JXB Advance Access originally published online on February 27, 2007
Journal of Experimental Botany 2007 58(6):1421-1432; doi:10.1093/jxb/erm005
RESEARCH PAPER |
Cloning and transcriptional analysis of Crepis alpina fatty acid desaturases affecting the biosynthesis of crepenynic acid
Department of Chemistry, Washington University in St Louis, One Brookings Drive, Campus Box 1134, St Louis, MO 63130-4899, USA
* To whom correspondence should be addressed. E-mail: kappock{at}wustl.edu
Received 25 August 2006; Revised 13 December 2006 Accepted 22 December 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Crepis alpina acetylenase is a variant FAD2 desaturase that catalyses the insertion of a triple bond at the
12 position of linoleic acid, forming crepenynic acid in developing seeds. Seeds contain a high level of crepenynic acid but other tissues contain none. Using reverse transcriptase-coupled PCR (RT-PCR), acetylenase transcripts were identified in non-seed C. alpina tissues, which were highest in flower heads. To understand why functional expression of the acetylenase is limited to seeds, genes that affect acetylenase activity by providing substrate (FAD2) or electrons (cytochrome b5), or that compete for substrate (FAD3), were cloned. RT-PCR analysis indicated that the availability of a preferred cytochrome b5 isoform is not a limiting factor. Developing seeds co-express acetylenase and FAD2 isoform 2 (FAD2-2) at high levels. Flower heads co-express FAD2-3 and FAD3 at high levels, and FAD2-2 and acetylenase at moderate levels. FAD2-3 was not expressed in developing seed. Real-time RT-PCR absolute transcript quantitation showed 104-fold higher acetylenase expression in developing seeds than in flower heads. Collectively, the results show that both the acetylenase expression level and the co-expression of other desaturases may contribute to the tissue specificity of crepenynate production. Helianthus annuus contains a 12 acetylenase in a polyacetylene biosynthetic pathway, so does not accumulate crepenynate. Real-time RT-PCR analysis showed relatively strong acetylenase expression in young sunflowers. Acetylenase transcription is observed in both species without accumulation of the enzymatic product, crepenynate. Functional expression of acetylenase appears to be affected by competition and collaboration with other enzymes. Key words: Acetylenase, Crepis alpina, developing seed, fatty acid desaturases, flower head, Helianthus annuus, real-time RT-PCR, transcriptional analysis
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
The fatty acid desaturase FAD2 is the
12-oleic acid desaturase that introduces a double bond at the 12 position of oleic acid (9c-18:1), forming linoleic acid (9c,12c-18:2) in the endoplasmic reticulum (ER; Fig. 1) (Miquel and Browse, 1992; Okuley et al., 1994). Functionally variant forms of FAD2 perform fatty acid hydroxylation, epoxygenation, and conjugation (Voelker and Kinney, 2001). Crepis alpina acetylenase is a variant FAD2 that catalyses the insertion of a triple bond at the 12 position of linoleic acid to form crepenynic acid (9c,12a-18:2). As a minor side reaction, C. alpina acetylenase can also convert oleic acid into a mixture of 18:2 isomers (9c,12c and 9c,12t) (Carlsson et al., 2004). The acetylenase substrate linoleic acid can also be converted to linolenic acid (9c,12c,15c-18:3) by FAD3. FAD2, FAD3, and acetylenase are all membrane-bound fatty acid desaturases (Shanklin and Cahoon, 1998) that require reductase partners. In the ER, electrons from NADH are moved to desaturases via the flavoprotein cytochrome b5 reductase and the electron carrier cytochrome b5 (cyt b5) (Shanklin and Cahoon, 1998). These transfers are summarized in Fig. 1.
|
It is challenging to reconstruct fatty acid biosynthetic pathways in transgenic plants. While cDNAs for a wide variety of unusual fatty acid biosynthetic enzymes have been identified, it has not yet been possible to use these genes to produce large amounts of unusual fatty acids in seeds of transgenic plants (Jaworski and Cahoon, 2003). This difficulty points to the need for more insight into the biochemistry of fatty acid biosynthesis and metabolism in oilseeds.
Some or all naturally occurring acetylenes are thought to derive from crepenynic acid (Bu'Lock and Smith, 1967). Polyacetylenic compounds have been detected in at least 15 plant families (Bohlmann et al., 1973). In the Compositae, more than 1100 species out of 267 genera from 13 tribes have been studied, and (poly)acetylenes are present in all 13 tribes. In one of the largest tribes, Heliantheae, polyacetylenes are generally found in roots or leaves. The whole-plant (poly)acetylene content of three Crepis species has been reported (Bohlmann et al., 1973): C. foetida contains only crepenynic acid, located only in the seeds; C. biennis contains trace amounts of polyacetylenes in the leaves; and C. sibirica does not contain any (poly)acetylenes. Crepis alpina contains crepenynic acid in seeds but has not been reported to contain polyacetylenes.
The Seed Oil Fatty Acids Database (http://www.bagkf.de/sofa/) identifies 32 plant species from Compositae, Leguminosae–Caesalpinioideae, and Rubiaceae that contain crepenynic acid in their seed oil, at levels ranging from 0.1% to 75% fatty acyl content. Among these plants, C. alpina accumulates crepenynate as 74% of the acyl content in seed triacylglycerols. Lee et al. (1998) cloned and characterized C. alpina acetylenase, and showed by northern blot that its expression was seed-specific, as was the presence of crepenynic acid. Cahoon et al. (2003) reported that acetylenase genes occur widely and that, in developing sunflower seed, expression was triggered by a fungal elicitor. However, no crepenynic acid accumulated in sunflower seeds expressing acetylenase; this was attributed to further metabolism of crepenynic acid.
The present studies of acetylenase expression in C. alpina were initiated using a sensitive and potentially quantitative detection method, RT-PCR. As expected, acetylenase transcription in developing seeds correlated with crepenynic acid accumulation. However, it was found that acetylenase was also transcribed in non-seed tissues of C. alpina that do not accumulate crepenynic acid. The lack of accumulation could be due to many factors: low acetylenase protein production; poor activity of the acetylenase enzyme (possibly due to incomplete assembly or mislocalization); insufficient or improperly presented substrate; deficiencies of co-factors or required accessory proteins; or consumption of the newly synthesized crepenynic acid by further metabolism.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Plant materials
Crepis alpina seeds were obtained from the Plant Germplasm Introduction and Testing Research Station of the US Department of Agriculture. Crepis alpina plants were grown in a growth chamber at 10 °C or 21 °C, or in a greenhouse (Department of Biology, Washington University). The growth chamber photoperiod was 16 h (150 µmol m–2 s–1) and humidity was 70%. Flower head stages were defined by size (5 mm increments, stages 1–8, S1–S8). Developing seeds were harvested from plants grown at 21 °C or in the greenhouse, 16–20 d after flowering.
Helianthus annuus seeds were purchased from Target. Plants were grown from seeds in a growth chamber at 10 °C or in a greenhouse. The whole plant except roots was harvested when indicated.
Nucleic acid isolation, cloning vectors, and bacterial strains
Plant genomic DNA was isolated from leaves of C. alpina or H. annuus using Plant DNeasy from Qiagen (Valencia, CA, USA). Plant total RNA was prepared using Qiagen Plant RNeasy with on-column DNase treatment to remove contaminating genomic DNA. Plasmid DNA was isolated using a Qiagen QIAprep Spin Miniprep Kit. The cloning vector pDrive (Qiagen) was used for cloning PCR-amplified fragments. Before cloning, PCR products were purified with the Qiagen QIAquick gel extraction kit. Escherichia coli strains DH5
or TOP10F' were used for cloning.
Isolation of the acetylenase genes from genomic DNA
The open reading frame (ORF) for the C. alpina acetylenase gene was amplified by PCR from leaf genomic DNA. [Note: FAD2 and variant FAD2 genes typically do not contain introns within their ORFs (Cahoon et al., 2003).] The primers used to amplify the acetylenase ORF were #630 and #631 (oligodeoxynucleotide sequences are given in Table S1 in the Supplementary data at JXB online). These primers were based on the reported C. alpina acetylenase sequence [GenBankTM accession number (AN) Y16285
[GenBank]
] (Lee et al., 1998). Two independent PCRs were conducted using Taq polymerase (Promega, Madison, WI, USA), and the sequences of the cloned products from each reaction were found to be identical. However, the sequence differed from the previously reported sequence in two of 1128 bp, one of which results in a Gly365
Glu substitution. This C. alpina acetylenase sequence was deposited with AN DQ289485.
The H. annuus acetylenase gene was amplified by PCR from genomic DNA and primers #1048 and #1049 (Cahoon et al., 2003), using Pfu DNA polymerase (Stratagene, La Jolla, CA, USA). The amplified product was digested and ligated into the NotI site of pYES2, yielding pJK349. The DNA sequence differed from the previously reported H. annuus acetylenase gene sequence (AN AY166773
[GenBank]
) in 18 of 1134 bp, one of which results in a Thr280Ser substitution. This H. annuus acetylenase sequence was deposited with AN DQ338433.
Degenerate primers for cyt b5 and FAD3/7
Degenerate primers for rapid amplification of cDNA 3' ends (3'-RACE) PCR amplification of an internal portion of the cyt b5 and FAD3/7 genes were designed using the CODEHOP (consensus-degenerate hybrid oligonucleotide primers) program (http://blocks.fhcrc.org/blocks/codehop.html) (Rose et al., 1998).
The cyt b5 input protein sequences were from Nicotiana tabacum (ANs P49098 [GenBank] , P49099 [GenBank] , S46306 [GenBank] , and CAA48240 [GenBank] .1), Olea europaea (ANs CAA04702 [GenBank] .1 and CAA04703 [GenBank] .1), Borago officinalis (AN O04354 [GenBank] ), Cuscuta reflexa (AN P49097 [GenBank] ), and Petuniaxhybrida (AN AAD10774 [GenBank] .1). The algorithm suggested sequences represented by degenerate primers #720 and #721.
The FAD3 (ER-type) or FAD7 (chloroplast-type) input sequences were from Nicotiana tabacum FAD3 (AN P48626 [GenBank] ) and FAD7 (AN P93350), Petroselinum crispum FAD7 (AN P93452), Sesamum indicum FAD7 (AN P48620 [GenBank] ), Solanum tuberosum FAD7 (AN O82068), Lycopersicon esculentum (=Solanum lycopersicum) FAD7 (AN AAP82170 [GenBank] ), Arabidopsis thaliana FAD3 (AN NP_180559 [GenBank] ), Glycine max FAD3 (AN BAB18135 [GenBank] ), and Brassica napus FAD3 (AN P48624 [GenBank] ). The algorithm suggested sequences represented by degenerate primers #911, #912, and #913.
Cloning of FAD2 gene fragments
Degenerate primers for amplification of FAD2 genes were from Cahoon et al. (2003). PCRs using degenerate primers #822 and #823 with an annealing temperature of 45 °C and 40 cycles of amplification yielded a 
1 kbp fragment from cDNA of stage 2 flower heads (S2) of plants grown at 21 °C. The products recovered contained two different partial FAD2 sequences (FAD2-2 and FAD2-3).
3'-RACE
3'-RACE was performed using Invitrogen (Carlsbad, CA, USA) kits according to the manufacturer's instructions. This method requires gene-specific primers (GSPs) for PCR. Degenerate oligodeoxynucleotides were used as GSPs to clone cyt b5 and FAD3/7 genes. GSPs for FAD2 were designed using the available internal sequence. GSP1 was used for the first PCR, and GSP2 was used for the nested confirmatory PCR. In some cases, GSP3 was used for a second, nested PCR. PCRs were performed using Taq polymerase. PCR amplification products with GSP1 were cloned in pDrive. Before sequencing, the insert fragments were subjected to nested PCR with GSP2 or GSP3 to confirm they were likely to be the targeted genes.
cDNA from flower head stage 2 of plants grown at 10 °C (10S2), or developing seeds, was used as a PCR template for 3'-RACE of cyt b5 genes. PCRs were performed with primers #720 (GSP1) and #721 (GSP2) using an annealing temperature of 60 °C with 30 cycles of amplification. Four isoforms (cyt b5-06, cyt b5-11, cyt b5-28, and cyt b5-56) were identified.
FAD2-2 PCRs used 10S2 cDNA template, with primers #876 (GSP1) and #877 (GSP2). FAD2-3 PCRs used the same template, with primers #876 (GSP1) and #891 (GSP2). Both used an annealing temperature of 58 °C with 30 cycles of amplification.
FAD3/FAD7 PCRs used 10S2 cDNA template, primers #911 (GSP1), #912 (GSP2), and #913 (GSP3), and an annealing temperature of 60 °C with 35 cycles of amplification. Two isoforms (FAD3/7-1 and FAD3/7-2) were identified.
5'-RACE
5'-RACE was performed using Invitrogen kits according to the manufacturer's instructions. This method requires antisense GSPs (GSP1, GSP2, and GSP3) designed from sequences obtained by 3'-RACE. GSP1 was used for the first PCR, and GSP2 and GSP3 were used for nested PCR. PCRs were performed using Taq polymerase.
Cyt b5 isoform PCRs used 10S2 cDNA template. For cyt b5-06, GSP1 was #801, GSP2 was #752, and GSP3 was #753; for cyt b5-11, GSP1 was #803, GSP2 was #774, and GSP3 was #771; for cyt b5-28, GSP1 was #805, GSP2 was #754, and GSP3 was #755; and for cyt b5-56, GSP1 was #895, GSP2 was #949, and GSP3 was #950.
FAD2-2 PCRs used 10S2 cDNA template with primers #879 (GSP1), #880 (GSP2), and #890 (GSP3). FAD2-3 PCRs used cDNA from flower head stage 2 of plants grown at 21 °C (21S2) as template, with primers #892 (GSP1) and #893 (GSP2).
FAD3/7-1 PCRs used cDNA from young leaves of plants grown at 21 °C as the template, with primers #934 (GSP1), #932 (GSP2), and #933 (GSP3). Analysis of the deduced amino acid sequence with ChloroP (http://www.cbs.dtu.dk/services/ChloroP/) (Emanuelsson et al., 1999) identified a chloroplast transit peptide of 59 amino acids. FAD3/7-2 PCRs used cDNA from the roots of plants grown at 21 °C as template, with primers #934 (GSP1), #935 (GSP2), and #936 (GSP3). By sequence similarity and the presence of a transit peptide, FAD3/7-1 was identified as a chloroplast form (FAD7), and FAD3/7-2 was identified as a microsomal form (FAD3).
Yeast transformation and growth
The acetylenase gene was amplified by PCR from C. alpina genomic DNA using primers #806 and #807. The FAD2-2 gene was amplified from C. alpina cDNA with primers #943 and #944. The FAD2-3 gene was amplified from cDNA with primers #946 and #947. The FAD3 gene was amplified from cDNA with primers #976 and #977. Amplified genes were cloned into the BamHI and XbaI sites of the yeast expression vector pYES2 (Invitrogen; galactose-inducible promoter GAL1), yielding pJK272 (acetylenase), pJK273 (FAD2-2), pJK274 (FAD2-3), and pJK277 (FAD3).
Saccharomyces cerevisiae INVSc1 (MATa his31 leu2 trp1-289 ura3-52) was transformed by the lithium acetate method according to the supplier's instructions (Invitrogen). Viable cells were selected on minimal medium lacking uracil. A single colony was used to inoculate cultures (3 ml) that were grown for 3 d at 20 °C in induction medium (containing 2% galactose) with (i) 0.1% Tergitol type NP-40, (ii) 0.03% oleic acid and 0.1% Tergitol, or (iii) 0.03% linoleic acid and 0.1% Tergitol.
Fatty acid methyl ester (FAME) analysis
FAMEs were prepared from C. alpina tissues by transesterification in 1% (w/v) sodium methoxide in methanol (Cahoon et al., 1999). FAMEs were analysed using an Agilent Technologies 6890N gas chromatograph (GC) fitted with an Innowax column (30x0.25 mm inner diameter; Agilent Technologies, Santa Clara, CA, USA) and a flame-ionization detector. The oven temperature programme was an initial 1 min hold at 170 °C, followed by a ramp to 210 °C at a rate of 5 °C min–1.
Yeast cultures (3 ml) expressing C. alpina desaturase genes were harvested by centrifugation, washed three times with water (2 ml), and dried under vacuum for 30 min. Transesterification was performed using 0.4 ml of 2.5% sulphuric acid in methanol containing 10 mg ml–1 heptadecanoic acid as an internal standard (80 °C for 1 h). Peak areas were referenced to the heptadecanoic acid standard. FAMEs were extracted with hexane (0.4 ml) after adding brine (0.4 ml) to the cooled reaction. FAMEs were resolved by HPLC on the C18 column at a flow rate of 0.5 ml min–1 with detection at 210 nm, using a mixture of acetonitrile and water as the solvent system. The gradient programme was 0–25 min, 15% acetonitrile (isocratic); 25–35 min, 15–0% acetonitrile (linear gradient); 35–90 min, 0% acetonitrile (isocratic). Retention time standards were a FAME mix (AccuStandard, New Haven, CT; 34% methyl linolenate, 36% methyl linoleate, 18% methyl oleate, 5% methyl stearate, 7% methyl palmitate), or selected individual components.
Qualitative RT-PCR
First-strand cDNA synthesis used DNase-treated total RNA template (1 µg), Superscript II RNase H– Reverse Transcriptase (Invitrogen), an oligo-d(T)15 primer, and Taq polymerase.
Different primer pairs were used to amplify the C. alpina acetylenase transcript: pair 1 (#630 and #631), pair 2 (#806 and #807), and pair 3 (#874 and #875). To confirm the specificity of acetylenase amplification, various annealing temperatures (50, 53, and 55 °C) and cycle numbers (30, 35, and 40) were used. The primers used to amplify cyt b5 isoforms were designed to match unique sequences in each cyt b5 gene: cyt b5-06, #800 and #801; cyt b5-11, #802 and #803; cyt b5-28, #804 and #805; and cyt b5-56, #894 and #895. PCR amplification of the cyt b5-06, cyt b5-11, and cyt b5-28 genes used an annealing temperature of 55 °C and 25 cycles. PCR amplification of the cyt b5-56 gene used an annealing temperature of 53 °C and 35 cycles.
Routine analysis of gene expression used for C. alpina acetylenase employed primers #630 and #631, an annealing temperature of 53 °C, and 35 cycles of amplification; for FAD2-2, primers #943 and #944, an annealing temperature of 50 °C, and 30 cycles of amplification; for FAD2-3, primers #946 and #947, an annealing temperature of 55 °C, and 35 cycles of amplification; for FAD3, primers #976 and #977, an annealing temperature of 50 °C, and 30 cycles of amplification; and for actin, primers #886 and #887, an annealing temperature of 50 °C, and 30 cycles of amplification. The partial DNA sequence of the actin gene was obtained from C. alpina EST analysis (J-W Nam, A Ransome, RK Wilson, and TJ Kappock, unpublished data). The H. annuus acetylenase gene was amplified with primers #1048 and #1049, an annealing temperature of 53 °C, and 30 cycles of amplification.
Quantitative real time RT-PCR analysis
First-strand cDNA preparations were performed as above and used as the PCR template (at a 1:5 dilution), except that the total RNA preparation was treated twice with RNase-free DNase (Qiagen). TaqMan minor groove binder (MGB) probes and primers were designed using PrimerExpress software (Applied Biosystems, Foster City, CA, USA). Serial dilutions of plasmid DNAs (pJK272, pJK273, pJK274, and pJK277) were copy number standards. The standards were processed in parallel with the cDNA samples and were used to generate a standard curve. All reactions were performed in triplicate.
Absolute quantitation assays were conducted using a 7500 real-time PCR system (Applied Biosystems). In real-time quantitative PCR, the threshold cycle (CT) is the cycle at which a significant increase in fluorescence takes place. Primers were present at a final concentration of 300 nM each, and TaqMan MGB probes were present at a final concentration of 250 nM in a total volume of 25 µl. Cycling conditions were 50 °C for 2 min, 95 °C for 10 min, followed by 40, 45, or 50 cycles of 95 °C for 15 s, and 60 °C for 60 s. Crepis alpina acetylenase real-time RT-PCR absolute quantitation used primers #1135 and #1136 with probe #1177 for detection. FAD2-2 quantitation used primers #1115 and #1116 with probe #1178. FAD2-3 quantitation used primers #1117 and #1118 with probe #1179. FAD3 quantitation used primers #1119 and #1120 with probe #1180. Helianthus annuus acetylenase quantitation used primers #1121 and #1148 with probe #1181.
Sequence analyses
DNA sequencing was performed by the staff of the Protein and Nucleic Acid Chemistry Laboratory at the Washington University School of Medicine. DNA sequences were assembled using Phred/Phrap/Consed (Ewing and Green, 1998; Ewing et al., 1998; Gordon et al., 1998). Phylogenetic trees were assembled using ClustalW (Thompson et al., 1994), Phylip (Felsenstein, 1989), and TreeView (Page, 1996). Sequence alignments were constructed using ClustalW and EPSpript (Gouet et al., 1999).
Sequence data
ANs for the desaturase cDNA sequences reported here are: DQ289485 (C. alpina acetylenase), DQ289486 (FAD2-2), DQ289487 (FAD2-3), DQ176017 (FAD3), DQ176018 (FAD7), and DQ338433 (H. annuus acetylenase). Crepis alpina cytochrome b5 ANs are: DQ174104 (cyt b5-06), DQ174105 (cyt b5-11), DQ174106 (cyt b5-28), and DQ174107 (cyt b5-56).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Crepis alpina acetylenase gene expression analysis
A sensitive RT-PCR approach was used to detect acetylenase gene expression in C. alpina. Initial analysis showed acetylenase transcripts in developing seed, flower head, roots, and, at a reproducible low level, stems. In flower heads, the expression profile showed stronger expression in earlier flower head stages (Fig. 2).
|
Three different acetylenase primer pairs (pairs 1–3) showed similar amplification from flower heads. A typical RT-PCR result using primer pair 1 is shown in Fig. 2B. The PCR product was identified as acetylenase by DNA sequencing. Negative controls omitted RT to confirm that the products were not amplified from genomic DNA contaminants.
Since temperature is known to affect fatty acid profiles in plants, altering physical properties of the membrane and increasing the proportion of multiply unsaturated fatty acids (Uemura et al., 1995), plants were grown at two different temperatures to assess effects on desaturase expression. Greenhouse- or 21 °C-grown C. alpina plants contained 0–300 observable seeds per flower head after S7. Plants grown at 10 °C seldom contained seeds.
Fatty acid content of plant tissues
Plant materials that showed acetylenase RNA expression were analysed by GC for the presence of crepenynate methyl esters in esterified plant extracts. As previously reported (Lee et al., 1998), crepenynic acid was detected only in seeds (see Table S2 in the Supplementary data at JXB online). The relative amounts of linoleic acid and linolenic acid varied widely in different tissues, and were lowest in seeds. The linolenic:linoleic acid ratio was also low in flower heads relative to most other tissues. Qualitatively, flower heads appear to contain less linoleic acid than other tissues, as determined by GC (see Table S2 in the Supplementary data at JXB online) and HPLC (data not shown) analyses of FAMEs.
Cloning of cyt b5 genes and expression analysis
Detecting acetylenase transcription but not crepenynic acid in C. alpina flower heads suggests that one or more factor(s) required for acetylenase function are not present in flower heads. One candidate is the immediate electron transfer partner, cyt b5. While membrane-associated cyt b5 provides electrons, the timing of electron transfer relative to the catalytic cycle is not known for any membrane desaturase.
Using degenerate PCR primers and the 3'-RACE method, four different C. alpina cyt b5 isoforms were identified. Full sequences were obtained using 5'-RACE. The inferred cyt b5 ORFs for cyt b5-06, cyt b5-11, cyt b5-28, and cyt b5-56 cDNAs encode proteins of 131, 136, 145, and 136 residues, respectively. The predicted cyt b5 amino acid sequences contain most of the conserved residues in cyt b5, including the two histidine residues that are axial haem ligands (His37 in the motif HPGG and His61, cyt b5-06) (Mathews, 1985).
Cyt b5-06 and cyt b5-11 proteins are 62% identical, and cyt b5-11 and cyt b5-56 are 57% identical, while other pairwise identities are 43–47% (see Fig. S1 in the Supplementary data at JXB online). Cyt b5-06 showed the highest identity (80%) to olive cyt b5-15 (AN CAA04702 [GenBank] ); cyt b5-11 is 77% identical to olive cyt b5-38 (AN CAA04703 [GenBank] ); cyt b5-28 is 68% identical to tobacco cyt b5 (AN CAA50575 [GenBank] ); and cyt b5-56 is 54% identical to olive cyt b5-38 (AN CAA04703 [GenBank] ). These relatively low identities among the C. alpina cyt b5s might indicate different specializations, such as specific spatial or temporal expression patterns. Cyt b5-06 is most closely related to other cyt b5s expressed in seeds or siliques (see Fig. S2 in the Supplementary data at JXB online).
RT-PCR analysis was used to analyse cyt b5 transcription in C. alpina; primers were carefully designed to be isoform-specific and the identity of all amplified products was confirmed by restriction mapping. RT-PCR detected transcription of all four cyt b5 isoforms in both flower heads and developing seed (Fig. 3). Constant cyt b5-11 and cyt b5-28 gene expression levels were observed in flower heads. Variable cyt b5-06 and cyt b5-56 gene expression levels were observed during flower head development. These variations generally correlated with variations in acetylenase gene expression levels. However, cyt b5-06 expression levels in developing seed were lower than those in flower heads, suggesting it is not especially partnered with acetylenase. In general, the expression of cyt b5 isoforms in developing seeds was not higher than in flower heads.
|
Since the four cyt b5 isoforms were first identified in flower head RNA, an attempt was made to find other cyt b5s in developing seed material. Restriction mapping of 96 developing seed cDNA clones showed only the original four isoforms.
Cloning of FAD2 and FAD3 genes
FAD2 desaturases provide the linoleic acid substrate to acetylenase, a variant FAD2. FAD3 desaturases potentially compete with acetylenase for linoleic acid. To find C. alpina FAD2 genes, PCRs were performed with a degenerate primer set designed for Asteraceae FAD2 genes, including variant forms (Cahoon et al., 2003). Two FAD2 isoforms were identified, and complete gene sequences were obtained by 3'- and 5'-RACE (FAD2-2 and FAD2-3). The intronless C. alpina acetylenase gene was cloned from genomic DNA. FAD2 isoforms were amplified using degenerate primers and both flower head and developing seed reverse transcript templates. While clones of both FAD2 isoforms were recovered from flower head, only FAD2-2 and acetylenase were found in developing seed. The FAD2-2 cDNA clone had a rather long 5' untranslated sequence of 973 bp. The 5' untranslated sequence of FAD2-3 was 120 bp. cDNA and genomic DNA sequences in the FAD2 coding regions were the same, confirming that they lack introns (Cahoon et al., 2003).
Acetylenase was 56% or 55% identical to FAD2-2 or FAD2-3 at the amino acid level, while FAD2-2 and FAD2-3 were 78% identical to each other (see Fig. S3 in the Supplementary data at JXB online). FAD2-2 had the highest identity (78%) with H. annuus FAD2-1 (12-oleate desaturase, AN AAL68981
[GenBank]
), and FAD2-3 had the highest identity (77%) with Olea europaea FAD2-1 (12-oleate desaturase, AN AAW63040
[GenBank]
). Comparisons of the acetylenase and FAD2s show that both FAD2-2 and FAD2-3 group with other typical FAD2s and each other, while acetylenase is in a distinct branch with other variant FAD2s (see Fig. S4 in the Supplementary data at JXB online). Acetylenases from C. alpina and H. annuus are closely related (77% identity).
FAD3 and FAD7 are the ER and plastid forms, respectively, of linoleate 15-desaturase. Degenerate primers were designed to amplify conserved sequence regions in FAD3/FAD7 and used for 3'-RACE. Two different putative FAD3 clones were obtained. Further 5' region sequence information from 5'-RACE revealed that one type was actually a chloroplast FAD7 (AN DQ176018) with a 59 amino acid targeting sequence. The other type was a microsomal FAD3 (AN DQ176017). Crepis alpina FAD3 and FAD7 are 68% identical and are most closely related to Betula pendula FAD3 (AN AAN17504
[GenBank]
, 74% identity) and Sesamum indicum FAD7 (AN P48620
[GenBank]
, 78% identity). Comparative analysis shows that C. alpina FAD3 groups with other FAD3s, and C. alpina FAD7 groups with other FAD7s (see Figs S5 and S6 in the Supplementary data at JXB online).
Functional characterization of FAD2 and FAD3 genes
The acetylenase can be functionally expressed in S. cerevisiae (Lee et al., 1998). The similarity among membrane desaturases requires confirmation that the new genes have the anticipated activities. The FAD2-2, FAD2-3, and FAD3 genes described here were cloned into a yeast vector, introduced into S. cerevisiae, and fatty acid profiles of the transformed cells were examined (Table 1). Saccharomyces cerevisiae cells containing the C. alpina FAD2-2 gene (pJK273) showed accumulation of linoleic acid with or without oleic acid supplementation. This shows the FAD2-2 gene was functionally expressed in yeast and converted endogenous oleic acid to linoleic acid. Saccharomyces cerevisiae containing the FAD2-3 gene (pJK274) showed accumulation of linoleic acid only when supplemented with oleic acid. Saccharomyces cerevisiae harbouring the FAD3 gene (pJK277) converted linoleic acid into linolenic acid. Linoleic and linolenic acid are not normally present in S. cerevisiae, and neither was formed in controls transformed with pYES2 grown under the same conditions. These results confirm that C. alpina FAD2-2, FAD2-3, and FAD3 have the expected activities, but do not represent a quantitative comparison of relative enzymatic activity.
|
Transcription analysis by RT-PCR
The expression of acetylenase and related genes was analysed by RT-PCR for a qualitative assessment of coordinated gene expression. Real-time RT-PCR was then used to quantitate transcript numbers in selected tissues. Since crepenynate accumulation has the potential to affect the physical properties of lipids, expression in plants grown at two temperatures was examined.
Simultaneous RT-PCR expression analysis of FAD2-2, FAD2-3, FAD3, and acetylenase was measured in a variety of C. alpina materials (Fig. 4). Young plants did not express acetylenase but had relatively high FAD2-2 and FAD3 gene expression levels. FAD2-2 gene expression was higher in young plants grown at 21 °C than at 10 °C, and higher in developing seed, flower head, and leaf than in root. FAD2-3 transcripts were not detected in young plants, but they were found in adult plant flower heads, and they were more abundant in plants grown at 21 °C than at 10 °C. FAD3 gene expression was observed in all tissues tested including developing seed. In adult plants, FAD3 transcript levels were relatively high in stem and flower heads. Crepis alpina actin transcription was used as a positive control for RT-PCR and gel loading. Actin transcript levels were high and constant in most tissues except developing seeds, which showed unexpectedly low actin transcription in the same quantity of input RNA. (In developing seeds, acetylenase transcripts served as an internal check on the success of RT-PCR.) These observations show that the acetylenase is co-expressed with FAD2-2 in developing seeds, and all tested desaturases are expressed in flower heads.
|
Transcript numbers were measured by quantitative expression analysis of the acetylenase and the desaturases by real-time RT-PCR with gene-specific TaqMan probes (Table 2). Real-time RT-PCR results agree with the qualitative RT-PCR analysis, with the exception of FAD3 in flower heads (in Fig. 4 and Table 2 compare stage 2 expression levels from plants grown at 10 °C and at 21 °C). Flower heads from plants grown at 10 °C had higher acetylenase transcript copy numbers than those grown at 21 °C. While the level of FAD2-2 expression is not significantly different in flower heads and developing seeds, there were clear differences in the expression levels of acetylenase, FAD2-3, and FAD3: acetylenase transcripts were 104-fold more abundant in developing seed; FAD2-3 was not expressed in developing seed; and FAD3 transcripts were 102-fold more abundant in flower heads. Flower heads contained many FAD2-2, FAD2-3, and FAD3 transcripts, while the acetylenase copy number was reproducibly low but not negligible. By contrast, developing seeds contained more FAD2-2 transcripts and a very high acetylenase transcript number, consistent with efficient processing of crepenynic acid precursors during seed filling.
|
Helianthus annuus acetylenase: gene cloning and real-time RT-PCR
Helianthus annuus contains an inducible
12-acetylenase but does not accumulate crepenynic acid, even in seeds (Cahoon et al., 2003). Helianthus annuus plants were tested for acetylenase expression outside seeds. The H. annuus acetylenase gene was obtained from a genomic DNA PCR product, cloned into the pYES2 vector, and used as a copy number standard. Qualitative RT-PCR analysis showed acetylenase expression in young plants (data not shown). DNA sequencing confirmed the PCR products were derived from the acetylenase gene. Real-time quantitative RT-PCR analysis showed high-level expression of acetylenase genes in young seedlings grown for 2 or 3 weeks in a greenhouse, 4 weeks at 10 °C, or those moved to 10 °C after 2 weeks in a greenhouse (Table 3). Whether plants were grown for some time at 10 °C did not affect transcript levels in 3-week-old seedlings. As expected, FAME analysis showed no crepenynic acid. This result demonstrates that ‘cryptic’ (i.e. uncorrelated with crepenynate accumulation) acetylenase gene expression is not unique to C. alpina.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Acetylenase and related proteins
Crepis alpina acetylenase is a potential model for FAD2-class membrane desaturases. With its unique and chemically unprecedented reaction, an easily detectable product, and apparent efficiency (at least in C. alpina seeds), this enzyme should provide an excellent system for studying the biochemical mechanisms of membrane fatty acid desaturases. In studies of acetylenase gene expression in C. alpina, acetylenase mRNAs were unexpectedly found in non-seed tissues without corresponding accumulation of crepenynic acid. This prompted us to examine what limits crepenynic acid production to the seeds of C. alpina plants. Several candidate limiting factors were examined in the current work.
The first aim was to identify cyt b5 isoforms and to characterize their expression, because cyt b5 directly provides the electrons needed for desaturase reactions (Fig. 1). For the O2-dependent desaturation reaction, two externally provided electrons are consumed during the formation of each double (or triple) bond. Two different electron transport systems supply these reducing equivalents, one in the plastid (ferredoxin) and the other in the ER (cyt b5) (Shanklin and Cahoon, 1998). Electron transport is a known limiting factor in some plastid desaturase systems. Ferredoxin provides electrons to acyl-acyl carrier protein desaturases (‘soluble’ desaturases), and some isoforms of ferredoxin support higher activity when co-expressed with desaturases (Cahoon et al., 1996; Schultz et al., 2000). The soluble desaturases are evolutionarily unrelated to the FAD2 membrane desaturases (Shanklin and Cahoon, 1998), but their electron transport needs are similar. Four isoforms of cyt b5 from flower heads were identified, and the same four isoforms were the only ones detected in developing seeds.
Some cyt b5s seem to have specific roles. Tobacco contains two isoforms of cyt b5, including one (AN X80008 [GenBank] ) specifically expressed in seeds (Napier et al., 1995). The transcripts of two cyt b5 isoforms from Arabidopsis (AN AB007801 [GenBank] and AB007802 [GenBank] ) accumulated to lower levels in the silique (containing developing seeds) than in the other organs (root, leaf, stem, or flower) (Fukuchi-Mizutani et al., 1999). Expression analysis of olive cyt b5-15 (AN CAA04702 [GenBank] ) showed that low levels of mRNA were detected in tissues characterized by high rates of lipid accumulation, such as young expanding leaves, maturing seeds, and ripening mesocarp (Martsinkovskaya et al., 1999). Considering the important role of cyt b5 in fatty acid biosynthesis, the low expression levels of the Arabidopsis and olive cyt b5 genes in lipid-accumulating tissues suggest that these plants have other cyt b5 isoform(s) that are predominantly involved in the biosynthesis of storage lipids. Some desaturase systems have little cyt b5 specificity: Trypanosoma brucei oleate desaturase may even use a cyt b5-like domain in another desaturase as an electron donor (Petrini et al., 2004). However, no evidence was found for a seed- or acetylenase-specific form of C. alpina cyt b5. Even if other C. alpina cyt b5 isoforms are identified, the available evidence indicates that cyt b5 availability does not limit acetylenase activity in C. alpina.
Because the formation of crepenynate by acetylenase would be affected by FAD2 (supplying the acetylenase substrate, linoleic acid) and FAD3 (competing for linoleic acid), the second aim was to clone C. alpina FAD2 and FAD3 and characterize their expression. Substrate availability is a limiting factor for at least one variant FAD2 desaturase system, Arabidopsis transformants containing Crepis palaestina 12-epoxygenase (Rezzonico et al., 2004). The accumulation of epoxy fatty acids was increased by supplying the substrate, linoleic acid. Co-expression of C. palaestina 12-desaturase and 12-epoxygenase yielded a modest improvement in total epoxy fatty acid content in Arabidopsis (Singh et al., 2001).
In the cloned genes of C. alpina FAD2-2, FAD2-3, FAD3, and FAD7, the eight histidine residues required for membrane desaturase activity (Shanklin et al., 1994) are present. These residues are part of a HX(3–4)HX(7–41)HX(2–3)HHX(61–189)(H/Q)X(2–3)HH motif found in almost all membrane desaturases (Shanklin and Cahoon, 1998), and at least some are likely to be ligands to the active site iron atoms.
In C. alpina flower heads, strong expression of FAD3 and a linolenic acid:linoleic acid ratio higher than that in seeds were observed (see Table S2 in the Supplementary data at JXB online). These results suggest that linoleic acid, the acetylenase substrate, is mainly diverted into linolenic acid production (Fig. 4; Table 2). In developing seed, strong expression of both acetylenase and FAD2-2 was observed, consistent with efficient conversion of oleic acid to crepenynic acid (Fig. 4; Table 2). As expected, the developing seed linolenic:linoleic acid ratio is low, while the crepenynic acid:linoleic acid is high (see Table S2 in the Supplementary data at JXB online). In Fig. 5, the predicted fluxes through unsaturated fatty acid biosynthesis in developing seed and in the flower head are illustrated. This analysis requires two assumptions. (i) It is assumed that FAD3 and acetylenase are co-expressed in one or more of the various tissues that make up the flower head, so they actually compete for substrate. The higher linolenic:linoleic acid ratio in flower heads is consistent with FAD2 and FAD3 co-expression in the same cell. (ii) An assumption is made common to almost all transcription analyses that enzyme activity correlates at least roughly with transcript levels. Some membrane desaturases are post-translationally regulated (Heinemann et al., 2003; Tang et al., 2005).
|
The final limiting factor considered is the acetylenase gene expression level, detected by real-time RT-PCR. Acetylenase is expressed at an extremely high level in developing seeds (Table 2), which may be required to produce sufficient product. If crepenynic acid is produced in flower heads, it may be produced in amounts simply too small to detect, or it may be converted to other compounds.
Additional plausible limiting factors are an essential accessory protein, such as a substrate-modifying enzyme or a desaturase assembly protein. This possibility is being examined by EST-profiling several C. alpina tissues (J-W Nam, RK Wilson, and TJ Kappock, unpublished observations).
Relevance to heterologous desaturase expression
Acetylenase is unable to produce detectable crepenynate in C. alpina flower heads, apparently because of two synergistic effects: acetylenase transcript copy numbers and substrate levels are low, due to diversion into linolenic acid biosynthesis (Fig. 5). The inability of acetylenase expression to alter fatty acid composition in C. alpina flower heads is reminiscent of attempts to express variant FAD2s in transgenic plants outside seeds (Broun et al., 1998; Rezzonico et al., 2004). An unidentified factor may limit the accumulation of unusual fatty acids in foreign cells, including cells in different parts of the producer plant, if the unusual fatty acids interfere with other enzymes or disrupt membrane structure and function (Millar et al., 2000).
Even transgenic constructs with strong seed-specific promoters give disappointing yields of unusual fatty acids in transgenic plants, generally substantially below the level found in native plants (Voelker et al., 1996; Jaworski and Cahoon, 2003). These observations suggest that the components needed for high level unusual fatty acid production are missing: Suh et al. (2002) suggest that a necessary multi-component enzyme association fails to occur, while Jaworski and Cahoon (2003) suggest that it may be necessary to add genes that channel the fatty acid into triacylglycerols. Very high level gene expression also appears to be important.
Possible roles of localized FAD2 expression
Crepis alpina FAD2-related desaturases have unique expression patterns, with spatial and temporal specificity (Fig. 4). For instance, C. alpina FAD2-3 expression was only observed in flower heads. Flower head-specific FAD2 isoform expression has not been observed previously. In soybeans, FAD2-1 is strongly expressed in developing seeds and FAD2-2 is expressed in both vegetative tissues and developing seeds (Heppard et al., 1996). Arabidopsis FAD2 is expressed throughout the plant (Okuley et al., 1994). Olive FAD2-1 is expressed in leaves as well as seeds (Hernandez et al., 2005).
When FAD2 expression at different temperatures was examined, plants generally exhibited a significant increase in degree of fatty acid unsaturation at lower temperature; however, there is no apparent increase in the rate of transcription or the stability of fatty acid desaturase mRNAs at lower temperatures (Okuley et al., 1994; Heppard et al., 1996), apart from the Arabidopsis FAD8 gene (Gibson et al., 1994). The present results show a similar pattern: FAD2 expression levels are either similar in plants grown at 10 °C or 21 °C, or slightly higher at 21 °C. FAD2-2 expression was higher at 21 °C than 10 °C. FAD2-3 expression, observed only in flower head, was higher at 21 °C than 10 °C. Tang et al. (2005) examined the temperature dependence of protein stability for two soybean FAD2 isoforms expressed in yeast. One isoform (FAD2-1A) showed enhanced protein degradation at high growth temperatures, due in part to a destabilizing portion of its sequence. These results imply opposite temperature responses for FAD2 isoform activities in C. alpina flowers and in soybeans, and highlight the likely importance of post-translational regulatory mechanisms in the functional expression of desaturases.
Acetylenase gene expression in flower heads is higher in plants grown at 10 °C, an interesting exception among the C. alpina desaturases studied here. Any change in lipid chemical properties induced by the acetylenase reaction would seem to be counterintuitive for low-temperature induction: where cis-double bonds introduce ‘kinks’ into a fatty acid chain and enhance membrane fluidity by decreasing the efficiency of intramolecular packing, acetylenic moieties are rigid and relatively straight, comparable with saturated fatty acids or fatty acids with trans-double bonds. It is not known if acetylenase gene expression affects lipid or membrane properties in flower heads.
It will be necessary to determine which parts of the flower heads express FAD2-2 and FAD2-3 to understand why these genes are transcribed at 21 °C but not 10 °C. FAD2-2 and FAD2-3, together with FAD3, seem to be actively involved in linolenic acid production in flower heads at 21 °C. The limited expression of FAD2-3 in flower heads might regulate linolenic acid biosynthesis. Linolenic acid plays an important role as a precursor of jasmonic acid, which regulates several critical steps in pollen development (McConn and Browse, 1996). Crepis alpina plants grown at 21 °C include functional reproductive organs and produce seeds, while plants grown at 10 °C only sporadically produce seeds.
Why is acetylenase expressed in flower heads, when it is not expressed in other tissues (e.g. leaves)? Acetylenase expressed in flower heads may have a different function: C. alpina acetylenase can convert oleic acid into a mixture of 18:2 isomers (9c,12c and 9c,12t) as a minor reaction (Carlsson et al., 2004). The analytical methods used here are not expected to discriminate between these isomers.
Acetylenase expression in Helianthus annuus
Many different plants contain an acetylenase gene (a variant form of FAD2) (Cahoon et al., 2003). Gene expression was induced by treatment with a fungal elicitor in developing seeds of H. annuus. However, the level of mRNA expression was not correlated with acetylenic fatty acid production or acetylenase protein accumulation.
In this study, acetylenase gene expression was examined in young H. annuus seedlings. As anticipated (Cahoon et al., 2003), acetylenase transcription is not associated with crepenynic acid accumulation in seedlings. Real-time quantitative RT-PCR revealed relatively abundant acetylenase gene transcripts in H. annuus seedlings (Table 3), actually higher than the levels observed in C. alpina flower heads. The expression of H. annuus acetylenase outside seed (and without fungal elicitation) was unexpected, considering that the plant does not accumulate crepenynic acid, even in seed tissues. This cryptic acetylenase expression is analogous to what was observed in C. alpina flower heads.
Although acetylenase expression in non-seed tissues is observed in both C. alpina and H. annuus, the expression level in young plants is much higher (
104 fold) in the latter (Tables 2, 3). Considering that polyacetylenic compounds are formed in H. annuus roots (Bohlmann et al., 1973), it should contain enzymes that use crepenynic acid and its derivatives as substrates. [While crepenynic acid is a precursor for (poly)acetylenes and probably other compounds, many of these ‘downstream’ biosynthetic genes have yet to be identified (Cahoon et al., 2003).] Any crepenynic acid formed in non-root H. annuus tissues may be consumed by an uncharacterized biosynthetic pathway, which would account for the acetylenase expression observed. No such downstream pathway is known in C. alpina seeds, which accumulate high levels of crepenynic acid. Therefore, it appears that the two plant acetylenase systems are regulated differently, and employ different mechanisms to control acetylenic compound accumulation: C. alpina acetylenase function appears to be regulated by suppressing transcript copy number and substrate availability (substrate level regulation), while in H. annuus any crepenynic acid produced would be consumed (product level regulation). Other plants known to contain variant FAD2 acetylenases (parsley, English ivy, and Calendula) (Cahoon et al., 2003) might also show cryptic acetylenase expression.
|
Supplementary data |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
The following supplementary data can be found at JXB online.
Figure S1. Deduced C. alpina cyt b5 amino acid sequences and alignment using Clustal W.
Figure S2. Unrooted Neighbor–Joining phylogenetic comparison of the cyt b5 isoforms
characterized in this study with cyt b5s from other systems.
Figure S3. Deduced amino acid sequences of C. alpina acetylenase and two types of FAD2, and their alignment using Clustal W.
Figure S4. Unrooted Neighbor–Joining phylogenetic comparison of acetylenase, FAD2-2, and FAD2-3 from C. alpina with members of FAD2 family of enzymes.
Figure S5. Deduced amino acid sequences of FAD3 and FAD7 and their alignment.
Figure S6. Unrooted Neighbor–Joining phylogenies of CaFAD3 and CaFAD7 with other FAD3/7 enzymes.
Table S1. Oligodeoxynucleotides used in this study.
Table S2. Levels of selected fatty acids in C. alpina tissues measured by gas chromatography.
|
Acknowledgements |
|---|
We thank E Cahoon for kind assistance, expert advice, and access to GC equipment. We thank A Ransome for excellent technical assistance. We thank E Mebrahtu, J Roose, J Francois, and H Marella for assistance, H Yi, S Preuss, and Y You for advice, and RS Quatrano for the use of equipment. We are grateful for support of this work by the Monsanto/Washington University Plant Science Agreement.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionSupplementary dataReferences |
|---|
Bohlmann F, Burkhardt T, Zdero C. (1973) Naturally occurring acetylenesLondon Academic Press.
Broun P, Boddupalli S, Somerville C. (1998) A bifunctional oleate 12-hydroxylase:desaturase from Lesquerella fendleri. The Plant Journal 13 201–210.[CrossRef][ISI][Medline]
Bu'Lock JD and Smith GN. (1967) The origin of naturally-occurring acetylenes. Journal of the Chemical Society (C) 332–336.
Cahoon EB, Carlson TJ, Ripp KG, Schweiger BJ, Cook GA, Hall SE, Kinney AJ. (1999) Biosynthetic origin of conjugated double bonds: production of fatty acid components of high-value drying oils in transgenic soybean embryos. Proceedings of the National Academy of Sciences, USA 96 12935–12940.
Cahoon EB, Mills LA, Shanklin J. (1996) Modification of the fatty acid composition of Escherichia coli by coexpression of a plant acyl-acyl carrier protein desaturase and ferredoxin. Journal of Bacteriology 178 936–939.
Cahoon EB, Schnurr JA, Huffman EA, Minto RE. (2003) Fungal responsive fatty acid acetylenases occur widely in evolutionarily distant plant families. The Plant Journal 34 671–683.[CrossRef][ISI][Medline]
Carlsson AS, Thomaeus S, Hamberg M, Stymne S. (2004) Properties of two multifunctional plant fatty acid acetylenase/desaturase enzymes. European Journal of Biochemistry 271 2991–2997.[ISI][Medline]
Emanuelsson O, Nielsen H, von Heijne G. (1999) ChloroP, a neural network-based method for predicting chloroplast transit peptides and their cleavage sites. Protein Science 8 978–984.[Abstract]
Ewing B and Green P. (1998) Base-calling of automated sequencer traces using Phred. II. Error probabilities. Genome Research 8 186–194.
Ewing B, Hillier L, Wendl MC, Green P. (1998) Base-calling of automated sequencer traces using Phred. I. Accuracy assessment. Genome Research 8 175–185.
Felsenstein J. (1989) PHYLIP – Phylogeny Inference Package. Cladistics 5 164–166.
Fukuchi-Mizutani M, Mizutani M, Tanaka Y, Kusumi T, Ohta D. (1999) Microsomal electron transfer in higher plants: cloning and heterologous expression of NADH-cytochrome b5 reductase from Arabidopsis. Plant Physiology 119 353–362.
Gibson S, Arondel V, Iba K, Somerville C. (1994) Cloning of a temperature-regulated gene encoding a chloroplast omega-3 desaturase from Arabidopsis thaliana. Plant Physiology 106 1615–1621.[Abstract]
Gordon D, Abajian C, Green P. (1998) Consed: a graphical tool for sequence finishing. Genome Research 8 195–202.
Gouet P, Courcelle E, Stuart DI, Metoz F. (1999) ESPript: analysis of multiple sequence alignments in PostScript. Bioinformatics 15 305–308.
Heinemann FS, Mziaut H, Korza G, Ozols J. (2003) A microsomal endopeptidase from liver that preferentially degrades stearoyl-CoA desaturase. Biochemistry 42 6929–6937.[CrossRef][Medline]
Heppard EP, Kinney AJ, Stecca KL, Miao GH. (1996) Developmental and growth temperature regulation of two different microsomal -6 desaturase genes in soybeans. Plant Physiology 110 311–319.[Abstract]
Hernandez ML, Mancha M, Martinez-Rivas JM. (2005) Molecular cloning and characterization of genes encoding two microsomal oleate desaturases (FAD2) from olive. Phytochemistry 66 1417–1426.[CrossRef][ISI][Medline]
Jaworski J and Cahoon EB. (2003) Industrial oils from transgenic plants. Current Opinion in Plant Biology 6 178–184.[CrossRef][ISI][Medline]
Lee M, Lenman M, Banas A, et al. (1998) Identification of non-heme diiron proteins that catalyze triple bond and epoxy group formation. Science 280 915–918.
Mathews FS. (1985) The structure, function and evolution of cytochromes. Progress in Biophysics and Molecular Biology 45 1–56.[CrossRef][ISI][Medline]
Martsinkovskaya AI, Poghosyan ZP, Haralampidis K, Murphy DJ, Hatzopoulos P. (1999) Temporal and spatial gene expression of cytochrome B5 during flower and fruit development in olives. Plant Molecular Biology 40 79–90.[CrossRef][ISI][Medline]
McConn M and Browse J. (1996) The critical requirement for linolenic acid is pollen development, not photosynthesis, in an Arabidopsis mutant. The Plant Cell 8 403–416.[Abstract]
Millar AA, Smith MA, Kunst L. (2000) All fatty acids are not equal: discrimination in plant membrane lipids. Trends in Plant Science 5 95–101.[CrossRef][ISI][Medline]