structure and expression of an arabidopsis acetyl …each of 'i'orula yeast rna, with or...

Plant Physiol. (1994) 105: 611-617

Structure and Expression of an Arabidopsis Acetyl-Coenzyme A Carboxylase Gene'

Keith R. Roesler, Basil S. Shorrosh, and John B. Ohlrogge*

Department of Botany and Plant Pathology, Michigan State University, East Lansing, Michigan 48824

Acetyl-coenzyme A carboxylase (ACCase) catalyzes the formation of malonyl-coenzyme A, which is used in the plastid for fatty acid synthesis and in the cytosol for severa1 pathways including fatty acid elongation and flavonoid synthesis. Two overlapping Arabidopsis genomic clones were isolated and sequenced to determine the entire ACCase-coding region. Thirty introns with an average size of 94 bp were identified by comparison with an alfalfa ACCase cDNA sequence. l h e 10-kb Arabidopsis ACCase gene encodes a 251-kD polypeptide, which has 80% amino acid sequence identity with alfalfa ACCase and about 40% identity with ACCase of rat, chicken, yeast, and the diatom Cyclotella. No chloroplast transit peptide sequence was observed, suggesting that this Arabidopsis gene encodes a cytosolic ACCase isozyme. ACCase gene transcripts were detected by RNase protection assays in Arabidopsis root, leaf, silique, and seed. Cenomic DNA blot analysis revealed the presence of a second related Arabidopsis ACCase gene.

ACCase (EC 6.4.1.2) catalyzes the ATP-dependent carboxylation of acetyl-COA to produce malonyl-COA. This reaction occurs in two steps: carboxylation of a biotin prosthetic group using HC03- as carboxyl donor, followed by transfer of the carboxyl group from biotin to acetyl-COA. The biotin carboxylase, carboxyl transferase, and biotin components of ACCase are each associated with different polypeptides in prokaryotes (Samols et al., 1988). In contrast, ACCase of nonplant eukaryotes is composed of multimers of a single multifunctional polypeptide.

In plants, evidence for both a prokaryotic type ACCase (Kannangara and Stumpf, 1972; Nikolau et al., 1993; Sasaki et al., 1993) and a eukaryotic type (Hanvood, 1988) has been obtained. The malonyl-COA produced by ACCase is used in a wide variety of reactions and pathways in plants, including fatty acid synthesis and elongation (Harwood, 1988), flavonoid synthesis (Ebel and Hahlbrock, 1977; Ebel et al., 1984), malonation of the ethylene precursor aminocyclopropane- 1 - carboxylate (Liu et al., 1983; Kionka and Amrhein, 1984), and malonation of amino acids and glycosides.

Malonyl-COA must be available in multiple subcellular locations because some of these reactions, such as fatty acid synthesis, occur in the plastid and others, such as flavonoid

' This research was supported in part by grants from the National Science Foundation (DCB 90-05290) and the Midwest Plant Biotech- nology Consortium. Acknowledgement is also made of the Michigan Agricultura1 Experiment Station for its support of this work.

* Corresponding author; fax 1-517-353-1926. 61 1

synthesis, occur in the cytosol. Malonyl-COA must also be available in greatly differing amounts with respect to time and tissue. For example, increased amounts of malonyl-COA are needed for fatty acid synthesis in developing seeds of species that store large quantities of triacylglycerols (Post- Beittenmiller et al., 1993). In floral tissue, malonyl-COA is used in the chalcone synthase reaction for synthesis of the flavonoid pigments, which constitute up to 15% of the dry weight of this tissue (Goodwin and Mercer, 1983). In some tissues, ACCase might provide malonyl-COA constitutively to produce fatty acids for membrane synthesis and mainte- nance while providing a 'burst" of malonyl-COA for only a short period to synthesize flavonoids during exposure to UV light (Ebel and Hahlbrock, 1977) or during funga1 pathogen attack (Ebel et al., 1984).

How ACCase provides malonyl-COA in such diverse locations, times, and tissues is not clearly understood. Presum- ably, multiple ACCase genes, transcriptional regulation, and posttranscriptional regulation could a11 be involved. To gain insight into which of these possibilities might occur, we are investigating ACCase gene structure, organization, and expression in Arabidopsis. In this report we describe the isolation and expression of an ACCase gene that probably encodes a cytosolic isozyme.

MATERIALS AND METHODS

lsolation and Sequence Analysis of Cenomic Clones

To obtain an ACCase probe, PCR was camed out with a coriander endosperm cDNA library (Cahoon et al., 1992). Degenerate primers were prepared to two peptides conserved in ACCase of rat, chicken, and Cyclotella: VEIKFR and FADLHD, corresponding to residues 2038 to 2043 and 2102 to 2107 of Figure 2. A 207-bp PCR product with identity to known ACCase sequences was obtained and used to screen an Arabidopsis thalinna ecotype Columbia Agem 11 genomic library (kindly provided by Came Schneider and Chris So- merville). One positive clone was obtained by screening 1.6 X 105 plaques. An EcoRIISalI restriction fragment of this clone, corresponding to residues 1520 to 1586 of Figure 2, was used to reprobe the same filters, and four additional positive clones were obtained. The same genomic library was subsequently screened with restriction fragments of an alfalfa ACCase cDNA (probes named "5'" and '3'ACC" by Shor- rosh et al., 1994) and six more positive clones were obtained.

Abbreviation: ACCase, acetyl-coenzyme A carboxylase

www.plantphysiol.orgon April 30, 2018 - Published by Downloaded from Copyright © 1994 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

612 Roesler et al. Plant Physiol. Vol. 105, 1994

v

Probe 1 Probe 2 t7zzzza EE4 HEBH S E S B B S B B EBH

I I

2.0 4.0 6.0 8.0

I1 I 1 1 1 111 I 11,-

Kb

Figure 1. Map of the Arabidopsis ACCase gene. lntrons are shown as solid bars. ACC-2 and ACC-7 designate the two genomic clones used to determine the ACCase-coding region. Probes 1 2nd 2 were used in genomic DNA blot analysis or in RNase protection assays, respectively. Letters represent sites of restriction enzymes used in genomic DNA blot analysis: 6, Bglll; E, EcoRI; H, HindIII; S, Sacl.

Restriction mapping plus partia1 sequence analysis of overlapping regions revealed that these 11 clones a11 likely represented the same gene and that none contained the entire ACCase-coding region (not shown). Two overlapping clones of approximately 12 and 14 kb (ACC-2 and ACC-7, respectively, of Fig. 1) were then sequenced extensively to determine the coding region. A11 of the coding region included in ACC-7 was sequenced, and the remainder of the coding region was obtained from ACC-2. Both strands of the coding region were sequenced in their entirety as subclones in pBluescript KS+ (Stratagene) using either dideoxy chain ter- mination with the Sequenase kit (United States Biochemical) or a dye-primer method through the Michigan State Univer- sity sequencing facility.

cDNA Synthesis

To confirm the identity of the ACCase start Met, a cDNA that included the surrounding region was synthesized and sequenced. First-strand cDNA was synthesized with Moloney murine leukemia virus reverse transcriptase using 5 pg of Arabidopsis total RNA and a 17-mer primer 00177) corresponding to the region encoding Asn354 to Va1359 of Figure 2. Double-stranded cDNA was then synthesized by PCR with the first-strand cDNA as template, using JO177 and a primer 00190) from the 5' nontranslated region of the ACCase gene. An aliquot of this PCR product was used in a second round of PCR with JOl90 and a 3' primer 00191) corre- spondmg to the region encoding Leu'77 to Ser'sz of Figure 2. First-strand cDNA synthesis and PCR reactions were done under conditions similar to those described elsewhere (Shor- rosh et al., 1994). The resulting PCR product of 575 bp was sequenced and found to be identical with the corresponding genomic DNA sequence except that the first intron of Figure 2 was missing in the cDNA sequence as expected. In both the genomic DNA and cDNA sequences, an in-frame stop codon was observed 15 bp upstream from the start Met of Figure 2.

Cenomic DNA Blot Analysis

Arabidopsis genomic DNA (10 pg) was digested with BglII, EcoRI, HindIII, or SacI, electrophoresed in a 0.8% agarose gel, and blotted to Zetaprobe nylon membrane (Bio-Rad) in 0.4 N

NaOH. The probe was a random hexamer-labeled 1316-bp SacI fragment of the ACCase gene (probe 1 of Fig. 1). Hy- bridization was canied out in 5X SSC, 0.05X Bbotto (Sam- brook et al., 1989) at 55OC for 16 h. The blot was washed twice for 30 min each in 0.2X SSC, 0.1% SDS at Ei5OC.

RNase Protection Analysis

Plasmid pBluesaipt KS+ containing a 3345-bp SacI fragment of the ACCase gene was linearized with BglII, and a 643-nuclleotide 32P-labeled RNA probe was synthesized with T3 polymerase using the materials and procedure of a Max- iscript kiit (Ambion, Austin, TX). The RNA probe contained 527 nucleotides from the ACCase gene (corresponding to probe 2 of Fig. 1) with the remainder from the vector polylinker. RNase protection assays were done using an RPA I1 kit (Ambion). The labeled RNA probe was hylxidized at 45OC for 16 h with 5 pg of total RNA from Arabidopsis root, leaf, silique (including seed), or seed. After the probe was hybridized, digestion with RNases A and T1 was done, and the labekd, protected RNA was resolved in a 5% polyacryl- amide, 8 M urea gel. Control assays were done with 10 pg each of 'I'orula yeast RNA, with or without the RNase digestion. The RNA was isolated from 5- to 7-week-old.4rabidopsis plants grown in soil in continuous light. The isolation procedure of Hall et al. (1978) was followed except that developing seed was homogenized in a microfuge tribe with a minipestlle.

RESULTS

ACCase Sequence Charaderization

Eleven Arabidopsis ACCase clones were obtained by screening a genomic library with a coriander ACCase PCR product or by screening with restriction fragments of an alfalfa ACCase cDNA. Two overlapping clonej were sequenced to determine the entire ACCase-coding yegion (Fig. 1). To erisure that these clones represented the same gene, 940 bp lof the overlapping region, including four introns, were sequenced from both clones and found to be identical. Thirty introns were identified by comparing the Arabidopsis gene with an alfalfa ACCase cDNA sequence (Figs. 1 and 2). The Arabidopsis ACCase amino acid sequence was identical with the alfalfa sequence across most introns. Furthermore,



Arabidopsis Acetyl-COA Carboxylase

A r a MAGSV---NGNHSAVGPGINYETVSQVDEFCKALRGKRPIHSILIANNGMAAVKFIRSVRTWAYETFGTEKAILLVGMATPEDMRINAEHIRIADQFVEVPGGTNNNNYANVQLIV~AEV ..-.. GRG . . YLNS.L.SRHPA.TTE ... Y.N . . G.NK ...................... S ............... A......................................L.I.. 1 A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

A r a A1 f

TRVDAVWPGWGHASENPELPDALDAKGI I FLGPPASSMAALGDKIGSSLIAQAADVPTLPWSGSHKI PPNSNLVTI PEEIYRQACVYTTEEAIASCQVV' .H ..................... K....V......I..................E...............E.D.I...D....A...............

HNDDEVRALFKQVQGEVPGSPI FIMKVASqSRHLEVQLLCDKHGNVSALHSRDCSVQRRHQuI EEGPITVAPPETVKKLEQAARRLAKSVNYVGAATI EY LYSMDTGEYY FLELNPRLQ .. I . . Q . . . F A ............................... E ................ . .Q.. .FA. . . ..........

YEHPVTEWI A E I NLPAAQVAVGMGI PLWQI P E I RRFYGI EHGGGYDSWRKTSVVAFPFDFDKAQS I R .................. E ................... M . . . . . N . G . K . . . . L . T . . . . . E . . . T K

DSQF~HVFAFGESRnLAIANMVLGLKEIQIRGEIRTNVDYTIDLL~SDYRDNKIHTGWLDSRIAMRVRAERPPWYLSVVGGAL~SATSAAVVSDYVGYLEKGQIPP~ISLVHSQVS ............................................. N ............................. ......... S. . .L .......................... LN I EGSKYUDVVRGGSGTY RLRMNKSEVVAEI HTLRDGGLLM~DGKSHVIYAEEEAAGTRLLIDGRTCLL~DHDPSKLMAETPCKLMRY LISDNSN IDADTPYAEVEVMKMCMPLLS .S.. ...... . M . . . . P.S.K.KL.Q.. I E . . .............. .N.. ....... .............. D. .... I G . ..... L.. . VA. D.Q. .................... PASGVIHFKMSEGQAM~GELIANLDLDDPSAVRKAEPFHGSFPRLGLPTAISGRVHQRCAATLNAARMI LAGY EHKVDmVQDLLNCLDSPELPFLQWQECFAVLATRLPKNLRNfiE .... I ... R.A ............ K.....G.........T....I..P......K...K...S.............NI..V..KS............................D...E.. SKYREFESISRNSLTTDFPAKLLKGILEAHLSSCDEKERGALERLIEPLMSLAK RESHARVIVHSLFEEYLSVEELFNDNMCADVIERMRQLYKKDLLKIVDIVLSH~IKNKNK .. S .. I Q ...... L.LQ ................. V.S ... A .. K...I..-S.Q.I........A.........P.N.K......V...T.. V.

.I .... DK ........... Q . . . . . Q . . . I . . . .s.. . I D.M P. . . . . . . . . . . . . . . . . . . . . . . . . . . .

VAVRVTSEDPDDGFKPTSGRV~LSFKSKPNVWAYFSV .............. T . . G.K ..................

*

..... HK . . Q...... LVLRLMEQLVYPNPAAYRDKLIRFSTLNHTNYSELALKASQLLEQTKLSELRSNIARSLSELEMFTEDGENMDTPKRKSAINERIEDLVSASLAVEDALVGLFDHSDHTLQRRVVETYIR

........ .......... . . . . . . RLY~YVVKDSVRMQWHRSGLLASWEFLEEHMERKNIGLDDPDTSEKGLVEKRSKR~GAMVIIKSLQFLPSIISAALRETKHN-----DYETAGAPLSGNMMHIAIVGINNQMSLLQD~

I... YV ....-. VE T .... H.EK ... V..V........A........ATNFHDPLKSGSGDSSNH.......GL............. GDEDQAQERVNKLAKILKEEEVSSSLCSAGVGVISCIIQRDEGRTPMRHSFHWSLEKQYYVEEPLLRHLEPPLSIYLE~KLKGYSNIQYTPSRDRQWHLYTVTD-KPVPIKRMFLRSLV ........ . I D . . ... .R.Q.IG. I I H A . . . . D . . ........ .A. . . . .... . S . .L.. ..... .L . . ............. .C.E. . R . . ........... .V.T. .Q. .Q

.. P.T.E.YSSY.RL.AET.R.QLA.SY .. R S I F .... G ........ S..TTI.SE....Y.Y.I.EQ.......YSKKINIE.GQ......A....L.Q...S...........FV.. I RQATMNDGFI LQQGQDKQLSQTLI SMAFTSKCVLRSLMDAMEELELNAHNAAMKPDHAHMFLCI LRDEQI DDLVPFPRRVEVNAEDEETTVEMI LEEAAREI HRSVGVRMHRLGVCEWEV

RLWLVSSGLACGAWRVVVANVTGRTCTVHlYREVETPGRNSLIYHSITKKGPLHETPISDQYKPLGYLDRQRLAARRSNTTYCYDFPCAFGTALELLWASQHPGVKKPYKDTLINVKELV K. . 1TAC.Q. N. .... I. N.. . . H. ........ M. DATTHKVV. S .V.V. . . . .GV.VNEN.Q.. . G I . . K. .... KNS. ..... .Q. S. . QS.SI .QT. 1QRANDKD.LK.T. . K

FSKPEGSSGTSLDLVERPPGLNDFGMVAWCLDMSTPEFP . . E U . . W. ... VPA. . L. ... .V.. ... LME.C. . K. .

SPEDHERIGSSVIAHEVKLSSGETRWVIDTIVGKEDGIGVENLTGSGAIAGAYSKAYNETFTLTFVSGRTVGIGAYLARLGMRCIQRLDQPIILTGFSTLNKLLGREVYSSHMQLGGPKI T ... YA ...... M . . . L . . E . . . . . . . . . . . . . . . . . L . . . . . S . . . . . . . . . . R . . K . . . . . . Y . T . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A . . . . . . . . . . . . . . . . .

MGTNGVVHLTVSDDLEGVSAILNWLSYIPAYVGGPLPVLAPLDPPERIVEYVPENSCDPRAAIAGVKDNTG~LGGIFDKNSFIETLEG VETQTVMQII . . . S . .K . . . .Y.SH.. .A.. I V K . . .... .E.. .L.. ........ .S.TL.VN.. ... ............

PADPGQLDSHERVVPQAGQVWFPDSAAKTAQALMDFNREELPLFILANWRGFSGGQRDLFEGILQAGSTIVENLRTYRQPVFVYIPMMGELRGGAWVVVDSQINSDYVEMYADETARGNV ......................... .T.. . . . IL . . ......... I.. ............................ .K. .I.. ............ LEPEGTI E 1 KFRTKELLECMGRLDQKLISLKAKLQDAKQSEAYANI ELLQQQI KAREKQLLPVY IQIATKFAELHDTSMRMAAKGVI KSVVEWSGSRSFFYKKLNRRIAESSLVKNVREA

. .. .. . . R . . H I . . ER.. K..

....... M ....... R......R....Q..N..E..SE..SNKD.GAYDS...RF.......L.T.............L..K.....RE.LD.RK...V..QR.H...G.H..INI.. D.

SGDNLAYKSSMRLIQDWFCNSDI AKGKEEAWTDDQVFFTWKDNVSNY E L K L S E L R A Q K L L N Q L A E I G N S S - D L Q A L P Q G L A N L L N K V R K V L G A . . Q.S.V.A.N.LKE.YL ....... R . D .. L..EA..R.R.DPA...D..K...V.R..L..TN..D.AL..........A..S.L.A.S.DK.ISEL.....

Figure 2. Comparison of deduced amino acid sequences of Arabidopsis and alfalfa ACCase. The alfalfa sequence (Shorrosh et al., 1994) is shown only where different from Arabidopsis. Adjacent underlined residues indicate an intron located between codons. Single underlined residues indicate an intron located within a codon. An asterisk at position 71 1 marks the biotin-binding site. The three shaded regions are the proposed ATP-, carboxybiotin-, and acetyl-COA- binding sites, respectively, from N terminus to C terminus.

61 3

118

238

358

478

598

718

837

957

1077

1192

1311

1431

1551

1671

1791

1911

2031

2151

2254

the exon/intron border junctions fit the consensus sequence n/gt.. .a& (Goodall and Filipowicz, 1991) for 29 of the 30 introns. The remaining intron, the 15th from the 5' end, used gc rather than gt at the 5' junction. This border sequence, although rare, has been observed previously in other Arubi- dopsis introns (evident in a table of 569 Arubidopsis introns compiled by Mike Cherry and posted in Arubidopsis E-mail network, September 13, 1993). The introns ranged in size from 73 to 180 bp and averaged 94 bp.

The Arabidopsis ACCase gene encodes a 2254-amino acid

polypeptide with a calculated molecular mass of 251 kD and an isoelectric point of 6.0. In severa1 previous studies, ACCase purified from plants comprised a homodimer of >200 kD subunits (Egin-Buhler and Ebel, 1983; Slabas and Hellyer, 1985; Charles and Cheny, 1986; Bettey et al., 1992; Egli et al., 1993; Gomicki and Haselkom, 1993), consistent with the deduced molecular mass of the Arubidopsis polypeptide determined here. Biotin is covalently bound to a Lys residue flanked by Met residues in most biotin-containing polypeptides so far sequenced. This MKM consensus sequence was



614 Roesler et al. Plant Physiol. Vol. 105, 1994

Table I. ACCase am/no acid sequence comparisonsThe Arabidopsis amino acid sequence was compared with that

of alfalfa (Shorrosh et al., 1994), yeast (Al-Feel et al., 1992), rat(Lopez-Casillas et al., 1988), chicken (Takai et al., 1988), and Cyc/o-te//a (Roessler and Ohlrogge, 1993). The CCC Gap program (Dev-ereux et al., 1984) was used with values of 5.0 and 0.3 for gapweight and gap length, respectively. Results are the percentageidentities versus Arabidopsis ACCase.________________

Arabidopsis Amino Acid No.

AlfalfaYeastRatChickenCyc/ote//a

1-762

8950515147

763-1546

7327252421

1547-2254

7848474646

8042414039

identified in Arabidopsis ACCase at residues 710 to 712 (Fig.2). A Val residue immediately precedes this tripeptide in alleukaryotic ACCase sequences so far determined, in agree-ment with the Arabidopsis ACCase but in contrast to theprokaryotic enzyme or to other biotin carboxylases.

Pro residues were observed 27 and 35 positions upstreamfrom this biotin-binding site, similar to previous observationswith ACCase of other eukaryotes. These double Pro residuesare proposed to form a hinge that allows the HCOs'-bindingsite to approach the biotin-binding site, thus facilitating car-boxyl transfer (Samols et al., 1988). Regions of the ACCaseprimary structure proposed to be involved in the binding ofATP, carboxybiorin, and acetyl-CoA (Al-Feel et al., 1992; Liand Cronan, 1992a, 1992b) were also located in the Arabi-dopsis sequence and are shaded in Figure 2. The start Metshown in Figure 2 was initially identified based on its sur-rounding nucleotide sequence (ACAATGGCT), which fitsthe consensus sequence for higher plant start Mets (Joshi,1987; Lutcke et al., 1987). Sequencing 560 bp upstreamrevealed no other Mets that conformed well to the consensussequence.

To confirm the identity of the start Met, a cDNA thatincluded the surrounding region was synthesized and se-quenced. An in-frame stop codon was observed 15 bp up-stream from the start Met in both the genomic and cDNAsequences. The position of the start Met is conserved withthat of the alfalfa ACCase start Met, which is also known tobe authentic because of in-frame upstream stop codons inthe cDNA sequence (Shorrosh et al., 1994). Features of higherplant chloroplast transit peptides (Keegstra et al., 1989) werenot evident in the Arabidopsis sequence. Acidic residues, rarein transit peptides, were observed at positions 19, 25, and26. Sequence identity with cytosolic ACCase of chicken andrat was observed as near as 20 residues from the start Met,with too few residues remaining to comprise a typical transitpeptide of 30 to 70 residues. Other characteristics of transitpeptides, such as very abundant Ser and Thr residues, werealso absent.

The Arabidopsis ACCase amino acid sequence was com-pared with ACCase sequences of diverse organisms (Table I).Substantial identity was found in the N-terminal regioncontaining the biotin carboxylase domain and the biotin-

binding site. Considerable identity was also observed in theC-terminal region, which includes the carboxyl transferasedomain. In contrast, much less identity was found inthe central one-third of the primary structure. ArabidopsisACCase had 80% amino acid sequence identity overall incomparison with alfalfa ACCase and about 40% identitywith ACCase of rat, chicken, yeast, and the alga Cyclotella.Rat liver ACCase is regulated by reversible phosphorylation(Kirn et al., 1989). None of the seven Ser residues known tobe phosphorylated in the rat enzyme are present in Arabidop-sis ACCase. Partial sequences of 1306 and 546 residues,respectively, for maize and wheat ACCase are also available(A.R. Ashton, C.L.V. Jenkins, P.R. Whitfield, unpublisheddata; GenBank sequence Z24449; K.M. Elborough, J.W.Simon, R. Swinhoe, A.R. Ashton, A.R. Slabas, unpublisheddata; GenBank sequence Z23038). These monocot ACCasesequences have 62 and 69% identity, respectively, with thecorresponding regions of Arabidopsis ACCase. Over thesesame regions, alfalfa ACCase has 74 and 78% identity withthe Arabidopsis enzyme.

Genomic ONA Blot Analysis

An Arabidopsis genomic DNA blot was probed with a 1316-bp Sad fragment from the biotin carboxylase region of theACCase gene (probe 1 of Fig. 1). The probe contained oneinternal EcoRI site. Expected bands of the correct size wereobserved in each lane (Fig. 3). In addition, a less prominent,unexpected band was clearly visible in the Bg/II, Hindlll, andSad lanes, suggesting the presence of a second ACCase gene.With the Hmdlll digest, the second band was smaller in sizethan the expected band, thus ruling out any artifact resultingfrom a partial digest. To further confirm the presence of asecond Arabidopsis ACCase gene, a second blot (not shown)was prepared with the same restriction enzymes and probedwith a 1152-bp Bglll/Sall fragment from the carboxyl trans-

B E H S23.1 -

9.4-6.6-

4.4-

.1

0.6-

Figure 3. Arabidopsis genomic DNA blot analysis. The probe was a1316-bp Sacl fragment of the Arabidopsis ACCase gene (probe 1 ofFig. 1). Restriction enzymes used were Bg/II (B), fcoRI (E), H/ndlll(H), and Sacl (S). Approximate sizes in kb are given on the left.Hybridization conditions are described in "Materials and Methods." www.plantphysiol.orgon April 30, 2018 - Published by Downloaded from

Copyright © 1994 American Society of Plant Biologists. All rights reserved.


Arabidopsis Acetyl-CoA Carboxylase 615

ferase region, corresponding to the region encoding Ser1585 toAsp1969 of Figure 2. One extra band was again observed insome lanes, further suggesting the presence of a second gene.Similar evidence for two Arabidopsis ACCase genes has beenobtained independently (Yanai et al., 1993).

RNase Protection Assays

The presence of two cross-hybridizing Arabidopsis ACCasegenes seemed likely to complicate RNA blot analysis. There-fore, RNase protection assays rather than RNA blots weredone to assess tissue-specific expression of the clonedACCase gene (Fig. 4). Only RNA transcripts from the clonedgene should be detected with this assay, since even single-base mismatches in the hybrid would be cleaved duringRNase treatment (Myers et al., 1985). RNA from Arabidopsisroot, leaf, silique, and seed all showed protection by theACCase probe. The protected fragment was smaller than theprobe as expected, since the probe included an additionalsequence from the vector polylinker. The yeast control RNAshowed no protection.

DISCUSSION

The present study provides two lines of evidence for mul-tiple ACCase genes in Arabidopsis. First, the cloned Arabidop-sis gene does not appear to have a transit peptide sequence,suggesting that it encodes a cytosolic ACCase isozyme. Be-cause fatty acid synthesis occurs primarily in the plastid andisolated chloroplasts possess ACCase activity sufficient tosupport in vivo rates of fatty acid synthesis (Roughan andLaing, 1982), another gene(s) encoding a plastidial ACCaseisozyme must exist. Consistent with this expectation, ACCasehas been partially purified from isolated Ricinus plastids(Finlayson and Dennis, 1983). Second, the genomic DNAblot analysis presented here suggests the presence of tworelated Arabidopsis ACCase genes. Whether the additionalgene encodes a plastidial ACCase isozyme or another cyto-solic isozyme is yet to be determined.

The observation of multiple ACCase genes is consistentwith previous biochemical studies of the maize enzyme. Twomaize ACCase isozymes were purified, only one of whichwas detected in chloroplasts (Egli et al., 1993). It seems likely

Sd S L R Y1 Y2

Figure 4. RNase protection analysis. Arabidopsis total RNA (5 ^g)from root (R), leaf (L), silique (S), or seed (Sd) was used. The labeledRNA probe was from the carboxyl transferase region of the Arabi-dopsis ACCase gene (probe 2 of Fig. 1). Controls were 10 /tg ofTorula yeast RNA with (Y1) or without (Y2) the RNase treatment.Film exposure times were 2 h for Y2 and 6 d for all other samples.Sizes in nucleotides are given on the right.

that the maize isozymes are encoded by two genes becausepolyclonal antibodies to one isozyme did not cross-react wellwith the other isozyme. Other possible explanations for thetwo maize isozymes, such as proteolytic processing of a singlegene product, alternate splicing of RNA from the same gene,or use of alternate start codons to generate two polypeptidesfrom the same gene would result in polypeptides sharingmuch structural identity, and substantial antibody cross-reactivity would thus have been observed. An ACCase com-plex of 91-, 87-, and 35-kD subunits was recently proposedfor pea chloroplasts (Sasaki et al., 1993). If a similar complexis present in Arabidopsis, then genes in addition to theone described here would be needed to encode the smallerpolypeptides.

The Arabidopsis ACCase gene message was detected in alltissues examined, including both vegetative and reproductivetissues. This apparently ubiquitous expression is not surpris-ing considering the need for malonyl-CoA in the cytosol ofall cells. For example, very long chain fatty acids are com-ponents of plasma membrane lipids (Cahoon and Lynch,1991) and are also needed for synthesis of cuticular waxes tocover the surface of both aerial and underground tissues(Harwood, 1988). These very long chain fatty acids aresynthesized outside the plastid by elongation of 16- or 18-carbon fatty acids exported from the plastid. Malonyl-CoAfor the elongation reactions must be present in the cytosoland is presumably provided by a cytosolic ACCase. Detectionof the cytosolic ACCase gene message in developing seedalso seems reasonable. Arabidopsis, like numerous Brassicaceasp., contains the very long chain fatty acids eicosenoic (20:1)and erucic (22:1) in seed storage triacylglycerols (James andDooner, 1990), and these fatty acids are also synthesized byelongation of oleic acid exported from the plastid (Pollardand Stumpf, 1980). Antisense RNA experiments with tissue-specific promoters may help define the precise functions ineach tissue for this cytosolic ACCase isozyme. Somers et al.(1993) also reported that one ACCase gene product waspresent in both leaf and seed in maize; it was the majorACCase isozyme in both tissues and was concluded to beinvolved in fatty acid biosynthesis for both membranes andseed embryo triacylglycerol. If this is so, this maize genewould encode a plastidial ACCase isozyme and would notcorrespond to the cytosolic ACCase gene isolated here.

The ubiquitous expression of this Arabidopsis ACCase genemay differ from that of an alfalfa cytosolic ACCase gene.The alfalfa gene message was not detected in alfalfa suspen-sion culture cells except when induced with fungal elicitors(Shorrosh et al., 1994). Perhaps the primary role of the alfalfagene is to provide malonyl-CoA for isoflavonoid synthesiswhen needed for the plant defense system, and at other timesthis gene is not expressed. Alternatively, the alfalfa genecould be expressed in unelicited cells at a basal level too lowto be detected readily by RNA blot analysis, which is lesssensitive than the RNase protection method used here. Ex-tensive analyses and comparisons of the promoters of thesetwo genes may reveal differences that explain the differentexpression patterns.

The amino acid sequence comparisons revealed that Ara-bidopsis ACCase has lower sequence identity with wheat ormaize ACCase than with alfalfa ACCase. Since the wheat



61 6 Roesler et al. Plant Physiol. Vol. 105, 1994

and maize sequences are not complete, we do not know whether they represent plastidial or cytosolic ACCase isozymes. Therefore, it is not known whether the lesser sequence identity primarily reflects differences between dicot versus monocot cytosolic ACCase or, rather, reflects structural differences between plastidial and cytosolic ACCase isozymes. Because of the different environments (e.g. different pH and [Mg2+]) in the plastid versus the cytosol of a plant cell, substantial structural differences in the isozymes from these locations might be expected.

ACCase appears to have an important regulatory role in plant fatty acid synthesis. Analysis of substrate and product pool sizes implicated ACCase in the light/dark regulation of fatty acid synthesis in spinach leaves and chloroplasts (Post- Beittenmiller et al., 1991, 1992). ACCase may also be the site of feedback inhibition of fatty acid synthesis in tobacco suspension cells supplemented with exogenous fatty acids (Shintani and Ohlrogge, 1993). Furthermore, ACCase activity increases in association with lipid deposition in developing seeds of oilseed crops (Simcox et al., 1979; Tumham and Northcote, 1983; Charles et al., 1986; Deerburg et al., 1990). These observations have prompted speculation that ACCase may be a rate-limiting enzyme for oilseed fatty acid synthesis, but conclusive evidence supporting this hypothesis has not been obtained. For example, the activities and/or levels of a11 examined lipid biosynthetic proteins, and not just ACCase, increase during seed development (Post-Beittenmiller et al., 1993). The availability of the ACCase gene described here, when modified to encode a plastid transit peptide, may allow a direct test in transgenic plants of the hypothesis that ACCase expression levels are limiting for seed oil production.

In conclusion, we have characterized an Arabidopsis gene that is transcribed in diverse tissues and appears to encode a cytosolic ACCase isozyme. We also have evidence for at least one additional Arabidopsis ACCase gene. These results plus future experiments manipulating the expression of the ACCase genes should help clarify how ACCase provides its product in varying amounts, times, and locations in the plant.

ACKNOWLEDCMENTS

We thank Carrie Schneider and Chris Somerville for providing the genomic library and Steven Demler for assistance with the amino acid sequence comparisons. We also thank Paul Roessler for providing the Cyclotella ACCase sequence before publication, and Rache1 Granger for isolating developing Arabidopsis seed.

Received December 21, 1993; accepted February 28, 1994. Copyright Clearance Center: 0032-0889/94/105/0611/07. The GenBank accession number for the sequence reported in this

article is L27074.

LITERATURE ClTED

Al-Feel W, Chirala SS, Wakil SJ (1992) Cloning of the yeast FAS3 gene and primary structure of yeast acetyl-COA carboxylase. Proc Natl Acad Sci USA 8 9 4534-4538

Bettey M, Ireland RJ, Smith AM (1992) Purification and characterization of acetyl COA carboxylase from developing pea embryos. J Plant Physioll40 513-520

Cahoon EB, Lynch DL (1991) Analysis of glucocerebrosides of rye (Secale cereale L. cv Puma) leaf and plasma membrane. Plant Physiol95 58-68

Cahoon EB, Shanklin J, Ohlrogge JB (1992) Expression of a coriander desalturase results in petroselinic acid production in transgenic tobacco. Proc Natl Acad Sci USA 8 9 11184-11188

Charles DJ, Cherry JH (1986) Purification and characterization of acetyl-CoA carboxylase from developing soybean seeds. Phyto- chemisbry 2 5 1067-1071

Charles DJ, Hasegawa PM, Cherry JH (1986) Characterization of acetyl-CoA carboxylase in the seed of two soybean genotypes. PhytochLemisty 2 5 55-59

Deerburg S, von Twickel J, Forster H-H, Cole T, Fuhrmann J, Heise K-P (1990) Synthesis of medium chain fatty acids and their incorpoi:ation into triacylglycerols by cell free fractions from Cu- phea em,bryos. Planta 180 440-444

Devereux J, Haeberli P, Smithies O (1984) A comprehmsive set of sequence analysis programs for the VAX. Nucleic Acids Res 12 387-39!5

Ebel J, Hahlbrock K (1977) Enzymes of flavone and flavonol gly- coside biosynthesis. Coordinated and selective induction in cell- suspension cultures of Petroselinum hortense. Eur J Biochem 7 5 201-20!)

Ebel J, Schmidt WE, Loyal R (1984) Phytoalexin synthesis in soybean cells: elicitor induction of phenylalanine ammonia-lyase and chalcone synthase mRNAs and correlation with phytoalexin ac- cumulaiion. Arch Biochem Biophys 232: 240-248

Egin-Buhler 8, Ebel J (1983) Improved purification and further characterization of acetyl-COA carboxylase from cultured cells of parsley (Petroselinum hortense). Eur J Biochem 133 335-339

Egli MA, Gengenbach BG, Gronwald JW, Somers DA, Wyse DL (1993) Characterization of maize acetyl-coenzyme A carboxylase. Plant Plhysiol 101: 499-506

Finlayson SA, Dennis DT (1983) Acetyl-coenzyme A carboxylase from the developing endospenn of Ricinus cominunis. Arch Biochem Biophys 225 576-585

Goodall GJ, Filipowicz W (1991) Different effects of intron nucleotide cornposition and secondary structure on pre-mR.NA splicing in monocot and dicot plants. EMBO J 1 0 2635-2644

Goodwin TW, Mercer E1 (1983) Introduction to Plant Biochemishy, Ed 2. Pergamon Press, New York, p 545

Gornicki P, Haselkorn R (1993) Wheat acetyl-COA carboxylase. Plant Mo1 Biol 22 547-552

Hall TC, Ma Y, Buchbinder BU, Pyne JW, Sun SM, Bliss FA (1978) Messenger RNA for G1 protein of French bean secds: cell free translation and product characterization. Proc Natl Acad Sci USA

Harwood JL (1988) Fatty acid metabolism. Annu Rev r'lant Physiol Plant h4ol Biol 3 9 101-138

James DVV, Dooner HK (1990) Isolation of EMS-induced mutants in Arabidopsis altered in seed fatty acid composition. Theor Appl Genet E10 241-245

Joshi CP (1987) An inspection of the domain betwtten putative TATA box and translation start site in 79 plant genes. Nucleic Aads R.es 1 5 6643-6653

Kannangiara CG, Stumpf PK (1972) Fat metabolism in higher plants. A procaryotic type acetyl COA carboxylase in spinach chloroplasts. Arch Biochem Biophys 152 83-91

Keegstra K, Olsen LJ, Theg SM (1989) Chloroplastic precursors and their transport across the envelope membranes. Anriu Rev Plant Physiol Plant Mo1 Biol40 471-501

Kim K-H[, Lopez-Casillas F, Bai D-H (1989) Role of reversible phosphorylation of acetyl-COA carboxylase in long-chain fatty acid synthesis. FASEB J 3 2250-2256

Kionka (2 , Amrhein N (1984) The enzymatic malonation of 1- aminocyclopropane-1-carboxylic acid in homogenatc-s of mung- bean hypocotyls. Planta 162 226-235

Li S-J, Cronan JE (1992a) The gene encoding the biotiri carboxylase subunit of Escherichia coli acetyl-COA carboxylase. 1 Biol Chem

Li S-J, Cr'onan JE (1992b) The genes encoding the two c;irboxyltrans- ferase subunits of Escherichia coli acetyl-COA carboxylase. J Biol Chem :!67: 16841-16847

Liu Y, HLoffman NE, Yang SF (1983) Relationship between the malonation of 1 -aminocyclopropane-1-carboxylic acid and D- amino (acids in mung-bean hypocotyls. Planta 158 437-441

7 5 31915-3200

267: 855-863



Arabidopsis Acetyl-COA Carboxylase 61 7

Lopez-Casillas F, Bai D-H, Lu0 X, Kong I-S, Hermodson MA, Kim K-H (1988) Structure of the coding sequence and primary amino acid sequence of acetyl-coenzyme A carboxylase. Proc Natl Acad

Lutcke HA, Chow KC, Mickel FS Moss KA, Kern HF, Scheele GA (1987) Selection of AUG initiation codons differs in plants and animals. EMBO J 6 43-48

Myers RM, Larin 2, Maniatis T (1985) Detection of single base substitutions by ribonuclease cleavage at mismatches in RNA:DNA duplexes. Science 230 1242-1246

Nikolau BJ, Wurtele ES, Caffrey J, Chen Y, Crane V, Diez T, Huang J-Y, McDowell MT, Shang X-M, Song J, Wang X, Weaver LM (1993) The biochemistry and molecular biology of acetyl-COA carboxylase and other biotin enzymes. In N Murata, C Somerville, eds, Biochemistry and Molecular Biology of Membrane and Storage Lipids of Plants. American Society of Plant Physiologists, Rock- ville, MD, pp 138-149

Pollard MR, Stumpf PK (1980) Biosynthesis of C20 and C22 fatty acids by developing seeds of Limnanthes alba. Plant Physiol 66

Post-Beittenmiller D, Jaworski JG, Ohlrogge JB (1991) Zn vivo pools of free and acylated acyl camer proteins in spinach. Evidence for sites of regulation of fatty acid biosynthesis. J Biol Chem 266

Post-Beittenmiller D, Ohlrogge JB, Jaworski JG (1993) Regulation of plant lipid biosynthesis: an example of developmental regulation superimposed on a ubiquitous pathway. In DPS Verma, ed, Con- trol of Plant Gene Expression. CRC Press, Boca Raton, FL, pp

Post-Beittenmiller, D, Roughan PG, Ohlrogge JB (1992) Regulation of plant fatty acid biosynthesis: analysis of acyl-COA and acyl- ACP substrate pools in spinach and pea chloroplasts. Plant Physiol

Roessler PG, Ohlrogge JB (1993) Cloning and characterization of the gene that encodes acety-coenzyme A carboxylase in the alga Cyclotella cyptica. J Biol Chem 268 19254-19259

Sci USA 8 5 5784-5788

649-655

1858-1865

157-174

100 923-930

Roughan PG, Laing WA (1982) Activation of spinach chloroplast acetyl-coenzyme A carboxylase by coenzyme A. FEBS Lett 144:

Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning. A Laboratory Manual, Ed 2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Samols D, Thornton CG, Murtif VL, Kumar GK, Haase FC, Wood HG (1988) Evolutionary conservation among biotin enzymes. J Biol Chem 263 6461-6464

Sasaki Y, Hakamada K, Suama Y, Nagano Y, Furusawa I, Matsuno R (1993) Chloroplast-encoded protein as a subunit of acetyl-coA carboxylase in pea plant. J Biol Chem 268 25118-25123

Shintani DK, Ohlrogge JB (1993) Feedback regulation of fatty acid synthesis in tobacco cell suspension cultures (abstract No. 54). Plant PhysiollO2 S-11

Shorrosh BS, Dixon RA, Ohlrogge JB (1994) Molecular cloning, characterization, and elicitation of acetyl-COA carboxylase from alfalfa. Proc Natl Acad Sci USA 91: 4323-4327

Simcox PD, Garland W, DeLuca V, Canvin DT, Dennis DT (1979) Respiratory pathways and fat synthesis in the developing castor oil seed. Can J Bot 57: 1008-1014

Slabas AR, Hellyer A (1985) Rapid purification of a high molecular weight subunit polypeptide form of rape seed acetyl COA carboxylase. Plant Sci 39 177-182

Somers DA, Keith RA, Egli MA, Marshall LC, Gengenbach BG, Gronwald JW, Wyse DL (1993) Expression of Accl gene-encoded acetyl-coenzyme A carboxylase in developing maize (Zea mays L.) kemels. Plant Physiol 101: 1097-1101

Takai T, Yokoyama C, Wada K, Tanabe T (1988) Primary structure of chicken liver acetyl-coenzyme A carboxylase deduced from cDNA sequence. J Biol Chem 263 2651-2657

Turnham E, Northcote DH (1983) Changes in the activity of acetyl- COA carboxylase during rape-seed formation. Biochem J 212

Yanai Y, Mitsukawa N, Liu Y-G, Whittier RF, Shimada H (1993) RFLP mapping of an Arabidopsis acetyl-COA carboxylase gene (abstract No. 382). Plant Physiol102 S-70

341-344

223-229



structure and expression of an arabidopsis acetyl …each of 'i'orula yeast rna, with or...

Documents