Cloning, sequencing and functional expression of a cDNA encoding a NADP-dependent malic enzyme from human liver

Gene, 159 (1995) 255 260 © 1995 Elsevier Science B.V. All rights reserved. 0378-1119/95/$09.50

255

G E N E 08723

Cloning, sequencing and functional expression of a cDNA encoding a NADP-dependent malic enzyme from human liver (Cytosolic protein; recombinant DNA)

C o n s u e l o G o n z f i l e z - M a n c h 6 n , M i l a g r o s Ferrer, M a t i l d e S. A y u s o a n d R o b e r t o P a r r i l l a Department of Physiopathology and Human Molecular Genetics, Centro de Investigaciones Biol6gicas (CSIC ), Veldzquez 144, 28006 Madrid, Spain Received by M. Salas: 24 June 1994; Revised/Accepted: 7 November/15 November 1994; Received at publishers: 23 December 1994

SUMMARY

This work reports the structure of a cDNA (ME) encoding a human malic enzyme (ME) (malate NADP oxidoreductase, EC 1.1.1.40) elucidated by joining several overlapping fragments amplified by PCR from human hepatic cDNA or from cDNA libraries. The full-length cDNA has an open reading frame (ORF) of 1719 bp that encodes a 572-aminoacid protein of 64 113 Da, similar to the native monomeric, cytosolic, NADP-dependent ME isolated from human liver. The comparison of the structure of this cDNA with that of the human mitochondrial NAD(P)-dependent ME (EC 1.1.1.39) shows a homology of 63%, suggesting that these two forms originated from the same gene. The expression of the cDNA in Escheriehia coli as a translational fusion (glutathione S-transferase::ME) protein yielded a product of the predicted mass. The recombinant protein shows NADP-dependent malate oxidoreductase activity and is virtually inactive with NAD. It also shows other distinct features of the native cytosolic NADP-dependent ME, like Mn 2+ dependence, similar substrate (Km = 117 gM) and cofactor affinity ( g m = 2 gM) constants, and a lack of allosteric regulation. In human proliferative cells, the NADP-dependent ME activity is poorly expressed and barely inducible by thyroid hormones.

INTRODUCTION

The activity of the malic enzyme (ME) Was initially shown to catalyze the NADP-linked oxidative decarboxylation of malate (L-malate: NADP (decarboxylating), EC 1.1.1.40) (Ochoa et al., 1948). The higher content of Correspondence to: Dr. C. Gonz~dez-Manch6n, Centro de Investigaciones Biol6gicas, Velhzquez 144, 28006 Madrid, Spain. Tel./Fax (34-1) 562-8025; e-mail: [email protected] Abbreviations: aa, amino acid(s); bp, base pair(s); cDNA, DNA complementary to RNA; ct, cytosolic; GST glutathione S-transferase; IPTG, isopropy113-D-thiogalactoside; kb, kilobase(s) or 1000 bp; Kin, Michaelis constant; ME, malic enzyme; ME, gene (DNA) encoding ME; rot, mitochondrial; NAD, nicotinamide-adenine dinucleotide; NADP, NAD phosphate; nt, nucleotide(s); oligo, oligodeoxyribonucleotide; ORF, open reading frame; PCR, polymerase chain reaction; re-, recombinant; SDS, sodium dodecyl sulfate; [], denotes plasmid-carrier state; ::, novel junction (fusion or insertion). SSD1 0378-1119(95)00004-6

this enzyme in lipogenic tissues and the large variations in its activity in response to the nutritional and hormonal status led to the conclusion that it could serve the function of supplying reducing power (NADPH) to the cytosol for the synthesis of fatty acids from acetyl CoA (Tepperman and Tepperman, 1964; Wise and Ball, 1964). MEs occur in mammalian cells in at least three molecular forms. The NADP-linked enzyme (EC 1.1.1.40) is found in both the cytosolic (ct) and mitochondrial (mt) compartments (Simpson and Estabrook, 1969). Another mt protein, active with both NAD or NADP redox systems (EC 1.1.1.39), has been isolated from different sources (Simpson and Estabrook, 1969) and its functional significance has not been yet ascertained. Despite the reports involving ME in the pathogenesis of certain genetic disorders (Stumpf et al., 1982), the information concerning this enzyme in humans is very limited (Swierczyfiski et al., 1982; DiDonato et al., 1986;

256 Taroni et al., 1988). The structure of the human mt NAD(P)-dependent ME gene has been recently reported (Loeber et al., 1991). However, the structure of the ct form has not been yet elucidated although its activity may account for more than 90% of the total activity (Zelewski and Swierczyflski, 1991). Therefore, we found of interest to elucidate the coding sequence of the human ct ME. The aim of the present study was to clone and to asses the primary structure, functional expression, and kinetic properties of the re-enzyme.

EXPERIMENTAL AND DISCUSSION

(a) Cloning and sequencing of the human NADPdependent M E The search for human NADP-dependent ME cDNA by conventional screening methods has been elusive. We prepared human cDNA libraries from liver, hepatocarcinoma HepG2 cells, and placenta. Two other human eDNA libraries were obtained from commercial sources (Clontech, Palo Alto, CA, USA). We failed to detect positive clones after intensive screening of these eDNA libraries with an approx, l-kb fragment probe derived from the 3' end of the eDNA encoding the rat NADP-dependent ME (Strait et al., 1989). Then, we designed a pair of degenerated oligos (Fig. 1, oligos ME-3) complementary to highly conserved regions in rodent (Magnuson et al., 1986; Bagchi et al., 1987), human mt (Loeber et al., 1991), or plant (Rothermel and Nelson, 1989) malic enzymes. The nt sequence of the 558-bp fragment obtained by PCR-amplification of human liver eDNA with these primers showed a high degree of identity to the rat NADP-dependent ME and could be distinguished from the human mt ME eDNA by restriction analysis. The screening of cDNA libraries with this homologous fragment was unsuccessful; therefore, we adopted the strategy of reconstructing a full-length cDNA by joining overlapping PCR-amplified fragments from human hepatic eDNA. The scheme in Fig. 1 illustrates the different steps involved in the process. This approach yielded a eDNA fragment comprising a 1719-bp ORF (Fig. 2). The start codon is preceded three nt upstream by a purine and the immediately adjacent 3' base is a dG. These are characteristic structural features in cDNAs from eukaryotic mRNAs (Kozak, 1984). The sequence is 92% identical to the rat ct NADP-dependent ME (Magnuson et al, 1986) and 63% identical to the human mt, NAD(P)-dependent ME (Loeber et al., 1991), suggesting that they may have originated from the same gene. In view of the structural differences with the mt NAD(P)-linked form and the remarkable similarity to the rodent NADP-dependent ME, it is reasonable to

ME-1 - -

-

M ~ 1 - 2 - -

ME-2

-

-

\ M

E

-

3

ME-4

/

M~1-2-3

M~5

ME-4-5

human

ME

Fig. 1. Construction of a eDNA encoding a human NADP-dependent ME by the 'splicing by overlap extension' procedure. Methods: Variable amounts of human liver eDNA or eDNA libraries were used as templates for the PCR. We used the following pairs of oligos, designed to hybridize highly conserved regions of eDNA, to amplify overlapping fragments of eDNA: ME-l: sense (~,gtl0-S), 5'-CTTTTGAGCAAGTTCAGCCTGGTTAAG,

antisense, 5'-GTTCAATTGCTGTCTCTCTTCCA; ME-2: sense, 5'-AACAAGGRMWTGGCWTTTA,

antisense, 5'-AATAAAGAGACCTCTTGGCTTC; ME-3: sense, 5'-GTWTAYACWCCSACSGTKGGTCTTGCC,

antisense, 5'-MGCTGTYCCTTGAATATCATCRTTRAA; ME-4: sense, 5'-AAGTATCGAAACCAGTATTGC,

antisense, 5'-TCCATCTGGGAGAGTGACTGG; ME-5: sense, 5'-TTTGCCAGYGGCAGTCCWTTT,

antisense 0~gtl0-AS), 5'-ATGGGACCTTCTTTATGAGTATTCG. The amplifications were carried out with Taq polymerase according to the protocol recommended by Perkin-Elmer Cetus (Norwalk, CT, USA). MgCI2 concentration and annealing temperatures were optimized for each pair of primers. The PCR products were cloned in a T-vector (Marchuk et al., 1991), and their primary nt sequence determined. The nt coding sequence of the human ME eDNA was constructed by joining the PCR-amplified fragments by the 'splicing by overlap extension' (SOE) procedure (Horton et al., 1989). Human hepatic eDNA libraries in bacteriophage )~gt10 were prepared according to previously described procedures (Gubler and Hoffman, 1983) ( K = G or T; M = A or C; R = A or G; S = C or G; W = A or T; Y = C or T)

assume that our eDNA encodes a c t , NADP-dependent form of human ME. The structure of the mt NADPlinked ME gene has not been yet determined. However, since the mass of the native protein differs from that of the ct form (Taroni et al., 1988), it is unlikely that our cDNA encoded a mt, NADP-dependent, ME. Moreover, our cDNA lacks a 5' mt leader sequence.

(b) Comparison of the human ct ME with other MEs The ct ME has a predicted sequence of 572 aa (64.1kDa), while the mt, NAD(P)-dependent, ME (Loeber et al., 1991) has 584 aa (65.4 kDa). The identities and similarities of the ct ME with other MEs were determined according to Altschul et al. (1990). The aa sequence of the human ct ME shows a 92% identity and 96% similarity (identical plus similar) with the rodent NADP-dependent ME, while only 54% and 71% with other NADP-dependent MEs, like the plant ct ME. It

257 A M

T ~ ( ] C ~ C L ~ E P E A P R R R

~ H

C ~ T C ~ A C L ~ C T H Q R G Y L L T

O R

C ~ C C ~ C L ~ C N P H L K K D L

~ A

' F

I T I ~ t - ~ ~ T L E E R Q Q

~ L

N

I

120 40

~ T G G A T I ~ ] C C A C C ~ C A G ' I V _ A ( I ~ q ~ A C ~ A A A A A A T ] ~ T C ~ C ~ w I T I ~ C A ( 3 G T A T C ~ ~ ~ T H G L L P P S F N S Q E I Q v L R V V K N F E H L N S D F D R Y L L L M D L Q

D

240 80

AC~TC~'I'I'I'I~TAC~~TCTCA~q~C, R N E K L F Y R V L T S

D

I

E

AAAqTCAT(]CCTA~ITI'ATAC~ K F M P I V Y T T

A C ~ L T I•t~ T T A C T A T C C A C G A T C ~ T A T I ~ R G L F I T I H D R G H I

A

S

V

G A C ~ A A T ~ T ~ q ~ S C D L G C N G M G I

V

G

L

A

C

Q

~ T A T ~ I ' I ' I ~ Q Y S L V F RK P

C T C A A T S C A T ~ T G ' T C A T C ~ ~ ~ ~ L N A W P E D V I K A I V V T D G E R

CI~AAAT~ATATACAGC~(X]C~TC~TC~ P V G K L A L Y T A C G

G/~ACITAAA~TCEACTCTACATI~CTAC~C~GTAAC~TATC~ATC~'I E L L K D P L Y I G L R Q R R V R G S

E

Y

G

M

N

P

Q

D

D

T I"IT i~ L D

I

L

G

L

V

T T C ~ T ~ C ~ T C 4 k G I L D V G T E N E

~ E F

~ M

E

A

V

IT I~ A S S K

~ Y

G

~ M

N

C

720 240

G

840 280

E

960 320

~ N

C A A G T A T C ~ C C A G K Y f{ N Q Y C T F N D D

T I

A Q

~ G

~ T

~ T A ~ ~ ~ A S V A V A

L

Q

T

I

L

T N

K

L

S

D

R

I

T

E

A

A

L

G

I

A

H

L

I

V

G~I'I'I~A G L P K

E

K

A

TC-AA~TAT(3C~TTCA~fTAAT~O'ITAA~C~C'I'FAACA I K K I W L V D S K G L I V K G R

A

S

L

T

Q

E

K

E

K

'i'I'iL.-TCC(~TC~AACATC~ 1080 F A H E H E E 360

I

G

V

G

A

A

A

G

2"I~q'IL~CAAATIL~TAq
CCTATTA'IT I ' I ' I L q C ~ G T A A q L ~ C T A C - C ~ ~ G C ~ T ~ ( g C ~ A C A A A A T A A C C A ~ C ~ P I I F A L S N P T S K A E C S A E Q C Y K I T K G R A C~T~C~C~I~I'I<. L P D G R T L

F

P

G

Q

ACT~IATA T A E V

Q

V

S

D

I

A

R

1200 400

I'I'i'i'I t . ~ T l ' l ' l ~ ~ F A S G S P F D P V T

1320 440

A

F

N

E

C'I~CT~CACATCCATCATA~'q~CC C G L R H I D D K V

F

L

T

1440 480

A T C ~ T A C C A ~ C < 3 I"I"It - ~ q q ~ G C A G T A A ~ q ~ T P P L N T I R D V S L K I A V K I

V

Q

D

1560 520

G C A T A C A A ~ T ~ C ' I I 3 1 " i ' i ~ T C ~ C C C C A A A A C A ~ T I ' I ~ C I ~ T G T A C A ~ A ~ ~ A ~ ~ A ~ A Y K E K M A T V Y P E P Q N K E E F V S S Q M Y S T N Y D Q I L P D C Y P W

P

1680 560

I

S

Q

G

N

C~TI~A N S Y

M

~ E K

L

L

Q

( L

A

ATCaAACaAACCT/~qq~C~TAAAACC~CTSCCCTCAT M K N L E A I V Q E I K P T A

F

600 200

P

C L

K

480 160

L

L

A

G

C

E

C I T A T ' I ~ IT IC - ~ T ~ C A A T G ~ T G C A T I ' I ~ L I Q F E D F A N V N A F R

L

360 120

TAAACACCT~ K H L Q

E

V

F

P

G

G

R

L

Y

V

A

L

G

V

V

A

C~IAAAATACAC~CCAAAG'I~CC~AACGCAAC~;XL]GAqq~AAcITI'ATTAGTAAAATCI~q'kI'I~~~~~A~ A E V Q K I Q T K V N Q

1800 572

7'IC_.ATAGTATfX.ZTJ~TA;~

Fig. 2. Nucleotide sequence of the human NADP-dependent ME. The primary structure of the cDNA encoding the human hepatic NADP-dependent ME was obtained as described under Fig. 1. Restriction mapping and sequence analysis were carried out as described by Marck (1988). Underlined regions correspond to the putative malate binding domain (aa 92-117), ADP+-binding domain (aa 150 174) and NAD(P)-binding domain (aa 301-333). This human ct ME sequence has been deposited with GSDB (accession No. L34035).

shows 56% identities and 74% similarities with the human NAD(P)-dependent form (Loeber et al., 1991). The first 80 aa and, approximately, the same number at the C' terminus show little homology. Fig. 3 shows the primary structures of the putative substrate or NAD binding domain (Kulkarni et al., 1993), the ADP-binding

Malate or NAD-bindingdomain 97

118

I ~ dinucleotide fold (Wierenga et al., 1985) and the NAD(P)-binding fold (Scrutton et al., 1990) of human NADP-dependent and other MEs. The sequence of these domains is compared with the corresponding regions of the rat (Scrutton et al., 1990) and plant ct (van Doorsselaere et al., 1991) NADP-dependent MEs and NAD(P)-binding domain

ADP-binding domain 150

174

301

333

Human NADP Rat NADP Plant ct NADP

FMPIVYTP~JGLACQQYSLVFR FM ......... LA-QQYSLAFR LL-F ....... E A - Q K Y G S I F G R

V T D G E R I LGLGDLGCNGMG I PVGKL -E .... C N - M - - P V - --E . . . . . . . CQ-I PV

G A G E A A L G I A H L I % ~ A L E K .... EGLPKEKAIKKIWL - - E - A L - I A H L I V M A M E K .... EGLSKEKARQKIWL ---E-GT-IAELIALEMSR...RSKTPLEETRKKIWL

107

160

311

Human NAD Nematode NAD Potato (62kDa) NAD

LM . . . . . LA-SQYGHIF~ LM LA-QNFGYIYR YA . . . . . . . . LV-QKYSGLFR

128

184

-- -E ...... VY M--PV .... E ..... AN I -PV .... S ......... IQ-I--AI --

343

---E-AL-IANLII%MS~gE . . . . N G L S E Q E A Q K K I W M A ST I A E M I V H Q M Q N . . . E G I S K E E A C N R I Y L M ---S-GI-VLNAARKTMA~MLGNTEIAFESARSQF~g

Fig. 3. Comparison of the putative malate, ADP and NAD(P)-binding domains in different MEs. Comparison of the putative aa (single letter code) consensus binding domains of malate (Kulkarni et al., 1993), A D P (Wierenga et al., 1985) and NAD(P) (Scrutton et al., 1990) from the rat (Magnuson et al., 1986), plant cytosolic (Van Doorsselaere et al., 1991), human mitochondrial (Loeber et al., 1991), ascaris (Kulkarni et al., 1993) and the 62-kDa subunit of potato (Winning et al., 1994) MEs. The numbers indicate the position of the aa in the human sequences. Dashes indicate complete identity.

258 with the human mt .(Loeber et al., 1991), nematode (Wierenga et al., 1985), and 62-kDa subunit of plant (Wining et al., 1994) NAD-dependent, enzymes. In the malate/NAD(P) binding domain, the Cys n° and the preceding motif VYTPTVG (aa 101-107 in the human NADP-dependent ME) are conserved in all species. The homology of this region with the nt-binding domains of goose fatty acid synthetase and human glyceraldehyde3-phosphate dehydrogenase suggested its involvement in the nicotinamide nt binding (Poulouse and Kolattukudy, 1983). However, the impairment of malate binding by alkylating the enzyme with the substrate analog bromopyruvate without altering the nt binding, indicates that this segment is primarily involved in the binding of substrate (Wierenga et al., 1985; Satterlee and Hsu, 1991). Other residues conserved in all sequences are Arg n8 (Fig. 3) and Arg sl (not shown), that seem to be essential for enzyme activity (Rao et al., 1987). The aa 150 to 174 delimit the [3~13fold that is thought to be involved in the ADP binding of dinucleotides (Wierenga et al., 1985) (Fig. 3). This segment contains a distinct arrangement of Gly, hydrophilic and non polar aa, and is conserved in all species. The Cys 164 m a y contribute to confer cofactor specificity as far as it is present only in the NADP-dependent (ct) but not in the NAD-dependent (mt) enzymes (Cushman, 1992; Winning et al., 1994). The region delimited by aa 301 to 333 resembles the putative N A D ( P ) binding domain proposed by Scrutton et al. (1990) (Fig. 3). The number of identical aa in all sequences is very limited. It is not apparent the existence of cofactor specific recognition motifs. Variations from the consensus sequence have been noticed before (Wierenga et al., 1985; Loeber, 1991; Cushman, 1992). Thus, it seems plausible to conclude that other regions of the protein, perhaps some of the highly conserved stretches, must contribute to determine the nicotinamide nt-binding specificity.

(c) Production of human NADP-dependent ME in Escherichia coli We subcloned the NADP-dependent M E cDNA into the pGEX-4T-2 vector to produce a translational GST fusion protein in Escherichia coil. The fusion protein has a mass of 96 kDa (Fig. 4); the size of the ME protein agrees with the calculated 64 kDa and with that reported for the monomeric form of the human hepatic ct ME (Zelewski and Swierczyfiski, 1991). The specific activity of the purified re-protein, 40 I~mol/min per mg protein, is similar to that reported for the enzyme purified from human liver (Zelewski and Swierczyfiski, 1991) and to the human mt ME synthesized in E. coli (Loeber et al., 1991). The re-ME, but neither the GST produced by expres-

M

1

2

3

4

kDa 200

-

-

116 97

--

66

--

45

--

w

31

B

I***,

~

Fig. 4. Purificationof a ME-GST fusion protein. Methods:The coding region of the ME was obtained by PCR amplification of the cDNA constructed by the 'splicing by overlap extension' procedure (Fig. 1), using the followingprimers: upper oligo, 5'-ATGGAGCCCGAAGCCCCCCG; lower oligo, 5'-ATGGGACCTTCTTTATGAGTATTCG. The PCR product was treated with Klenow DNA polymerase,cloned in-frame into the SmaI site of the E. coil expressionvector pGEX-4T2 (Pharmacia Biotech, Upsala, Sweden) and its sequence verified. JM105 [pGEX-4T-2-ME] cells were induced with IPTG for 3 h. After lysis of the cells with sonication the soluble fraction was separated by centrifugation. The ME-GST fusion protein was purified by affinity chromatography using a glutathione-Sepaharose-4Bcolumn. Lanes: 1, GST purified from parental JM105[pGEX-4T-2] cells; 2, translation of the antisense strand of ME; 3, E. coli JM 105[pGEX-4T-2-ME] sonicate; 4, ME-GST fusion protein eluate from glutathione-Sepharose4B column; M, markers. sion of a vector without the M E cDNA, nor the product of M E cDNA cloned in the reversed orientation, showed malate oxidoreductase activity. The activity of this re-ME is strictly dependent on Mn 2+. It is, at least, 100-fold more active with N A D P than with NAD under the same conditions. Since these are properties shared by all the ct malic enzymes, this observation adds further support in favour of the identification of our sequence as a c t form of the human ME. Moreover, the activity of our protein was neither inhibited by ATP nor stimulated by fumarate, that are allosteric regulators of the rot, NAD(P)-dependent, ME (Mandela and Sauer, 1975). In contrast to the human mt NAD(P)-dependent ME, which Krn for malate is 10 m M in the absence of allosteric effectors (Loeber et al., 1991), the K m of the recombinant ct ME was 112 ~tM. This value agrees very well with those (120 131 IaM) reported for isolated human NADPdependent MEs (Zelewski and Swierczyfiski, 1991; Chang et al., 1992). These observations indicate that the catalytic properties of ME do not rely on postranslational modifications.

(d) Expression of human NADP-dependent M E mRNA We detected transcripts of approx. 2.4 and 3.8 kb (Fig. 5) in Northern blot analysis of human hepatic or

259 < I-n" UJ >

U5 0

._1

13..

(e) Conclusions

._1

kb 6.2ii!iii~i~ 3.9

kb

-


3.8


'" 2 . 4

2.8-

!il ii!i!i 1.8

-

Fig. 5. Northern blot hybridization of ME mRNA from human liver and placenta. Methods: Total RNA from frozen samples of human liver and placenta was obtained by the guanidinium thiocyanate procedure (Chomczynski and Sacchi, 1987). Poly(A)+RNA was isolated by oligo (dT)-cellulose chromatography, size fractionated on 1.2 % denaturing agarose gels and then transferred to positively charged nylon membranes. The filters were prehybridized in 50% formamide/0.12mM Na2HPO4/0.25 mM NaCI/7% (w/v) SDS at 42°C for 4 h. Hybridization with the a2p-labelled full-length human NADP-dependent ME cDNA (106 cpm/ml) was carried out overnight under the same conditions as prehybridization. The membranes were washed twice for 15 min at room temperature in 1 × SSC/0.01% SDS, followed by one wash with 0.1 × SSC/0.1% SDS (SSC is 0.15 M NaC1/0.15 M Na3.citrate pH 7.6).

placental poly(A)+RNA probed with human NADPdependent ME cDNA. In rodents, co-equal expression of 3.1 and 2-kb ME mRNAs in various tissues and cells has been reported (Bagchi et al., 1987). The existence of a single gene encoding ME suggests that these mRNAs are the result of postranscriptional processing. Unlike previous observations made in primary cultured rat liver cells (Fukuda et al., 1992; Molero et al., 1993), in human hepatic proliferative HepG2 cells, the activity of the NADP-dependent ME is almost negligible. The ME mRNA was poorly represented, and it was not significantly induced by thyroid hormone stimulation (results not shown). In contrast, the mt form of ME, that under physiological conditions accounts for no more than 10% of the total activity (Zelewski and Swierczyfiski, 1991 ), is found at high level in cells undergoing high rates of division (Loeber et al., 1991). The reason for the reciprocal relationship between the expression of NAD and NADP-dependent forms of M E and the proliferative status seems to be a subject of great interest.

(1) A cDNA encoding a ME from human liver has been constructed by the "splicing by overlap extension" PCR procedure (Horton et al., 1989). This cDNA contains an ORF of 1719 bp that encodes a 572-aa protein of 64 kDa. (2) The following observations indicate that the herein reported cDNA encodes act, NADP-dependent ME (EC 1.1.1.40): (i) The predicted protein encoded by this cDNA differs in size from that of the human native mt form of NADP-linked ME; (ii) It shows high homology with NADP-dependent enzymes from other species; (iii) The re-enzyme, produced in E. coli as a GST fusion, shows the characteristic kinetic features of the ct, NADP-dependent, native ME; (iv) We detected transcripts of 2.4 and 3.8 kb in Northern blot analysis of human liver and placenta RNAs probed with this cDNA. (3) In contrast to previous work carried out in primary cultures of rat liver cells, in human proliferative liver cells the NADP-dependent ME is poorly expressed and it is not significantly induced by thyroid hormone.

ACKNOWLEDGEMENTS

Authors wish to express their gratitude to Mrs. Ana Rodriguez-Monje for her excellent technical assistance. This work has been supported in part by grants from the Spanish Plan Nacional de Investigaci6n Cientifica y T6cnica (SAL 509/91 and SAF 93-0788), Fondo de Investigaciones Sanitarias (93/0162) and Direcci6n General de Investigaci6n Cientifica y T6cnica (PB93/163).

REFERENCES Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J.: Basic local alignment search tool. J. Mol. Biol. 215 (1990) 403 410. Bagchi, S., Wise, L.S., Brown, M.L., Bergman, D., Sul, H.S. and Rubin, C.S.J.: Structure and expression of murine malic enzyme mRNA. J. Biol, Chem. 262 (1987) 1558 1565. Chang, G.-G., Huang, T.-M., Wang, J.-K., Lee, H.-J., Chou, W.-Y. and Meng, C.-L.: Kinetic mechanism of the ct malic enzyme from the human breast cancer cell line. Arch. Biochem. Biophys. 296 (1992) 468-473. Chomczynski, P. and Sacchi, N.: Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162 (1987) 156-159. Cushman, J.C.: Characterization and expression of a NADP-malic enzyme cDNA induced by salt stress from the facultative crassulacean acid metabolism plant, Mesembryahthemum crystallinum. Eur. J. Biochem. 208 (1992) 259 266. Di Donato, S., Taroni, F. and Gellera, C.: Evidence for the presence of two mitochondrial and one cytosolic malic enzymes in human skeletal muscle. Fed. Proc. 45 (1986) 1490.

260 Fukuda, H., Katsurada, A. and Iritani, N.: Nutritional and hormonal regulation of mRNA levels of lipogenic enzymes in primary cultures of rat hepatocytes. J. Biochem. 111 (1992) 25-30. Gubler, U. and Hoffman, B.J.: A simple and very eficient method for generating cDNA libraries. Gene 25 (1983) 263-269. Horton, R.M., Hunt, H.D., Ho, S.N., Pullen, J.K. and Pease, L.R.: Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. Gene 77 (1989) 61-68. Kozak, L.P.: Compilation and analysis of sequences upstream from the translational start site in eukaryotic mRNAs. Nucleic Acids Res. 12 (1984) 857-872. Kulkarni, G.,Cook, P.F. and Harris, B.G.: Cloning and nucleotide sequence of a full-length cDNA encoding Ascaris suum malic enzyme. Arch. Biochem. Biophys. 300 (1993) 231 237. Loeber, G.L., Infante, A.A., Maurer-Fogy, I., Krystek, E. and Dworkin, M.B.: Human NAD-dependent mitochondrial malic enzyme, cDNA cloning, primary structure, and expression in Escherichia coli. J. Biol. Chem. 266 (1991) 3016-3021. Magnuson, M.A., Morioka, H., Tecce, M.F. and Nikodem, V.M.: Coding nucleotide sequence of rat malic enzyme mRNA. J. Biol. Chem. 261 (1986) 1183 1186. Mandella, R.D. and Sauer, L.A.: The mitochondrial malic enzymes, I. Sumitochondrial localization and purification and properties of the NAD(P)-dependent from adrenal cortex. J. Biol. Chem. 250 (1975) 5877-5884. Marchuk, D., Drumm, M., Saulino, A. and Collins, F.S.: Construction of T-vectors, a rapid and general system for direct cloning of unmodified PCR products. Nucleic Acids Res. 19 (1991) 1154. Marck, C.: DNA Strider. A "C" program for the fast analysis of DNA and protein sequences on the Apple Macintosh family of computers. Nucleic Acids Res. 16 (1988) 1829-1836. Molero, C., Benito, M. and Lorenzo, M.: Regulation of malic enzyme gene expression by nutrients, hormones and growth factors in fetal hepatocyte primary cultures. J. Cell. Physiol. 155 (1993) 197--203. Ochoa, S., Mehler, A.H. and Kornberg, A.: Biosynthesis of dicarboxylic acids by carbon dioxide fixation, I. Isolation and properties of an enzyme from pigeon liver catalyzing the reversible oxidative decarboxylation of L-malic acid. J. Biol. Chem. 174 (1948) 979-1000. Poulouse, A.J. and Kolattukudy, P.E.: Sequence of a tryptic peptide from the NADPH binding site of the enoyl reductase domain of fatty acid synthase. Arch. Biochem. Biophys. 220 (1983) 652-656. Rao, G.S.J., Kong, C.-T., Benjamin, R.C., Harris, B.G. and Cook, P.F.: Modification of an arginine residue essential for the activity of NAD-malic enzyme from Ascaris suum. Arch. Biochem. Biophys. 255 (1987) 8-13.

Rothermel, B.A. and Nelson, T.: Primary structure of the maize NADPdependent malic enzyme. J. Biol. Chem. 264 (1989) 19587-19592. Satterlee, J. and Hsu, R.Y.: Duck liver malic enzyme: sequence of a tryptic peptide containing the cysteine residue labeled by the substrate analog bromopyruvate. Biochim. Biophys. Acta 1079 (1991) 247 252. Scrutton, N.S., Berry, A. and Perham, R.N.: Redesign of the coenzyme specificity of a dehydrogenase by protein engineering. Nature 343 (1990) 38-43. Simpson, E.R. and Estabrook, R.W.: Mitochondrial malic enzyme: the source of reduced nicotinamide adenine dinucleotide phosphate for steroid hydroxylation in bovine adrenal cortex mitochondria. Arch. Biochem. Biophys. 129 (1969) 384-395. Strait, K.A., Kinlaw, W.B., Mariash, C.N. and Oppenheimer, J.H.: Kinetics of induction by thyroid hormone of the two hepatic mRNAs coding for cytosolic malic enzyme in the hypothytoid and euthyroid states. Evidence against an obligatory role of S14 protein in malic enzyme gene expression. J. Biol. Chem. 264 (1989) 19784 19789. Stumpf, D.A., Parks, J.K., Eguren, L.A. and Haas, R.: Friedreich ataxia, III. Mitochondrial malic enzyme deficiency. Neurology 32 (1982) 221-227. Swierczyfiski, J., Scistowski, P., Aleksandrowicz, Z. and Zelewski, L.: NADP+-dependent malic enzyme activity in human term placental mitochondria. Biochem. Med. 28 (1982) 247 255. Taroni, F., Gellera, C. and Di Donato, S.: Evidence for two distinct mitochondrial malic enzymes in human skeletal muscle: purification and properties of the NAD(P)-dependent enzyme. Biochim. Biophys. Acta 916 (1988) 446-454. Tepperman, H.M. and Tepperman, J.: Patterns of dietary and hormonal induction of certain NADP-linked liver enzymes. Am. J. Physiol. 206 (1964) 357 361. van Doorsselaere, J., Villarroel, R., van Montagu, M. and Inz6, D.: Nucleotide sequence of a cDNA encoding malic enzyme from poplar. Plant Physiol. 96 (1991) 1385 1386. Wierenga, R.K., De Maeyer, M.C.H. and Hol, W.G.J.: Interaction of pyrophosphate moieties with a-helixes in dinucleotide binding proteins. Biochemistry 24 (1985) 1346 1357. Winning, B.M., Bourguignon, J. and Leaver, C.J.: Plant mitochondrial NAD-dependent malic enzyme, cDNA cloning, deduced primary structure of the 59 and 62-kDa subunits, import, gene complexity and expression analysis. J. Biol. Chem. 269 (1994) 4780-4786. Wise, E.M. and Ball, E.G.: Malic enzyme and lipogenesis. Proc. Natl. Acad. Sci. USA 52 (1964) 1255 1263. Zelewski, M. and Swierczyfiski, J.: Malic enzyme in human liver. lntracellular distribution, purification and properties of cytosolic enzyme. Eur. J. Biochem. 201 (1991) 339 345.