S10). The fibre lengths of RNAi lines were obviously shorter at each stage from 10 DPA through to maturity compared with the controls (Figure 3a,b). In GbEXPA2 overexpression lines, the fibre lengths were not evidently altered at 10 DPA, 20 DPA and in mature fibre, although they were slightly longer at 15 DPA (Figure 3a,c). The fibre of GbEXPATR overexpression lines began to elongate more rapidly from 15 DPA to 20 DPA, and as a result, the mature fibres were significantly longer than controls (Figure 3a,d).

Table 1.

Fibre quality analysis of field‐grown transgenic lines and controls in 2014

Fibre samples	UHML(mm)	MIC	STR(cN/tex)
3′ UTR‐RNAi (T₆)
IE6	24.13 ± 0.63^e	5.73 ± 0.22^bc	23.40 ± 0.64^d
IE9	26.49 ± 0.72^cd	6.00 ± 0.13^ab	24.35 ± 0.40^cd
IE11	25.77 ± 0.48^d	6.09 ± 0.12^a	24.32 ± 0.76^cd
CDS‐RNAi (T₆)
IE5	25.55 ± 0.50^d	6.17 ± 0.14^a	22.99 ± 0.85^d
IE27	23.81 ± 0.47^e	6.27 ± 0.12^a	23.05 ± 0.48^d
Controls
YZ1	27.96 ± 0.22^b	5.23 ± 0.08^d	25.56 ± 0.37^bc
138	27.50 ± 0.76^bc	5.23 ± 0.07^d	25.52 ± 0.55^bc
GbEXPATR OE (T₅)
OE2	29.60 ± 0.57^a	5.14 ± 0.16^d	26.54 ± 0.66^ab
OE3	29.79 ± 0.71^a	4.78 ± 0.25^e	27.03 ± 0.90^a
OE4	30.11 ± 0.86^a	4.61 ± 0.17^e	27.01 ± 0.62^a
GbEXPA2 OE (T₃)
OE25	27.39 ± 0.57^bc	5.53 ± 0.09^c	25.16 ± 0.96^bc
OE29	27.55 ± 0.56^bc	6.04 ± 0.10^a	25.25 ± 0.35^bc
OE30	26.59 ± 0.15^cd	5.98 ± 0.07^ab	24.43 ± 0.31^cd

Open in a new tab

UHML, upper half mean length (mm); MIC, micronaire; STR, strength (cN/tex) determined by High Volume Instrumentation. The data were assayed for samples of three biological replicates. Values are mean ± SD. In each column, values that are not followed by the same letters are significantly different based on the Tukey's multiple comparison test (P < 0.05).

Fibre length measurements in transgenic lines and controls. (a) Final mature fibre phenotype of transgenic lines. For RNAi lines, mature fibre length were decreased; *GbEXPA2* overexpression (OE) did not obviously affect mature fibre length; however, *GbEXPATR* overexpression enhanced fibre length. (b–d) Temporal changes in fibre length from 10 DPA to maturity in *EXPA* RNAi, *GbEXPA2* overexpression, *GbEXPATR* overexpression and controls lines (YZ1 and 138), respectively (2012). Generally, five to eight bolls were measured for each line in a biological replicate, and two independent measurements were conducted. The values are the mean ± SD. Within each fibre development stage (10 DPA, 15 DPA, 20 DPA and mature), different letters in b–d indicate statistically significant differences at P < 0.05 based on ANOVA (Tukey's multiple comparison test).

Micronaire is commonly used to assess cotton fineness and maturity (Montalvo, 2005), and 3.7–5.0 is regarded as the preferred value range for spinning. A moderately lower micronaire value is an indicator of fibre fineness with adequate maturity and is a desirable fibre property. The RNAi lines had higher micronaire values, which were an increase of 9.6%–19.9% compared with wild type (Table 1). GbEXPA2 overexpression lines showed a slightly higher micronaire value, increasing 5.7%–15.5%. However, the GbEXPATR overexpression lines showed a lower micronaire value, a decrease of 1.7%–11.9% relative to wild type, improving micronaire value to within the preferred commercial range. The fibre strength of the transgenic lines was also investigated. The mature fibre of RNAi lines had a lower strength than the one of wild type, with a 4.7%–10.1% reduction (Table 1). GbEXPA2 overexpression lines did not show obvious change in fibre strength. However, GbEXPATR overexpression lines increased in fibre strength (3.8%–5.8%) compared to wild type.

The full‐length α‐expansin can be functionally complemented by GbEXPATR in fibre elongation

Homozygous lines of the 3′ UTR‐RNAi plants (IE6, IE9 and IE11) were crossed with the highest expressing 35::GbEXPATR homozygous line (OE3; Figure 4a). GbEXPATR did not contain the same UTR present in the full‐length α‐expansins, so should not be silenced by the 3′ UTR construct. RT‐PCR analysis confirmed that GbEXPATR was indeed expressed in F₁ plants (Figure 4b). The short fibre phenotype of EXPA 3′ UTR‐RNAi at 15 DPA and in the mature fibre was completely restored in F₁ lines IOE6, IOE9 and IOE11 (Figure 4c), indicating that GbEXPATR can substitute for the full‐length EXPA in enhancing fibre elongation.

Complementation analysis of fibre length by crossing a *GbEXPATR* overexpression line with different *EXPA* 3′ UTR‐RNAi lines. (a) A schematic diagram of the crossing strategy. The *GbEXPATR* OE line (OE3) was transferred into *EXPA* 3′ UTR‐RNAi lines (IE6, IE9 and IE11) via crossing. (b) RT‐PCR showed that *GbEXPATR* was highly expressed in F₁ plant leaves, confirming the presence of *GbEXPATR* transcripts in the RNAi lines. (c) The short fibre phenotype of *EXPA* 3′ UTR‐RNAi lines were completely restored in F₁ lines. Fibre length analysis at 15 DPA and in mature fibres of F₁ lines (IOE6, IOE9 and IOE11), *EXPA* RNAi lines (IE6, IE9 and IE11) and controls (YZ1 and 138), respectively. Generally, five to eight bolls were measured for each line in a biological replicate, and two independent measurements were conducted. The values are the mean ± SD. Within each fibre development stage (15 DPA and mature), different letters indicate statistically significant differences at P < 0.05 based on ANOVA (Tukey's multiple comparison test).

Ectopic expression of GbEXPATR alters the expression of secondary cell wall‐associated genes

Digital gene expression profiling (Solexa) of 10 DPA fibre from 3′ UTR‐RNAi lines did not detect any meaningful gene expression differences between the RNAi and wild‐type lines at this developmental stage, other than the silenced α‐expansins (Table S5). However, as there were changes of fibre properties in some lines, which are determined by the secondary cell wall, we also examined genes characteristically expressed at later stages during secondary cell wall deposition including two chitinase‐like genes (CTLs), GhCTL1 and GhCTL2 (Zhang et al., 2004), two secondary cell wall‐specific cellulose synthesis genes (CESAs), GhCESA1 and GhCESA2 (Pear et al., 1996), and one thaumatin‐like protein gene, GbTLP1 (Munis et al., 2010). The expression levels of CTLs, CESAs and TLP genes at 15 and 20 DPA were not consistently altered in RNAi lines compared with controls (Figure 5), and there was no correlation between level of EXPA suppression and expression of secondary cell wall genes. In contrast, a significant reduction in the expression levels of all three types of secondary cell wall genes was often observed at 15 DPA in GbEXPATR and GbEXPA2 overexpression lines. Surprisingly, the CTL genes were the most affected of the secondary cell wall‐specific genes in the GbEXPA2, but particularly the GbEXPATR overexpression lines. At 20 DPA, the transcripts of these secondary cell wall genes also showed slightly decreased expression in the GbEXPATR overexpression lines, but not in the GbEXPA2 overexpression lines. In addition, GhSuSy and GhGluc1 transcript levels were also detected in transgenic lines at 20 DPA. GhSuSy showed a relatively higher expression level in RNAi and GbEXPA2 overexpression lines. GhGluc1 expression was significantly decreased in GbEXPATR overexpression lines (Figure 5).

Expression of secondary cell wall‐related genes in transgenic lines and controls. qRT‐PCR analysis of *GhCTL1*, *GhCTL2*, *GhCESA1*, *GhCESA2* and *GhTLP1* expression in 15 and 20 DPA detached fibres, respectively, and qRT‐PCR analysis of *GhSuSy* and *GhGluc1* in 20 DPA fibres. Expression of secondary wall‐related genes in *GbEXPATR* and *GbEXPA2* overexpressed lines was significantly decreased at 15 DPA. Particularly, the *CTL* genes were the most affected of the secondary wall‐specific genes in the *GbEXPATR* overexpression lines. At 20 DPA, the transcripts of these secondary wall genes also showed slightly decreased expression in the *GbEXPATR* overexpression lines, but not in the *GbEXPA2* overexpression lines. All expression levels were normalized to *UBQ7*. The error bars represent the SD of three biological replicates.

To provide further insight into the relationship between EXPA and CTLs, we cloned two CTLs genes, GbCTL1 (DQ912958) and GbCTL2 (DQ912959), in our normalized cDNA library of G. barbadense 3‐79 fibre. qRT‐PCR showed that GbCTL1 and GbCTL2 were preferentially expressed from 18 to 25 DPA in fibre (Figure 6a). We obtained CTL1 and CTL2 RNAi and overexpression transgenic lines (Figure S11a,b). Mature fibre quality analysis showed that CTL1 and CTL2 RNAi lines resulted in a 0%–3.4% increase in fibre length; however, the one CTL2 overexpression line we were able to obtain showed a 5.1% decrease compared with wild type in Wuhan (T₃ generation; Figures 6b and S11c). Furthermore, we found that EXPA expression levels were increased in CTL RNAi lines, but no change was found in the one overexpressor (Figure 6c). A previous study has shown that CTL mutations in Arabidopsis cause a cellulose‐deficient phenotype and are important for cellulose synthesis (Sánchez‐Rodríguez et al., 2012). Consistent with the above findings, GhCESA1 expression had a tendency towards lower levels in RNAi lines in 18 DPA fibres, which facilitated fibre cell elongation (Figure 6c).

Down‐regulation of *CTLs* expressed predominantly in fibre secondary wall stage could increase fibre elongation and *EXPA* expression. (a) *GbCTL1* and *GbCTL2* were expressed predominantly during secondary wall synthesis of fibre development. qRT‐PCR analysis of *GbCTL1* and *GbCTL2* expression in root (R), hypocotyl (Hy), leaf (L), anther (An), ovule (0 DPA) with fibre attached and isolated fibre (3 to 25 DPA). (b) The morphological differences of mature fibre between the wild‐type, *CTL1* and *CTL2* transgenic cotton lines. *CTL1* (IC1‐3, IC1‐17 and IC1‐19) and *CTL2* (IC2‐7, IC2‐11 and IC2‐65) RNAi lines increased fibre length in contrast to the reduction in length observed in the *GbCTL2* overexpression line (OEC2). Scale bar, 1 cm. (c) *EXPA* expression was increased and *GhCESA1* expression was decreased in 18 DPA fibres of *CTLs* RNAi lines. All expression levels were normalized to *UBQ7*. The error bars represent the SD of three biological replicates.

Ectopic expression of GbEXPATR and GbEXPA2 alters secondary cell wall structure and polymer composition

Ectopic expression of both GbEXPATR and GbEXPA2 down‐regulated secondary cell wall‐related gene expression during the early transition stage, which prompted us to evaluate whether secondary cell wall structure, wall thickness and crystalline cellulose content were also altered in the fibres of the two overexpression transgenics. Obvious differences in the secondary cell wall morphology of mature fibre were found between the transgenic lines and controls in cross sections of resin‐embedded fibres (Figure 7a,b). RNAi and GbEXPA2 overexpression lines caused a thicker cell wall, which is consistent with the higher micronaire results from these lines. GbEXPA2 overexpression lines also caused higher crystalline cellulose content measured by acetic nitric digestion (Figure 7c). A thinner secondary cell wall was observed in GbEXPATR overexpression lines as expected from their lower micronaire; however, less change was found in their crystalline cellulose content.

Morphological and compositional analysis of the fibre cell wall in transgenic and control lines. (a) The microscope images of mature fibre cross sections of the control, RNAi and *GbEXPATR* and *GbEXPA2* overexpression lines. RNAi and *GbEXPA2* overexpression lines produced fibres with thicker wall, but *GbEXPATR* overexpression lines resulted in thinner wall compared with controls. Two‐year replicates showed the same attributes. (1–4) RNAi lines of IE6, IE9, IE5 and IE27. (5) Wild‐type control (YZ1). (6) *35S::GUS* control (138). (7–9) *GbEXPATR* overexpression lines of OE2, OE3 and OE4. (10–12) *GbEXPA2* overexpression lines of OE25, OE29 and OE30. Scale bar, 50 μm. (b) Measurement of mature fibre cell wall thickness from transgenic and control lines determined by at least 200 fibres per image and three biological replicates. Different letters indicate statistically significant differences at P < 0.05 based on ANOVA (Tukey's multiple comparison test). The error bars indicate the standard deviation. (c) Crystalline cellulose contents in mature fibre of transgenic lines and controls. *GbEXPA2* overexpression lines caused higher crystalline cellulose content in mature fibre. (d) Cellulose crystalline index for transgenic lines and controls from 20 DPA fibre detected by X‐ray diffraction. *GbEXPATR* overexpression lines resulted in lower cellulose crystallinity compared with controls. The error bars represent the SD of three biological replicates. Crystalline index: CrI. (e) Polysaccharide composition by monosaccharide linkage analysis (mol %) for 20 DPA fibres in transgenic lines and controls. RNAi and *GbEXPA2* overexpression lines increased cellulose levels with decreasing levels of other polysaccharides compared with controls, contrary to *GbEXPATR* overexpression lines. Arabinogalactan: AG.

To investigate whether the EXPA RNAi and overexpression lines contained altered wall polymer composition, monosaccharide linkage analysis (Pettolino et al., 2012) was performed on 20 DPA fibres. Cell wall polysaccharide composition estimation revealed that at 20 DPA, EXPA suppression increased cellulose levels and decreased levels of other polysaccharides (pectin, hemicellulose, callose, etc.) compared with wild type (Figure 7e and Tables S6, S7). The GbEXPA2 overexpression lines also resulted in increased cellulose deposition at 20 DPA with a corresponding decrease of other polysaccharides, similar to RNAi lines. In contrast, the amount of cellulose in GbEXPATR overexpression lines was reduced in parallel with low cellulose crystallinity (Figure 7d). GbEXPATR overexpression lines had more callose, xyloglucan and heteromannan and perhaps more pectin, arabinan and type I arabinogalactan compared to wild type.

Discussion

The role of α‐expansin in fibre cell wall remodelling

In this study, the suppression of fibre‐specific α‐expansin in cotton caused the production of shorter and thicker fibres compared with wild‐type plants (Figures 3b, 7a,b and Table 1) and highlighted the importance of these cell wall‐loosening proteins in fibre cell wall production and growth. A similar phenotype has been reported in sweet potato fibrous roots of IbEXP1‐antisense plants, which produced a thicker and shorter fibrous root (Noh et al., 2013). The detailed examination of fibre development dynamics (Figures 3b and S10a), secondary cell wall‐associated gene expression levels (Figure 5) and cell wall composition (Figure 7e) revealed that the detrimental fibre phenotypes of the cotton RNAi lines resulted primarily from an early reduction (by 10 DPA) in fibre elongation and cellulose content increase in secondary cell wall synthesis stage. Previous work has shown that in the secondary cell wall stage of fibre development, sucrose synthase plays an important role in cellulose synthesis and ectopic expression of GhSuSy can increase cellulose content in hybrid poplar (Coleman et al., 2009; Ruan et al., 1997). So, GhSuSy showed a relatively higher expression level at 20 DPA in RNAi lines, indicating the possible availability of more substrates for cellulose synthases (Figure 5).

GbEXPA2 overexpression increased Arabidopsis primary root and root hair length (Figures S3b–d, S4) demonstrating that it was a functional α‐expansin, but in cotton GbEXPA2, overexpression was not very effective and only transiently increased fibre length at the end of the rapid primary cell wall elongation stage (15 DPA; Figures 3c and S10b). The enhanced total fibre α‐expansin expression achieved in cotton fibre was modest (approximately double in the best lines) even from the strong 35S promoter, suggesting that there might be some feedback regulation on these genes. The higher total α‐expansin expression levels throughout fibre development did, however, have some impacts on the fibre cell wall beyond the elongation period and resulted in a slightly thicker fibre and higher micronaire values at maturity, without impacting on final fibre length. In GbEXPA2 overexpression lines at 15 DPA, secondary cell wall‐related gene expressions were lower than normal expression of wide type; however, at 20 DPA, the expressing of these genes had normal levels, which was not consistent to increased total cellulose content and lower levels of other matrix polysaccharides. The cause of the altered secondary cell wall in GbEXPA2 overexpression might be biochemical or biophysical rather than just transcriptional. Reducing the velocities of CESA complexes can result in reduced cellulose production (Sánchez‐Rodríguez et al., 2012), so higher velocities might give the opposite effect. The higher cell wall‐loosening activities in GbEXPA2 overexpression lines may sufficiently alter the noncovalent cross‐linking between polysaccharides, presumably through an uncharacterized mechanism to allow more fluid movement of CESA complexes leading to higher final cellulose contents. Similar to our results, two α‐expansin genes (ClEXPA1 and ClEXPA2) from Chinese fir have been expressed in transgenic tobacco where they caused a cell‐type‐specific increase in cellulose deposition during secondary cell wall formation resulting in an increase in stem diameter (Wang et al., 2011). There is growing evidence that α‐expansins are not only key cell wall‐loosening agents in plant elongation but also participate in secondary cell wall deposition by regulating cellulose synthesis and assembly by mechanisms yet to be elucidated. Overexpression of GhRDL1 alone was shown to increase mature fibre length; however, co‐expression with GhEXPA1 did not affect mature fibre length (Xu et al., 2013a). Unfortunately, the authors did not show the results of the overexpression of GhEXPA1 alone in cotton. According to our results, we can deduce that GhEXPA1 overexpression neutralized GhRDL1 promotion of fibre elongation; therefore, the co‐expression of GhRDL1 and GhEXPA1 had no obvious effect on fibre.

Overexpression of the truncated GbEXPATR had some similar impacts to overexpression of GbEXPA2, such as in delaying the onset of expression of secondary cell wall‐related genes at 15 DPA (Figure 5), but had a very different outcome on fibre development as the deposition of the secondary cell wall never really recovered. GbEXPATR overexpression lines by 20 DPA had significant reduction in cellulose content and cellulose crystallinity index with accompanying higher callose, hemicellulose and pectin contents (Figure 7d,e) leading to thinner fibre walls at maturity. The later entry into secondary cell wall formation allowed the primary cell wall stage to be longer for fibre cell extension and produced longer fibres.

The expression of the 1,3 β‐glucanase, GhGluc1, involved in callose degradation, and plasmodesmata opening has been shown to correlate with the transition from fibre elongation to secondary cell wall deposition (Ruan et al., 2004). In GbEXPATR overexpression lines, GhGluc1 expression was significantly decreased (Figure 5), suggesting that those fibres would have maintained high callose content and closure of plasmodesmata for longer to maintain turgor for further fibre elongation. Furthermore, different from other secondary cell wall polysaccharides, callose may play an important role as a type of lubricating agent during the fibre cell wall transition stage (DeLanghe, 1986), and Gb has been reported to have higher callose production during secondary cell wall synthesis compared with Gh (Avci et al., 2013; Li et al., 2013). The effect on callose may be one of the ways that GbEXPATR promoted fibre elongation.

We also found that the down‐regulation of CTL1 and CTL2 increased EXPA expression, which promoted fibre length (Figures 6 and S11). It is interesting to note that the high expression of GbEXPATR together with the substantially lower CTL levels compared with overexpression GbEXPA2 lines involved in co‐regulating cell elongation may be one important reason that GbEXPATR overexpression lines produced superior fibre quality compared to GbEXPA2. Taken together, these results showed that GbEXPA2 and GbEXPATR performed similarly in primary cell wall and differently in secondary cell wall during fibre development, presumably as a consequence of their different domain structures and inherent cell wall‐loosening activities.

GbEXPATR may be one player accounting for better fibre quality in G. barbadense

The plant cell wall is composed of a complex polysaccharide matrix and determines many aspects of plant morphogenesis and different physiological processes. Cotton fibre cell wall properties determine qualities, such as length, fineness and strength. Recently, cell wall‐associated studies, including the differences in cellulose, callose and de‐esterified pectin levels in cotton fibre from primary cell wall to secondary cell wall (Avci et al., 2013; Li et al., 2013; Liu et al., 2013) and comparative expression profiling for Gb fibre development (Chaudhary et al., 2008), have all provided novel insights into the differences in Gb and Gh fibre development mechanisms. Gb fibre appears to enter into secondary cell wall synthesis later and has lower cellulose and higher callose content than Gh, resulting in a longer elongation period and hence longer fibres. So the Gb fibre is longer, finer and stronger than Gh and is commercially more valuable in textile manufacture. However, allotetraploid Gb has a poor ecological adaptability and can only be planted in certain limited growing areas. High fibre quality is a major goal of cotton breeding, and introducing genes that confer fibre quality from Gb to Gh is an important breeding strategy. By traditional breeding, introducing genes from Gb to Gh is difficult; however, it will be easy by transgenic biotechnology. For example, targeted expression of iaaM, overexpression of WLIM1a and down‐regulation of PHYA1 gene have successfully improved cotton fibre quality (Abdurakhmonov et al., 2014; Han et al., 2013; Zhang et al., 2011). So we need to clone superior fibre genes from Gb and GbEXPATR should be a good candidate gene.

According to our study, the evolution of the GbEXPATR gene lacking domain 2 may be one of the factors specifying the superior fibre quality of Gb. Some evidences support this conclusion. First, GbEXPATR is specifically and highly expressed during the fibre elongation and transition phases (Figure 1b), suggesting that GbEXPATR is involved in cell enlargement during fibre development. Second, amongst the Gossypium species examined, GbEXPATR occurs only in the Gb fibre and is not present in Gh or either of the diploid progenitors of the tetraploid cottons (Figure 1b,d and Tables S2, S3), all of which have poorer fibre quality. Because Gb and Gh underwent completely independent evolution and domestication (Wendel et al., 2009), the spontaneous deletion that gave rise to GbEXPATR on the A_T genome would be specific to that lineage and might have been maintained under continued selection for better fibre quality during domestication. Third, compared with GbEXPA2, overexpression of GbEXPATR results in longer, finer and stronger fibres in Gh plants, which are similar characteristics to Gb fibres (Table 1). Fourth, overexpression of GbEXPATR decreased the expression of certain secondary cell wall‐associated genes to retard secondary cell wall formation, decreased the cellulose content and increased the callose content (Figures 5 and 7e), consistent to all molecular events which are observed during Gb fibre development (Avci et al., 2013; Li et al., 2013). Introducing the Gb fibre‐specific truncated form of EXPA into Gh should, therefore, provide an effective approach for understanding the mechanism by which this gene altered fibre development and hence improves fibre quality.

In summary, GbEXPATR is a truncated member of the α‐expansin superfamily that has utility in the manipulation of cell wall composition and hence cell wall physical properties. The two homoeologous GbEXPA2 and GbEXPATR genes can play similar in primary cell wall and distinct in secondary cell wall roles in fibre development to affect fibre quality. Therefore, our results provide valuable insights into fibre development and the mechanisms whereby α‐expansins influence cell elongation and particularly secondary cell wall production in plants. GbEXPATR is a novel genetic resource for improving Gh fibre quality through biotechnology.

Experimental procedures

General materials and methods are presented here. Details are provided in supporting experimental procedures (Methods S1–S11).

Plants materials and growth conditions

For analysis of GbEXPA2 and GbEXPATR expression patterns, four Gb varieties and five Gh varieties were used and grown in the experiment field under standard farming conditions. Gossypium hirsutum YZ1 was used for transforming (more details are listed in supporting experimental procedures).

Polysaccharide composition and cellulose crystallinity of cotton cell walls

Polysaccharide composition was conducted according to Pettolino et al. (2012), and cellulose crystallinity was performed as described by Li et al. (2013) (more details are provided in supporting experimental procedures).

Supporting information

Figure S1 The nucleotide sequences, amino acid sequences and intron sequences of GbEXPA2 and GbEXPATR were highly similar in their overlapping regions, but different in their 3′ untranslated regions.

Figure S2 Localization of GbEXPA2 and GbEXPATR through stable expressions of GbEXPA2‐G3GFP and GbEXPATR‐G3GFP fusion proteins.

Figure S3 GbEXPA2 and GbEXPATR over‐expression in Arabidopsis increased primary root length.

Figure S4 Root hairs on the primary roots of GbEXPA2 and GbEXPATR over‐expression lines showed an increase in length compared with that of controls.

Figure S5 The 15 α‐expansion contigs from Gh were expressed in different cotton tissues and one of them (Gh‐contig29) was expressed specially in the developing fiber cell.

Figure S6 Nucleotide sequence identity and SNP loci from different cotton fiber α‐expansins.

PBI-14-951-s003.pdf^{(9.2MB, pdf)}

Figure S7 Transgenic lines with single or two‐copy inserts were identified by Southern blotting.

Figure S8 Expression analysis from different transgenic cotton lines and controls in cotton fiber.

Figure S9 GbEXPA2 and GbEXPATR expression levels were strongly increased compared with controls at 0 DPA ovules and leaves.

Figure S10 Temporal changes in fiber length from 15 DPA to maturity in EXPA RNAi, GbEXPA2 over‐expression, GbEXPATR over‐expression and controls lines (YZ1 and 138), respectively (2013).

Figure S11 Fiber properties of CTL1 and CTL2 transformants.

Figure S12 A schematic diagram of the pCAMBIA 2301 ^m vector used for cotton transformation.

PBI-14-951-s004.pdf^{(5.5MB, pdf)}

Table S1 Accession numbers of cotton α‐expansin ESTs from Gh, Gb, Ga and Gr available from TIGR.

Table S2 The sequences of 15 cotton α‐expansins contigs assembled from EST sequences from Gh available from TIGR.

Table S3 The sequences of 4 cotton α‐expansins contigs assembled from EST sequences from Gb, Ga and Gr available from TIGR.

Table S4 Fiber quality analysis of the transgenic cotton lines of different generations grown in Wuhan, Hubei Province.

Table S5 Gene IDs and annotations of differentially expressed genes between two RNAi lines (IE6 and IE9) and two wide type lines (YZ1‐1 and YZ1‐2) in 10 DPA fiber.

Table S6 Monosaccharide linkage composition of 20 DPA fiber cell walls preparation and possible polysaccharide assignments in RNAi, GbEXPA2 and GbEXPATR over‐expression lines.

Table S7 Polysaccharide composition based on linkage analysis in Table S6.

Table S8 Primers used for quantitative real‐time PCR and for gene cloning and vector construction.

PBI-14-951-s002.pdf^{(297.3KB, pdf)}

Methods S1 Plant materials and growth conditions.

Methods S2 Sequences cloning and analysis.

Methods S3 Expression analysis.

Methods S4 Southern and Northern blotting analysis.

Methods S5 Vector construction and transformation.

Methods S6 Subcellular localization of GbEXPA2 and GbEXPATR.

Methods S7 Measurement of Arabidopsis primary root and root hair length.

Methods S8 Immature fiber length measurement and mature fiber quality analysis.

Methods S9 Microscopic observation of cell wall thickness.

Methods S10 X‐Ray diffraction analysis.

Methods S11 Polysaccharide composition of cotton cell walls.

PBI-14-951-s001.pdf^{(180.7KB, pdf)}

Acknowledgements

We are grateful to Dina Yulia (CSIRO, Plant Industry, Australia) for cell wall composition analysis. We thank Dr. Deyi Xie (Henan Academy of Agricultural Sciences, China) for providing field assistance. We also thank Elizabeth Dennis for her critical comments on the manuscript (CSIRO, Plant industry, Australia). This project is supported by National Natural Science Foundation of China (30871560), the Programme of Introducing Talents of Discipline to Universities in China (the 111 Project no. B14032), and the Fundamental Research Funds for the Central Universities (2014PY004).

References

Abdurakhmonov, I.Y. , Buriev, Z.T. , Saha, S. , Jenkins, J.N. , Abdukarimov, A. and Pepper, A.E. (2014) Phytochrome RNAi enhances major fibre quality and agronomic traits of the cotton Gossypium hirsutum L. Nat. Commun. 5, 3062. [DOI] [PubMed] [Google Scholar]
Avci, U. , Pattathil, S. , Singh, B. , Brown, V.L. , Hahn, M.G. and Haigler, C.H. (2013) Cotton fiber cell walls of Gossypium hirsutum and Gossypium barbadense have differences related to loosely‐bound xyloglucan. PLoS ONE, 8, e56315. [DOI] [PMC free article] [PubMed] [Google Scholar]
Brummell, D. , Harpster, M. , Civello, P. , Palys, J. , Bennett, A. and Dunsmuir, P. (1999) Modification of expansin protein abundance in tomato fruit alters softening and cell wall polymer metabolism during ripening. Plant Cell, 11, 2203–2216. [DOI] [PMC free article] [PubMed] [Google Scholar]
Chaudhary, B. , Hovav, R. , Rapp, R. , Verma, N. , Udall, J.A. and Wendel, J.F. (2008) Global analysis of gene expression in cotton fibers from wild and domesticated Gossypium barbadense . Evol. Dev. 10, 567–582. [DOI] [PubMed] [Google Scholar]
Choi, D. , Lee, Y. , Cho, H. and Kende, H. (2003) Regulation of expansin gene expression affects growth and development in transgenic rice plants. Plant Cell, 15, 1386–1398. [DOI] [PMC free article] [PubMed] [Google Scholar]
Coleman, H.D. , Yan, J. and Mansfield, S.D. (2009) Sucrose synthase affects carbon partitioning to increase cellulose production and altered cell wall ultrastructure. Proc. Natl Acad. Sci. USA, 106, 13118–13123. [DOI] [PMC free article] [PubMed] [Google Scholar]
DeLanghe, E.A.L. (1986) Lint development. In Cotton Physiology( Mauney, J.R. and Stewart, J.M. , eds), pp. 325–349. Memphis, TN: The Cotton Foundation. [Google Scholar]
Georgelis, N. , Yennawar, N.H. and Cosgrove, D.J. (2012) Structural basis for entropy‐driven cellulose binding by a type‐A cellulose‐binding module (CBM) and bacterial expansin. Proc. Natl Acad. Sci. USA, 109, 14830–14835. [DOI] [PMC free article] [PubMed] [Google Scholar]
Goh, H.H. , Sloan, J. , Dorca‐Fornell, C. and Fleming, A. (2012) Inducible repression of multiple expansin genes leads to growth suppression during leaf development. Plant Physiol. 159, 1759–1770. [DOI] [PMC free article] [PubMed] [Google Scholar]
Guo, W. , Zhao, J. , Li, X. , Qin, L. , Yan, X. and Liao, H. (2011) A soybean β‐expansin gene GmEXPB2 intrinsically involved in root system architecture responses to abiotic stresses. Plant J. 66, 541–552. [DOI] [PubMed] [Google Scholar]
Haigler, C.H. , Betancur, L. , Stiff, M.R. and Tuttle, J.R. (2012) Cotton fiber: a powerful single‐cell model for cell wall and cellulose research. Front. Plant Sci. 3, 104. [DOI] [PMC free article] [PubMed] [Google Scholar]
Han, L.B. , Li, Y.B. , Wang, H.Y. , Wu, X.M. , Li, C.L. , Luo, M. , Wu, S.J. , Kong, Z.S. , Pei, Y. , Jiao, G.L. and Xia, G.X. (2013) The dual functions of WLIM1a in cell elongation and secondary wall formation in developing cotton fibers. Plant Cell, 25, 4421–4438. [DOI] [PMC free article] [PubMed] [Google Scholar]
Harmer, S.E. , Orford, S.J. and Timmis, J.N. (2002) Characterisation of six α‐expansin genes in Gossypium hirsutum (upland cotton). Mol. Genet. Genomics, 268, 1–9. [DOI] [PubMed] [Google Scholar]
Hutchison, K. , Singer, P. , McInnis, S. , Diaz‐Sala, C. and Greenwood, M. (1999) Expansins are conserved in conifers and expressed in hypocotyls in response to exogenous auxin. Plant Physiol. 120, 827–832. [DOI] [PMC free article] [PubMed] [Google Scholar]
Kim, H.J. and Triplett, B.A. (2001) Cotton fiber growth in planta and in vitro. Models for plant cell elongation and cell wall biogenesis. Plant Physiol. 127, 1361–1366. [PMC free article] [PubMed] [Google Scholar]
Lee, D. , Ahn, J. , Song, S. , Do Choi, Y. and Lee, J. (2003) Expression of an expansin gene is correlated with root elongation in soybean. Plant Physiol. 131, 985–997. [DOI] [PMC free article] [PubMed] [Google Scholar]
Lee, J. , Burns, T. , Light, G. , Sun, Y. , Fokar, M. , Kasukabe, Y. , Fujisawa, K. , Maekawa, Y. and Allen, R. (2010) Xyloglucan endotransglycosylase/hydrolase genes in cotton and their role in fiber elongation. Planta, 232, 1191–1205. [DOI] [PubMed] [Google Scholar]
Li, A. , Xia, T. , Xu, W. , Chen, T. , Li, X. , Fan, J. , Wang, R. , Feng, S. , Wang, Y. , Wang, B. and Peng, L. (2013) An integrative analysis of four CESA isoforms specific for fiber cellulose production between Gossypium hirsutum and Gossypium barbadense . Planta, 237, 1585–1597. [DOI] [PubMed] [Google Scholar]
Li, F. , Fan, G. , Wang, K. , Sun, F. , Yuan, Y. , Song, G. , Li, Q. , Ma, Z. , Lu, C. , Zou, C. , Chen, W. , Liang, X. , Shang, H. , Liu, W. , Shi, C. , Xiao, G. , Gou, C. , Ye, W. , Xu, X. , Zhang, X. , Wei, H. , Li, Z. , Zhang, G. , Wang, J. , Liu, K. , Kohel, R.J. , Percy, R.G. , Yu, J.Z. , Zhu, Y.X. , Wang, J. and Yu, S. (2014) Genome sequence of the cultivated cotton Gossypium arboreum . Nat. Genet. 46, 567–572. [DOI] [PubMed] [Google Scholar]
Liu, Q. , Talbot, M. and Llewellyn, D.J. (2013) Pectin methylesterase and pectin remodelling differ in the fibre walls of two Gossypium species with very different fibre properties. PLoS ONE, 8, e65131. [DOI] [PMC free article] [PubMed] [Google Scholar]
Lü, P. , Kang, M. , Jiang, X. , Dai, F. , Gao, J. and Zhang, C. (2013) RhEXPA4, a rose expansin gene, modulates leaf growth and confers drought and salt tolerance to Arabidopsis . Planta, 237, 1547–1559. [DOI] [PubMed] [Google Scholar]
Ma, N. , Wang, Y. , Qiu, S. , Kang, Z. , Che, S. , Wang, G. and Huang, J. (2013) Overexpression of OsEXPA8, a root‐specific gene, improves rice growth and root system architecture by facilitating cell extension. PLoS ONE, 8, e75997. [DOI] [PMC free article] [PubMed] [Google Scholar]
McQueen‐Mason, S. , Durachko, D.M. and Cosgrove, D.J. (1992) Two endogenous proteins that induce cell wall extension in plants. Plant Cell, 4, 1425–1433. [DOI] [PMC free article] [PubMed] [Google Scholar]
Montalvo, J.G. (2005) Relationships between micronaire, fineness, and maturity. Part I. Fundamentals. J. Cotton Sci. 9, 81–88. [Google Scholar]
Munis, M.F.H. , Tu, L. , Deng, F. , Tan, J. , Xu, L. , Xu, S. , Long, L. and Zhang, X. (2010) A thaumatin‐like protein gene involved in cotton fiber secondary cell wall development enhances resistance against Verticillium dahliae and other stresses in transgenic tobacco. Biochem. Biophys. Res. Commun. 393, 38–44. [DOI] [PubMed] [Google Scholar]
Noh, S.A. , Lee, H.S. , Kim, Y.S. , Paek, K.H. , Shin, J.S. and Bae, J.M. (2013) Downregulation of the IbEXP1 gene enhanced storage root development in sweetpotato. J. Exp. Bot. 64, 129–142. [DOI] [PMC free article] [PubMed] [Google Scholar]
Paterson, A.H. , Wendel, J.F. , Gundlach, H. , Guo, H. , Jenkins, J. , Jin, D. , Llewellyn, D. , Showmaker, K.C. , Shu, S. , Udall, J. , Yoo, M.J. , Byers, R. , Chen, W. , Doron‐Faigenboim, A. , Duke, M.V. , Gong, L. , Grimwood, J. , Grover, C. , Grupp, K. , Hu, G. , Lee, T.H. , Li, J. , Lin, L. , Liu, T. , Marler, B.S. , Page, J.T. , Roberts, A.W. , Romanel, E. , Sanders, W.S. , Szadkowski, E. , Tan, X. , Tang, H. , Xu, C. , Wang, J. , Wang, Z. , Zhang, D. , Zhang, L. , Ashrafi, H. , Bedon, F. , Bowers, J.E. , Brubaker, C.L. , Chee, P.W. , Das, S. , Gingle, A.R. , Haigler, C.H. , Harker, D. , Hoffmann, L.V. , Hovav, R. , Jones, D.C. , Lemke, C. , Mansoor, S. , ur Rahman, M. , Rainville, L.N. , Rambani, A. , Reddy, U.K. , Rong, J.K. , Saranga, Y. , Scheffler, B.E. , Scheffler, J.A. , Stelly, D.M. , Triplett, B.A. , Van Deynze, A. , Vaslin, M.F. , Waghmare, V.N. , Walford, S.A. , Wright, R.J. , Zaki, E.A. , Zhang, T. , Dennis, E.S. , Mayer, K.F. , Peterson, D.G. , Rokhsar, D.S. , Wang, X. and Schmutz, J. (2012) Repeated polyploidization of Gossypium genomes and the evolution of spinnable cotton fibres. Nature, 492, 423–427. [DOI] [PubMed] [Google Scholar]
Pear, J.R. , Kawagoe, Y. , Schreckengost, W.E. , Delmer, D.P. and Stalker, D.M. (1996) Higher plants contain homologs of the bacterial celA genes encoding the catalytic subunit of cellulose synthase. Proc. Natl Acad. Sci. USA, 93, 12637–12642. [DOI] [PMC free article] [PubMed] [Google Scholar]
Pettolino, F.A. , Walsh, C. , Fincher, G.B. and Bacic, A. (2012) Determining the polysaccharide composition of plant cell walls. Nat. Protoc. 7, 1590–1607. [DOI] [PubMed] [Google Scholar]
Qin, Y.M. and Zhu, Y.X. (2011) How cotton fibers elongate: a tale of linear cell‐growth mode. Curr. Opin. Plant Biol. 14, 106–111. [DOI] [PubMed] [Google Scholar]
Ruan, Y.L. , Chourey, P.S. , Delmer, D.P. and Perez‐Grau, L. (1997) The differential expression of sucrose synthase in relation to diverse patterns of carbon partitioning in developing cotton seed. Plant Physiol. 115, 375–385. [DOI] [PMC free article] [PubMed] [Google Scholar]
Ruan, Y.L. , Xu, S.M. , White, R. and Furbank, R.T. (2004) Genotypic and developmental evidence for the role of plasmodesmatal regulation in cotton fiber elongation mediated by callose turnover. Plant Physiol. 136, 4104–4113. [DOI] [PMC free article] [PubMed] [Google Scholar]
Sampedro, J. and Cosgrove, D. (2005) The expansin superfamily. Genome Biol. 6, 242. [DOI] [PMC free article] [PubMed] [Google Scholar]
Sánchez‐Rodríguez, C. , Bauer, S. , Hématy, K. , Saxe, F. , Ibáñez, A.B. , Vodermaier, V. , Konlechner, C. , Sampathkumar, A. , Rüggeberg, M. , Aichinger, E. , Neumetzler, L. , Burgert, I. , Somerville, C. , Hauser, M.‐T. and Persson, S. (2012) CHITINASE‐LIKE1/POM‐POM1 and its homolog CTL2 are glucan‐interacting proteins important for cellulose biosynthesis in Arabidopsis . Plant Cell, 24, 589–607. [DOI] [PMC free article] [PubMed] [Google Scholar]
Son, S.H. , Chang, S.C. , Park, C.H. and Kim, S.K. (2012) Ethylene negatively regulates EXPA5 expression in Arabidopsis thaliana . Physiol. Plant. 144, 254–262. [DOI] [PubMed] [Google Scholar]
Tu, L.L. , Zhang, X.L. , Liang, S.G. , Liu, D.Q. , Zhu, L.F. , Zeng, F.C. , Nie, Y.C. , Guo, X.P. , Deng, F.L. , Tan, J.F. and Xu, L. (2007) Genes expression analyses of sea‐island cotton (Gossypium barbadense L.) during fiber development. Plant Cell Rep. 26, 1309–1320. [DOI] [PubMed] [Google Scholar]
Wang, H. , Guo, Y. , Lv, F. , Zhu, H. , Wu, S. , Jiang, Y. , Li, F. , Zhou, B. , Guo, W. and Zhang, T. (2010) The essential role of GhPEL gene, encoding a pectate lyase, in cell wall loosening by depolymerization of the de‐esterified pectin during fiber elongation in cotton. Plant Mol. Biol. 72, 397–406. [DOI] [PubMed] [Google Scholar]
Wang, G. , Gao, Y. , Wang, J. , Yang, L. , Song, R. , Li, X. and Shi, J. (2011) Overexpression of two cambium‐abundant Chinese fir (Cunninghamia lanceolata) α‐expansin genes ClEXPA1 and ClEXPA2 affect growth and development in transgenic tobacco and increase the amount of cellulose in stem cell walls. Plant Biotechnol. J. 9, 486–502. [DOI] [PubMed] [Google Scholar]
Wendel, J. , Brubaker, C. , Alvarez, I. , Cronn, R. and Stewart, J. (2009) Evolution and natural history of the cotton genus. In Genetics and Genomics of Cotton ( Paterson, A. , ed), pp. 3–22. New York, US: Springer. [Google Scholar]
Xu, B. , Gou, J. , Li, F. , Shangguan, X. , Zhao, B. , Yang, C. , Wang, L. , Yuan, S. , Liu, C. and Chen, X. (2013a) A cotton BURP domain protein interacts with α‐expansin and their co‐expression promotes plant growth and fruit production. Mol. Plant, 6, 945–958. [DOI] [PubMed] [Google Scholar]
Xu, W. , Zhang, D. , Wu, Y. , Qin, L. , Huang, G. , Li, J. , Li, L. and Li, X. (2013b) Cotton PRP5 gene encoding a proline‐rich protein is involved in fiber development. Plant Mol. Biol. 82, 353–365. [DOI] [PubMed] [Google Scholar]
Yu, Z.M. , Kang, B. , He, X.W. , Lv, S.L. , Bai, Y.H. , Ding, W.N. , Chen, M. , Cho, H.T. and Wu, P. (2011) Root hair‐specific expansins modulate root hair elongation in rice. Plant J. 66, 725–734. [DOI] [PubMed] [Google Scholar]
Zenoni, S. , Reale, L. , Tornielli, G. , Lanfaloni, L. , Porceddu, A. , Ferrarini, A. , Moretti, C. , Zamboni, A. , Speghini, A. and Ferranti, F. (2004) Downregulation of the Petunia hybrida α‐expansin gene PhEXP1 reduces the amount of crystalline cellulose in cell walls and leads to phenotypic changes in petal limbs. Plant Cell, 16, 295–308. [DOI] [PMC free article] [PubMed] [Google Scholar]
Zenoni, S. , Fasoli, M. , Tornielli, G.B. , Dal Santo, S. , Sanson, A. , de Groot, P. , Sordo, S. , Citterio, S. , Monti, F. and Pezzotti, M. (2011) Overexpression of PhEXPA1 increases cell size, modifies cell wall polymer composition and affects the timing of axillary meristem development in Petunia hybrida . New Phytol. 191, 662–677. [DOI] [PubMed] [Google Scholar]
Zhang, D. , Hrmova, M. , Wan, C.H. , Wu, C. , Balzen, J. , Cai, W. , Wang, J. , Densmore, L.D. , Fincher, G.B. , Zhang, H. and Haigler, C.H. (2004) Members of a new group of chitinase‐like genes are expressed preferentially in cotton cells with secondary walls. Plant Mol. Biol. 54, 353–372. [DOI] [PubMed] [Google Scholar]
Zhang, M. , Zheng, X. , Song, S. , Zeng, Q. , Hou, L. , Li, D. , Zhao, J. , Wei, Y. , Li, X. , Luo, M. , Xiao, Y. , Luo, X. , Zhang, J. , Xiang, C. and Pei, Y. (2011) Spatiotemporal manipulation of auxin biosynthesis in cotton ovule epidermal cells enhances fiber yield and quality. Nat. Biotechnol. 29, 453–458. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials