New clues concerning pigment biosynthesis in green colored fiber provided by proteomics-based analysis

LI Yan-jun, SUN Shi-chao, ZHANG Xin-yu, WANG Xiang-fei, LIU Yong-chang, XUE Fei, SUN Jie

1 Key Laboratory of Oasis Eco-agriculture, College of Agriculture, Shihezi University, Shihezi 832003, P.R.China

2 College of Pharmacy, Shihezi University, Shihezi 832003, P.R.China

RESEARCH ARTICLE

New clues concerning pigment biosynthesis in green colored fiber provided by proteomics-based analysis

LI Yan-jun1, SUN Shi-chao1, ZHANG Xin-yu1, WANG Xiang-fei2, LIU Yong-chang1, XUE Fei1, SUN Jie1

1 Key Laboratory of Oasis Eco-agriculture, College of Agriculture, Shihezi University, Shihezi 832003, P.R.China

2 College of Pharmacy, Shihezi University, Shihezi 832003, P.R.China

To separate the proteins related to pigment synthesis in green colored fiber (GCF), we performed a comparative proteomic analysis to identify the differentially expressed proteins between green cotton fiber and a white near-isogenic line (NIL).One differential spot identi fied as phenylocumaran benzylic ether redutase-like protein (PCBER) was expressed only in GCF, but was not found in white colored fiber (WCF) at any time points. Since PCBER was a key enzyme in lignans biosynthesis, total lignans were extracted from GCF and WCF and their content was determined by using a chromotropic acid spectrophotometric method. The results showed that total lignans content in GCF was signi ficantly higher than that in WCF. The qPCR analysis for two PLR genes associated with lignans biosynthesis showed that the expression level of two genes was much higher in GCF than that in WCF at 24 and 27 days post anthesis (DPA), which may be responsible for the higher lignans content in GCF. Our study suggested that PCBER and lignans may be responsible for the color difference between GCF and WCF. Additionally, p-dimethylaminocinnamaldehyde (DMACA) staining demonstrated that the pigment in GCF was not proanthocyanidins, and was different from that in brown colored fiber (BCF). This study provided new clues for uncovering the molecular mechanisms related to pigment biosynthesis in GCF.

green colored cotton, proteomics, upland cotton, pigment biosynthesis, phenylocumaran benzylic ether redutaselike protein

1. Introduction

Naturally colored cotton is a type of cotton with brown or green colored fiber. It is unique and exceptionally different from common white cotton, as the textile requires no dyeing process. Therefore, the use of naturally colored cotton can eliminate the cost of chemical dyeing, including dye water disposal and water and energy consumption, and can protect the population from harmful pollution generated by chemical dyeing (Yatsu et al. 1983; Dutt et al. 2004; Xiao et al. 2007).

Great interest has been focused on cultivating naturally colored cotton, but signi ficant advances have not been made due to the limitations of this cotton. Naturally colored cotton yields too little lints, and the fiber is short and weak,and not suitable for machine spinning (Murthy 2001).Additionally, naturally colored cotton fibers can only be found in brown and green colors, which is an insuf ficient number of colors for clothing versatility. Dull colors, poor quality, and low yield issues have become the bottleneck for the development of colored cotton for the textile industry(Qiu 2004). In recent years, studies have greatly improved the fiber qualities and yields through traditional breeding methods, including continuous directional selection, distant hybridization, and composite hybridization (Murthy 2001; Qiu 2004). It is dif ficult to alter the fiber colors using traditional breeding methods due to a lack of genetic resources for the generation of color. With the rapid development of modern biotechnology, genetic engineering has made it possible to create new colored cotton varieties, although the molecular mechanism of pigment biosynthesis in colored cotton fiber is still unclear. However, this uncertainty has resulted in slow progress in the improvement of fiber color through transgenic technology. Therefore, isolation and identi fication of genes related to pigment synthesis and clari fication of the molecular mechanisms of pigment biosynthesis in colored cotton fiber will provide essential theoretical knowledge and applied value for the improvement of the current colored cotton varieties by using genetic engineering methods.

Proteomic analysis is a powerful tool for identifying proteins on a large scale. Comparative proteomic analyses have been performed on common white fiber and brown fiber to identify proteins related to pigment biosynthesis (Li et al.2013b). We used a proteomics-based analysis to generate a snapshot of the proteome in green colored fiber (GCF),but found no additional proteins associated with pigment biosynthesis (Li et al. 2013a). To separate the proteins associated with pigment synthesis in the green colored fiber,we performed two-dimensional electrophoresis (2-DE) and matrix assisted laser desorption ionization time-of- flight/time-of- flight (MALDI-TOF/TOF) mass spectrometry (MS)analyses to compare the protein pro files between GCF and white colored fiber (WCF) in this study. A total of 25 differential protein spots were identi fied. A spot unique to GCF was found to be a cotton PCBER, which is a crucial enzyme involved in lignans biosynthesis. Therefore, total lignans contents were determined, and real-time quantitative PCR (qPCR) analysis for 2 PLR genes related to lignans synthesis were conducted to study the involvement of lignans in GCF and WCF. This study provided new clues concerning pigment biosynthesis in GCF and found a theoretical basis for changing the color of colored cotton fibers through genetic engineering methods.

2. Materials and methods

2.1. Plant materials and treatments

Two upland cotton genotypes, Xincaimian 7 (GCF) and NIL-7 (WCF) were used in this study. NIL-7 is a nearisogenic line (NIL) of Xincaimian 7 that was obtained from continuous backcross breeding by using Xincaimian 7 as the recurrent parent. Cotton bolls were labeled on the anthesis day. The ages of cotton bolls harvested for RNA extraction were 3, 6, 9, 12, 15, 18, 21, 24, and 27 days post anthesis(DPA). The time-points used for proteomic analysis were 12, 15, and 24 DPA representing the fiber elongation and the secondary wall deposition stage. A part of flowers were picked on the anthesis day. Fiber cells stripped from ovules by sterile forceps and flowers picked were stored at –80°C refrigerator after frozen in liquid nitrogen. Seeds coat was stripped from Xincaimian 7 and NIL-7. The kernels were sterilizated by 0.1% HgCl2for 10 min, washed three times with sterile water and then sown on Murashige and Skoog(MS) medium. The roots, stems, and leaves were collected when the seedlings grow true leaves, and stored at –80°C after immediately frozen in liquid nitrogen.

2.2. Protein 2-DE and MALDI-TOF/TOF analysis

Total protein was extracted from GCF and WCF according to Yao et al. (2006). The protein was dissolved in protein solution containing 7 mol L–1urea, 2 mol L–1thiourea, 4%CHAPS, 40 mmol L–1dithiothreitol (DTT), and the protein concentration was determined with the 2-D Quant Kit (GE Healthcare, USA). A total of 1 mg of the sample protein was loaded onto non-linear 24 cm immobilized pH gradient(IPG) strips (pH 4–7) with 450 μL of isoelectric focusing(IEF) buffer. The IEF, two-phase electrophoresis and gels staining were performed according to our previous report (Li et al. 2013b). The stained gels were scanned for image analysis with the ImageScanner III System (GE Healthcare, USA) and analyzed with ImageMaster 2D Platinum software ver. 5.0 (GE Healthcare). The protein spots were quanti fied using its normalized volume which was calculated as %volume provided by ImageMaster 2D Platinum Software (GE Healthcare). Differential protein spots between WCF and GCF were selected according to protein spot abundance with a minimum change of 1.5-fold and analysis by MALDI-TOF/TOF. The mass spectrometry(MS) and MS/MS mass spectrometry data were matched against the GenBank protein database with Mascot online software (http://www.matrixscience.com).

2.3. RNA extraction and qPCR

The total RNA was extracted from cotton fibers using a modi fied cetyltrimethyl ammonium bromide (CTAB) method according to Jiang and Zhang (2003). First-strand cDNA was synthesized from 1 mg of total RNA using the M-MLV reverse transcriptase (TaKaRa Bio., Japan) according to the manufacturer’s instructions. The cDNA derived from different tissues including roots (R), stems (S), leaves (L),flowers (F), and cotton fibers at different developmental stages were used as template for PCR ampli fication. The speci fic primers for PCBER, PLR1, and PLR2 genes were listed in Table 1. The cotton His3 gene was used as an internal control. The cycling conditions were as follows:1 min at 95°C followed by 45 cycles of 15 s at 95°C, 20 s at 57°C, 30 s at 72°C, which were performed with SYBR Premix Ex Taq (TaKaRa Bio.) on LightCycler?480 qPCR System(Roche, Switzerland). Each data represents the average of three independent experiments, and data analysis was conducted by Microsoft Excel ver. 2003.

2.4. The extraction of lignans and determination of its content

Total lignans were extracted from mature fiber and determined using a chromotropic acid spectrophotometric method with schisantherin A as a standard (Chen et al.2010). In brief, 1 g mature fiber was soaked in 30 mL of 95%ethanol for 12 h at 50°C, and then treated with ultrasonic for 30 min (500 W, 4×104Hz). Then, 1 mL of the supernatant was collected and evaporated at 50°C water bath. The dried extract was dissolved in 0.5 mL of 10% chromotropic acid,and then, 3 mL of concentrated sulfuric acid and 1.5 mL of distilled water were added into the solution. After 30 min incubation at boiling water bath, the reaction mixture was set to 5 mL with distilled water, and the absorbance at 570 nm was determined by using a UV-2100 spectrophotometer(Shimadzu, Japan). The analysis was performed with three independent biological replicates. A total of 20 mg of schisantherin A was dissolved in 200 mL of ethanol. Then,0, 0.5, 1.0, 1.5, 2, and 3 mL of this solution were collected and evaporated. The dried schisantherin A was dissolved in 0.5 mL of 10% chromotropic acid and then treated as above mentioned. The absorbance data of the reaction mixture were used to make the standard curve.

2.5. DMACA staining of the cotton fiber

The presence of proanthocyanidins in cotton fiber was monitored by DMACA staining according to Xiao et al. (2007).In brief, 12, 15, and 24 days post anthesis (DPA) and mature GCF were stained for 10 min in 6 N HCl:95% ethanol (1:1)containing 0.1% DMACA and washed three times with distilled water. The fibers were immersed in 6 N HCl:95% ethanol(1:1) without DMACA as a control. WCF and BCF used as controls were stained as the same method.

3. Results

3.1. Analysis of the proteome pro files and differential spots between GCF and WCF

The total protein extracted from Xincaimian 7 fibers and NIL-7 fibers was used for 2-DE. Proteomic analysis was performed with two independent biological replicates and two technical replicates for each sample of GCF and WCF.The 2-DE gels of samples were acquired and compared to identify the differential protein spots using the ImageMaster 2D Platinum software. The differential protein spots were excised from gels, digested with trypsin, and then subjected to MAIDI TOF/TOF MS analysis. The mass spectrometry data obtained with the Mascot online software were matched to the proteins in GenBank. The representative 2-DE gel images of total WCF and GCF proteins at 12, 15, and 24 DPA were showed in Fig. 1. The proteome comparisons revealed fewer differential spots between GCF and WCF than expected. Only 25 differential spots were found at three time points, and were listed in Appendix A. In order to find more differential spots, we also performed the comparative proteomic analysis between GCF and WCF at 18, 21, and 27 DPA, but found no additional differential spots (2-DE images were not shown).

One protein spot (spot no. 19) identi fied as PCBER was interesting because it was expressed only at 15, 18, 21,24, and 27 DPA in GCF, but was not found in WCF at any time points (Fig. 2), suggesting that PCBER may play an important role in pigment synthesis process of green fiber.Furthermore, it was the only protein spot identi fied that showed a signi ficant change in abundance at each time point and was strongly associated with phenylpropanoid pathway which has been reported to be related to pigment biosynthesis (Weisshaar and Jenkins 1998; Tanaka et al.2008). The GenBank accession number of the matched protein was ABN12322, and its corresponding mRNA sequence was EF217320.

Table 1 Sequences of the primers used for qPCR analysis

Fig. 1 Representative 2-DE gel images of total white colored fiber (WCF) and green colored fiber (GCF) proteins at 12, 15, and 24 days post anthesis (DPA). The differential protein spots were marked with arrows and numbers. The spots with lower abundance in GCF were shown in A, C, and E, and the spots with higher abundance were shown in B, D, and F. The area with black box will be shown in closed view.

3.2. The qPCR analysis of PCBER in developing cotton fiber

To determine whether the differences were present at the mRNA levels of PCBER between green and white colored cotton (GCC and WCC) and to investigate the expression patterns of PCBER in different tissues of cotton, qPCR analysis was conducted by using cDNA of different tissues,including roots, stems, leaves, flowers, and fibers from green and white cotton as template. The qPCR results indicated that PCBER was expressed in all of the examined tissues but most highly in cotton fiber. The expression pattern of PCBER showed a similar trend in GCF and WCF at different developmental stages. The expression level of PCBER increased gradually at early fiber development. PCBER is weakly expressed at 12 DPA in GCF but highly expressed at 12 DPA in WCF. Then, PCBER maintained high expression level from 15 to 24 DPA, reached a maximum at 21 DPA,and then declined gradually. Strikingly, the expression of PCBER was much higher in WCF than in GCF at different stages of cotton fiber development (Fig. 3). The transcript pattern and protein expression of PCBER showed in 2-DE results (Fig. 2) con flicted, which may be caused by following reasons such as post-transcriptional modi fication, stability of mRNA, alternatively splicing, post-translational modi fication,and protein half-lives, etc. (Anderson and Scilhamer 1997;Gygi et al. 1999; Pradet-Balade et al. 2001; Greenbaum et al. 2003).

3.3. Determination of total lignans content in GCF and WCF

PCBER is a crucial enzyme involved in the lignans biosynthesis. Because the protein spot identi fied as PCBER was unique to GCF, the end products lignans were extracted from mature fiber and used to detect the difference of its content between GCF and WCF. The results showed that total lignans content in GCF was signi ficantly higher than that in WCF (Fig. 4). Based on the difference of total lignans content between GCF and WCF, it is possible that lignans were responsible for the color difference between GCF and WCF.

3.4. Detection of the expression level of two PLR genes in cotton fibers

Fig. 2 The closed view of protein spot identi fied as phenylocumaran benzylic ether redutase-like protein (PCBER)at different stages of cotton fiber development. The arrowhead pinpoints the protein spot (spot no. 19) identi fied as PCBER which is only present in green colored fiber (GCF), but not in white colored fiber (WCF). DPA, days post anthesis.

The biosynthesis of lignans involves PCBER and pinoresinollariciresinol-lariciresional reductase (PLR) which are phylogenetically related reductases (Dinkova-Kostova et al.1996; Gang et al. 1999). Reaction catalyzed by PCBER and PLR lead to 8–5′ linked neolignans and 8–8′ linked lignans, respectively. Since PLR and PCBER exhibited high homology in gene sequence, two PLR genes were found from cotton genome database by using PCBER as a probe sequence and named as PLR1 (Gh_A08G1368) and PLR2 (Gh_D08G1661). The qPCR analysis for two PLR genes was conducted to further study the involvement of lignans synthesis. The results showed that the expression level of PLR1 was much higher in GCF than that in WCF at 18, 24, and 27 DPA, and the expression level of PLR2 was much higher in GCF than that in WCF at 24 and 27 DPA(Fig. 5), which may be responsible for the higher total lignans content in GCF.

3.5. Detection of proanthocyanidins in cotton fibers

Fig. 3 Relative expression of PCBER in various tissues from white colored cotton (WCC) and green colored cotton(GCC). The 3–27 represent fiber cells at 3, 6, 9, 12, 15, 18,21, 24, and 27 days post anthesis (DPA), respectively. R, S,L, and F represent samples from root, stem, leaf, and flower,respectively. Bars are SD.

Fig. 4 Total lignan content of mature white colored fiber (WCF)and green colored fiber (GCF). Experiments were performed with three independent biological replicates. Different uppercase letters indicate signi ficant differences at the 1% probability level (t-test). Bars are SD.

In our previous studies, a proteomic analysis revealed that protein spots associated with flavonoid pathway showing higher abundance in BCF at all three time points, and these altered spots abundance re flected changes in the end products proanthocyanidins (Li et al. 2013b). However,proteins associated with flavonoid pathway were not found in the differential spots between WCF and GCF, implying that proanthocyanidins may not abundantly exist in GCF.DMACA staining was used to detect the presence of proanthocyanidins in GCF, and WCF and BCF were stained as controls. As shown in Fig. 6, all of the GCF and WCF turned light blue at 12, 15, and 24 DPA, no color change was observed in the mature fiber, but all of the BCF turned dark blue after DMACA staining as reported before (Li et al. 2013b). Our results demonstrated that only a small amount proanthocyanidins in premature GCF and none in mature GCF, suggesting that the pigment in GCF were not proanthocyanidins, and were different from that in BCF.

4. Discussion

4.1. The proteome comparisons between GCF and WCF

The composition of the pigment in GCF is complicated and inconclusive. The pigment identi fication was affected by different extraction methods, which can result in different results. The pigment in GCF extracted by methanol was found to be flavone and flavonols (Zhao and Wang 2005).The chloroform-methanol extract of the green fibers contained one colourless compound identi fied as caffeic acid, and several yellow compounds whose ultraviolet and visible spectrum is very similar to that of caffeic-acid ester(Schmutz et al. 1993).

Fig. 5 Relative expression of two PLR genes in developing white colored fiber (WCF) and green colored fiber (GCF). The 12–27 represent fiber cells at 12, 15, 18, 24, and 27 days post anthesis (DPA), respectively. Bars are SD.

The comparative proteomic analysis in our study can intuitively re flect the differences at protein level between GCF and WCF, find the differential proteins that may play key roles in pigment biosynthesis, and simultaneously avoid the problem existed in the process of pigment extraction.This is the first time that a comparative proteomic analysis revealed the difference between GCF and WCF. We found that a protein spot identi fied as PCBER was expressed only at 15, 18, 21, 24, and 27 DPA in GCF, but was not found in WCF at any time points. The protein spot showed the most signi ficant change in abundance at each time point and was strongly associated with phenylpropanoid pathway which has been reported to be related to pigment biosynthesis (Weisshaar and Jenkins 1998; Tanaka et al.2008). Therefore, PCBER was paid more attention and expected to be related to pigment synthesis in GCF.

4.2. PCBER may be responsible for the color difference between WCF and GCF

Fig. 6 Detection of proanthocyanidins in cotton fibers. –represents cotton fibers treated with the solution of 6 N HCl:95% ethanol (1:1) without p-dimethylaminocinnamaldehyde(DMACA); + represents cotton fibers treated with the solution of 6 N HCl:95% ethanol (1:1) containing 0.1% DMACA. DPA,days post anthesis. WCF, white colored fiber; GCF, green colored fiber; BCF, brown colored fiber.

PCBER was abundantly existed in the xylem of plants(Vander Mijnsbrugge et al. 2000). Recent research found that PCBER transcripts were present in almost all cotton tissues, such as cotyledon, stem, leaf, fiber, and root (Haigler et al. 2005; Shi et al. 2006; Taliercio et al. 2006; Udall et al.2006), suggesting that it is an important ingredient in the development of cotton. A comparative proteomic analysis between fiberless and wild cotton fiber found that PCBER was only abundant in wild cotton fiber (Turley 2008).Because of the rich abundance of PCBER in cotton tissues,the expression difference at protein and RNA levels of PCBER may be responsible for the color difference between WCF and GCF. The PCBER gene will be a novel candidate gene for improving the fiber color of GCF, and further efforts to study its functions were needed.

4.3. Lignans may be responsible for the color generation in GCF

The lignans are a group of chemical compounds derived from phenylalanine found in a variety of plant materials, and have many biological activities such as antioxidant, antivirus, anti-insect, plant toxicity and cytotoxicity, and growth regulation. However, most of the studies focus on their antioxidant roles in basic research of human diseases (Rios et al. 2002; Touillaud et al. 2007), and the studies on their roles in other aspects, especially plant growth regulation were rarely reported (Binns et al. 1987; Lynn et al. 1987).Recent studies reported that there are a large number of secondary metabolites in cotton fiber, which have an effect on the development of cotton fibers (Fan et al. 2009; Han et al. 2013). As one of the secondary metabolites catalyzed by PCBER, lignans may present in cotton fiber and play a role in cotton fiber development. We first demonstrated that lignans were present in cotton fibers, and its content in GCF was signi ficantly higher than that in WCF. Therefore, lignans may be one of the main substances that were responsible for the color generation in GCF.

Lignans, including lignan and neolignan, are phenylpropanoid dimmers synthesized through the coupling of phenylpropanoid monomers such as coniferyl alcohol.Lignan linked through 8–8′ carbons in side chains of the monomers, and neolignan linked through 8–5′ bonds or 5–5′bonds. The biosynthesis of lignans involves phylogenetically related NADPH-dependent reductases: pinoresinollariciresinol reductase (PLR) and phenylcoumaran benzylic ether reductase (PCBER) (Dinkova-Kostova et al. 1996;Gang et al. 1999). The higher expression of PLR genes in GCF than that in WCF may directed the phenylpropanoid pathway toward to 8–8′ linked lignans biosynthesis in GCF.

4.4. The pigment in GCF was not proanthocyanidins

In our previous study, a proteomic analysis revealed that proteins related to flavonoid synthesis showed higher abundance in BCF at 3 important time points (12, 18, and 24 DPA) (Li et al. 2013b). However, flavonoid biosynthesisrelated protein spots showed no changes in abundance in this study (Appendix A), suggesting that the pigment in GCF may not belong to flavonoid. Furthermore, DMACA staining (Fig. 6) in this study suggested that only a small amount of proanthocyanidins in premature GCF and none in mature GCF, suggesting that the pigment in GCF were not proanthocyanidins. Therefore, our results demonstrated that the pigment in GCF were different from that in BCF.

5. Conclusion

This is the first time that a comparative proteomic analysis revealed the difference between GCF and WCF. A protein spot identi fied as PCBER was expressed only in GCF, but was not found in WCF at any time points. Lignans catalyzed by PCBER and PLR were present in cotton fibers, and its content in GCF was signi ficantly higher than that in WCF.Lignans may be one of the main substances that were responsible for the color generation in GCF.

Acknowledgements

The research was supported by the National Natural Science Foundation of China (31460360), the National Key Research and Development Program, China (2016YFD0101900),and the Foundation Research Funds for Advanced Talents of Shihezi University, China (RCZX201316). The authors thank Dr. Yang Chunlin and Dr. Wang Fuxin (Institute of Microbiology, Chinese Academy of Sciences) for their technical assistance with 2-DE and gels analyses. The authors also thank AB Sciex Co. (Shanghai, China) for the kind assistance in MS/MS analysis.

Appendixassociated with this paper can be available on http://www.ChinaAgriSci.com/V2/En/appendix.htm

Anderson L, Scilhamer J. 1997. A comparison of selected mRNA and protein abundances in human liver. Electrophoresin,18, 533–537.

Binns A N, Chen R H, Wood H N, Lynn D G. 1987. Cell division promoting activity of naturally occurring dehydrodiconiferyl glucosides: Do cell wall components control cell division?Proceedings of the National Academy of Sciences of the United States of America,84, 980–984.

Chen Y, Shi Z W, Huang X M. 2010. Ultrasonic-assisted extraction of total lignans and its content analysis in soybeans. Soybean Science,29, 168–173. (in Chinese)

Dinkova-Kostova A T, Gang D R, Davin L B, Bedgar D L,Chu A, Lewis N G. 1996. (+)-Pinoresinol/(+)-lariciresinol reductase from Forsythia intermedia - Protein puri fication,cDNA cloning, heterologous expression and comparison to iso flavone reductase. The Journal of Biological Chemistry,271, 29473–29482.

Dutt Y, Wang X D, Zhu Y G, Li Y Y. 2004. Breeding for high yield and fibre quality in coloured cotton. Plant Breeding,123, 145–151.

Fan L, Shi W J, Hu W R, Hao X Y, Wang D M, Yuan H,Yan H Y. 2009. Molecular and biochemical evidence for phenylpropanoid synthesis and presence of wall-linked phenolics in cotton fibers. Journal of Integrative Plant Biology,51, 626–637.

Gang D R, Kasahara H, Xia Z Q, Mijnsbrugge K V, Bauw G,Boeruan W, Montafu M V, Davin L B, Lewis N G. 1999.Evolution of plant defense mechanisms. Relationships of phenylcoumaran benzylic ether reductases to pinoresinollariciresinol and iso flavone reductases. The Journal of Biological Chemistry,274, 7516–7527.

Greenbaum D, Colangelo C, Williams K, Gerstein M. 2003.Comparing protein abundance and mRNA expression levels on a genomic scale. Genome Biology,4, 117.

Gygi S P, Rochon Y, Franza B R, Aebersold R. 1999. Correlation between protein and mRNA abundance in yeast. Molecular and Cellular Biology,19, 1720–1730.

Haigler C H, Zhang D, Wilkerson C G. 2005. Biotechnological improvement of cotton fibre maturity. Physiologia Plantarum,124, 285–294.

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, Xia G X. 2013. The dual functions of WLIM1a in cell elongation and secondary wall formation in developing cotton fibers. The Plant Cell,25, 4421–4438.Jiang J X, Zhang T Z. 2003. Extraction of total RNA in cotton tissues with CTAB-acidic phenolic method. Cotton Science,15, 166–167. (in Chinese)

Li Y J, Wang F X, Wang Y Q, Liu Y C, Zhang X Y, Sun Y Q, Sun J. 2013a. The identi fication of the proteins in green cotton fiber using a proteomics-based approach. Biotechnology Letters,35, 1519–1523.

Li Y J, Zhang X Y, Wang F X, Yang C L, Liu F, Xia G X, Sun J.2013b. A comparative proteomic analysis provides insights into pigment biosynthesis in brown colored fiber. Journal of Proteomics,78, 374–388.

Lynn D G, Chen R H, Manning K S, Wood H N. 1987. The structural characterization of endogenous factors from Vinca rosea crown gall tumors that promote cell division in tobacco cells. Proceeding of the National Academy of Sciences of the United States of America,84, 615–619.

Murthy M S S. 2001. Never say dye: The story of coloured cotton. Resinance,11, 29–35.

Pradet-Balade B, Boulme F, Beug H, Mullner E W, Garcia-Sanz J A. 2001. Translation control: Bridging the gap between genomics and proteomics. Trends in Biochemical Sciences,26, 225–229.

Qiu X M. 2004. Research progress and prospects on naturallycolored cotton. Cotton Science,16, 249–254. (in Chinese)

Rios J L, Giner R M, Prieto J M. 2002. New findings on the bioactivity of lignans. Studies in Natural Products Chemistry,26, 183–292.

Schmutz A, Jenny T, Amrhein N, Ryser U. 1993. Caffeic acid and glycerol are constituents of suberin layers in green cotton fibres. Planta,189, 453–460.

Shi Y H, Zhu S W, Mao X Z, Feng J X, Qin Y M, Zhang L,Cheng J, Wei L P, Wang Z Y, Zhu Y X. 2006. Transcriptome pro filing, molecular biological, and physiological studies reveal a major role for ethylene in cotton fiber cell elongation. The Plant Cell,18, 651–664.

Taliercio E, Allen R D, Essenberg M, Klueva N, Nguyen H,Patil M A, Payton P, Millena A C, Phillips A L, Pierce M L,Schef fler B, Turley R, Wang J, Zhang D, Schef fler J. 2006.Analysis of ESTs from multiple Gossypium hirsutum tissues and identi fication of SSRs. Genome,49, 306–319.

Tanaka Y, Sasaki N, Ohmiya A. 2008. Biosynthesis of plant pigments: anthocyanins, betalains and carotenoids. The Plant Journal,54, 733–749.

Touillaud M S, Thiebaut A C, Fournier A, Niravong M, Boutron-Ruault M C, Clavel-Chapelon F. 2007. Dietary lignan intake and postmenopausal breast cancer risk by estrogen and progesterone receptor status. Journal of the National Cancer Institute,99, 475–486.

Turley R B. 2008. Expression of a phenylcoumaran benzylic ether reductase-like protein in the ovules of Gossypium hirsutum. Biologia Plantarum,52, 759–762.

Udall J A, Swanson J M, Haller K, Rapp R A, Sparks M E,Hat field J, Yu Y, Wu Y, Dowd C, Arpat A B, Sickler B A,Wilkins T A, Guo J Y, Chen X Y, Schef fler J, Taliercio E,Turley R, McFadden H, Payton P, Klueva N, et al. 2006. A global assembly of cotton ESTs. Genome Research,16,441–450.

Vander Mijnsbrugge K, Beeckman H, De Rycke R, Van Montagu M, Engler G, Boerjan W. 2000. Phenylcoumaran benzylic ether reductase, a prominent poplar xylem protein, is strongly associated with phenylpropanoid biosynthesis in lignifying cells. Planta,211, 502–509.

Weisshaar B, Jenkins G I. 1998. Phenylpropanoid biosynthesis and its regulation. Current Opinion in Plant Biology,1,251–257.

Xiao Y H, Zhang Z S, Yin M H, Luo M, Li X B, Hou L, Pei Y.2007. Cotton flavonoid structural genes related to the pigmentation in brown fibers. Biochemical and Biophysical Research Communications,358, 73–78.

Yao Y, Yang Y W, Liu J Y. 2006. An ef ficient protein for proteomic analysis of developing cotton fiber. Electrophoresis,27,4559–4569.

Yatsu L Y, Espelie K E, Kolattukudy P E. 1983. Ultrastructural and chemical evidence that the cell wall of green cotton fiber is suberized. Plant Physiology,73, 521–524.

Zhao X Q, Wang X D. 2005. Composition analysis of pigment in colored cotton fiber. Acta Agronomica Sinica,31, 456–462.(in Chinese)

27 December, 2016 Accepted 11 May, 2017

LI Yan-jun, E-mail: lyj20022002@sina.com.cn; Correspondence SUN Jie, Tel: +86-993-2057999, E-mail: sunjie@shzu.edu.cn

? 2018 CAAS. Publishing services by Elsevier B.V. All rights reserved.

10.1016/S2095-3119(17)61692-7

Section editor ZHANG Xue-yong

Managing editor WANG Ning