Eur. J. Biochem. 204,491 -499 (1992) Q FEBS 1992

Cloning and expression of a human ATP-citrate lyase cDNA Nabil A. ELSHOURBAGY Joseph C. NEAR', Peter J. KMETZ', Timothy N. C. WELLS', Pieter H. E. GROOT', Barbara A. SAXTY2, Stcphen A. HUGHES3, Michelle FRANKLIN3 and Israel S. GLOGER3

' Department of' Molecular Genetics, SmithKline Beecham Pharmaceuticals, King of Prussia, USA ' Atherosclerosis Programme and Molecular Genetic Section SmithKline Beecham Pharmaceuticals, The Frythe, Herts, England (Received August 27,1991) - EJB 91 1151

A full-length cDNA clone of 4.3 kb encoding the human ATP-citrate lyase enzyme has been isolated by screening a human cDNA library with the recently isolated rat ATP-citrate lyase cDNA clone [Elshourbagy et al. (1990) J. Biol. Chem. 265, 14301. Nucleic-acid sequence data indicate that the cDNA contains the complete coding region for the enzyme, which is 1105 amino acids in length with a calculated molecular mass of 121419 Da. Comparison of the human and rat ATP-citrate lyase cDNA sequences reveals 96.3% amino acid identity throughout the entire sequence. Further sequence analysis identified the His765 catalytic phosphorylation site, the ATP-binding site, as well as the CoA binding site. The human ATP-citrate lyase cDNA clone was subcloned into a mammalian expression vector for expression in African green monkey kidney cells (COS) and Chinese hamster ovary cells (CHO) cells. Transfected COS cells expressed detectable levels of an enzymatically active recombinant ATP-citrate lyase enzyme. Stable, amplified expression of ATP-citrate lyase in CHO cells was achieved by using coamplification with dihydrofolate reductase. Resistant cells expressed high levels of enzymatically active ATP-citrate lyase (3 pg/cell/d). Site-specific mutagenesis of His765+Ala diminishes the catalytic activity of the expressed ATP-citrate lyase protein. Since catalysis of ATP-citrate lyase is postulated to involve the formation of phosphohistidine, these results are consistent with the pattern of earlier observations of the significance of the histidine residue in catalysis of the human ATPcitrate lyase.

ATP-citrate lyase is the primary enzyme responsible for the synthesis of cytosolic acetyl-CoA in many tissues. The enzyme is a tetramer (molecular mass of 440000 Da) of four apparently identical subunits [l]. The synthesized acetyl-CoA serves several important biosynthetic pathways, including lipogenesis and cholesterogenesis[2]. Catalysis of ATP-citrate lyase involves the participation of a phosphoenzyme intermediate [3] resulting from the phosphorylation of the enzyme at the catalytic site by the substrate Mg . ATP in the first step of the reaction [4]. A maximum of two phosphate groups are incorporated at the catalytic site/ enzyme tetramer [5]. The resulting phosphoenzyme reacts with citrate and CoA to yield the products acetyl-CoA and oxaloacetate [6]. In addition to the catalytic site, serine residues can be phosphorylated at two chemically distinct regulatory sites located on two tryptic peptides [7]. Phosphorylation of these sites has been observed in vitro in response to both CAMP-dependent(A site) [8] and CAMP-independent (B sites) protein kinases [9]. The phosphorylation of the enzyme is enhanced by glucagon, insulin, vasopressin and transforming growth factor PI, however, phosphorylation of the regulatCorrespondence lo N. A. Elshourbagy, Dept. of Molecular Genetics, SmithKline Beecham Pharm., P.O.Box 1539, L-48 King of Prussia, PA 19406 USA AbbrPviations. CHO, Chinese hamster ovary cells; COS,African grcen monkey kidney cells; DHFR, dihydrofolate rcductase. Enzyme. ATP-citrate Lyase (EC4.1.3.8).

ory sites does not affect the enzyme activity to any appreciable extent [lo]. The physiological function of the regulatory sites is therefore unclear. Since ATP-citrate lyase is a potential target for hypolipidemic intervention, we have recently isolated and characterized a cDNA clone of the rat-liver enzyme [ll]. This study reports the complete structure of the human ATP-citrate lyase and the finding of an extensive similarity to the rat enzyme. In addition, the sequence information plus site-specific mutagenesis allowed us to speculate on the catalytic phosphorylation site and regulatory elements in the human ATP-citrate lyase. Finally, the human ATP-citrate lyase cDNA was subcloned into a mammalian expression vector and used to produce a stable Chinese hamster ovary (CHO) cell line that expresses high levels of catalytically active enzyme.

EXPERIMENTAL PROCEDURES Screening of the human cDNA library Human ATP-citrate lyase cDNA clones were selected from a AZAP liver cDNA library by screening approximately 5 x lo5 clones [12] at a reduced stringency with a previously characterized rat ATP-citrate lyase cDNA [l I]. The hybridization probe was a 1.2-kbp Pstl- Sac1 restriction endonuclease fragment of the rat cDNA [l 11that was 32Plabeled by random priming [13]. Plaque hybridizations were carried out at 42°C in a buffer containing20% deionized formamide, 0.9 M NaCl,

@C U g C C I F C U g ctg 8 t C 8.8 V a l V a l ~ y mP r o Amp C l n &u 11. Gym Arg Arg -8

888

ctt ggt CkC gtt #C Q . C c t C act C t g gat &C mag t C C tgg ctg 8.9 CCa -9 a Gly u h u V a l cly V a l u p L.u T h r Lou k p cly V a l ~ y msrr ~ r Lou p Lym P r o &g

cly ~ y m~

ctg Wa a 9 L.u C l y C l n C l u

756 816

996

1116

1236

1356

1476

1596

1716

1836

1956

2076

2196

t a t gtc ctt gac t t g gcq gcc sag gtg gac gcc act gcc gac tac a t c tgc u a qtg u g tqg ggt gac atc gag ttc cct ccc ccc t t c ggg ogg gtg gca t a t cca g8g gaa gcc tac Tyr V a l b u Asp Lou Ala A h Lym V a l Asp Ala T h r Ala Amp Tyr 11. C y m Lym V a l L y a Trp C l y Amp 11. C l u P h r P r o P r o P r o P h r C l y Arg V a l Ala Tyr P r o C1u ClU A h Tyr

a t t gca gac ctc gat gcc 81 . aqt ggg g u agc ctg aag ctg acc t t g c t g aac ccc u r qgg agg a t c tgg acc atg q t g gcc ggg ggt ggc gcc tct g t c gtg tac agc gat acc atc 11. Ala Amp I a u Asp Ala Lym S r r C l y Ala S r r h u Lym Lou T h r Lou L.u k n P r o Lym C l y &g 11. T r p T h r llrt V a l Ala C l y C l y C l y Ala S i r V a l V a l T y r Sor Amp T h r I10

t g t gat cta 999 ggt gtc a8c gag ctg g u u c t a t gqg gag t a c tca ggc gcc ccc agc qag u g u g acc t a t gac t a t gcc aag act atc ctc t c c c t c atg acc cga gag 8.9 U C C y m Asp Lou G l y C l y V a l k n C l u Lou Ala Amn Tyr G l y C l u Tyr S r r C l y Ala P r o Srr C l u C l n Cln T h r Tyr Amp T y r Ala Lyy. T h r 110 Lou Sir Xau Mmt T h r A t g ClU Lym Hi.

cca gat ggc aag atc ctc a t c a t t qga gqc rgc a t c gca u c ttc acc aac gtg gct qcc acq ttc aag w c atc gtg aga gca a t t cqa gat tac Crg ggc ccc C t g aag gag cac gaa P r o Asp C l y Lym I10 Lou 11. 110 C l y G l y S i r 11- Ala Amn P h r T h r Amn V a l Ala Ala T h r P h r Lym C l y 110 V a l Arg A h 11. Arg &p Tyr C l n C l y P r o Lou LYs C l u H i m ClU

@C 8- atC ttt # C CQa 89' ggt WC CCC IaC t a t -9 989 ggC tt8 -9 @g atg gga C C 889 aCC act 8 t C CCC atC cclt @C ttt QQC aC8 plQ U C atg SCg V a l T h r 11. P h r V a l Arg Arg C l y C l y P r o Asn Tyr G l n Clu C l y Iau Arg V a l Mot C l y C l u V a l C l y t y a T h r Thr C l y 110 P r o Ilr H i m V a l P h r C l y Thr C l u Thr His Not Thr

gCC a t t gtg gQC atg gCC tgg CCg gCC 8 t C CCC U C C8g Ec. CCC 8 U gCg gCC U C act aaC ttt &C &C aaC gCC U g Ogg gag 8Ca tCg act C C I gCC CCC agC Ala 11. V a l C l y llrt Ala T r p Ala P r o Ala 11. P r o Amn C l n P r o P r o T h r Ala Ala Him T h r Ala Amn P h r L.u Lou Amn Ala C l n Arg C l u Thr E r r Thr P r o Ala P r o Ilrr Arg T h r

gca tct ttt t a t gag tcc atg gtc gat gag gtc agg gcc gat gag qtg gcg cct gcr aag sag gcc u g cct gcc atg cca u a gat tea g t c cca agt cca aga t c c Ctg u a gga mag Ala 9.1 P h r Tyr C l u Srr Not V a l Amp C l u Val Arg Ala Amp C l u V a l Ala P r o Ala Lym Lym Alr Lym P r o Ala llrt P r o C l n Amp Sor V a l P r o E r r P r o Arg E r r Lou C l n C l y Lya

agc acc acc ctc

t t c agc cgc cac acc aag gcc a t t gtg tgg ggc atg cag acc cgg gcc gtg car ggc atg ctg gac ttt gac t a t gtc tgc t c c cga gac gag ccc tca gtg gct gcc S r r T h r T h r I m u P h r Srr Arg H i m T h r Lyy. 111. 11. V a l T r p C l y Mot G l n T h r Arg Ala V a l C l n C l y Y.t Lou Asp Phm Amp Tyr V a l C y m Srr Arg Amp C l u P r o 8.r V a l Ala Ala

atg gtc t a t c c t t t c act ggg gac u c u g u g u p ttt t a c tgg ggg cac u a gag a t c ctg atc cct gtc ttc 8.9 aac atg gct gat gcc atg agg aag u c ccg gag gta gat qtg Mot V a l Tyr P r o Ph- T h r G l y Aap Him Lym C l n Lym P h r Tyr T r p Gly Him Lym Glu 11. Lou 11. P r o V a l P h r Lym Amn Mot Ala Amp Ala fit Arg Lym Hi. P r o C l u V a l Amp V a l

ctc a t c u c ttt gcc tct ctc cgc tct gcc t a t gac agc acc at9 gag acc atg aac t a t gcc cag atc cgg acc a t c gcc a t c a t a gct gaa ggc a t c cct gag gcc c t c acg aga aag Lou 11. Asn P h r Ala S r r Lou Arg S r r Ala Tyr Amp S r r T h r bbt C l u Thr Xot Amn Tyr Ala C l n 11. Arg T h r 11. Ala 11. 110 Ala C l u C l y I10 P r o C l u Ala Lbu Thr Arg L y a

ctg a t c u g aaq gcg g.c u g rag gga gtg acc a t c a t c gga cct gcc act g t t gga ggc a t c aag cct ggg tgc ttt aag a t t ggc aac aca ggt ggg atg ctg gac aac atc ctg gcc Lou 11. Lym Lym Ala Amp C l n Lya C l y V a l T h r 11. 11. C l y P r o Ala T h r V a l C l y C l y 11. Lym P r o C l y C y m P h r Lym 110 C l y Amn T h r C l y C l y Mot L o u Asp Amn 110 I a u A h

tcc aaa ctg t a c ccc c r g q u gct gtg gcc t a t g t c tu cgt tcc gga ggc atg tcc aac gag c t c aac aat a t c a t c t c t cgg acc acg gat ggc g t c t a t gag ggc gtg gcc a t t ggt S r r Lya Lou Tyr P r o C l n Ala Ala V a l Ala Tyr V a l S r r Arg 3-r C l y C l y bbt S r r Amn C l u Lou Amn Amn Ilr 11. Srr &g T h r "hr Amp C l y V a l T y r C l u C l y V a l Ala Ilr C l y

ggg gac 899 t i c c ~ ggc g t c c 8 U t t c atg gat ut gtq t t a cgc t a t cag gac act cca gqa qtc aaa atg a t t gtg g t t ctt gqa gag a t t ggg ggc act gag gaa t a t aag a t t t c c C l y Amp Arg T y r P r o G l y S-r T h r P h r M o t Asp H i . V a l Lou Arg Tyr C l n Asp T h r P r o Cly V a l Lym Xot 11- V a l V a l Lou Cly C l u 11- C l y C l y T h r C l u C l u Tyr Lym 11- hr

cgg ggc a t c a-9 gag ggc CQC ctc act aag ccc a t c g t c tgc tgg t g c a t c ggg acg tgt gcc a c c atg t t c t c c tct gag gtc c.g ttt ggc ut gct gga gct t g t gcc aac u g gct Arg C l y 11- LYm ClU G l y Arg Lou T h r Lym P r o 11- V a l C y m Trp Cy8 110 Cly T h r Cy8 Ala T h r J8et P h r Srr S i r Clu V a l C l n P h r G l y H i m Ala C l y N a C y m A8n C l n Ala

637

757

877

997

1117

1237

1357

1477

1597

1717

1837

1957

2077

2 1 97

2316

636

gcc cct gac qac arg aaa 9.8 a t t ctg gcc agt ttt a t c tcc qgc ctc ttc ut ttc t a c gag gac t t g tac ttc rcc t a c ctc gag atc aat ccc ctt gta gtg acc aaa gat gga gtc A h P r o Asp Asp Lym Lym C l u 11- L o u ' A l a S r r Ph- Ilr arr C l y L . u P h r &n Pho Tyr C l u Amp Lou Tyr P h r Thr Tyr Lou C l u 111 A8n P r o Lou V a l V a l Thr l y m Amp C l y V a l

517

3 96

276

516

aca gtt ggc aag gcc a m qgc ttc c t c aaq mc ttt ctg a t c gag ccc t t c gcc ccc u c agt cag gct gag gag ttc t a t gtc tgc atc t a t gcc acc cga gaa ggq gac tac gtc T h r V a l G l y L y a Ala T h r C l y P h o I a u Lym Amn P h r Xau 11- G l u P r o P h r Ala P r o Him Sot C l n A h Clu C l u P h r Tyr V a l C y m 11. Tyr Ala T h r Arg C l u C l y A8p T y l V a l

LOU

156

35

P C @g ggt 9.t @g P C gCC a89 gCC Crg 8.q Ckg &t gtt WC @cJ gat 9.9 U a ctg art C c t 989 gaC atC 8.. &g t t C C8C U C Q.9 WQ C . C &g t t g *C U c Lou P h r Him Mia Clu C l y C l y V a l Amp V a l C l y Amp V a l Amp Ala L y m Ala C l n L y m L o u Lou V a l C l y V a l Asp C l u Lym Iau Amn P r o C l u Amp Z l r Lym Lys H i m L o u Lou V a l H i m

fla

QCC

c l n Amn

SaC t t g

c t c ctt t a c aag t t c atc tgt acc acc tu gcc a t c u g aat cgg ttc u g t a t gct cgg gtc act cct gac a u g4c tgg gcc cgc t t g ctg cag gac CIC ccc tgg ctg ctc Clu Lou Lou T y r Lym P h r 11- Cya T h r T h r Srr Ala I l r C l n Amn Arg P h r Lya T y r Ala A r g V a l T h r P r o Amp T h r Amp T r p Ala Arg Lou Lou C l n Amp a i r P r o T r p Lou Lou

9.8

CCcgg8ttttg~~gtt~cgggcdgtgQ.8g8rgCCCcgCC~Cgg~cttcggC8g~ggtt.g~gUg~CtctC atg t g utCg gCC gCC a.9 gCa 8tt tCa gag U g aCg ggC a.8 Mot S r r Ala Lym Ala 11. S i r C l u C l n T h r C l y Lym

397

277

157

37

-84

P h,

u3

2556

2676

2196

2916

3036

3156

3276

3423

3582 3741 3900 4059 4213

2431

2557

2677

2797

2917

3037

3157

3217

3424

3583

3742

3901

4060

Fig. 1. Nucleotide and deduced amino acid sequence of ATP-citrate lyase cDNA and protein, respectively. Nucleotide residues are numbered in the 5‘ to 3’ direction, beginning with the first residue of the ATG triplet, encoding the initiation codon for Metl. The nucleotides at the 5’ side of residue 1 are indicated by negative numbers; the number of the nucleotide residue is given at the right and left of each line. Deduced amino acid residues are indicated, beginning with the initiation methionhe. The translation termination codon is indicated.

2436

2311

P

w

\D

494 50 mM sodium phosphate, pH 7.0, 5 mM EDTA, 0.1'/0 SDS and 200 pg/ml denatured herring sperm DNA. The cDNA inserts from positive recombinants were subcloned into pBluescript for nucleotide sequence determinations by the dideoxynucleotide chain-termination method [14]. Sequence data was analyzed using the Wisconsin system [15]. Site-specific mutation and expression of human ATP-citrate lyase in African green monkey kidney, (COS) cells

The EcoRI fragment containing the entire human ATPcitrate lyase cDNA sequence was subcloned into the mammalian expression vector pRJB4, driven by the Rous-sarcomavirus promoter, and the resulting construct (pRJB4HCL) ; Fig. 4a) was used for mutogenesis studies. Site-specific mutation of His765-tAla in human ATP-citrate lyase was constructed using the polymerase-chain reaction; four oligonucleotide probes were constructed as shown.

96-well plates in F12 medium (Gibco) supplemented with 10% fetal calf serum. After 48 h, the cells were grown in the same media containing neomycin (400 pg/ml) for 2 weeks. Surviving cells were assayed for the production of ATP-citrate lyase. Cells that expressed the highest level of the enzyme were selected, then grown in Dulbeceo's modified Eagle's medium without nucleosides and supplemented with both neomycin and 10 nM methotrexate with 7.5% dialyzed fetal calf serum. After 2 weeks, cells surviving the methotrexate selection were allowed to grow and assayed for ATP-citrate lyase using an enzymatic assay and Western-blot analysis. For amplification, 1 x lo6 cells, resistant to 10 nM methotrexate, were plated in 90-mm plates and exposed initially to 50 nM methotrexate for 2 weeks. Surviving cells (1 x lo6)were replated and exposed to 100 nM methotrexate. After 10 d, resistant clones were pooled and subjected to stepwise increments in methotrexate concentrations up to 5 pM. Single-cell clones, resistant to 1 mM methotrexate, were isolated by end-limited dilution.

KpnI

1

(a) GGGGACAGGTACCCGGGCTCCACATTCATG. (b) GTCCAGTTTGGCGCCGCTGGAGCTTGTGCC.

t

NarI (c) GGCACAAGCTCCAGCGGCGCCAAACTGGAC.

7

NarI

(d) TCCATTGGCCACGAGATCTTCGTATACAGAC.

t

Xcal

Probe b contains a His765 4Ala, which is underlined. Probe a and c were used to generate KpnI -NarI fragments and probes b and d were used to generate NarI - XcaI fragments. These fragments were synthesized using the polymerase-chain-reaction technique followed by digestion of each fragment with the above-designated restriction enzyme. The two fragments were then subcloned into pRJB4HCl that was previously digested to delete the KpnI - XcuI fragment. The pRJB4, pRJB4HCL and pRJB4HCL (His -+ Ala) mutant were then transfected into COS cells. The day prior to transfection, cells were seeded at 1 x lo6 cells/100 mm tissue culture dish and allowed to grow until they reached 75% confluency. The pRJB4HCL vector was introduced into COS cells by the dextran-sulfate - DEAE method [I61 and the cells were collected 48 h later. Transiently transfected cells were examined for the expression of human ATP-citrate lyase by Western-blot analysis using rabbit anti-(rat ATP-citrate lyase) serum, produced by injecting rabbits with rat purified protein [l 11 and by measuring the activity of the enzyme as described below. Stable expression and amplification of human ATP-citrate lyase in CHO cells

The EcoRI fragment containing the entire human ATPcitrate lyase cDNA sequence was subcloned into the mammalian expression vector RLDN (Fig. 4b) which contains both the G418 and dihydrofolate reductase (DHFR) resistance genes. The CHO cells were transfected with RLDNHCL vector using electroporation 1171, then the cells were plated in

Analysis of ATP-citrate lyase protein synthesis

Cells, selected for resistance to methotrexate at different concentrations, were mechanically detached from the dishes in cold 140mM NaCl, 2.7mM KCl, 8 m M Na2HP04 and 1.5 mM KH2P04 (NaCl/Pi). Enzymatic-activity assays were performed as described [ll]. Briefly, citrate cleavage by ATPcitrate lyase is coupled to the malate dehydrogenase reaction and measures the conversion of NADH to NAD' at 340 nm [18].The reaction is performed in buffer containing 10 mM 2-mercaptoethanol. SDS/PAGE was performed according to Laemmli, et al. [19]. Cells, in NaC1/Pi, were centrifuged and resuspended in 50 mM Tris/HCl, pH 6.8, 100 mM dithiothreitol, 2% SDS, 0.1 % bromophenol blue and 10% glycerol. Western-blot analysis was performed using a polyclonal antibody raised in rabbits against native rat ATP-citrate lyase [Ill. Purified rat ATP-citrate lyase was used as a positive control in all gels. Northern-blot analysis of ATP-citrate lyase RNA synthesis

Total cytoplasmic RNA from CHO cells was extracted using the guanidinium isothiocyanate method and purified through a CsCl cushion [20]. Poly(A)-rich RNA was isolated using oligo(dT) columns [21]. 10 pg total RNA or 5 pg poly(A)-rich RNA were electrophoresed through 1% agarose gels containing formaldehyde. The gel was blotted to a Hybond nylon membrane (Amersham, VIC) by pressure blotting in 20 x 0.3 M sodium citrate and 3 M sodium chloride. RNA was cross-linked to membranes by baking for 2 h at 80"C. Prehybridization and hybridization were carried out as described previously [I 13. RESULTS AND DISCUSSION

We screened approximately 500000 recombinants of a human liver cDNA library with a 1.2-kbp PstI - Sac1 restriction endonuclease fragment of the rat ATP-citrate lyase cDNA clone [Ill. This screening yielded six candidate clones. Restriction analysis indicated that the insert sizes were all approximately 4.3 kbp in length. The DNA from four of these candidates were used to probe human liver mRNA by Northernblot analysis and were found to hybridize to an mRNA of approximately 4.3 kb (data not shown). Thus, it appeared that

49 5 the recombinant cDNA species were essentially full-length copies of the corresponding RNA. Primary structure of human ATP-citrate lyase The complete sequences of both strands were determined. There was only one open reading frame, beginning at a methionine codon (nucleotide 1) and ending at the stop codon, TAA (nucleotide 3318; Fig. 1). The cDNA insert also contained 5’ and 3’ non-translated regions of 84 and 896 nucleotides, respectively. The latter region contains the polyadenylation signal, AATAAA, beginning at nucleotide 41 82. The deduced amino acid sequence contains 1105 residues with a calculated M , of 121419. This value is in good agreement with the M , of 120000 estimated for the ATP-citrate lyase by gel electrophoresis. The human ATP-citrate lyase amino acid sequence was compared to that of the rat and found to be identical in 96.3% of the overall positions (Fig. 2). The inclusion of conservative replacements increases the degree of similarity to 98%. To achieve an optimal alignment, a 5-amino-acid-residue gap in the rat sequence was postulated (amino acids 459 -463 of the human sequence). In the rat, ATP-citrate lyase is phosphorylated in vivo in response to insulin, glucagon, p-adrenergic agonists and transforming growth factor pl [22] and in vitro by CAMP-dependent protein kinase [8]. This latter phosphorylation site is the first serine in the peptide TASFSES (amino acids 452 -458) with Ser454 being the phosphorylation target. This sequence is highly conserved in the human, but a 5-amino-acid gap needed to be inserted in the rat sequence to achieve optimal alignment. Whether this sequence has significance with respect to the human ATP-citrate lyase phosphorylation is at present unknown. As postulated earlier with the rat ATP-citrate lyase 1111, the human ATP-citrate lyase amino acid sequence exhibited strong similarity with the a chain of succinyl CoA synthetase of Escherichia coli 1231. Analysis using the FASTA program 1241 indicates that the ATP-citrate lyase (amino acids 550800) and bacterial succinyl-CoA synthetase a chain (amino acids 45 -283) shows a statistically significant sequence identity of 33% (Fig. 3). The inclusion of conservative replacement increases the degree of similarity to 77%. The two enzymes catalyze similar reactions. Succinyl-CoA synthase involves autophosphorylation of the enzyme by ATP, phosphorylation of succinate by phosphoenzyme, and attack of CoA on succinate phosphate to form succinyl-CoA. With ATP-citrate lyase, the pattern is identical as far as the formation of citryl-CoA. However, this unstable compound undergoes a retroClaisen reaction to produce acetyl-CoA and oxaloacetate. Comparison of the two sequences reveals a 5-amino-acid identity, GHAGA (labeled ATP A in Fig. 3), which surrounds the catalytic histidine residue in succinyl-CoA synthetase. This histidine is autophosphorylated by ATP in the first step of the enzyme reaction. Since the ATP-citrate lyase reaction also involves a phosphohistidine intermediate, it seems reasonable to suggest that His765 is the site of catalytic phosphorylation in ATP-citrate lyase. Evidence for this supposition was obtained by mutagenesis experiments (see below). Further comparisons revealed that, within the regions of similarity between the two enzymes, there is a sequence (labeled ATP B in Fig. 3) which is similar to the consensus sequence for part of the ATP-binding pocket. This region is predicted to form part of the ATP-binding site [Ill. A model for a consensus CoA-binding site has been proposed based on the available X-ray crystallographic data from citrate synthase and hom-

1 MSAKAISEQTGKELLYKYICTTSAIQNRFKYARVTPDTDWAHLLQDHPWL5 0

lIllIllIlllollll:llIIlIIIIIIIIl111111III:11111111

1 MSAKAISEQTGKELLYKFICTTSAIQNRFKYARVTPDTDWARLLQDHPWL 50

51 LSQSLVVKPDQLIKRRGKLGLVGVNLSLDGVKSWLKPRLGHEATVGKAKG 1 0 0 /ll.lll/l//llllIlllIIlIl:l.lllllllllllll:lllllll.l 51 LSQNLVVKPDQLIKRRGKLGLVGVDLTLDGVKSWLKPRLGQEATVGKRTG 1 0 0 101FLKNFLIEPFVPHSQAEEFYVCIXATREGDYVLFHHEGGVDVGDVDTKAQ 1 5 0

llllllllll.lllllllllllllllllllllllllllIIIIIIII.III

101FLKNFLIEPFAPHSQAEEFYVCIXATREGDYVLFHHEGGVDVGDVDAKAQ 1 5 0

151 KLLVGVDEKLNAEDIKRHLLVHAPEDKKEILASFISGLFNFYEDLYFTYL IIIIIIIIIII:IIII:IIlIIII:IIIII/IIIIIlIl/IIIIIIIIII

200

1 5 1 KLLVGVDEKLNPEDIKKHLLVHAPDDKKEILASFISGLFNFYEDLYFTYL2 0 0 250 2 0 1 EINPLVVTKDGVYILDLAAKVDATADYICKVKWGDIEFPPPFGREAYPEE

lllllllllllll:IllllIlIlIllllIIIlllllIlllIIll

lllll

2 0 1 EINPLWTKDGVYVLDLAAKVDATADYICKVKWGDIEFPPPFGRVAYPEE 250

251 AYIADLDAKSGASLKLTLLNPKGRIWTMVAGGGASWYSDTICDLGGVNE 300 Illllllllll/lllllIl/Il/llll/II/l//l/lllllll//lIl11 2 5 1 AYIADLDAKSGASLKLTLLNPKGRIWTMVAGGGASWYSDTICDLGGVNE 300 3 0 1 LANYGEYSGAPSEQQTYDYAKTILSLMTREKHPDGKILIIGGSIANFTNV

350 l//llllllllllllllllIllllllllllllllll/lIll/llllllll 301 LANYGEYSGAPSEQQTYDYAKTILSLMTREKHPDGKILIIGGSIANFTNV 350 351 AATFKGIVRAIRDYQGSLKEHEVTIFVRRGGPNYQEGLRVMGEVGKTTGI 400 Illl11lIIIlIIIlI.11111l1111111111lIIIIIIIIIIlIIIII 351 AATFKGIVRAIRDYQGPLKEHEVTIFVRRGGPNYQEGLRVGEVGKTTGI 400 4o1

HVFGTETHMTAIVGMAWAPAIPNQPPTMHTANFLLNASGSTSTPAPS

PI 4 50 l1/l1llll0llllllllllllIllllllllllllI0l .lllllIl 4 0 1 PIHV_GTETHMTAIVGMAWAPAIPNQPPTAAHTANFLLNAQRETSTPAPS 4 5 0 r

.....

451 RTASFSES RADEVAPAKKAKPAMPQDSVPSPRSLQGKSATLFSRH 495 IIIII II l1111l1111111l1lllllllllllllll.llllll 451 RTASFYESMVDEVRADEVAPAKKAKPAMPQDSVPSPRSLQGKSTTLFSRH 500 496 TKAIVWGMQTRAVQGMLDFDYVCSRDEPSVAAMVYPFTGDHKQKFYWGHK 545 IlllllIII/IIlIllIIIIl1111l11l1l11111l111111/IIIIII 501 TKAIVWGMQTRAVQGMLDFDYVCSRDEPSVAAMVYPFTGDHKQKFYWGHK550

546 EILIPVFKNMADAMKXHPEVVLINFASLRSAYDSTMETMNYAQIRTIAI 5 9 5 llolIlIlIlllI:IlIllIIIIIIlIIIIIlIIlllII/lIIIIIIIl 551 EILIPVFKNMADAMRKHPEVDVLINFASLRSAYDSTMETMNYAQIRTIAI 6 0 0 5 9 6 IAEGIPEALTRKLIKKADQKGVTIIGPATVGGIKPGCFKIGNTGGMLDNI645

1ll1l1lll11llIIllllllllllilllllllIllllillltlllllll 6 0 1 IAEGIPEALTRKLIKKADQKGVTIIGPATVGGIKPGCFKIGNTGGMLDNI 650 64 6 LASKLYRPGSVAYVSRSGGMSNELNNIISRTTDGVYEGVAIGGDRYPGST

695

llIII/..:.lIlIIIIIIIIIIIIIIIIIIIIIlIlIIIIIIllllIll 6 5 1 LASKLYPQAAVAYVSRSGGMSNELNNIISRTTDGVYEGVAIGGDRYPGST 7 0 0 6 9 6 FMDHVLRYQDTPGVKMIWLGEIGGTEEYKICRGIKEGRLTKPWCWCIG

745

ll1lll1111l111lIIIIIIIIlI/IlI/I:IIIII/lllll:IIIIIl 7 0 1 FMDHVLRYQDTPGVKMIWLGEIGGTEEYKISRGIXEGRLTKPIVCWCIG

750

7 4 6 TCATMFSSEVQFGHAGACANQASETAVAKNQALKEAGVFVPRSFDELGEI 795

lIlIlIIlIllIlIlllllllll1llIllIIlllIlIlIIIIIIIIIII/ 751 TCATMFSSEVQFGHAGACANQASETAVAKNQALKEAGVFVPRSFDELGEI 800 796 IQSVYEDLVAKGAIVPAQEVPPPTVPMDYSWARELGLIRKPASFMTSICD llllllllll.l.lllllllllllllllllllllllllllllllllllll

845

8 0 1 IQSVYEDLVANGVIVPAQEVPPPTVPMDYSWARELGLIRKPASFMTSICD 850

846 ERGQELIYAGMPITEVFKEEMGIGGVLGLLWFQRRLPKYSCQFIEMCLMV 895 l111111111l11llIIIlllll/l.lllllll:llllIIIIIIllIlII 900 851 ERGQELIYAGMPITEVFKEEMGIGGALGL~I~FQKRLPKYSCQFIEMCLMV 896 TADHGPAVSGAHNTIICARAGKDLVSSLTSGLLTIGDRFGGALDAAAKMF 9 4 5 lllllllllllllllllll.: :IlIIIIllIIIIIllIIll111111l1 9 0 1 TADHGPAVSGAHNTIICARTAVELVSSLTSGLLTIGDRFGGALDAAAKMF 9 5 0 9 4 6 SKAFDSGIIPMEFVNKMKKEGKLIMGIGHRVKSINNPDMRVQILKDFVKQ

995

Illllllllllllllllll1IIIII1I1IIIIIIl1lI11111111:1:1 9 5 1 SKAFDSGIIPMEFVNKMKKEGKLIMGIGHRVKSINNPDMRVQILXDYVRQ 1000 9 9 6 HFPATPLLDYALEVEKITTSKKPNLILNVDGFIGVAFVDMLRNCGSFTRE 1045

IIlI/lllllllllllllll1/11111111/:II/lll/1ll11llllll

1 0 0 1 HFPATPLLDYALEVEKITTSKKPNLILNVDGLIGVAFVDMLRNCGSFTRE 1050

1 0 4 6 EADEYVDIGALNGVFVLGRSMGFIGHYLDQKRLKQGLYRHPWDDISYVLP

1095

IIIII:1111111:1l1111111111111111111111IIIIIIIIIIII 1051 EADEYIDIGALNGIFVLGRSMGPIGIIYLDQKRLKQGLYRHPWDDISYVLP 1100 1096 EHMSM 1100 IIIII 1101 EHMSM 1105

Fig. 2. Alignment of the amino acid sequences of rat and human ATPcitrate lyase. The optimal alignment of the deduced amino acid sequences were made with the use of the Wisconsin program [15]. The location of one of the regulatory phosphorylation sites identified earlier [Ill is indicated (P).

496 520 530 540 550 560 570 DFDYVCSRDEPSVAAMVYPFTGDHKQKFYWGHKEILIPVFKNMADAMRKHPEVDVLINFA 1 ::: : ) I ) : : : : I : ::: : I :: S y e c s a GFTGSQGTFHSEQAIAYGTKMVGGVTPGKGGTTHLGLPVFNTVREAVAATGATASVI-YV

Phcl

20

30

580

50

40

600

590

60

610

620

70

630

SLRSAYDSTMETMNYAQIRTIAIIAEGIPEALTRKLIKKADQKGVTIIGPATVGGIKPGC : : 1I::I::: I I: I : l : l l l l : :: I I : I I : I l l : I l:lI S y e c s a PAPFCKDSILEAID-AGIKLIITITEGIPTLDMLTVKVKLDEAGVRMIGPNCPGVITPGVITPGE Phcl

80 640

90 650

110

100 660

670

120 680

130 690

Phcl

FKIGNTGGMLDNILASKLYPQAAVAYVSRSGGMSNELNNIISRTTDGVYEGVAIGGDRYP Ill I : : : : : : : I : I I I I I : : : I : :: : I : I:IIIl: I S y e c s a CKIG--------IQPGHIHKPGKVGIVSRSGTLTYEAVKQTTDYGFGQSTCVGIGGDPIP 140

160

150

170

180

A T D ‘#Rq’ -” Y

700

Phcl

}

710 720 \730 740 750 GSTFMDHVLRYQDTPGVKMIVVLGEIGGTEEYKISRGIKEGRLTKPIVCWCIGTCATMFS 1 l : I : l : :::::I :: l l : : l l l l l : : l : : I l l : : l l l : l : I: 1 :

S y e c s a GSNFIDILEMFEKDPOTEAIVMIGEIGGSAEEEAAAYIKE-HVTKPWGYIAGVTA---P 190 200 210 220 230 240 ATP “ A CoA ’760 770 780 ” 790 800 810 Phcl SEVQFGHAGACANQASETAVAKNQALKEAGVFVPRSFDELGEIIQSVYEDLVANGVIVPA :: : : I I / I I : :::I1 :I :ll::lll : 1l::::Il :::I S y e c s a KGKRMGHAGAIIAGGKGTADEKFAALEAAGVKTVRSLADIGEALKTVLK

’

1250

260

270

280

Fig. 3. Comparison of the amino acid sequences of the human ATP-citrate lyase (pHCL) and the a subunit of E. coli succinyl CoA synthetase (SYECSA) 123).The bold lines indicate the possible ATP-binding and CoA-binding sites. The lines and dots indicate residues on either identical or conscrvative substitutions in the bacterial and mammalian enzymes. The arrow indicates the location of a phosphohistidine residue at the E. roli succinyl-CoA synthetase active site. One line, identical; two dots. conserved.

ologies between citrate synthase, acetyl-CoA carboxylase and propionyl-CoA carboxylase [25, 261. One section of the ATPcitrate lyase sequence, marked in Fig. 3, conforms to this pattern.

citrate lyase. An alternative explanation might be that His765 is removed from the active site and inactivation is caused by conformational distortion. Expression of ATP-citrate lyase in CHO cells

Expression of human ATP-citrate lyase by COS cells The recombinant plasmid containing the human ATPcitrate lyase cDNA (pRJB4CL, Fig. 4A) was transfected into subconfluent cultures of COS cells. RNA was extracted 48 h after transfection and Northern-blot hybridization revealed the presence of a single transcript with an estimated molecular mass of 4.3 kb (data not shown). To detect the translation products immunologically, the cell extracts as well as the medium from 48 h transfected COS cells were assayed by Western blotting, using a specific polyclonal antibody against the rat ATP-citrate lyase. A protein of M , 120000 was identified in the transfected cells but not in the media, confirming that this product is expressed as a non-secreted product. Furthermore, the recombinant ATP-citrate lyase is expressed as an enzymatically active protein. This was confirmed by measurements of ATP-citrate lyase activities in extracts of control COS cells (pRJB4, Fig. 5) and COS cells transfected with recombinant ATP-citrate lyase (pRJB4HCL). The pRJB4HCL construct was then used to confirm the significance of His765 in the catalysis. A single mutation of His765-tAla resulted in the expression of a protein with exactly the same molecular mass (120 kDa) as ATP-citrate lyase. However, this protein was catalytically inactive (Fig. 5). It has been postulated earlier that catalysis of the ATP-citrate lyase involved the participation of a phosphoenzyme intermediate, resulting from phosphorylation of the enzyme at the catalytic site by the substrate Mg . ATP in the first step of the reaction [3]. This catalytic site is phosphorylated by a maximum of two acid-labile phosphate groups/enzyme tetramer and has been shown to contain phosphohistidine (41. Our specific singlemutation studies confirm and extend these observations and identify the location of the active-site histidine in human ATP-

ATP-citrate lyase was expressed using a coamplification expression system based on DHFR expression and cell resistance to methotrexate. A clone resistant to 10 nM methotrexate showed detectable amounts of enzymatically active enzyme and was selected for further amplification. A stepwise incrementation in methotrexate concentration was performed. We selected the strategy of pooling resistant clones at each methotrexate concentration and selecting individual clones by end-point dilution at high methotrexate concentration 1271. Cell growth in CHO-resistant cells was not affected by concentrations up to 500 nM methotrexate showing a doubling time of approximately 20 h (data not shown). Cells resistant to 1 pM methotrexate grew slower, with a doubling time of approximately 30 h, while in cells resistant to 2 pM and 5 pM methotrexate, the growing rate was greatly reduced to a doubling time of approximately 50-60 h (data not shown). As a result of the slow growth of 2 pM and 5 pM methotrexate-resistant cells, individual clones were selected from cells growing in 1 pM methotrexate. Fig. 6 shows the analysis of ATP-citrate lyase mRNA synthesis in the different cell populations. A 4.3-kb band, consistent with the expected size of the ATP-citrate lyase mRNA, is clearly detectable in CHO cell resistant to methotrexate (Fig. 6, lanes B-E) and barely detectable in unamplified DHFR-CHO cells. The synthesis of ATP-citrate lyase mRNA increases at each increased methotrexate concentration. Densitometer analysis of the intensity of this band (after equalization to the loading factor using the actin band) shows that at 1-pM methotrexate concentrations, the levels of ATP-citrate lyase mRNA are at least 30-times greater than in unamplified DHFR-CHO cells. SDSi PAGE (Fig. 7) of cells extracts shows a similar pattern of

AatII-5

497

Rjb4hc1 7064

L A

OnJ-3168

XCaI-3492

Ldfnker. SV-SV40 early PolyrC 5-5' noncoding. 3-3' noncoding. Bw.66H Polyr

Fig. 4. Construction of the mammalian expression vectors containing the human ATPcitrate lyase cDNA (pBRJWHCL and pRLDNHCL). The human ATP-citrate lyase cDNA was subcloned into the expression cassette of pRJB4 or pRLDN vectors. The pRJB4 vector containing the Simian virus 40 replication origin, Rous LTR promoter and a bovine growth hormone polyadenylation signal. The RLDN vector contains a Simian virus 40 replication origin, Rous LTR promoter, a bovine growth hormone polyadenyhtion signal, Neo-resistant and DHFR-resistant genes.

pRJB4

pRJB4 HCI

pRJB4 HCfH,

A,aj

Fig.5. Expression of recombinant human ATP-citrate lyase in mammalian cells. The expression plasmid containing the human ATPcitrate lyase cDNA (pRJB4HCL) was constructed as described in Fig. 4. The pRJB4HCL His -+ Ala mutant was constructed using the polymerase-chain reaction; the following four oligonuleotide probes wcre designed:

(a)GGGGACAGGTACCCGGGCTCCACATTCATG; (b)GTCCAGTTTGGCuGCTGGAGCTTGTGCC;

(c)GGCACAAGCTCCAGCGGCGCCAAACTGGAC; (d)TCCATTGGCCACGAGATCTTCGTATACAGAC. Probe (b) contains a His-+Ala mutation which is underlined. Probes a and c were used to generate KpnI-NarI fragments and probes b and d were used to generate NarI-XcaI fragments. These fragments were constructed using the polymerase-chain reaction [33]. The two fragments were subcloned into pRJB4HCL, previously digested to delete a KpnI-XcnI fragment. The pRJB4HCL and pRJB4HCL (His+Ala) mutant were then transfected into COS cells as described in Materials and Methods. The cell extract was examined by Western-blot analysis using rabbit antibody against native ATPcitrate lyase and compared to cells transfected with the vector clone. The cell extract, as well as the media from cells that were transfected with pRJB4, pRJB4HCL or pRJB4HCL (His-Ala mutation) were assayed for ATP-citrate lyase enzyme activity as described in Materials and Methods. ACL, ATP-citrate lyase.

RNA. Coomassie-blue staining shows a protein band of M , approximately 120000, which is clearly overexpressed in the methotrexate-resistant cells (Fig. 7, lanes 2 - 6). This band comigrates with purified rat ATP-citrate lyase (Fig. 7, lane R) and is poorly expressed in non-amplified DHFR-CHO cells (Fig. 7, lane 1). Again, comparison of ATP-citrate lyase protein levels in 1 pM and 2 pM methotrexate-resistant cells with unamplified cells resembles the mRNA pattern seen in unamplified cells. In order to further characterize specifically the ATP-citrate lyase protein synthesis, a Western-blot analysis was performed using a polyclonal rabbit antibody raised against rat ATP-citrate lyase. A major band corresponding to a complete ATP-citrate lyase protein, comigrating with immunoblotted purified rat ATP-citrate lyase, was observed. This band increases in intensity with increasing concentrations of methotrexate and is virtually non-existent in untransfected cells. In order to check that the active form of the ATPcitrate lyase was produced, assays for enzymic activity were performed on the different cell populations (Fig. 8). High levels of enzymatically active ATP-citrate lyase are produced in transfected cells. As seen in mRNA and SDSjPAGE analysis, these levels increase concomitantly with increases in methotrexate concentrations. Cells resistant to 1 pM methotrexate show approximately 40-times higher levels than unamplified cells. Levels of protein production were up to 3 pg/ce11/24 h (data not shown). Single clones were selected

Fig. 6. Northern-blot analysis of ATP-citrate lyase mRNA synthesis. 3 x lo7 cells were grown a t 37°C in 5% CO,. RNA was extracted using the gudnidinium isothiocyanate method [19]. 10 pg aliquots of total RNA were electrophoresed in 1 YOagarose gels and blotted onto nylon membranes. After cross-linking for 2 h at 80 C, blots were prehybridized and hybridized to 32P-labelledhomologous cDNA probes, Lane A, untransfected CHO-DHFR cells; lane B, transfected CHO cells resistant to 10 nM methotrexate; lane C, transfected CHO ceik resistant to 100 nM methotrexate; lane D, transfected CHO cells resistant to 500 nM methotrexate; lane E, transfected CHO cells resistant to 1 pM methotrexate; lane F, liver RNA extracted from rats fed a high-carbohydrate, low-fat diet for 2 d [l I]. The position of the 28s and 18s RNA molecular mass markers are indicated. ACL, ATP-citrate lyase.

Fig. 7. SDS/PAGE of ATP-citrate lyase synthesis. Confluent 90-mm dishes of CHO cells were prepared. Cells were detached from the dishes in NaC1/Pi and centrifuged a t 5 0 0 0 ~for 5 min. Cell pellets were resuspended in loading buffer (Materials and Methods). 12.5% polyacrylamide gels were electrophoresed at 100 V for 6 h. Gels were then fixed in 15% acetic acid for 1 h, stained in 0.5% Coomassie blue for 1 h and destained in 10% acetic acid/50% methanol. Lane R, pure rat ATP-citrate lyase: lane 1, CHO-DHFR cells; lanes 2-6, CHO cells resistant to 10 (lane 2), 100 (lane 3), 500 (lane 4), 1000 (lane 5 ) and 2000 (lane 6 ) nM methotrexate. Molecular mass markers are chicken egg albumin (43 kDa), bovine serum albumin (67 kDa) and phosphorylase /? (94 kDa). ACL, ATP-citrate lyase.

499

-

c-

3000

REFERENCES

i

Control

lOnM

lOOOnM (pool)

Clone1

Clone2

(100OnM) (1000nM)

Clone4 (1000nM)

Methotrexate (nM) Fig. 8. ATP-citratc lyase enzymatic activity. CHO cells were grown at 37°C in 90-mm dishes to confluence. Cells were then detached from

dishes in NaCI/Pi at 4'C. Enzymatic assays were performed by measuring changes in absorbance at 340 nM as a result of the conversion of oxalacctate and NADH to malate and NAD'. ACL, ATPcitrate lyase.

from cells resistant to 1 pM methotrexate. Analysis of the clones for enzymic activity showed similar levels of active ATP-citrate lyase production, and one clone showed levels 2.5-times greater than the combined pool of clones (Fig. 8). An important aspect of the amplification approach for gene expression is the stability of the cells resistant to methotrexate for the production of high levels of the desired protein. High stability has been obtained in the absence of methotrexate for heterologous DHFR genes [28]. However, other reports show instability in the absence of selection pressure [29]. In our case, the ATP-citrate lyase product from cells resistant to 1 pM methotrexate seem to be very stable in the presence of methotrexate for at least 3 months. However, in the absence of the drug, levels of ATP-citrate lyase production fell sharply after 3 weeks (results not shown). Amplification in CHO cells is usually associated with expanded chromosomal regions and not with double minute chromosomes [30]. Different types of alterations can contribute to instability, including the formation of tetraploid cells following amplification [27], the formation of amplified regions located in dicentric chromosomes [27, 30, 311 and insertion and integration of extrachromosomal elements [32]. The mechanism involved in the case of ATP-citrate lyase is unclear. Availability of large quantities of human ATP-citrate lyase will allow more detailed studies on the structural and kinetic properties of this enzyme. We would like to thank Dr Derk Bergsma for his assistance with the polymerase-chain reaction, Dr Terry-Francis for measuring the enzyme activity of ATP-citrate lyase, Dr Ganesh Sathe for oligonucleotide synthesis and Keith Deen for his computer assistance.

1. Singh, M., Richards, E. G., Mukherjee, A. & Srere, P. A. (1976) J . Biol. Chem. 251, 5242- 5250. 2. Sullivan, A. C., Triscari, J., Hamilton, J. G., Miller, 0. & Wheatley, V. R. (1973) Lipid.? Y, 121 - 128. 3. Patel, M. S. & Owen, 0. E. (1976) Biochem. J . 156, 603-607. 4. Spector, L. B. (1972) Enzymes (3rdedn) 7, 357-389. 5. Cottam, G. L. & Srere, P. A. (1969) J . Biol. Chem. 130, 304314. 6. Inoue, H., Suzuki, F., Tanioka, H. &Takeda, Y. (1966) Biochem. Biophys. Rex Commun. 26,602 - 610. 7. Ramakrishna, S., Murthy, K. S. & Benjamin, W. B. (1989) Biochemistry 28, 856 - 860. 8. Pierce, M. W., Palmer, J. L., Keutmann, H. T. & Avruch, J . (1981) J . Biol. Chem. 256,8867-8870. 9. Ramakrishna,S.&Renjamin, W.B.(1981)FERSLeft. 124,140144. 10. Linn, T. & Srere, P. A. (1979) J. Biol. Chem. 254, 1691- 1699. 11. Elshourbagy, N. A, Ncar, J. C., KmetL, P. J., Sathe, G. M., Southen, C., Strickler, J. E., Gross, M., Young, J. F., Wells, T. N. & Groot, P. H. (1990) J. Biol. Chem. 265,1430-1435. 12. Elshourbagy, N. A., Walker, D. W., Bogaski, M. J., Gordon, J. 1. & Taylor, J . M. (1986) J. Biol. Chem. 261, 1998-2OU2. 13. Feinberg, A. P. & Vogelstein, C. (1983) Anal. Biochem. 132, 6-13. 14 Sanger, F., Niclen, S. & Couslson, A. R. (1977) Proc. Nut1 Acud. Sci. USA 74,5463 - 5466. ,I .< Devereux, J., Hacherli, P. & Marquess, P. (1984) Nuleic Acids Rex. 12, 387-395. 16. McCutchan, J. H. & Pagano, J. S. (1968) J . Not1 Can. Inst. 41, 351 - 356. 17. Potter, H. (1988) Anal. Biochem. 174, 361. 18. Wells, T. N. C. (1991) Eur. J . Biochem. 199, 163-168. 19. Laemrnli, U. K. (1970) Nufure 227,680-685. 20. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J . & Rutter, W. J. (3979) Biochemistry 18, 5294-5299. 21. Aviv, H. & Leder, P. (1972) Proc. Nut1 Acud. Sci. 69,1408-1412. 22. Alexander, M. C., Kowaloff, E. M., Witters, L. A., Dennihy, T. & Avruch, J. (1979) J . Biol. Chem. 254, 8052-8056. 23. Buck, D., Spencer, M. E. & Guest, J. R. (1985) Biochemistry 24, 6252-6259. 24. Pai, E. F., Sachsenheimer, W., Schirmer, H. & Schulz, G . E. (1977) J . Mol. Biol. 114, 37-45. 25. Lopez-Casillas, F., Bai, D., Lou, X., Kong, I., Hermodson, M . & Kim, K. (1988) Proc. Nut1 Acad. Sci. USA 85,5784-5788. 26. Suissea, M., Suda, K. & Schatz, G. (1984) EMBO J . 3, 1777317781. 27. Kaufman, R. S . (1990) Methods Enzymol. 185, 537-566. 28. Neubcrg, S . H., Kaufman, R. J., Schimke, R. T., Urlaub, G. & Chasin, L. A. (1988) Proc. Nut1 Acud. Sci. USA 75, 55535556. 29. Weidlc, U. H., Buckel, P. & Weinberg, J. (1988) Gene ( A m s t . ) 66, 193- 203. 30. Stark, G. R. & Wahl, G. M. (1984) Annu. Rev. Biochem. 53,447481. 31. Kaufman, R. J., Sharp, P. A. & Latt, S. A. (1983) Mol. Cell. Bid. 3,699-711. 32. Ruiz, J. C. & Wahl, G. M . (1990) Mol. Cell. Bid. 10, 3056-3065. 33. Saiki, R., Scharf, S., Falsona, F., Mullis, K . , Horn, G., Erlich, H. A. & Arnheim, N. (1985) Science 230,1350-1354.