INTRODUCTION

Cotton fibers are seed trichomes of ovule epidermis that have significant economical importance in textile production. Fiber properties such as length, strength, uniformity and yarn spinning efficiency are no longer entirely satisfactory predictors as our society progresses forward. Instead, surface friction forces that resulted directly from the chemical constituents of fiber cells play an important role in modern costume industry 1. As derivatives of very long chain fatty acids (VLCFA, fatty acids >C18), waxes are major components of cotton fiber cuticle 2 and display inverse relationships with micronaire, a measure of fiber linear density and fineness 3. Recently, it has been reported that an ABC transporter from Arabidopsis thialiana mediated transport of lipids out of epidermal cells to the plant surface for biosynthesis of waxy cuticle 4.

Biosynthesis of fatty acids is a primary metabolic event representing a vital aspect for every cell of all organisms to survive 5. Plants synthesize fatty acids within the plastids, and use a discrete and highly conserved group of dissociated enzymes named fatty acid synthase system type 2 (FASII) 6. This system, similarily to the eukaryotic mitochondrial FAS 7, 8, 9, 10, differs from the mammalian cytosolic fatty acid synthase system type 1 (FASI) where all reactions are carried out by a single multifunctional polypeptide 11. Fatty acids synthesized within plastid are exported to endoplasmic reticulum (ER) for elongation, resulting in the formation of VLCFAs that are essential cellular precursors of waxes, seed storage lipids and sphingolipids 5. Biosynthesis of VLCFA in plants occurs on the cytosolic face of the ER membrane and is carried out by a membrane-bound fatty acid elongation system (FAE). The first reaction involves condensation of malonyl-CoA with a long chain acyl substrate producing a 3-ketoacyl-CoA, catalyzed by 3-ketoacyl-CoA synthase (KCS). Subsequent reactions include reduction to 3-hydroxyacyl-CoA catalyzed by 3-ketoacyl-CoA reductase (KCR) and dehydration to an enoyl-CoA by 3-hydroxyacyl-CoA dehydratase, followed by a second reduction catalyzed by trans-2-enoyl-CoA reductase to form the elongated acyl-CoA 12. A number of plant KCS(s) that are all specific to saturated and monounsaturated fatty acids have been identified and partially characterized 2. An orthologue of Saccharomyces cerevisiae trans-2-enoyl-CoA reductase was found in A. thialiana 13. To date, 3-hydroxyacyl-CoA dehydratase has not been identified from any sources.

3-Ketoacyl-CoA reductase encoded by S. cerevisiae YBR159w catalyzes the second important step of VLCFAs biosynthesis 14, 15 and is a member of the short-chain alcohol dehydrogenase reductase family (SDR) 16. Disruption of ScYBR159w (ybr159wΔ) resulted in significantly reduced VLCFA synthesis and accumulation of dihydrosphignosine, phytosphingosine and ceramides 15. Four plant KCRs including maize YBR159 (GL8) 17, 18, A. thaliana YBR159 14 and two isoforms of Brassica napus 19 were found to be orthologues of yeast Ybr159p. The GL8 mutant maize displayed defects in the biosynthesis of the cuticular waxes deposited on the outer epidermis, rendering the plant a characteristic “glossy” phenotype 17. In this study, we identified two cDNAs from developing cotton fiber cells. We characterized these cotton proteins by functional complementation of a yeast mutant, ybr159wΔ. Results obtained from biochemical assays indicated that the cotton gene products possess NADPH dependent 3-ketoacyl-CoA reductase activities that are important for VLCFAs biosynthesis.

MATERIALS AND METHODS

Library screen, sequencing and phylogenetic analyses

Complementary DNA library was constructed by using total RNA extracted from 10 dpa cotton (Gossypium hirsutum L.cv.Xuzhou 142) fiber cells 20. The library was screened using the AtYBR159w 14 coding sequence as a probe at low stringency as previously described 21. Putative full-length GhKCR1 and GhKCR2 cDNAs were obtained by sequencing all colonies that hybridized to the probe from both ends.

QRT-PCR analysis of GhKCR(s) expression patterns

Real time quantitative RT-PCR (QRT-PCR) was carried out using the SYBR green PCR kit (Applied Biosystems) in a DNA Engine Opticon-Continuous Fluorescence Detection System (MJ Research). Following PCR primers were used: GhKCR1, 5′-CACTTTGGGTTCTTTATCACTCTT-3′ (forward, F) and 5′-TTTATCTTCTTCACGCCTTCAT-3′ (reverse, R); GhKCR2, 5′- TATCCTTTCTCACCCGCTCC-3′ (F) and 5′- CACCACTTCCTCCTCCACCTC-3′ (R). Each pair of primers produced a single DNA fragment of appropriate size in any stages of fiber development, different cotton tissue and mutant cotton ovules. All QRT-PCRs were carried out with following parameters: initiation with a 10 min denaturation at 95°C, followed by 42 cycles of amplification with 10 sec denaturation at 94°C, 20 sec annealing according to the melting temperature of each pair of primers, 20-30 sec extension at 72°C, then read the plate for fluorescence data collection at 78-80°C. After a final extension at 72°C for 5-10 min, melting curve was performed from 65°C to 95°C (1 sec hold per 0.2°C increasing) to check the specificity of the amplified product. Samples were analyzed in triplicates using independent RNA samples and were quantified by the comparative cycle threshold method 22.

Functional complementation of yeast ybr159wΔ mutant strain by cotton enzymes

ORFs of GhKCR1 and GhKCR2 were amplified using the following primers: 5′- CACCACAAAATGGAAGCCTGCTTCTTCGATACT-3′ (F1) and 5′-TTCCTTCTTCCTAGAATCTTTCAT-3′ (R1); 5′- CACCACAAAATGCAGCCCACATGGTTGTTAGC-3′ (F2) and 5′-CGCAAGGACCTCTGCTCGCC TTCG-3′ (R2). The sequence “CACC” at the beginning of each forward primer is required for cloning into pENTR TOPO® vector (Invitrogen) and the sequence “ACAAA” is used to enhance gene expression in yeast. The constructed vectors, pENTR TOPO::GhKCR1 and pENTR TOPO::GhKCR2 were verified by sequence analysis. The genes were transferred from pENTR TOPO to yeast expression vector pYTV 23 via LR reaction, using homologous sequences present on both vectors, resulting in pYTV::GhKCR1 and pYTV::GhKCR2. Gal1 promoter was fused at the N-terminus and 3× FLAG, 6× His and 2× IgG were present to facilitate detection of the fusion protein 23.

The S. cerevisiae haploid strain BY4741 ybr159wΔ (MATa; his3Δ1; leu2Δ0; met15Δ0; ura3Δ0; ybr159w::kanMX4, EUROSCARF) was grown on YPD (1% [wt/v] yeast extract, 2% [wt/v] peptone, and 2% [wt/v] D-glucose) supplemented with 300 μg/ml of Geneticine. pYTV::GhKCR(s) were transformed into ybr159wΔ mutant cells. Wild-type strain BY4741 and ybr159wΔ mutant cells were transformed with plain pYTV separately and were used as controls. The transformants were selected on synthetic complete medium lacking uracil (Sc-Ura) plates. The growth rates of wild-type BY4741 transformed with pYTV, ybr159wΔ mutant transformed with pYTV, and those transformed with pYTV::GhKCR(s) for overexpression in ybr159wΔ were examined on synthetic complete medium lacking tryptophan (Sc-Trp) containing 2% [wt/v] galactose and 0.05% [wt/v] D-glucose.

Semi-quantitative RT-PCR analysis

Total RNA was extracted from yeast cells transformed by pYTV::GhKCR(s) using RNeasy Mini Kit (Qiagen) and cDNA was synthesized from 5 μg total RNA using the SUPERSCRIPT™ first-strand synthesis system for RT-PCR (Invitrogen). RT-PCR primers were the same as described in the previous section for ORF cloning. Yeast actin gene, ACT1 (Accession Number NP_116614), was used as internal control in parallel reactions.

Preparation of ER extracts from yeast cells

Yeast cells transformed by pYTV or pYTV::GhKCR(s) were grown to exponential phase (2-4 × 106cells/ml) in Sc-Ura medium containing 2% (wt/v) galactose and 0.05% (wt/v) D-glucose at 30°C. The cells were harvested by centrifuging for 15 min at 1300 g, 4°C. The cell pellets were washed with sterile H2O, and suspended in ice-cold lysis buffer containing 50 mM Tris, pH 7.5, 1 mM EGTA, 0.5 mM β-mercaptoethanol, 1 mM phenylmethlsulfonyl fluoride (PMSF) and 2 μg/ml pepstatin A. The cells were disrupted with glass beads and centrifuged for 15 min at 15000 g, 4°C to remove the cell debris. The supernatants were centrifuged for 90 min at 85,000 g in a Sorval Ti70 rotor at 4°C to generate pellet (P85) and supernatant (S85) fractions. P85 was suspended in lysis buffer and used for elongase assay. The protein concentrations were determined by the method of Lowry et al 24, using bovine serum albumin as the standard. Equivalent amount of total lysate, P85 and S85 were precipitated by adding trichloroacetic acid (TCA) to 10% final concentration and processed for immunoblotting.

Immunoblotting

Equivalent amount of total lysate, high speed supernatant (S85) and membrane pellet (P85) were separated on 12% SDS-PAGE and electro-transferred to PVDF HybondP membrane (Amersham Biosciences). Recombinant GhKCR(s) were detected by using mouse monoclonal antibody against His-tag as the primary antibody (Invitrogen) and affinity-purified goat anti-mouse IgG conjugated with horseradish peroxidase as the second antibody. S. cerevisiae ER marker protein Kar2p 25 was detected by using polyclonal rabbit antibody against ScKar2p 26, a gift from M. Rose, as the primary antibody and affinity-purified goat anti-rabbit IgG conjugated with horseradish peroxidase as the second antibody (Bio-Rad Laboratories). The recognized epitopes were detected by using the ECL Western blotting detection system (Amersham Biosciences).

Elongase assays of cotton 3-ketoacyl-CoA reductases overexpressed in yeast cells

Palmitoyl-CoA was used as acyl-CoA acceptor in all reactions. The elongation assays were divided into two groups. One was supplemented with both NADPH and NADH 15, and another was supplemented only with NADH. The elongation assays contained 50 mM Tris-HCl, pH 7.5, 1 mM MgCl2, 0.15 mM Triton X-100, 1 mM NADPH, 1 mM NADH, 10 mM β-mercaptoethanol, 40 μM palmitoyl-CoA, 60 μM [2-14C] malonyl-CoA (6.5 dpm/pmol, PerkinElmer Life Sciences) in a final reaction volume of 0.2 ml. To initiate the elongation reaction, 0.05 mg ER protein extracted from yeast cells was added. The reaction was incubated at 30°C and continued for the indicated time. The reactions were stopped by adding 0.1 ml of 75% KOH (w/v) and 0.2 ml of ethanol, saponified at 70°C for 1 h, and then acidified by adding 0.4 ml of 5 N HCl with 0.2 ml of ethanol. Fatty acids were collected using 1 ml of hexane. The extractions were dried under nitrogen, and separated by thin-layer chromatography (TLC) using hexane/diethyl ether/acetic acid (30:70:1). The TLC plates were exposed to a PhosphorImager screen, the resulting image was analyzed, and the lipids were quantified using a Bio-Imaging Analyzer with 2D Image master software (Amersham Biosciences).

Hydropathy analysis

Hydropathy plot for each GhKCR gene was produced according to the method described previously27.

RESULTS

Identification of two cotton 3-ketoacyl-CoA reductases

A total of 19 colonies were obtained as a result of library screen with 15 found to encode for GhKCR1 and 4 for GhKCR2 after sequencing of the full-length insert. GhKCR1 possessed a 963 bp ORF that encoded a protein of 320 amino acids with a predicted molecular mass of 36 kD. GhKCR2 contained a 924 bp ORF that encoded a protein of 307 amino acids with a predicted molecular mass of 34 kD. We submitted GhKCR1 and 2 to GenBank with accession numbers AY902466 and AY902467, respectively. As shown in Fig. 1, GhKCR1 shared 68% sequence identity to that of AtYBR159 and Bn-KCR2, 66.7% to BnKCR1, 65% to ZmYBR159, 30% to ScYbr159p, 39% to human HsKCR and 41% to mouse MuKCR. GhKCR2 shared 59% total sequence identity with that of GhKCR1 and showed similar or slightly lower amino acid identities to the above-mentioned enzymes. All these proteins belong to SDR family 16 characterized by the presence of Rossmann fold (the nucleotide binding site) and a triad of catalytically important and highly conserved Ser-Tyr-Lys residues. A canonical dilysine ER retention signal is present in the C-terminus of GhKCR1 but not in GhKCR2. These results suggested that GhKCR1 and 2 might be the orthologues of AtYBR159.

Figure 1
figure 1

Amino acid sequence alignment of GhKCR1 and GhKCR2. Amino acid sequence alignment of GhKCR1 (Genbank accession no. AY902466) and GhKCR2 (Genbank accession no. AY902467) with orthologues from B. napus (Genbank accession no. AAO43448 and AAO43449), A. thaliana (Genbank accession no. NP_564905), maize (Genbank accession No. AAB82767), S. cerevisiae (Genbank accession no. AAS56194), humans (Genbank accession no. AAP36605) and mouse (Genbank accession No. NM_00829). Black shading indicates strictly conserved residues, whereas gray shading marks regions of less strict conservation. An active triad of highly conserved S-Y-K residues is marked with an asterisk and canonical dilysine ER retention motif is indicated (▾).

Full size image

GhKCR(s) showed fiber-preferential expression patterns

Relative transcript levels of GhKCR(s) from 3, 5, 10, 15, 20 dpa wild-type cotton ovules together with their fibers (wt) were compared with either those of 0 dpa wt ovules or 10 dpa ovules of a fuzzless-lintless mutant (fl) by QRT-PCR analysis (Fig. 2). GhKCR1 transcripts increased about 30-fold and that of GhKCR2 increased about 20-fold only 5 d after flowering, whereas, very low levels of transcripts were observed in roots, stems and leaves for either gene (Fig. 2). Both GhKCRs were suppressed significantly in ovules of the fl mutant as well (Fig. 2).

Figure 2
figure 2

Both GhKCR genes are highly preferentially expressed in fast-elongating wild-type cotton fibers, but not in roots, stems, leaves and mutant cotton ovules. 0, +3, +5, +10, +15, +20 and + 10fl indicate that total RNA samples prepared from 0, 3, 5, 10, 15, 20 dpa wide-type or 10 dpa fl mutant cotton ovules were used as the template for QRT-PCR analysis. Error bars indicate means ± SE obtained from three independent experiments. Cotton ubiquitin gene, UBQ7(Genbank accession no. AY189972), was included as the template control. Fold increase or decrease was calculated using data obtained from mRNA samples prepared from ovules harvested around the day of anthesis (0 dpa).

Full size image

Functional complementation of yeast mutant ybr159wΔ cells by the cotton enzymes

As seen in Fig. 3A, without transformation of GhKCR(s), S. cerevisiae haploid ybr159wΔ mutant cells grew very slowly on Sc-Trp- plate. Wild-type growth rate was resumed in mutant cells expressing either GhKCR1 or 2 suggesting that heterologous expression of GhKCR(s) was able to complement the genetic deficiency. In agreement with the plate assay, ybr159wΔ mutant cells transformed by pYTV::GhKCR2 performed similarly with that of wild-type whereas cells expressing GhKCR1 grew even better than the wild-type (Fig. 3B). RT-PCR analysis of ybr159wΔ mutant cells transformed by either pYTV::GhKCR1 or pYTV::GhKCR2 showed that all the transgenes were actively transcribed during the log phase of yeast culture (Fig. 3C). Our data suggested that the genetic complementation phenomenon was related to the biochemical functions of individual GhKCR(s).

Figure 3
figure 3figure 3figure 3

Yeast ybr159wΔ mutant cells can be functionally complemented by overexpression of individual GhKCR(s). (A) Wild-type rate of cell division was obtained in ybr159wΔ mutant cells that overexpressed either GhKCR1 or GhKCR2. All cells were inoculated from overnight culture and were grown in SC-Trp , 2% Galactose liquid media to an OD600 of 0.5 before being plated out on SC-Trp solid media in a series of 10-fold dilutions with sterile H2O. Cells were then incubated at 30°C for 2 days before being examined with naked eyes. (B) Growth curves of individual yeast strains cultured in SC-Trp, 2% Galactose liquid media up to 50 h at 30°C. Cell numbers were determined by using a haemocytometer. Error bars indicate means ± SE from three independent cultures. (C) RT-PCR analysis of yeast cell lines expressing different GhKCR(s) harvested at the indicated time after liquid culture. We included only one actin gene in the figure as the template control since it was expressed in a very similar manner in all strains.

Full size image

Subcellular localization of GhKCR proteins expressed in yeast

Insoluble membranous fractions (P85) were separated from the high speed supernatant (S85) by differential centrifugation using total proteins extracted from ybr159wΔ cells transformed either with pYTV or pYTV::GhKCR(s). Equivalent amount of total lysate, P85 and S85 were separated on SDS-PAGE and analyzed by immunoblotting using anti-His and anti-ScKar2p separately. Both GhKCR1 (Fig. 4A) and GhKCR2 (Fig. 4B) were detected mainly in P85 fraction in a similar manner with that of ER marker ScKar2p 25, 26. These data agreed with predictions of hydropathy plots since both GhKCR1 and 2 showed at least one membrane-spanning domain at their N-termini (Fig. 4). In the total lysate (TL) and P85 fractions prepared from ybr159wΔ cells without being transformed by the GhKCR(s), only ScKar2p was detected (Fig. 4).

Figure 4
figure 4

GhKCRs expressed in yeast were associated with their ER fractions. Membrane-spanning topology for each GhKCR was plotted according to White and Wimley 27 and was placed in the upper panel of each subset labeled (A) and (B) representing analysis GhKCR1 and 2, respectively. Western blotting results were reported in the lower panel of each subset. Data on the left were obtained by using protein extracted from ybr159wΔ mutant cells transformed by the plain vector pYTV and data on the right were obtained by using protein extracted from ybr159wΔ mutant cells transformed by pYTV::GhKCR1 and pYTV::GhKCR2 accordingly.

Full size image

GhKCRs were active 3-ketoacyl-CoA reductases as confirmed by in vitro elongase assays

3-Ketoacyl-CoA reductase activity was measured by elongation of palmitoyl-CoA and was determined by measuring the amount of 14C, originating from [2-14C]-malonyl-CoA, incorporated into the elongated fatty acid product, stearoyl-CoA (Fig. 5A). As expected, using KCR deficient ybr159wΔ mutant cell extract lead to the accumulation of a significant amount of 3-ketostearate with a concomitant reduction in 3-hydroxystearate as well as the final elongated product, stearate (Fig. 5). Overexpression of GhKCR1 and 2 separately in ybr159wΔ mutant cells in the presence of NADH and NADPH resulted in enhanced production of 3-hydroxystearate, trans-2-stearate and stearate (Fig. 5B, Tab. 1) in the assay. In the absence of NADPH as a cofactor, however, no obvious differences in the amount of intermediates and the final product generated using extract from ybr159wΔ mutant with or without being transformed by GhKCR(s) were observed (Fig. 5C, Tab.1). When both NADH and NADPH were omitted, the reactions were stopped at the intermediate of 3-keto-stearate, indicating that the reduction step was cofactor-dependent (Fig. 5D). These data suggested that the both GhKCR(s) are NADPH-dependent 3-ketoacyl-CoA reductases.

Figure 5
figure 5figure 5figure 5figure 5

GhKCRs function as NADPH-dependent 3-ketoacyl-CoA reductase in ybr159wΔ mutant cells. (A) Schematic drawings of reactions from palmitoyl-CoA to stearoyl-CoA. (B) ybr159wΔ mutant yeast cells transformed with GhKCR1, 2 or 3 metabolized palmitoyl-CoA at the same rate as wild-type cells in the presence of both NADH and NADPH. (C) When NADPH was not included in the experiment, all cells metabolized palmitoyl-CoA very slowly in the same way as that of ybr159wΔ mutant. (D) 3-ketostearate was accumulated similarly in all cell lines if both NADPH and NADH were not added in the assay. All reactions were initiated by adding 50 μg of proteins extracted from wild-type (wt), ybr159wΔ or ybr159wΔ mutant cells expressing GhKCR1 or 2 (KCR1 or 2 respectively). For TLC analysis, the fatty acyl-CoA thioesters were extracted as free fatty acid. The assays were stopped at the indicated time and the fatty acids were extracted and separated by TLC. Positions labeled as 1, 2, 3, 4 and O denote 3-ketostearate, 3-hydroxystearate, trans-2-stearate, stearate and the origin of TLC respectively. Relative positions were determined by comparing the mobility with authentic standards as described 14.

Full size image
Table 1 Relative amount of 3-hydroxystearate, trans-2-stearate, and stearate of palmitate elongation catalyzed by GhKCR(s) in the presence or absence of NADPH. The values represent the relative amount of different intermediates at 30 min in relation to that measured in the yeast mutant cells transformed with plain vector that was arbitrarily defined as 1.
Full size table

DISCUSSION

In this report, we isolated two cDNAs, GhKCR1 and 2 from developing cotton fiber cells, and characterized their functions both in genetic as well as biochemical terms. The proteins were found to restore the growth rate of S. cerevisiae ybr159wΔ mutant to that of wild-type cells in a genetic complementation assay (Fig. 3). Recombinant GhKCR(s) displayed 3-ketoacyl-CoA reductase activities in an in vitro reaction that used palmitoyl-CoA as the substrate in the presence of malonyl-CoA (Fig. 5). Further studies showed that recombinant GhKCR1 and 2 were co-localized with the ER marker protein ScKar2p in S. cerevisiae mutant ybr159wΔ cells (Fig. 4), indicating that these cotton enzymes function as microsomal 3-ketoacyl-CoA reductases.

In S. cerevisiae, Ybr159p was shown to constitute majority of the KCR activity since extracts from ybr159wΔ mutant cells leads to a significant accumulation of 3-ketostearate in the elongase assay (Fig. 5B). Using extracts from both wild-type and mutant cells that overexpressed GhKCR(s), 3-ketostearate was not accumulated. Instead, production of 3-hydroxystearate and stearate was enhanced. Similar levels of 3-hydroxy-fatty acyl intermediates were found using extracts from mutant cells and those that overexpressed GhKCR(s) when NADPH was removed in the assay system (Fig. 5C) suggesting that GhKCR(s) were NADPH-dependent 3-ketoacyl-CoA reductases. NADPH-dependent 3-ketoreductase activities were previously detected from rat liver 28. The cofactor requirement for the mammalian microsomal KCR protein has not been settled yet 29, 30. The plant 3-ketoacy-[acyl carrier protein] reductase involved in plastid fatty acid synthesis was NADPH-dependent 31. NADPH binding was found to induce a conformational change that put three conserved amino acid residues into the active site to facilitate catalytic function 32.

In maize and Arabidopsis, only one gene homologous to yeast YBR159w was identified 14, 17. Recently, two highly homologous BnKCR(s) from Brassica genomes were identified 19 and they are homologues of AtYBR159. Three independent lines of evidence suggest that GhKCR1 and 2 cDNAs reported in the current work might represent two different KCR genes: (i) The GhKCRs show different expression patterns during various stages of cotton fiber development and also in different cotton tissue (Fig. 2). The expression profiles of the both genes are in accordance with our previous findings that several GhKCS genes encoding putative 3-ketoacyl-CoA synthases 33 and cotton genes encoding lipid transfer proteins 34, 35 were highly expressed at the same period. Our data indicate that long chain fatty acid biosynthesis and supply may be in high demand during the fast cell elongation period. (ii) GhKCR1 and 2 show significant sequence similarity to ScYbr159p and share 59% amino acid identity with each other. (iii) Although GhKCR2 protein is recovered from the microsomal fractions in the same way as GhKCR1, it is different from GhKCR1 since it lacks a predicted ER retention signal (Fig. 4).

In conclusion, we isolated two cDNAs encoding 3-ketoacyl-CoA reductases from developing cotton fibers. Genetic complementation and biochemical characterization of these cotton genes in yeast cells demonstrated that they could function as NADPH-dependent 3-ketoacyl-CoA reductases. Both theoretical and experimental analysis strongly suggested that GhKCRs are involved in ER-associated very long chain fatty acid elongation. Our work indicates that the VLCFA elongation pathway is upregulated during cotton fiber development and is similar to that found in yeast cells. Whether GhKCR(s) are essential for the elongation of VLCFAs in cotton need further investigation.