The Dihydrofolate Reductase Domain of PZasmodium fakiparum ...

14 downloads 13158 Views 2MB Size Report
dihydrofolate reductase (DHFR) domain of the thymid- ylate synthase-dihydrofolate reductase (TS-DHFR) bi- functional protein of Plasmodium fakiparum was de ...
Vol. 268. No . 29. Issue of Octobel’ 15, PP. 21637-21644.1993 Printed in U.S.A.

T H E JOURNAL OF BIOLOGICAL CHEMISTRY Q 1993 by The American Society for Biochemistry and Molecular Biology, Inc.

The Dihydrofolate Reductase Domain ofPZasmodium fakiparum Thymidylate Synthase-Dihydrofolate Reductase GENESYNTHESIS,EXPRESSION,

AND ANTI-FOLATE-RESISTANTMUTANTS* (Received for publication, March 25, 1993, and in revised form, June 18, 1993)

Worachart SirawarapornSB,Phisit PrapunwattanaS, Rachada SirawarapornS, Yongyuth YuthavongS, and Daniel V. SantinII From the $Departmentof Biochemistry, Faculty of Science, Mahidol University, Bangkok 10400, Thailand and the TDepartments of Biochemistry and Biophysics and of Pharmnceutical Chemistry, University of California, San Francisco, California 94143

A 693-base pair gene coding for the 27,132-dalton dihydrofolate reductase (DHFR) domain of the thymidylate synthase-dihydrofolate reductase (TS-DHFR)bifunctional protein of Plasmodium fakiparum was designed to have Escherichia coli codon preference and multiple unique restriction sites and was chemically synthesized. The gene was overexpressed (>50% total cellular protein) in E. coli as insoluble inclusion bodies which could be unfolded and refolded to recover soluble enzyme activity. The refolded DHFR was purified by methotrexate-Sepharose affinity chromatography to give the homogeneous enzyme. Activesite titration with methotrexate revealed that the purified protein was fully active. The purified DHFR migrates as a single band on sodium dodecyl sulfate-polyacrylamide gel electrophoresis with apparent mass of -30 kDa, and gel filtrationshowed that the protein is a monomer. The yield of purified enzyme was about 5-6 mg/liter of bacterial culture. Kinetic properties of the purified recombinant DHFR were similar to those reported for wild type bifunctional TS-DHFR. Cassette mutagenesis of the synthetic gene was performed to give the SlO8N and the NBlI+SlOSN mutants which provided DHFRs analogous to pyrimethamine-resistant mutants found in nature.

precursor of folate cofactors. In most organisms, DHFR is found as a monofunctional monomer of about 18-25 kDa (Blakley, 1984), whereas in protozoa, DHFRand TS are present on the same polypeptide chain. The bifunctional protein is a homodimer of two identical subunits (Ferone and Roland, 1980; Garrett et al., 1984). Analysis of the genes coding for the TS-DHFRs from protozoa revealed that the DHFR domain is at theamino terminus, the TSdomain is at the carboxyl terminus, and that thedomains are separatedby a junctional peptide (Ivanetich and Santi, 1990). The DHFR of Plasmodium falciparum has received considerable attention since it is the target of Pyr and Cyc, two of the few drugs effective for the prophylaxis and treatment of malaria. However, the use of these antifolates has been limited by the widespread emergence of resistant parasites (BruceChwatt et al., 1986). In recent years, it has been demonstrated that anti-folate resistance in P. falciparum is caused by point mutations of the enzyme which decrease inhibition by these drugs (Foote et al., 1990; Peterson et al., 1988,1990). Mutation of Ser-108 to Asn is believed to be the primary event leading to Pyr resistance in P. fakiparum,and subsequent mutations of Asn-51, Cys-59, and Ile-164 further increase drug resistance (Peterson et al., 1988). A correlation has also been demonstrated between the A16V mutation and Cyc resistance in malaria (Foote et al., 1990; Peterson et al., 1990). Our objective has been focused on employing a structureDihydrofolate reductase (DHFR)’ catalyzes the NADPH- based approach to developing inhibitors of P. fakiparum dependent reduction of Hpfolate to H4folate, an essential DHFR which might also be effective against anti-folate-resistant parasites. Since it is impractical to obtain significant * This work was supported in part by Tropical Disease Research/ amounts of enzyme from the parasite, we have cloned the Rockefeller Foundation Joint Venture Program Grant ID-880282 and United States Public Health Service Grant AI 19358. The costs of bifunctional TS-DHFR in a variety of expression vectors and publication of this article were defrayed in part by the payment of attempted to overexpress the protein in heterologous systems page charges. This article must therefore be hereby marked “adver- (Sirawaraporn et al., 1990). Unfortunately, the systems protisement’’ in accordance with 18U.S.C. Section 1734 solelyto indicate vided low amounts of enzyme. Wesuspected that theproblems this fact. in expression might be caused by the high A+T content of The nucleotide seqwnce(s) reported in thispaper hos been submitted to theGenBankTM/EMBL Data Bankwith accession number(s) the Plasmodium gene and/or that the gene product might be toxic to Escherichia coli (Hall et al., 1991). We therefore L17041. 8 Recipient of a Frohlich Research fellowship and a Rockefeller embarked on the chemical synthesis of the TS-DHFR gene Biotechnology career fellowship. to provide codons more favorable to E . coli and asystem more 11 To whom correspondence should be addressed Dept. of Biochem- amenable to mutagenesis. istry and Biophysics, University of California, San Francisco, CA In the present work we describe the design, synthesis, and 94143. Tel.: 415-476-1740;Fax: 415-476-0473. ’The abbreviations used are: DHFR, dihydrofolate reductase; TS, expression of the DHFR gene of the P. fakiparum. Under thymidylate synthase; RBS, ribosomal binding sequence; PAGE, control of the T7 promoter, the synthetic gene is highly polyacrylamide gel electrophoresis; H2folate, 7,8-dihydrofolate; expressed as inclusion bodies. The insoluble protein can be H,folate, 5,6,7,8-tetrahydrofolate; DTT, dithiothreitol; TES, N tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid; MTX, unfolded and refolded to give highly active protein. We also methotrexate; Cyc, cycloguanil; Pyr, pyrimethamine; IPTG, isopropyl describe the preparation and properties of two mutants of P. 1-thio-0-D-galactopyranoside. falciparum DHFR which confer resistance to Pyr and Cyc.

21637

21638

falciparum Plasmodium EXPERIMENTALPROCEDURES

Materials-Restriction endonucleases, T, DNA ligase, and other DNA-modifyingenzymes were products of New England Biolabs, Life Technologies, Inc., or Boehringer Mannheim. The Qiagen kit for plasmid isolation and purification was from Qiagen Inc. E. coli DH5acompetent cells were purchased from Life Technologies, Inc., and were used as host strain for plasmid-mediated transformations and general manipulation of recombinant plasmids. The expression vector PET-17b and the host strains E. coli BL21(DE3) and BL21(DES)pLysS were from Novagen. pKK223-2 and E. coli JM105 were from Pharmacia LKB Biotechnology Inc. Carbenicillin, Pyr, MTX, and NADPH were purchased from Sigma. Cyc was a gift from Burroughs Wellcome Co. The concentration of MTX was determined spectrophotometrically (Seeger et al., 1949). MTX-Sepharose CL-GB (-1 pmol/ml) (Meek et al., 1985) and Hzfolate (Friedkinet al., 1962) were prepared as described. Oligonucleotide synthesis, peptide sequencing, and automated DNA sequencing were performed at the Biomolecular Resource Center, University of California, San Francisco. Oligonucleotides were purified as described (Ivanetich et al., 1991) except those used for gene assembly were further purified by denaturing gel electrophoresis (Sambrook et al., 1989). Other general methods for DNA manipulations were performed as described (Sambrook et al., 1989). Construction and Mutagenesis of Synthetic DHFR Domain-The P. falciparum DHFR domain was assembled from seven complementary oligonucleotide fragments by sequential ligation into pUCm7 vector in which the polylinker of pUC18 was modified to contain all necessary cloning sites. Mutagenesis of the DHFR domain to give the S108N and N51I+S108N mutants was accomplished by cassette replacement of the gene fragments with synthetic oligonucleotide duplexes. The ligation reaction was used to transform E. coli DH5a, and thecells were plated on LB agar containing 100 pg/ml ampicillin. Plasmid DNA wascharacterized by restriction analysis and sequenced as described (Sanger et al., 1977). Expression Vectors-The vector pRBSrrnc carrying an RBS and transcription terminator was constructed by replacing the polylinker of pUC18 at EcoRI-Hind111 sites with a synthetic oligonucleotide adapter containing a modified 29-base 5"untranslated sequence of Lactobacillus caseiTS (Pinter et al., 1988) with three stop codons, an RBS sequence followed byNdeI-Not1 sites, and anrrnC transcription terminator (see Fig. 2). Expression constructscontainingDHFR fragments were transformed into the appropriate hosts. pUC-pfDHFR was transformed in E. coli DH5q pKK-pfDHFR and PETpfDHFR were originally propagated in DH5aand subsequently cloned intoJM105 (ladQ) and BL21(DE3) or BLPl(DEB)pLysS, respectively. Cell Growth and Induction-Bacterial clones weregrown in LB supplemented with 50 pg/ml thymine and 100 pg/ml carbenicillin. Fresh overnight cultures from a single colony of plasmid-containing E. coli BLZl(DE3)pLysS were used to inoculate 1-6 liters of LB. The cultures were grown at 37 "C until Am reached 0.5-0.6, when IPTG was added to a final concentration of 0.4 mM. The cultures were allowed to grow at 37 "C for an additional 3 h prior to harvesting by centrifugation at 10,000 X g for 10 min at 4 "C. The cell pellets were washed with 250 ml of cold phosphate-buffered saline (Cazf and M e free). Preparation of Inclusion Bodies, Unfolding, and Refolding-hclusion bodies were partially purified using the procedure of Lin and Cheng (1991). Unfolding and refolding were performed a t 4 "C. The partially purified inclusion bodies from -300ml of culture were dissolved in 10 ml of buffer A (20 mM potassium phosphate buffer, pH 7.0, 0.1 mM EDTA, 10 mM DTT) supplemented with 0.2 M KC1 and 6 M guanidine HCl. The mixture was stirred slowly at 4 "C for at least 1 h. Refolding was achieved by dropwise, 20-fold dilution of the denatured proteins into buffer A with 0.2 M KC1 and 20% glycerol. After 3 h, aggregates were pelleted by centrifugation at 10,000 X g for 20 min, and theclear supernatant was used for further purification. Protein Purification and Characterization-Unless specified, 20% glycerol was included in all buffers. The refolded protein (-200 ml; 0.04-0.07 mg/ml) was circulated twice through an MTX-Sepharose CL-GB column (1.5 X 5.0 cm) preequilibrated with buffer A with 0.2 M KC1 at a flow rate of -1 ml/min. The column was washed with buffer A containing 1 M KC1 (-50 ml) followed by the same buffer containing 50 mM KC1 (-50 ml). Thirty ml of 4 mM Hafolate in buffer B (50 mM TES, pH 7.8, 0.1 mM EDTA, 10 mM DTT) containing 50 mM KC1was applied to the column at the same flow rate. After 1 column volume, the column was equilibrated for 1 h prior to washing

3ihydrofolate Reductase with the same buffer. Fractions (4 ml) containing DHFR activity werepooled and concentrated. H2folate was removed by filtration through a prepacked Sephadex G-25 column (NAP-25), (Pharmacia) preequilibrated with buffer A. Enzyme Assay-The activity of DHFR was determined spectrophotometrically by monitoring the decrease in absorbance at 340 nm at 25 "C (Hillcoat et al., 1967). The standard assay (1 ml; 1-cm cuvette) contained 100 p~ Hzfolate, 100 p~ NADPH, 50 mM TES, pH 7.0,75 mM 2-mercaptoethanol, 1 mg/ml bovine serum albumin, and -0.01 unit of enzyme. Unless otherwise specified, the reaction was initiated with Hzfolate. One unit of enzyme activity is the amount of enzyme required to produce 1 pmol of product/min at 25 "C. For determination of the pH-initial velocity profile, a three-component, constant ionic strength buffer (Williams and Morrison, 1981) was used. The concentration of active enzyme upon refolding was determined by titration with MTX (Ackermann and Potter, 1949). For determination of steady-state kinetic parameters, the concentration of one substrate was varied between 0.7 and 50 p~ with the other substrate at 100 p ~ and , the reaction was initiated by addition of enzyme (-1-6 nM). Kinetic parameters were calculated using a nonlinear least squares fit of the data to the Michaelis-Menten equation. For determination of ICs0 values, reactions were initiated with 0.0060.01 unit ml"of enzyme, and initial velocities in the presence of varying amounts of inhibitors were fit to Equation 111-5 from Segel (1975). Protein Analysis-Protein determination(Read and Northcote, 1981) and SDS-PAGE (Laemmli, 1970) were performed as described. Proteins for amino-terminal sequencing were electroblotted onto Immobilon (Millipore) membranes (Matsudaira, 1987) and microsequenced on an Applied Biosystem 475 protein sequenator. RESULTS

Design of the Synthetic Gene-The strategy of the design of the synthetic P. falciparum DHFR domain was similar to that describe for the synthetic L. casei T S gene (Climie and Santi, 1990). The primary sequence of P. falciparum TSDHFR (Bzik et al., 1987) was reverse-translated using the program PROTORES (Martinez, 1985) to obtain degenerate DNA sequences, and a list of all potential restriction sites. The restriction enzymes in the search file were limited to those that were commercially available, recognize 6 bases or more, and were not present in pUC except in the polylinker. Using this approach, approximately 80 potential restriction sites were found to be present in the P. falciparum DHFR domain. Based on homology comparisons with other DHFRs and a previous report of a catalytically active P. falciparum DHFR domain (Hall etal., 1991),we surmised that theDHFR domain of the bifunctional TS-DHFR was represented by amino acids 1-231, and we placed an in-frame termination codon ( T U ) at codon 232. Whenever possible, restriction sites of the DHFR domain were chosen using codons of highly expressed genes of E. coli and to retain a maximal number and even distribution of unique restriction sites (Gouy and Gautier, 1982; de Boer and Kastelein, 1986) and avoid strong secondary structure. Suboptimal codons were sometimes used to allow introduction of desirable unique sites or destruction of redundant sites. The final design of the P. falciparum synthetic DHFR domain consisted of a 693-base pair coding sequence with 26 unique restriction sites. Construction of the Gene and Expression Vectors-Seven complementary oligonucleotide duplexes of about 100 nucleotides each were used to construct the synthetic DHFRcoding sequence (Fig. 1).To facilitate cloning, the EcoRI-Hind111 polylinker of pUC18 was replaced by a polylinker that contained the cloning sites present at theends of each the seven segments, i.e. EcoRI, SacII, AflII, ClaI, BglII,MluII, SpeI, and HindIII, to give pUCm7. The synthetic gene fragments were successively ligated into the modified vector. To shorten the time of gene assembly, each half of the synthetic gene was constructed simultaneously in two separate pUCm7 vectors. Fragment 1 (EcoRI-Sac 11), fragment 2 (SacII-AflII), and

a

21639

Plasmodium falciparum Dihydrofolate Reductase

pUCm7

R

S

Af

M Sf-H

C

R

Nd R

R

S

Af

C

B g M S p H

S A f C B g M S p H

I

Nd

Fd R S A f C

pfDHFR (1-3)

Nd

R

S

A

f

C

pfDHFR (4-7)

B

g

M

S

p

I

\

Nd

No

H

I

ptmm(1-7)

R

S

I Nd

Af I

C

B

I

I

g

pucpr~~

M

sp

I

I

H

No

FIG. 1. Assembly of the synthetic P. falciparum DHFR domain. The DHFR domain was assembled from a total of seven oligonucleotide fragments (FI-F7). The fragments were sequentially cloned into the modified pUC vector pUCm7. The coding sequence is shown as solid bars. Af, AflII; Eg, BglII; C, ChI; H, HindIII; M , M2uI; Nd, NdeI; No, NcoI; R, EcoRI; S , SacII; Sp, SpeI.

fragment 3 (AflIIIClaI) were successively ligated into pUCm7 to give pfDHFR (1-3) whereas fragment 4 (CluI-BglII), fragment 5 (BglII-MluI), fragment 6 (MluI-SpeI), and fragment 7 (SpeI-HindIII) were constructed in another pUCm7 to give pfDHFR (4-7). To create the entire DHFR domain in a single vector, pfDHFR (4-7) was restricted with ClaI and HindIII, and the fragment was ligated into the corresponding sites of the pfDHFR (1-3) to give pfDHFR (1-7) (Fig. 1). To create potential expression constructs, the NdeI-Not1 fragment of pfDHFR (1-7) was cloned into pRBSrrnc, which has an RBS at the5’ end followed by NdeI-Not1 cloning sites

and an rrnC transcription terminator at the 3’ end (Brodin et al., 1986; Young, 1979), to give pUC-pfDHFR. The DNA sequence of DHFR in pUC-pfDHFR is shown in Fig. 2. The EcoRI-Hind111fragment of pUC-pfDHFR was cloned into the corresponding sites of pKK223-3 to give pKK-pfDHFR. Mutation of C to G was made by cassette mutagenesis of pUCpfDHFR to destroy the BamHIsite silently, giving pUCpfDHFR Db. The NdeI-Hind111 fragment of pUC-pfDHFR Db was cloned into PET-17b to give PET-pfDHFR. Expression-Initial attempts to express the synthetic gene encoding the P. falciparum DHFR domain were by constitu-

21640

Plasmodium falciparum Dihydrofolate Reductase Em R I

RBS

Ndrl

Aar

I

4

I

4 h l 10

Ile GAC ATC C T GT A G

AJP

Ala

Tyr

CYJ

Ile

TAC GCT A T GC G A

ATTGCT A TA CG A

m a l.

LYJ AAA TTT

Val

Clu

Lp AAA TTT

Ser

G AA T GT A C C T CC A TG A T

* e. . I " "

"~

AJP GIu G AA GC A T T T CT C TT

Cly

LYJ AAA TTT

Lp AAA TTT

A J ~ Clu

Trp

LYJ

CYJ A m TGC AAC ACG T T G

5I

Tyr Glu LYJ T A C GAA A A A ATG CTT T T T

72

A J ~ Afp Mcl Pro A A C GAC A T G C C G T T G C T G T A C GGC

93

LYJ CCG AAA GGC T T T

II4

30

A GA AC A TCTTGT

Y I. ...

.Fbu r .

"

31

AJII

AJII Thr Tyr AAC AAC TAC ACC T C C G G T T T G TTG ATG TGG A CG G C iA P IG

2

Leu AJII Gly LYJ C T G GGC AAC A A A GAC C C G T T G T T T

"

Mer LYJ C ATG AAA G TAC TTT

52

AI/

Tyr Phe C ~ J Af ,Val TAC T T C TGC G G GTT A T G AAG ACG C C CAA

,

Pro

CCO TOG AAA GGC ACC T T T

T z l ,Tyrj Thr

Val

AC T

ACC T A TOG AT

GTA CAT

LYJ Clu Ser Lyr AAC GAA T C C A A A T T G C T T AGO T I T

Clu

Thr

Val

GAA CTT

ACC G T T GAC A C GT T G C T O G C A A ,AJP C T G {A V aAl i

Cly

Ar!

I

T h ll

"

73

Cyr Arl Lp T A C A A A CGC T G C A A A A T G T T T GCG ACG T T T Tyr

TT

AA

LYJ

Am

Scr

A A TCC C T TAGG G

Lyr LYJ AAA

TTT

LYJ AAA TTT

Lyr TTC AAA AAG T T T Phc

Alp Glu T T C GAC GAA AAG C T G C T T Phe

A mValValVal

Mer

A AG C TGT TGT TAT T G G C T T G CAA CA CA A T ACC CG

Alp ,Val G A T GTA C I A CAT

,

Ilr

Val

GI! G S llel G G A CG A T C C TATG T A

Trj

TA AT

BOl

Ile

Ilc A J ~ LYJ A T C ATC AAC AAA TAG TAG TTG T T T

Cya Phe Ilr Ilc Cly T G C T T C A T C A T C GGT ACG A A G T A G T A G CCA

Ara

$

Leu C CTG AA G GAC T T

Thi

T

I

Tyri

Ser

Leu

A

Pro

GTT CAA

ATC GTT T A G CAA

Val

AJP

I35

GAT CTA

L e j ,Lue Lys CTC CTA G C AAA G G GAT G TTT

Leu CTG GAC

I56

LYJ AAA TTT

177

Avr

Ilr

LYJ AAG TTC

AA TT

I

Val

am .. .n.

Lp AAC T A C T A C A A A TTG ATG ATG T T T

A m Tyr Tyr

2

C C ACC T G T G G T G AGC A

"

"

157

U

h

Em N I

Am Ara Ile AJn C G TA OA GAC CGA CTA CAG C TAT T C C T TGC C GG C G C GAC TC TTG CG TAG TTG CAA TAG AC AGC GCG

Pri

T

I

1102

Est 1107

136

Lyr TAC C T G AAC A A A A T G GAC T T G T T T

vi

epu II5

Am

Om

Leu Gln CTG C

AAA TTT

Len

I

Pst

94

Tyr

.I

I

;j

l. d.a l. .

ValVal

GTT CAA

Tyr Gln GIn

Phc

T C GA G T T T A C C A G GAG T T C C A A A T G G T C C T C A A G pG A uG C

LYJ AAA TTT

I78

I99

220

Hi, d Y

CTAAGGATTTTTTTTATC

FIG.2. DNA sequence of the synthetic P. fulciparum DHFR gene carried in pUC-pfDHFR. Lines with arrows indicate the ends of each gene fragment. Rectangular boxes indicate the cleavage sites of the restriction enzymes. The three stop codons 5' upstream to the RBS are indicated by an underline. The positions of amino acids are indicated on the left and right sides of the sequence. tive expression of the gene from pUC-pfDHFR or by TPTGinduced expression from pKK-pfDHFR. Neither gave detectable DHFR activity in crude extracts of appropriate transformed E. coli hosts. SDS-PAGE of soluble and insoluble fractions of the crude extract showed no evidence of DHFR expression (data not shown). The first successful expression of the P.fakiparum DHFR domain was with a PET vector under the control of T7 promoter (Studier et al., 1990). In this system, expression is induced by IPTG treatmentof transformed E. coli BL21(DE3)

which contains the T7RNA polymerase gene under the control of the lac promoter. After transformation of PET-pfDHFR into E. coli BL21(DE3) and BL21(DES)pLysS, and induction with IPTG, proteins were analyzed by SDS-PAGE. An intense band with a molecular mass of about 30 kDa accounting for over 50% of the protein was produced in the transformed E. coli BL21(DE3)pLysS (Fig. 3) but not in E. coli BL21(DE3) (data not shown). SDS-PAGE of the soluble and insoluble fractions of BL21(DE3)pLysS harboring PETpfDHFR revealed that the expressed product was mainly

Plasmodium falciparum Dihydrofolate Reductase kDaf

21641

the refolding buffer, the recovery of DHFR activity was achieved with specific activity of 0.8-1.2 units/mg. Aggregation of protein was still observed, but more than 80% of the 97.4 DHFR activity remained in the clear supernatant after cen66.2 trifugation of the aggregated proteins. The optimal concentrations of DTT and KC1 in both un45.0 folding and refolding buffer were investigated. With urea as denaturant, the highest specific activity of DHFR (-0.8-1.2 31.O units/mg) upon refolding was obtained in the presence of 10 mM DTT and0.2 M KCl. The specific activity of the refolded 21.5 enzyme was 8-10-fold higher when 6 M guanidine HC1 was employed as denaturant (Table I). The presence of 1 mM 14.4 Hzfolate in the refolding buffer did not improve the recovery of enzyme activity, but inclusion of 20% glycerolin the buffer FIG. 3. Expression of P. fulcipurum synthetic DHFR. Coomassie-stained SDS-10% PAGE of samplesobtained from IPTG minimized aggregation of the protein during the refolding induction of BLZl(DE3)pLysS harboring recombinantplasmid PET- step. The final refolding procedure utilized a 20-fold dilution pfDHFR. Lane 1, control pellet of BL21(DE3)pLysS harboring the into a refolding buffer containing 20% glycerol and yielded PET-17b vector; lune 2, control supernatant of BLZl(DE3)pLysS DHFR with a specific activity of 20-30 units/mg. However, harboring the PET-17b vector; lune 3, pellet of BL21(DE3)pLysS use of MTX as an active site titrantof properly folded enzyme harboring PET-pfDHFR lune 4, supernatant of BLPl(DE3)pLysS revealed that only 5-10%of the protein was active DHFR harboring PET-pfDHFR lane 5, partially purified inclusion bodies; lune 6, purified DHFR from MTX-Sepharose column. Molecular (Table I); from this, we estimated a specific activity of the pure pfDHFR tobe about 300-350 units/mg. are shown atthe left. masses ( X Purification and Characterization-The refolded DHFR present in the insoluble pellet. The soluble fraction of was purified about "-fold to homogeneity in about 80% yield BLPl(DE3)pLysS harboring PET-pfDHFR also showed de- by affinity chromatography on MTX-Sepharose (Table 11). tectable DHFR activity of-0.01 unit/mg. Induction of the SDS-PAGE showed a single protein band migrating with mass cells at lower temperature (25 "C)neither improved the solu- of 30 kDa (Fig. 3), andgel filtration on aSuperose 12 column bility nor increased the level of the expressed products (data revealed that the purified enzyme was a monomer (data not shown). Titration of the purified enzyme with MTX indicated not shown). To verify that the expressed product was pfDHFR, the that it was -100% active. The yield of the purified pfDHFR calculated for 1liter of E. coli culture was 5-6 mg, 50-60 times protein from inclusion bodies was electroblotted ontoan Immobilon P membrane, and the bandwas excised and sub- higher than reported previously for the wild type bifunctional jected to microsequencing analysis. The amino-terminal se- TS-DHFR (Sirawaraporn et al., 1990). The effects of pH on the purified pfDHFR and wild type quence of 10 residues matched the expected sequence M-ME-Q-V-C-D-V-F-D. There was aminor sequence present bifunctional TS-DHFR from P. fulciparurn clone 3D7 were (