Journal of Oil Palm Research (Special Issue) p. 1-l 7

CHARACTERIZA-TION AND

REGULATION OFTHE OIL PALM

: (Elaeis guineensis)STEAROYL-ACPDESATURASE

GENES

Keywords: Elaeis guineensis, stearoyl-ACPdesaturase and gene regulation.

SITI NOR AKMAR, A’; CHEAH, S c’;AMINAH, S’; LESLIE, C L 0’;SAMBANTHAMURTHI, R’ and

MURPHY, D J**

* Palm Oil Research lnstihrte of Malaysia, P.O. Box 10620.

60720 Ku& Lumpur. Malaysia.

“Depar!mnenl of Brassica and Oilseeds Re*earch. John lnnesCentre. Norwich Research Pah. Norwich NR4 7UH. UK.

1

UIO different stenroyl-ACP desaturase

1 genes are expressed in the mesocarp

of oil palm (E. guineensis) fruits.

Their nucleotide sequences share 93% and 76%

homologies within the coding and 3’ untranslated

regions, respectively. Southern blot analysis

showed that there are at least two copies of

stearoyl-ACP desaturase genes in the oil palm

genome. Northern blot analysis using gene

specific probes showed that the two genes from

the mesocarp are differentially regulated. One

gene is constitutiuely expressed with a high

leuel of expression in uarious oil palm tissues.

The othergene was found to be developmentally

regulated with its expression pattern correlating

with the pattern of oil synthesis in both the

mesocarp and kernel tissues. Polyclonal anti-

bodies were raised againts a peptide containing

the N-terminal sequence of the mature oil palm

stearoyl-ACP desaturase. The antibodies recog-

nized the predicted size protein of 37kDa in total

protein fractions of oil palm mesocarp after

separation on denaturing polyacrylamide gels.

Western blot analysis using the polyclonal

antibodies showed the enzyme level is high in

the mesocarp at late stage of ripening and

remains high in ripe fruits. The leaf form of the

enzyme appeared to be about 2kDa larger. High

levels of enzyme and gene expression were

detected in young mesocarp tissue consistent

with the requirement for high levels of unsatu-

rated fatty acids for membrane lipid biosyn-

thesis.

INTRODUCTION

I n plants, stearoyl-ACP desaturase is animportant fatty acid biosynthetic enzyme

responsible for the production of oleic acid. Itis a soluble enzyme in the plastid which intro-duces a cis double bond into saturated stearoyl-ACP (18:0-ACP) at the A9 position to producemonounsaturated oleoyl-ACP (18:1-ACP). Itscatalytic activity requires 02, NAD(P)H,NAD(P)H ferredoxin oxido-reductase andferredoxin (Nagai and Bloch, 1968). Thisenzyme is found in all plant tissues, where ithas a vital housekeeping role for producingunsaturated fatty acids required in membranelipid biosynthesis. In oil accumulating tissueslike anthers, seeds and mesocarp, it is alsoinvolved in the developmentally regulatedprocess of storage lipid biosynthesis (Murphyand Piffanelli, 19981. Olive oil and palm oil, forexample, contain high levels of oleic acid whichhas a great influence on the physical andnutritional properties of these vegetable oils.These properties are important in determiningtheir various applications as food product. Oleicacid is also a valuable feedstock for the rapidlygrowing oleochemical industry which uses plantoils as renewable resources. Oleic acid and itschemical derivatives are required for the pro-duction of high value-added products such ascosmetics, pharmaceutics and polymers (Prydeand Rothrus, 19891.

cDNA clones encoding stearoyl -ACPdesaturase have been isolated from variousplant species such as castor (Shanklin andSomerville, 19911, rape (Slocombe et al., 1992)and recently, the oil palm (Shah and Rashid,1996). The cDNAs encode precursor proteinscontaining N-terminal leader peptide requiredfor targeting to the plastid. The sequence of themature protein is highly conserved with greaterthan 80% identity between different plantspecies. The enzyme is a homodimer with asubunit mass of approximately 37kDA (Shanklinand Somerville, 1991; Thompson er al., 1991).A great advance in understanding the chemicalmechanism of double bond insertion wasachieved with the recent availability of thex-ray crystallographic structure of recombinanthomodimeric castor stearoyl-ACP desaturase

JOURNAL OF OIL PALM RESEARCH (SPECIAL ISSUE)

2

(Lindqvist et al., 1996). The three-dimensionalstructure serves as a model for analysing aminoacid residues required in catalysis and electrontransport for stearoyl-ACP desaturase and otherclosely related soluble plant desaturases. Fur-thermore, the overall conformation and, par-ticularly, the orientation of the iron-bindingresidues of this desaturase, are remarkablysimilar to those of microbial monooxygenasesand ribonucleotide reductases, implying thatall of these proteins form a related group ofdiiron-oxo enzymes with similar reaction mecha-nisms (see Murphy and Piffanelli, 1998 for arecent review).

The current oil palm genetic engineeringprogramme is aimed at increasing the level ofoleic acid in palm oil at the expense of palmiticacid (16:O). One strategy of genetic manipula-tion is to antisense palmitoyl-ACP thioesteraseand increase the expression of 8-ketoacyl-ACPsynthetase II in order to reduce palmitic pro-duction and channelling towards 18 carbon acylchains (Cheah et nl., 1995). Manipulation of thestearoyl-ACP desaturase gene may also berequired to cope with the possible accumulationof stearic acid so that more oleic acid can beproduced. Furthermore, there is also an inter-est in producing high stearate palm oil whichis achievable by introducing an antisense con-struct of the stearoyl-ACP desaturase gene ashas been reported in rape (Knutson et al.,1992). An understanding of the structure andregulation of stearoyl-ACP desaturase gene inthe oil palm would provide valuable back-ground knowledge for any genetic manipulationeffort. Such information is essential for alteringthe fatty acid composition of the storage oil withminimal perturbation to membrane lipids whichcan otherwise lead to detrimental agronomiceffect on the transgenic plants produced(Thompson and Li, .1996).

The aim of this study is to identify differentstearoyl-ACP desaturase genes from oil palmmesocarp and to study their regulation indifferent tissues with special emphasis on theoil-bearing mesocarp and kernel tissues. Inves-tigations were performed at both transcrip-tional and post-transcriptional levels by study-ing gene expression patterns and monitoringthe enzyme levels in these tissues.

ClUFNCTERlZA~ A+ REGULATION OF l+iE OIL PALM (Ekeis guineensis) STEAROYL-ACP DESATURASE GENES

MATERIALS AND METHODS

plant Materials

Different tissues horn E. guineensis. tenem(dum x pi&fern) variety were used. Inhres-cencea were tagged at anthesis and the fruitbunches were harvested at various weeks afteranthesis (WAA) for RNA and protein extrac-tied from the mesocarp and kernel tissues.Lea&s from 3-4 months polybag seedlings wereobtained for RNA and protein preparations.Germinated seedlings after one week undergo-ing germination process were used for RNAsxtraction.

Sreening of cDNA Library for Oil PalmStearoyl-ACP Desaturase Gene

Screening of a 15-week oil palm mesocarplambda ZAP II cDNA library was carried outbased on the methods described in Sambrooket al. (1989). Duplicate plaque lifts consistingof 200 000 recombinant clones were preparedfor screening using a heterologous probe, pRCD1(fall length stearoyl-ACP desaturase clone fromcastor). About 200 000 recombinant clones wereplated for isolating the full length clone usingpartial oil palm stearoyl-ACP desaturase cDNAclone as probe.

cDNA inserts to be used as probes werepurified and subsequently labelled using theMegaprime DNA labelling system fromAmersham. Hybridization of the probes tomembranes was carried out in hybridizationbuffer containing 5X SSPE (1X SSPE is0.18M NaCl, 1OmM NaHzPO,, pH 7.5, 1mMEDTA), 5X Denhardt’s solution (1X Denhardt’ssolution is 0.02% each Ficoll 400, bovineserum albumin, and polyvinylpyrrolidone),0.5% SDS, 1OOmg ml-*denatured salmon spermDNA, and 1X 106 - 5X 106cpm ml-i probes. Aflerthe hybridization using heterologous probes,the membranes were washed twice in 2X SSPE,0.1% SDS at 65°C for 15min. When oil palmprobes were used, an additional wash in0.1x SSPE, 0.1% SDS at 65°C for 20minwas carried out. The membranes were subse-quently exposed to x-ray films for 48hr at-80°C.

Sequence Analysis

Insert-carrying pBluescript SK-plasm&were in uiuo excised from lambda ZAP IIfollowing the method described by the supplier(Stratagene). Three and two overlappingsubclones were produced in order to get thecomplete nucleotide sequences of the classes 1and 2 oil palm stearoyl-ACP desaturase gene,respectively. Phagemid DNA for sequencing wasextracted using the Qiagen plasmid mini kit.DNA sequencing was carried out from bothdirections using the Automated Laser Fluores-cent (A.L.F.) sequencer (Pharmacia). TheDNASIS Sequence Analysis Software was usedfor sequence analysis and search for similaritybetween nucleotide and amino acid sequences.

Southern Blot Analysis

Cenomic DNAwas extracted fiorn leaf tissuefollowing the method of Dellporta et al. (1983).Digestion with restriction enzymes was carriedout according to the manufacturer’s instruc-tions. Electrophoresis of digested samples wasperformed using 0.9% agarose gel in TAE buffer(40mM Tris-acetate pH 7.9, 1mM EDTA). Fol-lowing electrophoresis, the gel was gently shakenin 0.25M HCl for 1Omin for depurination priorto transfer. The DNA was denatured duringovernight transfer onto Hybond N+ (Amersham)membrane under alkaline conditions using0.4M NaOH as the transfer buffer. The mem-brane~ was rinsed in 2X SSPE before continuingwith prehybridization.

Hybridization and preparation of labelledprobes were carried out as described above. Formedium stringency washes, the membrane waswashed twice in 4X SSPE and 0.1X SDS at50°C for 15min. The membrane was exposedto Kodak X-OMAT x-ray film for one week.

Northern Blot Analysis

Messenger RNA from various oil plam tis-sues were prepared as described in Siti NorAkmar et al. (1994). Four microlitres of mRNA(0.5pg nl1-1) was denatured in l&.11 of solutioncontaining 78% (v/v) deionized formamide,16% Wv) deionized glyoxal, and 1OmM NaHSOJ

3

NazHPOd (pH 7.0) by heating for 15min at 55°Cfollowed by immediate cooling. Denatured mRNAwas separated on 1.2% agarose gel using 40mMTris-acetate (pH 7.0) as electrophoresis buffer.Transfer to nylon membrane (Hybond-NAmersham) was carried out using a vacuumblotter (60cm HzO, 4hr) in 20X SSC (1X SSC is0.15M NaC1, 15mM t&odium citrate 2Hz0,pH7.0X Hybridization and preparation of label-led probes were carried out as described above.

Western Blot Analysis

Protein was extracted by homogenizing about1Og of sliced oil palm tissues in lOOmI of coldextraction buffer (40mM NaH2P0,, 60mMNaHPO,, O.lmM PMSF and 2mM dithiothreitol,pH ‘7.0). The supernatant was recovered bycentrifugation at 1OOOxg. Polyvinylpyrrolidonewas added until the solution became viscousand the mixture kept at 4°C overnight. Themixture was recentrifuged and the supernatantfiltered through glass wool. The clear solutionrecovered was vacuum dried. Protein concen-tration was determined based on the methodof Bradford (1976). Separation of total solubleprotein extract on denaturing polyacrylamidegel was carried out as described by Laemmli(1972). Transfer to nitrocellulose Hybond-Cmembrane was performed using transfer buffer(25mM Tris pH 8.3, 192mM glycine, 20%(v/v) methanol 0.02% (w/v) SDS at 4°C for3hr using a transblot apparatus (Biorad, USA).The membrane was blocked with 3% (w/v) drymilk powder in TBS (10mM Tris, 150mMNaCl, pH 7.4). Immunodetection of proteinswas performed using 1:lOOO dilution of thepolyclonal antibodies raised in rabbit againstmultiantigenic peptide (MAP) synthesizedusing a 14-mer N-terminal sequence of themature oil palm stearoyl-ACP desaturase. Ananti-rabbit IgG conjugated alkaline phosphatasewas used as secondary antibody. Colour deve-lopment was performed using NBT/BCIPsubstrate (Western Blue; Promega, USA) ac-cording to the manufacturer’s instructions.

MAP resin peptide synthesis (AppliedBiosystem) columns were used for synthesizingthe peptide antigen following the manufactur-er’s instructions. Two milligrams per millilitrepurified MAP in 50% (v/v) Freund’s complete

adjuvant were used in the first rabbit injection.Subsequent boostings were performed using2mg ml-1 MAP in 50% (v/v) Freund’s incompleteadjuvant. Antiserum was collected two weeksafter boosting. I

RESULTS AND DISCUSSIONI

Sequence Analysis of cDNA ClonesI

A full length stearoyl-ACP desaturase clonefrom castor (Shanklin and Somerville, 1991)was used as a heterologous probe to screen200 000 clones from a &week oil mesocarpcDNA library (Siti Nor Akmaret al., 1995). Thisresulted in the isolation of five partial geneswhose nucleotide sequences had greater than78% homology with the castor sequences. Based 1on their nucleotide sequences, the clones can !be divided into two classes, three in the firstand two in the second. The clones with thelongest insert in classes 1 and 2 were design- :ated pOPSN15 and pOPSN16, respectively. iThe overall sequence identity between pOPSN15and pOPSN16 was 86% with 93% identitywithin the coding and 76% identity within the3’ untranslated regions. pOPSN15 which hada longer insert of 1350kb, was used as a probein screening the same oil palm mesocarp cDNAlibrary. Out of the 40 putative clones isolated,three clones, designated pOPSN18, 19 and 20contained the longest inserts of approximately1.7kb. The three clones were identical andbelonged to class 1. They encoded a protein of393 amino acids. Their 3’ untranslated regionswere also identical with the exception ofpOPSN18 which contained 42 nucleotides lessjust before its poly (A)+ tail. pOPSN19 had thelongest 5’ untranslated region. These class Iclones were strongly homologous to the re-ported oil palm stearoyl-ACP desaturase se-quence (Shah and Rashid, 1996) with 99%identity at both the nucleotide and amino acidlevels. They are believed to be the same as thereported gene and sequencing errors haveprobably contributed to the slight differencesobserved.

When the nucleotide sequence of class 1,pOPSN19 was aligned with class 2, pOPSN16and translated in the same frame, it wasobserved that the identity between the encoded

ar

ofa ta ldatlt1II00ttc

JOURNAL OF OIL PALM RESEARCH (SPECIAL ISSUE)

I

c!

4

CHARACTERIZATION AND REGULATION OF THE OIL PALM (Elaeis guineensis) STEAROYL-ACP DESATURASE GENES

amino acids was about 95% (Figure 1). Severalof the variant nucleotides in pOPSN16 occurredat the third position of the codons without anyalteration to the encoded amino acids. Thededuced amino acid sequence of pOPSN19 hastwo occurrences of the D/E EXzH motif. Eventhough the sequence encoded by pOPSN16lacks 168 N-terminal amino acids, it has fiveout of six of these conserved residues lackingonly the conserved N-terminal E residue. Crys-tal’structure of stearoyl-ACP desaturase showedthat each enzyme subunit consists a diironcentre which forms the active site. The con-served E and H residues in the motif areinvolved in forming ligands with the diironcentre (Lindqvist et al., 1996).

Comparison of the deduced amino acidsequence of class 1 with the known sequencesof higher plant acyl-ACP desaturase genesshowed that it contains the complete sequenceof the chloroplast transit and mature peptides(Figure 2). Like sequences of most other stearoyl-ACP desaturases, the oil palm sequence con-tains the MAST sequence surrounding thecleavage site of the transit peptide (Piffanelli,1997). The proposed cleavage site is betweenthe methionine and alanine residues, removing31 N-terminal amino acids. The deduced aminoacid sequence of the mature peptide was highlyhomologous to stearoyl-ACP desaturase fromdiverse plant species with 86%, 83% and 81%identities with castor (Shanklin and Somerville,1991), rice (Akagi et al., 1995) and rape(Slocombe et al., 1992) sequences, respectively.If conservative substitutions are considered,the homology increases to 94%, 92% and 91%,respectively.

More recently, a few genes correspondingto variant soluble plant acyl-ACP desaturases,such as the A4-palmitoyl(16:O).ACP desaturaseof coriander (Coriandrum satiuum) seeds(Cahoon el al., 1992) and A9-16:0-ACP desa-turase of milkweed (Axlepius syriaca) seed(Cahoon et al., 1997) have been isolated fromlipid accumulating tissues. These enzymes acton different chain length saturated acyl-ACPsubstrates and some insert the double bond atdifferent position from stearoyl-ACP desaturase.Interestingly, their amino acid sequences havethe same conserved motif and share substantialoverall sequence homology of 60%-70% with A9

stearoyl-ACP desaturases from various plants.Similarly the oil palm sequence was found toshare considerable homology with about 66%identities with both the A4-16:0-ACP desaturasefrom coriander and A9-16:0-ACP desaturasefrom milkweed (Figure 2). All of these solubleplant desaturases are believed to be evolutiona-rily related to stearoyl-ACP desaturases. How-ever, the soluble desaturases form a separategroup from the putative membrance-boundplant stearoyl-CoA desaturase recently identi-fied in rose (Rosa hybrida) by Mizutani et al.(1995) which has very little sequence similarityas shown in the cluster alignment (Figure 3).The two oil palm sequences are closely relatedand they cluster together with the stronglyhomologous stearoyl-ACP desaturase sequencesfrom castor and the other monocot species, rice.

Gene Copy Number Determination

The entire insert containing class 1 gene(pOPSN19) was used as a probe in Southernanalysis to enable detection of closely relatedgenes with substantial homology in the codingregion. The hybridization signal was abolishedwith high stringency washing and, therefore,medium stringency washing was performedafter hybridization. Two bands were observedin lanes 1, 2 and 3 where the oil palm genomicDNA was digested with EcoRI, Hind III andXba I, respectively (Figure 4). One of the EcoRIand the Xba I bands was slightly more intensethan the second. This more intense band couldpossibly correspond to the class 1 gene and theless intense one to the class 2 gene. As a whole,this result suggests that there are at least twocopies of the stearoyl-ACP desaturase gene inthe oil palm genome.

Gene Expression Patterns

Gene-specific probes of approximately 300bpwere designed based on the 3’ untranslatedregions of classes 1 and 2 genes, excludingabout 70 of the 3’ terminal nucleotides near thepolyadenylation site. These were used to probeNorthern blots containing mRNA from sixdifferent stages of mesocarp (B-20 WAA), threedifferent stages of kernel (lo-14 WAA) deve-lopment and from vegetative tissues using high

5

XJRNAL OF OIL PALM RESEARCH (SPECIAL ISSUE)

T s w A ” w T R A W T A E E*N R H*G 0 I.

60

240

420

660

960

CHARACTERIZATION AND REGULATION OF THE OIL PALM (Elaeis guineensis) STEAROYL-ACP DESATURASE GENES

C T V L S F A DHMKKKIsH P A H L961 CGTACGGTGCTTTCTTTTGCTGACATGATG~G~G~GATCTCAATGCCTGCCCATCTG

Illll I I I I I I IIIIIIIIIIIIIIIIIIIIlIIIIIIlI IIIIIIIIIIIIIIIGGTACTGTTCTTGCCTTTGCTGACATGATGAAGAAGAAGATCTCGATGCCTGCCCATCTGM Y D G Q D D N L F E H F S A V A Q R L

1021 ATGTATGATGGTCAGGATGACAACCTCTCTTCGAGCACTTCTCGGCAGTAGCCCAGCGTTTG

I I I I I lIIIIIIIIIIIII IIlIIIIIIIIIIIIIIIII I I I I I IIIllIIIIIIIATGTACGATGGTCAGGATGATAACCTCTTCGAGCACTTCTCAGCAGTGGCCCAGCGTTTGG V Y T A K D Y A D I L E F L I N R W K

1081 GGTGTCTACACAGCCAAGGACTATGCCGACATACTTGAGTTCCTTATT~TAGGTGG~

I I I I I I I I I I Illlllllllllll IIIIllIIIIIIIIIII I I IlIlIIIIIIIGGTGTGTACACCGCCAAGGACTATGCTGACATACTTGAGTTCCTCCTTGATAGGTGG~V G E L T G F S G E G K R A Q D F V C T

1141 GTGGGTGAGTTAACCGGCTTCTCTGGGGAAGGTAAGAGAGCCCAGGACTTCGTCTGCACT

lllll I I I III1 I I I IIIIIIIIIIIIIIIIIII IIIIIIIIIII I I I I I I I I IGTGGGGGAGCTAACTGGCCTCTCTGGGGAAGGTAAGAGGTGAGGGCCCAGGACTTTGTCTGCACTLAPBIRRIEE R A Q E R A K Q A P

1201 CTTGCTCCCAGGATCAGGAGGATTGRAGAAAGAGCTCAGGAAAGGGCCAAGCAAGCACCA

lIIIIIIIIIlIIIIIIIIII III1 IlIIIIIIIIIII IIIIllIIIIIIIlIIIIlCTTGCTCCCAGGATCAGGAGGCTTGAGGAAAGAGCTCAGGGAAGGGGCCAAGCRAGCACCAR I P F S W I YGREVQL*

1261 CGCATACCTTTCAGCTGGATCTATGGCCGGGAAGTGCAACTCTGAGCATGACCAATATCT

I I I I I I I I I I I I IIIIIIIIIIIIIIIIIIII IIIIIIIIIIIII I I I ICGCTTTCCTTTCAGTTGGATCTATGGCCGGGRAGTACAAC~CTCTGAGCAT~TG~TGATA

1321 GTTTTGGTTTAAATACCAGTTCTTCAGGGTTGGCCAAAATTTTGCATCTGCATG..TATC

I I I I I I I I IIII I I I I I I I I I I I I I I I I I I I I I I I I I IIIIIIIIII I I IGTTTTGATTAAAATGGCAGTTCTTCTGGGTTGGCAARAATTTTACATCTGCATGTCTGTC

1381 AGCTATTTATAGGGTTTTGACGATTAGCTTAAAATCCATATATGCTTTGTTAGTAGG

111111111111111111111 I IIIIIII IIIIII I I IIIIIII I IITGCTATTTATAGGGTTTTGAG~~GCTTAAGATCCTGGG

1441 TCAGGTCAGGATGGGGAAAGCCGTAGATACTCTTTTAAGCTGTCACAGA..TCATCTTAT

I I IIIIIIIIIIII I II III II II II II IIIIIIIIITGA.GTCAGGATGGGGGAGGCTGTA . . . . . . . . . ..AAACTATCGTCGATGTCATCTTAT

1501 GAGCAGGTGAGGGATGCTGTTGAAGTGGTGTC..CAGTTGGAAACCTTCCTGTTTCATG

1111111111111 IIIIIIIIIIIIIIIIII I I I 11111111111111 IIIIGAGCAGGTGAGGGCTGCTGTTGAAGTGGTGTCTGGTGTTATAAACCTTCCTGTTTTATG

1561 TTTTCTTTCGTGTATTAGTACTCCTATGTAGGAAAATGGGCTGATTACTGTATCTGGAAC

I I II I III II IIIIIIIIIII IIIIIII IIIII IIlIIIIIIlITCTCTTTCATTTTAT..GTGCTCCTATGTAGCRARATGGCCTGATATACGTATCTGG~C

1621 CTAGTTCTTGCTCGCTTTTTTCAATGTCTGTATGCTTCATCTGGAATGTCTARATTTCAG

llllllllI I IIIIII IIIII I IIIIIII I llllllI III ITTAGTTCTTGTTATCTTTTTGAAATGTTTTTATGTTGCT

1681 T C T A T C A A T C T T A A G A T A T G T A A G G G G A T T A A T G T T A A A A 1 7 3 4

1020

1080

1140

1200

1260

1320

1 3 8 0

1 4 4 0

1 5 0 0

1 5 6 0

1 6 2 0

1 6 8 0

Figure 1. Nucleotide and deduced amino acid sequences of the fill length clone, pOPSNl9 (class I) andthe nucleotide sequence of the partial gene, pOPSNI6 fclass 2). Identical nucleotides between pOPSNl9 andpOPSNI6 are shown with matching lines. Dots have been introduced to optimize alignment. Proposed cleavagesite for plastid transit peptide is indicated by a downward arrow. The conserved E and H residues in thehigher plant acyl-ACP motif are indicated by asterisks.

arabidops176

rape174 "

oilpalm168

rice165

01 castor171

coriander159

milkweed157

arabidops236

rape234

oilpalm228

rice225

castor231

coriander219

milkweed217

CHARACTERIZATION AND REGULATION OF THE OIL PALM (E/ads guineensis) STEAROYL-ACP DESATURASE GENES

rmnoumm

~:~:::~

0: : : : : :

IY. ..lc.

IL.a

?r:. .t]

I

JOURNAL OF OIL PALM RESEARCH (SPECIAL ISSUE)

SOLUBLE

- Arabidopsis 1

- r a p e

-[1oil palm 1

oil palm 2

_ castor

- rice

A4C16coriander

A9C16milkweed

MEMBRANE-BOUND

Al2C18:lArabidopsis 2

A9C18-CoArose

Figure 3. Cluster alignment of the mature proteins of higher plant acyl-ACP desaturases and membrane ~bound desaturases. The sequences are soluble A!+stearoyl (IS:OI-ACP desaturase from A. thaliana, rape, oilpalm (class 1). oil palm (class 2), rice, castor, A4-palmitoyl (16:0)-ACP desaturase from coriander and~.916:0ACP desaturase from milkweed. The membrane-bound desaturases are putative plant stearoyl-CoAdesaturase from rose (Rosa hybrida) and A12-oleoyl (18:l) desaturase from A. thaliana.

1 0

12 3

Figure 4. Southern blot analysis of stearoyl-ACPdesaturase gem in the oil palm. Lanes 1, 2 and 3represent 1Opg E. guineensis genomic DNA digestedwith EcoRI, Hind III and Xba I, respectively afterseparation on 0.9% agarose gel, and probing with32P-labelled entire insert of pOPSNl9.

stringency conditions. It was shown that thetwo probes, each hybridized specifically to tran-scripts of about 1.7kb. The expression patternof class 1 was shown to be different from class2 (Figure 5). Class 2 is highly expressed andat approximately the same levels in theE. guineensis leaves, kernel at 12 and 14 WA4and mesocarp at the different stages of deve-lopment with the exception of 17-week mesocarpwhere the expression was approximately three-fold higher. Interestingly, negligible expressionlevels were detected in lo-week kernel andgerminated seedlings.

On the other hand, the expression of class1 was found to be much lower in general andappeared to be developmentally regulated. Thehighest level of expression in the mesocarptissue was around 15 WAA while the expressionlevel in ripe fruits was significantly reduced. Ahigh level of expression was also detected inyoung mesocarp tissue at eight WAA. The genewas expressed at very low levels in kernel

CHARACTERIZATION AND REGULATION OF THE OIL PALM (E/ads guineensis) STEAROYL-ACP DESATURASE GENES

tissues between 10 WAA and 14 WAA with themaximum at 12 WAA. The genes were alsoweakly expressed in leaves but the expressionwas not detected in germinated seedlings.

Both the class 1 and class 2 gene-specificprobes hybridized only very weakly with mRNAfrom 15-week mesocarp of E. oleifera. However,the 3’ untranslated regions of E. oleifera stearoyl-ACP desaturase genes may be different fromthose of E. guineensis, and it may not beaccurate to correlate the poor hybridizationsignals observed using both probes with lowexpression levels.

The different patterns of expression of thetwo E. guineensis genes indicates that theencoded proteins play different roles in the oilpalm. Class 2 which is constitutively expressed,most likely encodes an enzyme with housekeep-ing functions, possibly a stearoyl-ACPdesaturase involved in producing unsaturatedfatty acid for membrane lipid biosynthesis. Itseems likely that class 1 encodes a differentisoform of stearoyl-ACP desaturase found inlipid rich tissues which contain high levels ofoleic acid.

In developing oil palm fruits, accumulationof oil in the kernel occurs earlier at between11-14 WAA than in the mesocarp which startsat about 15 WAA and stops when the fruitsripen at 20 WAA (00 et al., 1986; Hartley,1988). The mesocarp oil contains 39% andkernel oil contains only 15% oleic acid (deManand dehlan, 1994). As the expression of class1 correlates closely with the pattern of oilsynthesis in the mesocarp and kernel, it wouldseem likely that these genes encode a desaturasedirectly involved in producing oleic acid forincorporation into triacylglycerol. The highlevel of expression in young mesocarp tissue isbelieved not to be related to oil synthesis butrather to production of oleic acid for membraneand cellular lipids for these actively dividingyoung tissues.

These results suggest the possibility thatthe oil palm has two differentially regulatedstearoyl-ACP desaturase genes. Occurrence ofdifferentially regulated genes for proteins in-volved in fatty acid synthesis has been reportedfor genes encording acyl carrier protein inArabidopsis thaliana (Baerson a n d Lamppa,1993) and Brossica napus (deSilva et al., 1992)

1 1

JOURNAL OF OIL PALM RESEARCH (SPECIAL ISSUE)

6, M K15M 8 I O I2 15 17 20 IO 12

I .8 kb- ,,,.;i :

I .?kb-

I 2 3 4 5 6 I 8 9

(a) class 1

1.8 kb-

I.3 kb-

I 2 3 4 5 6 7 8 9 1 0

(b) class 2

14

10

L G

,:::,

II 12

1 2

I 2 3 4 5 6 7 8 9 IO 11 I2

(c)

Figure 5. Northern blot analysis of oil palm stearoyl-ACP desaturase genes. Northern blot containing 2)poly (A)+RNA from various oil palm tissues was probed with 32P-labelled gene specific probes based onuntranslated regions of (a) pOPSNl9 and Cb) pOPSNI6. La ne I represents 15-week E. oleifera mewcarAll other samples were from E. guineensis tissues. Lanes (2-7) represent &IO. 12, 15,17and 20.week mesocarrespectively; lanes (S-10) represent IO. 12 and 14.week kernel, respectively; lanes 11 and I2 represent leavand germinated seedlings. Ethidium bromide stained gel (c) showing approximately equal loading of po(A)+RNA.

where the promoter of the first gene produced desaturase genes has been implicated iubiquitous expression while the second pro- Brassica naps where the promoter of a stearo)

!In iIl. I

duced loo-fold higher expression in seeds com-pared to leaves in transgenic tobacco plants.The results of the transgenic studies wereconsistent with the pattern of ACP transciptsin theArabidopsis thaliann and Brassicn napsplants. Differential regulation of stearoyl-ACP

ACP desature gene produced about 2-3 foldihigher expression in seeds compared to in’leaves of transgenic tobacco (Slocombe et al.,’1994). Previous studies using Northern analy-sis had shown the existence of seed-specificgene whose expression at maximal level was

1 2

lOO-fold higher in seeds than leaves (Slocombe&al., 1992). However, more recent studies haveshown that the Brossica napus stearoyl-ACPdesaturase genes involved in housekeepingfunctions are often upregulated to supply theadditional stage-specific and tissue-specificrequirements for unsaturated fatty acids(Piffanelli et al., 1997). This study showed thatthe situation may be different in the oil palmwhere one of the desaturase genes (class 1) maybs ‘induced to cater to the high demand forstearoyl-ACP desaturase activities in oil accu-mulating tissues at specific stages of develop-ment, The gene with the housekeeping role(class 2) may still be upregulated to someextent as shown by the three-fold higher ex-pression in oil palm mesocarp at 17 WAA.

Post-transcriptional Regulation

It was not possible to use the polyclonalantibodies previously raised against rapestearoyl-ACP desaturase (Piffanelli, 1997)because no cross reaction was detected withthe oil palm mesocarp extract. Based on thefull length oil palm stearoyl-ACP desaturaseamino acid sequence, a 14.residue peptide nearthe N-terminal of the mature proteinTVGPSTKVEIPKKP which contains threecharged amino acids (see Figure 1) was selectedfor synthesizing a multiantigenic peptide 0IAP)for raising antibodies. MAP consists of four oreight copies of the same peptide attached to abranchinglysine core matrix. The 14mer peptidewas synthesized directly onto the branchinglysine arms to form an immunogenic macro-molecular structure. The MAP approach provedcapable of producing high titre anti-peptideantibodies that recognized the native protein(Tam, 1992). Figure 6 shows that the polyclonalantibodies produced were highly specific for a37kDa protein from the oil palm mesocarpextract. This was the expected size for stearoyl-ACP desaturase consistent with that observedfrom other plant species (Shanklin andSomerville, 1991; Thompson et al., 1991). Therewas no cross reaction detected with the rapeembryo protein extract. This was not surprisingbecause the homology between the peptide usedfor raising the antibodies and rape sequencewas very weak (Figure 2).

The polyclonal antibodies were used todetect levels of stearoyl-ACP desaturase indifferent oil palm tissues (Figure 7). In themesocarp, the highest level observed was in theyoung tissue of eight weeks, similar to theNorthern blot data (Figure 5). Levels in 12- and15.week tissues were very low but it increasedagain at 17 weeks and remained high in theripe 20-week mesocarp. The enzyme was notdetectable in the kernel tissue while in theleaves, it appeared with a slightly higherapparent molecular weight of about 39kDa. Inthe photosynthetic tissues of Arabidopsisthaliana and Brassica aopus, two distinctisoforms of stearoyl-ACP desaturase were de-tected; a 38kDa isoform, the same size as thatfound in seeds and an additional and large 40kDa isoform (Piffanelli, 19971. There is noevidence as yet for a gene encoding stearoyl-ACP desaturase with specific expression inleaves which could lead to a distinct isoformfrom that in the seed. The studies in Brassicannpus suggest that the isoform in leaves withslower mobility on denaturing protein gels,could be the result of post-translational modi-fication of the enzyme found in both leaves andseeds. Since the oil palm leaf desaturase wasalso apparently 2kDa larger than the mesocarpand kernel isoforms, a similar post-transla-tional modification may also be occurring here.

The antibodies may recognize the encodedproducts of both oil palm classes 1 and 2 genesor just the former depending on their homologiesin the region used for antibody production,Comparison of the N-terminal amino acid se-quences encoded by the three different stearoyl-ACP desaturase genes isolated from Thunbergiadata (Cahoon et al., 1994) seed showed quitea high level of variability, but for two differentBrassica napus genes (Piffanelli, 19971, thesequences in this region are fully identical. Ifthe N-terminal sequence is also conserved inthe encoded products of both classes 1 and 2oil palm genes, we may expect quite high basallevels of enzyme in different tissues due to highand ubiquitous expression of the class 2 gene.There is a possibility that the class 2 gene maynot be efficiently translated and thus, the highabundance of the class 2 transcripts in differenttissues was not reflected in the levels of thegene product.

CHARACTERIZATION AND REGULATION OF THE OIL PALM (Hat-is guineensis) STEAROYL-ACP DESATURASE GENES

13

JOURNAL OF OIL PALM RESEARCH (SPECIAL ISSUE)

2 2 z8 w d z 2 8

45 kDa- 45 kDa-

29 kDa- 29 kDa-

1 2-3

Figure 6. Analysis ofpolyclonal antibodies raised against synthetic multiantigenic peptide from the oil palmstearoyl.ACP desaturase N-terminal sequence. The serum obtained after the third boost was diluted 1:lOOOand used for protein detection on a Western blot (a). Lanes I-3 respectively represent 20~ total protein fromoil palm mesocarp, rape embryo and of the multiantigenic peptide separated using 12% denaturingpolyacrylamide gel. For comparison, protein on a Western blot 0 were detected using polyclonal antibodiesraised against B. napus (rape) stearoyl-ACP desaturase. Lanes 1 and 2 represent 7pg total protein fromrape embryo and oil palm mesocarp, respectively.

I

M K L8 12 15 17 2 0 II 13

12 34 5 678a

12 3 4 5 6 78b

Figure 7. Western blot (4 was analysed using polyclonal antibodies specific for oil palm stearoyl-ACFdesaturase. Ten micrograms total protein from oil palm mesocarp and kernel at different stages ofdeuelopmemand from leaves were separated on 12% denaturing polyacrylamide gel and detected using I:1000 dilutionof the serum. Lanes (l-5) represent 8, 12, 15, 17 and 20.week mesocarp, respectively; lanes (6 and 7) represem11 and 13-week kernel and lane 8 represents leaue. Duplicate Coomassie Blue stained gel (b) showing equai

loading of protein.

14I

In young rapidly growing mesocarp tissues,there is a high requirement for unsaturatedfatty acids for membrane and cellular lipids.High levels of stearoyl-ACP desaturase ob-served in young mesocarp tissue are related toproduction of oleic acid which serves as thesubstrate for futher desaturation to polyun-saturated. linoleic (182) and linolenic (18:3)acids required for membrane lipid biosynthesis.Gene expression always precedes enzyme pro-duction. In the oil palm mesocarp, the enzymelevels start increasing at 17 weeks followinggene expression which has maximal expressionat 15 weeks for class 1 and 17 weeks for class2. Active oil accumulation in E. guineensismesocarp starts around 15 WAA and the fruitsripen at about 20 WAA. The major componentsof the final oil are palmitic and oleic acids.Studies by Aziz et al. (1986) showed that theoil composition at 19 WAA includes about42.5% palmitate and 36.5% oleate and has notyet reached the final composition of about 44%and 39%, respectively. Thus, the enzymes forproducing these fatty acids are still requiredeven at very late stages of ripening. Therefore,the observation that the enzyme level has notsignificantly dropped in ripe fruits is not totallyunexpected and the level may only eventuallydrop in over ripe fruit when there are no morechanges in the fatty acid composition and oilproduction has stopped.

CONCLUSION

Two differentially regulated stearoyl-ACP geneswere identified in the oil palm. Constitutiveexpression of one of the genes (class 2) suggestsa possible housekeeping role in membrane lipidbiosynthesis. The other gene (class 1) which isinduced in lipid-rich mesocarp and kernel tis-sues in phase with oil synthesis, is believed tohave a direct involvement in storage oil syn-thesis. This study also showed that regulationat transcriptional level is important in control-ling the levels of stearoyl-ACP desaturase inoil accumulating tissues. Based on these find-ings, it may be possible to manipulate the levelof stearoyl-ACP desature in the mesocarp toproduce oil with the desired composition byintroducing sense or antisense constructs of theclass 1 gene without interfering with mem-

CHARACTERlZATlON AND REGULATION OF THE OIL PALM (E/ask ouineensis) STEAROYL-ACP DESATURASE GENES

brane lipid biosynthesis. The promoter of thisgene would be useful in regulating expressionof heterologous genes required for oil modifi-cation. Furthermore, the polyclonal antibodiesspecific for oil palm stearoyl-ACP desaturaseproduced during this study would be a valuabletool to monitor successful production of trans-lational product encoded by the introduced oilpalm stearoyl-ACP desaturase gene, and alsoto check the effects produced on endogenotisstearoyl-ACP desaturase in transgenic oil palm.

REFERENCES

AKAGI, H; BABA, T; SHIMADA, H andFUJIMURA, T (1995). Nucleotide sequence ofa stearoyl-acyl carrier protein desaturase cDNAfrom developing seeds of rice. Plant Physiol,1089345-846.

AZIZ, A, ROSNAH, M S; MOHAMADIAH, Band WAN ZAILAN, W 0 (1986). Relationshipof the formation of storage oil with fatty acidcomposition in developing oil palm fruit. Proc.Mol. Biochem. Sot. Conf: 12:147-151.

BAERSON, S R and LAMPPA, G K (1993).Developmental regulation of an acyl carrierprotein gene promoter in vegetative and repro-ductive tissues. Plant Mol. Biol., 22:255-267.

BRADFORD, MM (1976). A rapid and sensitivemethod for the quantification of microgramquantities of protein utilizing the principle ofprotein-dye binding. Anal. Biochem., 72:142-146.

C A H O O N , E B ; S H A N K L I N , J a n dOHLROGGE, J B (1992). Expression of acorriander desaturase result in petroselenicproduction in transgenic tobacco. Proc. Natl.Acad. Sci. USA. 89:11184-11188.

CAHOON, E B; COUGHLAN, S J andSHANKLIN, J (1997). Characterization of astructurally and functionally diverged acyl-acylcarrier protein desaturase from milkweed seed.PZant Molecular Biol., 33:1105-1110.

CAHOON, E B; BECKER, C K; SHANKLIN,J and OHLROGGE, J B (1994). cDNAs for

1 5

JOURNAL OF OIL PALM RESEARCH (SPECIAL ISSUE)

isoforms of the A9-stearoyl-acyl carrier proteindesaturase from Thunbergia alata endosperm.Plant Physiol., IO6:807-808.

CHEAH, S C; SAMBANTHAMURTHI, R; SITINOR AKMAR, A; ABRIZAH, 0; MOHAMADARIF, A M; UMI SALAMAH, R and AHMADPARVEEZ, G K (1995). Towards genetic engi-neering of oil palm (E&is guineensis Jacq.). In(eds.’ Kader, J C and Mazliak, P). Plant LipidMetAboZism. Dordrecht: Kluwer Academic Pub-lishers, p. 570-572.

DELLAPORTA, S L; WOOD, J and HICKS,J B (1983). A plant DNA minipreparation:version II. Plant Mol. Biol. Report, 1:19-21.

deMAN, L and deMAN, J M (1994). Function-ality of palm oil, palm oil products and palmkernel oil in margarine and shortening. PORIMOccasional Paper No. 32:1-16.

deSILVA, J; ROBINSON, S J and SAFFORD,R (1992). The isolation and functional charac-terization of B. nnpus acyl carrier protein 5’flanking region involved in the regulation ofseed storage lipid synthesis. Plant Mol. Biol.,18:1163-1172.

HARTLEY, C W S (1988). The Oil Palm (Elaeisguineensis Jacq.). Third ed. New York. LongmanScientific and Technical.

KNUTZON, D S; THOMSON, G A; RADKE,S E and JOHNSON, W B (1992). Modificationof Brassica seed oil by antisense expression ofa stearoyl-acyl carrier protein desaturase gene.Proc. N&Z. Acad. Sci. USA, 89:2624-2628.

LAEMMLI, U K (1970). Cleavage of structuralproteins during assembly of the head ofbacteriophage T4. Nature, 227:680-685.

LINDQVIST, y; HUANG, W; SCHNEIDER, G,and SHANKLIN, J (1996). Crystal structure ofA9 stearoyl-acyl carrier protein desaturase fromcastor seed and its relationship to other di-ironproteins. The EMBO Journal, 15:4081-4092.

MIZUTANI, M S; SAVIN, K, CORNISH, E;TANAKA, Y; ASHIKARI, T; KUSUMI, T and

MURATA, N (19951. Senescence-induced expression of a homologue of A9 desaturase inrose petals. Plant Mol. Biol., 29:627-635.

MURPHY, D J and PIFFANELLI, P (19981,Fatty acid desaturases: structure, mechanisn:and regulation. In (ed. Harwood, J L). PIomLipid Biosynthesis: Recent Advances of Agricdtural Importance. Cambridge University Press,p. 95-130.

NAGAI, J and BLOCH, K (1968). Enzymaticdesaturation of stearoyl acyl carrier proteinJ. Biol. Chem., 243:4626-4633.

00, K C; LEE, K B and ONG, A S H (1986)Changes in fatty acid composition of the lipiiclasses in developing oil palm mesocarpPhytochemistry, 25:405-407.

PIFFANELLI, P (19971. Expression and regulation of lipid biosynthesis genes in developintissues of higher plants. Ph.D thesis. NorwichUniversity of East Anglia.

PIFFANELLI, P; ROSS, J H E and MURPHYD J (1997). Intra- and extracellular lipid cornposition and associated gene expression duringpollen development in Brassica napus. Plant J.11:549-562.

PRYDE, E H and ROTHFUS, J A (1989)Industrial and non-food uses of vegetable oilIn (eds. Robbelen, G R; Downey, K and AmranA). Oil Crops of the World: their Breeding antUtilization. New York, McGraw-Hill, p. 87-117

SAMBROOK, J; FRITSCH, E F and MANIATISi,:

T (1989). Molecular Cloning - A LaboratoryManual (second edition). New York, Cold SpringHarbor Laboratory Press.

SHANKLIN, J and SOMERVILLE, C (1991).Stearoyl-acyl-carrier protein desaturase fromhigher plants is structurally unrelated to theanimal and fungi homologs. Proc. Natl. Acad.Sci. USA, 88:2510-2514.

SHAH, F H and RASHID, 0 (1996). Nucleotidgsequence of a complementary DNA clone encoding stearoyl-acyl-carrier protein desaturaslfrom Elaeis guineensis var tenera (PGR96-110:Plant Physiol., 112:1399.

1 6

CHARACTERIZATION AND REGULATION OF THE OIL PALM (E/as-is guineensis) STEAROYL-ACP DESATURASE GENES

SITI NOR AKMAR, A; FARIDA, H S andCHEAH, S C (1994). Detection of differentiallyexpressed genes in the development of oil palmmesocarp. Asia Pacific Journal of MolecularBiology and Biotechnology No. 2:113-118.

SITI NOR AKMAR, A, FARIDA, H S andCHEAH, .S C (1995). Construction of oil palmqesocarp cDNA library and the isolation ofm,esocarp-specific cDNA clones. Asia PacificJournal ofMolecular Biology and Biotechnology,3, No. 2:106-111.

SLOCOMBE, S P; PIFFANELLI, P;FAIRBAIRN, D; BOWRA, S; HATZOPOULOS,P; TSIANTIS, M and MURPHY, D J (1994).Temporal and tissue-specific regulation of. aBmssica napus stearoyl-acyl carrier proteindesaturase gene. Plant Physiol., 104:1167-1176.

SLOCOMBE, S p; CUMMINS, I; JARVIS, R Pand MURPHY, D J (1992). Nucleotide sequenceand temporal regulation of a seed-specific B.

Napus cDNA encoding a stearoyl-acyl-carrierprotein (ACP), desaturase. Plant. Mol. Biol., 20:151-155.

TAM, J P (1992). In (eds. Koff, W L and Six,H). Vaccine Research Developments. MarcelDekker, New York, p. 51-87.

THOMPSON, G A and LI, C (1996). Alteredfatty acid composition of membrane lipids ‘inseeds and seedling tissues of high saturatecanolas. In (eds. Williams, J P, Mobasher, UK and Lem, N W). Physiology, Biochemistryand Molecular Biology of Plant Lipids. KluwerAcedemic Publishers, p. 313-315.

THOMPSON, G A; SCHERER, D E; FOXALL-VAN AKEN, S; KENNY, J W; YOUNG, H L;SHINTANI, D K; KRIDL, J C and KNAUF,V C (1991). Primary structure of the precursorand mature forms of stearoyl-acyl carrier pro-tein desaturase from sunflower embryos andrequirement of ferredoxin for enzyme activity.Proc. Natl. Acad. Sci. USA. 88:2578-2582.