Xanthine Phosphoribosyltransferase from Leishmania donovani

12 downloads 1024 Views 430KB Size Report
complex life cycle in which the extracellular, flagellated pro- mastigote is present in the .... ing 100 μg/ml ampicillin and 10 μg/ml streptomycin, and the lawn of ..... netic software package in a 100 mM Tris-HCl, pH 8.0, 10 mM MgCl2 buffer at 27 ...
THE JOURNAL OF BIOLOGICAL CHEMISTRY © 1999 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 274, No. 48, Issue of November 26, pp. 34403–34410, 1999 Printed in U.S.A.

Xanthine Phosphoribosyltransferase from Leishmania donovani MOLECULAR CLONING, BIOCHEMICAL CHARACTERIZATION, AND GENETIC ANALYSIS* (Received for publication, August 4, 1999, and in revised form, September 24, 1999)

Armando Jardim‡§, Susan E. Bergeson‡§, Sarah Shih‡, Nicola Carter‡, Randall W. Lucas‡, Gilles Merlin¶, Peter J. Myleri, Kenneth Stuarti, and Buddy Ullman‡** From the ‡Department of Biochemistry and Molecular Biology, Oregon Health Sciences University, Portland, Oregon 97201, the ¶Laboratoire de Parasitologie Mole´culaire, Universite´ Victor Se´gole`ne de Bordeaux II, UPRESA-5016 CNRS, 146 Rue Le´o Saignat, 33076 Bordeaux Cedex, France, and the iSeattle Biomedical Research Institute, Seattle, Washington 98109

Xanthine phosphoribosyltransferase (XPRT) from Leishmania donovani is a unique enzyme that lacks a mammalian counterpart and is, therefore, a potential target for antiparasitic therapy. To investigate the enzyme at the molecular and biochemical level, a cDNA encoding the L. donovani XPRT was isolated by functional complementation of a purine auxotroph of Escherichia coli that also harbors deficiencies in the prokaryotic phosphoribosyltransferase (PRT) activities. The cDNA was then used to isolate the XPRT genomic clone. XPRT encodes a 241-amino acid protein exhibiting ;33% amino acid identity with the L. donovani hypoxanthine-guanine phosphoribosyltransferase (HGPRT) and significant homology with other HGPRT family members. Southern blot analysis revealed that XPRT was a single copy gene that co-localized with HGPRT within a 4.3-kilobase pair (kb) EcoRI fragment, implying that the two genes arose as a result of an ancestral duplication event. Sequencing of this EcoRI fragment confirmed that HGPRT and XPRT were organized in a head-to-tail arrangement separated by an ;2.2-kb intergenic region. Both the 3.2-kb XPRT mRNA and XPRT enzyme were significantly up-regulated in Dhgprt and Dhgprt/Daprt L. donovani mutants. Genetic obliteration of the XPRT locus by targeted gene replacement indicated that XPRT was not an essential gene under most conditions and that the Dxprt null strain was competent of salvaging all purines except xanthine. XPRT was overexpressed in E. coli and the recombinant protein purified to homogeneity. Kinetic analysis revealed that the XPRT preferentially phosphoribosylated xanthine but could also recognize hypoxanthine and guanine. Km values of 7.1, 448.0, and >100 mM and kcat values of 3.5, 2.6, and ;0.003 s21 were calculated for xanthine, hypoxanthine, and guanine, respectively. The XPRT gene and XPRT protein provide the requisite molecular and biochemical reagents for subsequent studies to validate XPRT as a potential therapeutic target.

* This work was supported in part by NIAID Grant AI23682 from the National Institutes of Health and in part by a grant from The Burroughs Wellcome Fund. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The nucleotide sequence(s) reported in this paper has been submitted to the GenBankTM/EBI Data Bank with accession number(s) AF170105. § Both authors contributed equally to this work. ** Burroughs Wellcome Fund Scholar in Molecular Parasitology. To whom correspondence should be addressed: Dept. of Biochemistry and Molecular Biology, Oregon Health Sciences University, Portland, OR 97201. Tel.: 503-494-8437; Fax: 503-494-8393; E-mail: ullmanb@ohsu. edu. This paper is available on line at http://www.jbc.org

Leishmania donovani is a protozoan parasite that is the causative agent of visceral leishmaniasis, a devastating and invariably deadly disease if untreated. The parasite exhibits a complex life cycle in which the extracellular, flagellated promastigote is present in the phlebotomine sandfly vector, and the intracellular amastigote form is found within the phagolysosome of macrophages and other reticuloendothelial cells of the mammalian host. The current arsenal of drugs used to treat leishmaniasis was arrived at empirically and is far from ideal. Chemotherapy is complicated both by drug toxicity and resistance (1), and the need for more efficacious and less toxic agents, particularly rational drugs that exploit targets unique to the parasite, is acute. Perhaps the metabolic pathway that is most discrepant between Leishmania and the mammalian host is that by which purine nucleotides are synthesized. Whereas mammalian cells synthesize purine nucleotides de novo, all protozoan parasites lack this purine pathway (2). Consequently, each genus of parasite expresses a unique complement of purine salvage enzymes that enable the acquisition of host purines (2). Biochemical measurements in extracts of L. donovani revealed the existence of three biochemically distinct phosphoribosyltransferase (PRT)1 enzymes, hypoxanthine-guanine PRT (HGPRT), adenine PRT (APRT), and xanthine PRT (XPRT), all of which convert preformed purine bases to the nucleotide level (3). Whereas HGPRT and APRT have mammalian counterparts, XPRT is unique to L. donovani and is, therefore, a potential target for therapeutic exploitation. HGPRT and APRT have both been cloned and overexpressed in Escherichia coli and the native and recombinant HGPRT and APRT proteins purified to homogeneity and characterized (4, 5). The construction of viable Dhgprt and Daprt null mutants that can proliferate in defined growth medium supplemented with any purine base or nucleoside and salvage radiolabeled hypoxanthine, coupled with the ability of xanthine to obliterate hypoxanthine incorporation in Dhgprt but not wild type parasites, implicate a prominent role for XPRT in purine scavenge by L. donovani (4). In order to characterize XPRT and investigate its role in intact parasites, we have cloned XPRT by functional rescue of an E. coli purine auxotroph lacking prokaryotic PRT activities. Sequence analysis and molecular characterization of the XPRT 1 The abbreviations used are: PRT, phosphoribosyltransferase; HGPRT, hypoxanthine-guanine phosphoribosyltransferase; APRT, adenine phosphoribosyltransferase; XPRT, xanthine phosphoribosyltransferase; HGXPRT, hypoxanthine-guanine-xanthine phosphoribosyltransferase; LPI medium, low phosphate induction medium; PCR, polymerase chain reaction; PRPP, phosphoribosylpyrophosphate; DME-L Dulbecco’s modified Eagle’s medium, Leishmania; DME-L-FBS, DME-L plus 5% fetal bovine serum; PBS, phosphate-buffered saline; bp, base pair; kb, kilobase pair.

34403

34404

Xanthine Phosphoribosyltransferase from L. donovani

locus revealed that XPRT was a single copy gene located 2.2-kb downstream from HGPRT. To test XPRT function in intact parasites, a Dxprt knockout was created by double-targeted gene replacement and characterized for its ability to salvage a variety of purines. For biochemical characterization, XPRT was overexpressed in E. coli, and the recombinant enzyme was purified to homogeneity. Kinetic analysis revealed a novel enzyme of unique substrate specificity that phosphoribosylated xanthine most efficiently but also recognized, albeit inefficiently, hypoxanthine and guanine. EXPERIMENTAL PROCEDURES

Chemicals and Reagents—Restriction endonucleases and Tfl polymerase were purchased from either Life Technologies, Inc., New England Biolabs (Beverly, MA), or Epicentre Technologies (Madison, WI). Radiolabeled [14C]purine bases (50 – 60 mCi/mmol) were procured from Moravek (Brea, CA). [32P]dCTP was obtained from ICN Biomedicals (Costa Mesa, CA). All other reagents were of the highest quality commercially available. Isolation of the XPRT cDNA by Functional Complementation—A directionally cloned Uni-ZapXR L. donovani cDNA library was subjected to in vivo excision and electroporated into SF609 E. coli (Dprogpt-lac, thi, hpt, pup, purHJ), a strain that is auxotrophic for purines and lacks the bacterial PRT enzymes, hypoxanthine PRT, and xanthine-guanine PRT (6). The bacteria were plated on LB plates containing 100 mg/ml ampicillin and 10 mg/ml streptomycin, and the lawn of transformants was resuspended in 20 ml of phosphate-buffered saline (PBS) and washed twice with 20 ml of a low phosphate induction (LPI) medium. The cell suspension was diluted to 0.5 A550 with LPI, and 100-ml aliquots were plated on LPI agarose plates supplemented with 150 mM xanthine, 100 mg/ml ampicillin, and 10 mg/ml streptomycin and incubated at 37 °C for 72 h. Plasmid DNA was isolated from the resultant colonies, and the ends of the inserts were sequenced on a PerkinElmer Applied Biosystems 377 DNA sequencer using standard dyeterminator cycle sequencing methodology. Isolation of the L. donovani XPRT Genomic Clone—20,000 colonies from an L. donovani cosmid library were screened at high stringency using the XPRT cDNA as a hybridization probe using previously reported laboratory protocols (7). A 4.3-kb EcoRI fragment encompassing both HGPRT and XPRT was subcloned from one purified cosmid into the pBluescript vector KS1 (Stratagene, La Jolla, CA) for restriction mapping and DNA sequencing. Analyses of nucleotide and amino acid sequences were performed using the CLUSTAL X multiple sequence alignment program (8). Parasite Cell Culture—The construction and molecular characterization of the Dhgprt and Dhgprt/Daprt knockout parasites by targeted gene replacement has been described in detail (4, 5). DI700 is the wild type clone of L. donovani from which the Dhgprt and Dhgprt/Daprt clones were eventually derived. All L. donovani lines were propagated in a completely defined Dulbecco’s modified Eagle-Leishmania (DME-L) medium that was routinely supplemented with 100 mM purine. Selections of L. donovani transfectants were performed on DME-L-FBS plates (DME-L plus 5% fetal bovine serum) and 1.0% Noble agar. The purine source employed for routine propagation of all L. donovani strains was xanthine, except for the selection and maintenance of the Dxprt line in which the xanthine in the culture medium was replaced with hypoxanthine. The purine in DME-L was varied when assessing purine salvage capacities of wild type and mutant L. donovani strains. Northern Blot Analysis—Total RNA was isolated from wild type (DI700), Dhgprt, and Dhgprt/Daprt promastigotes in late exponential phase using the Qiagen RNeasy midi-kit. Thirty mg of total RNA were loaded onto each lane of a 0.8% formaldehyde-agarose gel and electrophoresed under standard conditions (7). The RNA was transferred onto a GeneScreen Plus nylon membrane (NEN Life Science Products) by capillary action using 40 mM NaOH and cross-linked with ultraviolet light. Membranes were prehybridized in 63 SSC, 50% formamide, 0.1% SDS solution containing Denhardt’s solution and 200 mg/ml herring sperm DNA for 2 h at 42 °C. Northern blots were hybridized at 42 °C for 20 h with a random prime labeled 598-bp fragment from the XPRT gene. Membranes were washed twice for 30 min in 13 SSC at 65 °C followed by autoradiography at 270 °C for 72 h. To normalize for RNA loading, the blot was stripped and reprobed with the L. enriettii 1.4-kb rRNA gene (9). Reverse Transcription PCR—mRNA was isolated from 1 3 108 log phase L. donovani DI700 promastigotes using the RNeasy midi kit (Qiagen Inc., Valencia, CA). cDNA was generated using random hex-

amer oligonucleotides and the Superscript reverse transcription kit (Life Technologies, Inc.). A nondegenerate sense primer, 59-CCAACGCTATATAAGTATCAGTTTCTGTACTTTATTG-39, was designed to the L. donovani mini-exon (10) that is trans-spliced onto the 59 terminus of all leishmanial mRNAs (11), and an antisense primer 59-GCGGGTTGTCGTAGCTGAT-39 to the XPRT coding sequence was used to amplify the 59 end of the XPRT mRNA. The polymerase chain reaction (PCR) was performed on an MJ Research thermocycler using 40 cycles of denaturation at 94 °C for 30 s, annealing at 50 °C for 45 s, and extension at 72 °C for 30 s with Tfl as the polymerase (Epicentre Technologies, Madison, WI). The PCR product was ligated into pBluescript KS1 and sequenced. Molecular Constructs for the Replacement of the XPRT Alleles—A sense oligonucleotide, 59-TCCCAAGCTTGTAGAGGCATGCCTCATC39, and an antisense primer, 59-ACGCGTCGACGTAATTTATCTATCGCTCCGTTGC-39, containing restriction sites for HindIII and SalI (underlined), respectively, were used to amplify a 970-bp 59 flank of XPRT by PCR. Similarly, a 620-bp sequence 39 to the XPRT coding region was amplified using the sense primer 59-TCCCCCGGGGGAAATGTGGAGGCGGCTGAG-39 and the antisense primer 59-GGAAGATCTTGAAGAGGGAGAGTGGCGAGGT-39 that encompassed SmaI and BglII restriction sites (underlined), respectively. To generate the knockout constructs the 970-bp 59 and 620-bp 39 flanks were cloned into the HindIII/SalI and SmaI/BglII sites of the pX63HYG and pX63NEO vectors, which contain the hygromycin phosphotransferase and neomycin phosphotransferase resistance markers, respectively (12). The presence of the XPRT flanking regions and their orientation were confirmed by restriction mapping. The drug resistance cassettes were designated pX63-NEO-Dxprt and pX63-HYG-Dxprt. Transfections—Parasites were transfected by electroporation using conditions identical to those reported previously (4). pX63-NEO-Dxprt and pX63-HYG-Dxprt were linearized by digestion with HindIII and BglII, and the 4.6- (pX63-NEO-Dxprt) and 4.8-kb (pX63-HYG-Dxprt) fragments were gel-purified prior to electroporation. The first wild type XPRT allele was replaced with pX63-HYG-Dxprt to create the XPRT/ xprt heterozygote, whereas the other wild type allele was supplanted with pX63-NEO-Dxprt to generate the homozygous Dxprt knockout strain. Electroporated parasites were maintained for 24 h in liquid medium prior to plating in semi-solid DME-L-FBS. The XPRT/xprt lines were selected in DME-L-FBS supplemented with xanthine containing 50 mg/ml hygromycin, whereas the Dxprt null mutant was selected in 20 mg/ml geneticin and 50 mg/ml hygromycin with hypoxanthine as the purine source. Hygromycin was added to the selective medium for the Dxprt knockouts to ensure selection of transfectants in which pX63-NEO-Dxprt had integrated into the remaining wild type allele in the XPRT/xprt line. Colonies isolated after transfection with either pX63-NEO-Dxprt or pX63-HYG-Dxprt were initially expanded in 1.0 ml of DME-L-FBS and then continually maintained in DME-L (no fetal bovine serum) containing the appropriate purine and selective agents. The allelic replacements in the genetically manipulated strains were established by Southern blotting, and the enzyme deficiencies determined by immunoblotting. Once the genetic lesions were confirmed, the purine salvage capacities of the wild type, XPRT/xprt, and Dxprt promastigotes were assessed by growth in DME-L supplemented with a single purine base or nucleoside. According to the recently adopted genetic nomenclature for genetically manipulated Leishmania strains (13), the heterozygote and homozygote null mutant are designated XPRT/Dxprt::NEO and Dxprt::NEO/Dxprt::HYG, respectively. For the purposes of simplicity and communication, these strains will be called XPRT/xprt and Dxprt throughout this paper. XPRT Measurements in DI700, Dhgprt, and Dhgprt/Daprt L. donovani—Mid-log phase cultures (160 ml) of DI700, Dhgprt, and Dhgprt/ Daprt promastigotes were harvested by centrifugation and washed twice with 5.0 ml of PBS. Cell pellets were suspended in 5.0 ml of 50 mM Tris-HCl, pH 8.0, 10 mM MgCl2, 1.0% Triton X-100 (Calbiochem) and subjected to three rounds of rapid freezing and thawing to lyse parasites. Lysates were clarified by centrifugation at 30,000 3 g for 30 min at 4 °C and brought to 40% saturation with solid (NH4)2SO4 and incubated on ice for 1 h. Precipitated proteins were removed, and the supernatant was brought to 80% saturation with (NH4)2SO4 and maintained at 4 °C for 16 h. The protein precipitates containing the bulk of the L. donovani PRT activities (3) were dissolved in 1.0 ml of 50 mM Tris-HCl, pH 8.0, 10 mM MgCl2, and aliquots were used for determining protein concentration (14). XPRT activity was measured by mixing 10 ml of the 80% (NH4)2SO4 protein fraction with 150 ml of an assay mixture containing 40 mM [14C]xanthine and 1 mM pyrophosphorylpyrophosphate (PRPP) in 100 mM Tris-HCl, pH 8.0, 10 mM MgCl2 and

Xanthine Phosphoribosyltransferase from L. donovani

34405

FIG. 1. Multiple amino acid sequence alignments of the L. donovani XPRT with HGPRT and HG(X)PRTs. The predicted amino acid sequences for the L. donovani XPRT (LdXPRT) and HGPRT (LdHGPRT), E. coli HPRT (EcHPRT) and XGPRT (EcXGPRT), human (HsHGPRT), C. fasciculata (CfHGPRT), Schistosoma mansoni (SmHGPRT) HGPRTs, and T. gondii HGXPRT (TgHGXPRT) were aligned using the CLUSTAL X multiple sequence alignment program (8), and sequence shading was performed the MacBox Shade 1.08 software. Identical amino acid sequence motifs are shown in reverse print with dark gray shading, and conserved residues are shaded with light gray boxes, and the glycosomal topogenic signal tripeptide in the XPRT sequence is bordered with a black box.

incubating at 37 °C. At various intervals, 15-ml aliquots were removed, spotted onto DEAE filters that were then washed extensively with H2O, and the levels of XMP quantitated by liquid spectrophotometry. Expression and Purification of Recombinant XPRT—The L. donovani XPRT coding sequence was amplified from the genomic DNA template by PCR using the 59 sense primer 59-TCTCATATGCTACCAACCCACAGTTGT-39 and the 39 antisense primer 59-TCTCTCTGCAGTCAGAGCTTGGCAGGGTAAC-39. An NdeI restriction site was created at the initiation methionine codon in the sense primer, and a PstI restriction site was introduced into the antisense primer downstream of the termination codon for subcloning into the pBAce (15) expression vector (restriction sites in primers described in previous sentence are underlined). The fidelity of the XPRT sequence generated by PCR was verified by DNA sequencing. The pBAce-XPRT construct was transformed into SF609 E. coli (6), and XPRT expression was induced in LPI medium as described (15). One-liter volumes of overnight bacterial cultures were harvested, and the cell pellets was resuspended in 20 ml of 20 mM Tris-HCl, pH 9.0, 10 mM MgCl2 (TM) buffer and lysed with two passes through a French press. Lysates were clarified by centrifugation at 31,000 3 g for 30 min, and the supernatants were applied to DEAEcellulose columns (2.7 3 10 cm) equilibrated with 100 mM Tris-HCl, pH 9.0, 10 mM MgCl2. Void volume fractions containing XPRT activity, as determined spectrophotometrically (see below), were pooled, and solid (NH4)2SO4 was added to 40% saturation. Following a 30-min incubation at 0 °C, the precipitated proteins were removed by centrifugation and discarded, and the supernatant was brought to 60% saturation with solid (NH4)2SO4 to precipitate XPRT. The protein pellet was dissolved in 5.0 ml of 1.0 M (NH4)2SO4 in 20 mM Tris-HCl, pH 9.0, 10 mM MgCl2 and applied to an octyl-Sepharose column (1.0 3 10 cm) equilibrated with 1.0 M (NH4)2SO4 in 50 mM Tris-HCl, pH 8.0, 10 mM MgCl2. XPRT enzyme was eluted with a 100-ml linear gradient from 1.0 M to zero (NH4)2SO4 in 50 mM Tris-HCl, pH 8.0, 10 mM MgCl2. XPRT fractions containing only a single 27-kDa polypeptide, as assessed by SDS-polyacrylamide gel electrophoresis (16), were pooled, dialyzed against 2 liters of 20 mM Tris-HCl, pH 8.0, 10 mM MgCl2, and concentrated to 10 mg/ml protein using a Centriprep concentrator (Amicon, Beverly, MA).

All XPRT preparations were rapidly frozen in a dry ice/acetone bath and stored at 270 °C. No significant loss in enzyme activity was observed after 10 months of storage. Generation of XPRT Antisera and Immunoblot Analysis—Polyclonal antisera against L. donovani XPRT were generated in rabbits by Cocalico Biologicals Inc. (Reamstown, PA) using recombinant XPRT protein as immunogen and standard injection protocols. Antisera titers were evaluated by immunoblotting after fractionation of crude promastigote lysates on a reducing 15% SDS slab gel and transfer of proteins onto a nitrocellulose membrane (17). Blots were incubated in a blocking buffer consisting of PBS containing 3% fetal calf serum and 0.1% Tween 20 and probed with XPRT antisera diluted 1:1000 in blocking buffer. Blots were incubated with a goat anti-rabbit second antibody horseradish peroxidase conjugate (Roche Molecular Biochemicals) and developed with the NEN Renaissance chemiluminescence reagent (NEN Life Science Products). Signals were quantitated by densitometric measurements on a Bio-Rad scanning densitometer (Bio-Rad). Complementation Analysis—The pBAce-XPRT construct was transformed into SF609 E. coli and plated on LB medium containing 100 mg/ml ampicillin and 50 mg/ml streptomycin. A single bacterial colony was then picked and resuspended in 200 ml of LPI medium supplemented with ampicillin and streptomycin, and 20-ml aliquots were then dispensed into 3.0 ml of liquid LPI medium containing both antibiotics and either adenine, adenine/hypoxanthine, adenine/guanine, adenine/ xanthine, or adenine/guanosine. Each nucleobase or nucleoside was present at a concentration of 150 mM. Cultures were incubated at 37 °C with vigorous shaking (300 rpm), and cell growth was monitored spectrophotometrically at 600 nm. Steady-state Kinetic Measurements of XPRT—Initial rate measurements for the forward reaction were determined using a 1.0-cm path length on a Beckman DU640 spectrophotometer equipped with a kinetic software package in a 100 mM Tris-HCl, pH 8.0, 10 mM MgCl2 buffer at 27 °C. Nucleobase substrate Km values were determined in 1.0 mM PRPP and either 1–100 mM xanthine, 1–100 mM guanine, or 1– 8000 mM hypoxanthine, whereas the Km value for PRPP was ascertained at 1.0 mM hypoxanthine using PRPP concentrations ranging between 7.5

34406

Xanthine Phosphoribosyltransferase from L. donovani

FIG. 2. Southern blot analysis of XPRT locus in DI700. A, cosmid DNA (2 mg) containing the L. donovani XPRT was digested overnight with the indicated restriction endonuclease and subjected to Southern blot analysis using the L. donovani HGPRT and XPRT coding regions as hybridization probes. B, schematic representation of the restriction map for the EcoRI 4.3-kb fragment containing the HGPRT and XPRT genes. and 1000 mM PRPP. XMP and GMP formation was monitored at 250 and 257 nm using extinction coefficients of 3900 and 4200 M21 cm21, respectively. IMP formation was monitored at 243 and 275 nm for hypoxanthine concentrations of 1–700 and 750 – 8000 mM, respectively. IMP concentrations were calculated using extinction coefficients of 2200 M21 cm21 at 243 nm and 1570 M21 cm21 at 275 nm. RESULTS

Molecular Cloning and Sequencing of the L. donovani XPRT cDNA and Gene—In the absence of any sequence information about the XPRT protein, a functional complementation strategy was implemented to isolate the L. donovani XPRT cDNA. A directionally cloned L. donovani cDNA library was transformed into SF609 E. coli and plated in minimal M9 induction medium supplemented with adenine and xanthine. Five colonies were obtained, and nucleotide sequence of the plasmids revealed that the inserts exhibited the codon bias of a leishmanial protein coding gene (18) and encoded a unique member of the purine PRT family, i.e. the insert predicted a polypeptide different from the L. donovani HGPRT (15) and APRT (19) with considerably more homology to the former. A 600-bp NcoI/SacI fragment from the protein coding region of the longest cDNA was then used as a probe to isolate an XPRT genomic clone contained within a 4.3-kb EcoRI fragment from a purified cosmid. The nucleotide sequences of the protein coding regions of the longest cDNA and genomic clone were identical and predicted a 241-amino acid protein with a molecular mass of 27.1 kDa (Fig. 1). An in-frame termination codon was located 16 –18 nucleotides upstream from assigned initiation codon (see GenBankTM accession number AF170105). A multiple sequence alignment of the L. donovani XPRT protein with members of the “HGPRT” family from phylogenetically diverse organisms revealed that the XPRT contained the following: (i), a Val-LeuLys-Gly-Ser pentapeptide implicated in binding the 59-phosphate group of IMP and GMP to HGPRTs (20), (ii) a Ser-Tyr dyad that is absolutely conserved among HGPRT family members and is essential for catalytic activity (21), (iii) a conserved Lys at position 186 that is implicated in stabilizing purine binding, and (iv) a fingerprint Lys-His-Val-Leu-Ile-Val-Glu-

Asp-Val-Cys-Asp-Ser-Gly-Arg-Thr PRPP binding motif that is found among almost all members of the PRT family, as well as PRPP synthetase (22) (Fig. 1). A pairwise alignment between the L. donovani HGPRT (15) and XPRT primary structures revealed a 33% amino acid identity and a similarity of 50% when conservative amino acid substitutions are considered, and similar pairwise alignments of XPRT with other HGPRTs displayed identities between 25 and 29% (Fig. 1). A similar comparison of the L. donovani XPRT with APRT (19) sequences showed essentially no homology except within the signature PRPP binding domain (data not shown). One other XPRT structural feature worth noting is the COOH-terminal tripeptide, Ala-Lys-Leu, a motif compatible with the topogenic signal for targeting to the glycosome (23), a peroxisomal-like organelle unique to trypanosomatid protozoa (24). Molecular Characterization of the XPRT Locus—Southern blot analysis of the XPRT locus using both cosmid (Fig. 2A) and genomic (data not shown) DNA gave identical restriction patterns and indicated that the gene was present as a single copy within the L. donovani genome, as BglII and SalI, both of which cut once within the XPRT coding region, each excised two hybridizing bands of unequal size. Strikingly, the restriction pattern of the XPRT locus was remarkably akin to that of HGPRT (Fig. 2A). Digestion of cosmid and L. donovani genomic (data not shown) DNA and probing with the XPRT coding region (1st panel) revealed a hybridization pattern consisting of a 7.5-kb BamHI, a 6.0-kb BglII, a 4.3-kb EcoRI, a 3.9-kb PstI, a 3.3-kb SacI, and a 2.9-kb SalI fragments common to those obtained after hybridization to HGPRT (2nd panel). Control experiments showed that XPRT and HGPRT do not crosshybridize under the high stringency hybridization and wash conditions employed in these Southern blot analyses. These data implied that HGPRT and XPRT were proximally located on the same L. donovani chromosome. The co-localization of HGPRT and XPRT was supported further by Southern blot analysis of L. donovani chromosomes fractionated by contour clamped homogeneous electric field gel electrophoresis (25) that revealed that the two genes were indeed located on the same chromosome (data not shown). Sequence analysis of the 4.3-kb EcoRI fragment subsequently verified that HGPRT (nucleotides 178 – 810) and XPRT (nucleotides 2968 –3670) were separated by ;2.2 kb of intergenic region that did not accommodate any significant open reading frames and that the two genes were oriented in the same direction. The longest open reading frames observed within this putative intergenic region were 201 and 312 bp and neither conformed to the leishmanial codon bias (18) nor displayed homology to any known sequences in the current protein data bases using the BLAST algorithm. A restriction map of the HGPRT/XPRT locus compiled from Southern blot and nucleotide sequence analysis is presented in Fig. 2B. Expression of XPRT in L. donovani—Northern blot analysis of total RNA revealed a 3.2-kb XPRT mRNA that was expressed at significantly higher levels in previously described (4) Dhgprt and Dhgprt/Daprt L. donovani strains (Fig. 3A). Dhgprt and Dhgprt/Daprt promastigotes contained approximately 3 and 10 times more XPRT transcript than wild type parasites, respectively. Normalization of these same blots with the Leishmania enriettii rRNA confirmed that equal amounts of RNA had been loaded onto each lane. The increased XPRT mRNA level observed in the Dhgprt/Daprt parasites corresponded to equivalently elevated levels of XPRT protein (Fig. 3B) and XPRT catalytic rates (Fig. 3C) in these strains as well. Creation of Dxprt Knockouts—In order to evaluate the role of XPRT in purine salvage by L. donovani, Dxprt null mutants were generated by sequentially targeting each wild type XPRT

Xanthine Phosphoribosyltransferase from L. donovani

34407

FIG. 4. Southern blot analysis of the XPRT loci wild type and mutant strains. Genomic DNA (20 mg) was isolated from DI700, XPRT/xprt, and Dxprt promastigotes, digested with EcoRI, fractionated on a 0.8% agarose gel, and blotted onto nylon membranes. Blots were hybridized under high stringency conditions with probes to either the XPRT coding region (A), to the 59- (B) or 39-flanking regions (C) of the XPRT. The location of these probes is indicated by the thick lines in Fig. 5A. Gels were stained with ethidium bromide prior to the transfer of the DNA to the nylon membranes in order to verify an equal loading in all lanes.

FIG. 3. Expression of XPRT in wild type and mutant L. donovani. Expression of the XPRT mRNA transcript and XPRT protein in DI700, Dhgprt, and Dhgprt/Daprt cell lines was assessed by Northern and Western blot analysis. A, for Northern blots, 30 mg of total RNA was loaded per lane, and the blot was probed with the XPRT coding region to detect XPRT transcript. RNA loading in each lane was normalized with an L. enriettii rRNA probe (9). B, for Western blots, the protein from 1 3 106 log phase promastigotes was resolved on a 15% reducing SDS-PAGE gel, transferred to a nitrocellulose membrane, and probed rabbit anti-XPRT antisera at a 1:1000 dilution. C, the XPRT enzymatic activity in DI700 (M), Dhgprt (f), and Dhgprt/Daprt (E) parasites was measured by a radiometric assay using 40 mM [14C]xanthine as described under “Experimental Procedures.”

allele with a drug resistance cassette. The first XPRT allele was replaced with pX63-NEO-Dxprt to create the XPRT/xprt (XPRT/Dxprt::NEO) heterozygote, and the second was displaced with pX63-HYG-Dxprt to generate the Dxprt (Dxprt::NEO/Dxprt::HYG) null mutant. The homozygote was selected in DME-L supplemented with hypoxanthine rather than xanthine as a purine source in order to circumvent a potential lethal deficiency in xanthine salvage capability. Southern blot analysis of the wild type, XPRT/xprt, and Dxprt lines revealed the new alleles at the XPRT locus that had been generated by homologous recombination with the drug resistance constructs (Fig. 4). The novel alleles could be effectively distinguished from the wild type XPRT counterpart by the presence of novel EcoRI restriction sites within the drug resistance cassettes. Restriction maps of the linearized pX63HYG-Dxprt and pX63-HYG-Dxprt fragments are shown in Fig. 5 beneath the restriction map of the XPRT locus showing the location of the probes used in the Southern blot experiments presented in Fig. 4. Western blot analysis confirmed the lack of

XPRT gene product in the Dxprt parasites (Fig. 6). Nutritional Requirements of Dxprt Parasites—The ability of wild type, XPRT/xprt, and Dxprt L. donovani to grow in completely defined growth medium and their inability to generate purine nucleotides de novo permitted an assessment of their nutritional requirements for exogenous purines when the extracellular purine source is varied. No differences in the growth rates of the wild type and genetically manipulated strains were observed when the exogenous purine was hypoxanthine, adenine, guanine, inosine, adenosine, or guanosine. The Dxprt strain, however, could not grow in medium in which xanthine was the sole purine source. Microscopic examination of Dxprt null parasites grown in DME-L containing xanthine, however, indicated that some parasites remained viable and motile even after 12 days in this medium. It should be noted that motile DI700 and Dxprt parasites were also observed in DME-L media lacking any exogenous purine, even 2–3 weeks after seeding the cultures. Complementation Analysis with XPRT—The XPRT protein coding region was subcloned into the NdeI and PstI sites of the pBAce expression vector (15) for biochemical evaluation of XPRT function. As shown in Fig. 7, SF609 E. coli transformed with the pBAce-XPRT plasmid was incubated in LPI medium containing either adenine alone or adenine plus either hypoxanthine, guanine, xanthine, or guanosine. After a 20-h incubation at 37 °C, growth was observed in adenine/xanthine, adenine/guanosine, and to a lesser extent in adenine/hypoxanthine containing LPI medium. No growth in medium containing adenine alone or adenine/guanine as purine sources was observed. Adenine alone cannot serve as a source of guanylate nucleotides for SF609 cells because of the presence of high histidine levels in the LPI medium (6). Purification and Characterization of Recombinant XPRT— The L. donovani XPRT in SF609 E. coli was overexpressed and purified by ion exchange and hydrophobic interaction chromatography to apparent homogeneity (Fig. 8). The extent of enzyme purification from the SF609 supernatant was ;4-fold, indicating that XPRT was ;25% of the soluble protein after XPRT induction. Using steady-state conditions, XPRT exhib-

34408

Xanthine Phosphoribosyltransferase from L. donovani

FIG. 5. Restriction maps of the XPRT locus and the pX63-NEO-Dxprt and pX63-HYG-Dxprt targeting constructs. Restriction maps of the wild type XPRT locus (A) and the pX63-NEO-Dxprt (B) and pX63-NEO-Dxprt (C) targeting constructs are shown. The sizes of the fragments expected from EcoRI digestion of the wild type XPRT allele or the rearranged xprt alleles after integration of the G418 and hygromycin resistance markers are marked. The locations of probes derived from the XPRT flanks and coding regions are indicated in A by thick black lines. HGPRT and XPRT coding regions are shown as black and white rectangular boxes, respectively, whereas corresponding 59 and 39 flanks of the XPRT locus that were amplified by PCR to create the gene targeting constructs are displayed as gray boxes. The antibiotic resistance markers, NEO and HYG, are indicated by thick white boxes and are appropriately designated, and the corresponding pXbased flanking regions from the Leishmania major dihydrofolate reductase-thymidylate synthase (dhfr-ts) gene are indicated by thin white rectangles.

FIG. 6. Immunoblot analysis of XPRT expression in wild type and Dxprt null mutants. Total cell lysates of exponentially growing DI700 (lane 1) and Dxprt (lane 2) promastigotes were characterized by Western blot analysis using anti-XPRT polyclonal antisera that was generated in rabbits using purified recombinant protein as immunogen.

ited a Km value of 7.1 6 2.3 mM for xanthine when PRPP was fixed at a concentration of 1.0 mM and a kcat value of 3.5 6 1.5 s21. Hypoxanthine and guanine were also recognized by the recombinant XPRT as substrates, although inefficiently. Km values of 448 6 97 and .100 mM were obtained for hypoxanthine and guanine, respectively, again at 1.0 mM PRPP. Interestingly, the kcat value obtained with hypoxanthine was 2.6 6 0.2 s21, a value comparable to that observed with xanthine. In contrast, the catalytic efficiency for guanine phosphoribosylation by XPRT was ;0.003 s21, ;3 orders of magnitude lower than the kcat values for either xanthine or hypoxanthine. It should be noted, however, that the kinetic measurements with guanine exhibited a significant degree of inaccuracy due to the poor solubility of guanine in aqueous solutions. DISCUSSION

The XPRT cDNA from L. donovani was isolated by functional complementation of an E. coli purine auxotroph that lacks xanthine phosphoribosylating activity (6) and subsequently used as a probe to isolate the corresponding genomic clone. This functional rescue cloning strategy was mandated by the lack of XPRT amino acid sequence information that precluded the design of specific oligonucleotides and the limited homology among members of the purine PRT family (26) that made PCR-based cloning schemes fundamentally inaccessible. The strict selection conditions employed required the ability of SF609 E. coli transformants to salvage xanthine from the

FIG. 7. Complementation analysis of SF609 E. coli transformed with L. donovani XPRT. SF609 E. coli transformed with the pBAceXPRT construct were inoculated into LPI medium containing either adenine (Ade), adenine/hypoxanthine (Ade/Hyp), adenine/guanine (Ade/Gua), adenine/xanthine (Ade/Xan), or adenine/guanosine (Ade/ Guo). All purines were present at a concentration of 100 mM. Cultures were incubated at 37 °C for 20 h with vigorous shaking and the optical densities measured spectrophotometrically at 600 nm. Each bar represents the mean and standard deviation error for three independent experiments.

minimal bacterial growth medium via expression of the parasite XPRT cDNA. The nucleotide sequences of the XPRT cDNA and genomic clones were identical, unsurprising in view of the fact that parasites of the Leishmania genus lack introns and that the cDNA and genomic clones were derived from the same leishmanial species. Amino acid sequence analysis revealed that XPRT encompassed all of the conserved motifs of PRTs that recognize 6-oxypurines as substrates and was most closely related to members of the HGPRT family. Homology to other members of the PRT family, including adenine and pyrimidine PRTs, was limited to the consensus PRPP binding motif (27). One noteworthy additional structural feature of the L. donovani XPRT is the COOH-terminal tripeptide, Ala-Lys-Leu, a sequence consistent with the degenerate tripeptide signal for import into the glycosome (23), a peroxisomal-like microbody

Xanthine Phosphoribosyltransferase from L. donovani

FIG. 8. Purification of the recombinant L. donovani XPRT. E. coli SF609 cells transformed with pBAce-XPRT were grown in LPI medium, and the recombinant XPRT was purified as outlined under “Experimental Procedures.” Lane A, clarified whole E. coli lysates; lane B, DEAE cellulose void volume fractions; lane C, XPRT purified after hydrophobic interaction chromatography on an octyl-Sepharose column.

unique to kinetoplastid parasites that accommodates glycolytic and other nutritional and biosynthetic enzymes (24). The L. donovani HGPRT also possesses a COOH-terminal glycosomal targeting signal, Ser-Lys-Val, and has been definitively localized to the glycosome by confocal and immunoelectron microscopy (28). APRT, the third of the L. donovani PRTs, lacks this glycosomal import motif (19). Preliminary cell fractionation studies in this laboratory have indicated that XPRT co-sediments with other known glycosomal markers, including HGPRT and glyceraldehyde-3-phosphate dehydrogenase. Kinetic analysis of recombinant L. donovani XPRT demonstrated that the enzyme is a unique member of the PRT family with a substrate predilection for xanthine. Although the kcat values of the L. donovani XPRT for xanthine and hypoxanthine are comparable, the lack of an exocyclic oxygen at C-2 of the purine ring causes a marked destabilization of the hypoxanthine binding affinity, as reflected by the .60-fold greater Km value for hypoxanthine than for xanthine. In contrast, substitution of an amino group at C-2 of guanine profoundly diminishes the kcat value (;1,000-fold) and increases the Km value to .100 mM. It should be noted, however, that exact calculations of kinetic parameters for XPRT with guanine were undermined by the fact that guanine is relatively insoluble in aqueous solutions at concentrations necessary for accurate determination of kinetic constants. Thus, the kinetic values reported here for guanine are approximations extrapolated from Hanes plots. In contrast to the more promiscuous XPRT enzyme, the L. donovani HGPRT recognizes hypoxanthine and guanine exclusively among the naturally occurring purine bases with Km values in the low micromolar range and does not phosphoribosylate either xanthine or adenine (15). The inability of HGPRT to recognize xanthine has now been substantiated genetically, as Dxprt parasites cannot grow when xanthine is the only purine in the culture medium. The substrate specificities (and kinetic parameters) ascribed to both the purified recombinant L. donovani XPRT (Fig. 8) and HGPRT (15, 21) are supported by the results of the complementation analyses in which only efficient substrates could support the growth of SF609 transformants in minimal medium. The existence of genetically and biochemically distinct HGPRT and XPRT activities is unusual among eukaryotes. For instance, mammalian cells only express an HGPRT activity with a strict substrate specificity for the bases hypoxanthine and guanine (29). Indeed, the lack of a mammalian XPRT is the foundation for the utility of the bacterial xanthine-guanine phosphoribosyltransferase gene as a dominant selectable marker in mammalian cells (30). The L. donovani HGPRT and XPRT genes appear to have arisen as a consequence of an

34409

ancestral gene duplication event, as Southern blot data of both cosmid and genomic DNA and nucleotide sequencing demonstrated that the two genes co-localized within the genome in a head-to-tail arrangement separated by ;2 kb of noncoding DNA. The precise selective pressure necessary to sustain a distinctive XPRT activity within insect vector and mammalian hosts is not known, although high concentrations of xanthine in human plasma (31) intimate a potential selective advantage for organisms expressing an XPRT. Separate HGPRT and XPRT activities are also found in Trypanosoma cruzi (2) and Trypanosoma brucei, although the trypanosome XPRTs and their genes have not been characterized at the molecular, chromosomal, or biochemical level. Xanthine phosphoribosylating proteins have, however, been thoroughly characterized in other protozoan parasites, but these activities are associated with the HGPRT enzyme. Thus, for example, Plasmodium falciparum (32), Toxoplasma gondii (33), Tritrichomonas fetus (34), and Eimeria tenella (35) all contain a common hypoxanthine-guanine-xanthine PRT (HGXPRT) protein. The L. donovani XPRT substrate profile, however, is markedly different from the plasmodial, toxoplasmal, tritrichomonal, and eimerial HGXPRTs, as the L. donovani XPRT exhibits a striking preference for xanthine, whereas xanthine is the least favored of the 3 bases for the parasite HGXPRT enzymes. Other members of the PRT family that display unusual substrate preferences for 6-oxypurines include the Giardia lamblia guanine PRT (36) and the prokaryotic hypoxanthine- and xanthine-guanine PRTs (37). Conversely, every APRT enzyme thus far characterized exhibits an invariant and exclusive preference for adenine. Despite the substrate preference for xanthine, the capacity of XPRT to recognize hypoxanthine as a substrate provides a biochemical mechanism for the previously unexplained observation that Dhgprt/Daprt knockout parasites, even those additionally deficient in adenosine kinase activity, can grow in completely defined DME-L supplemented with any of a variety of purine sources, including hypoxanthine (4). That XPRT might be capable of hypoxanthine salvage was conjectured from the ability of these Dhgprt/Daprt lines to incorporate [3H]hypoxanthine into nucleotides and from the observation that high concentrations of xanthine obliterate [3H]hypoxanthine uptake into Dhgprt parasites but allow residual [3H]hypoxanthine uptake into wild type parasites (4). The ability of other purines to support the growth of these Dhgprt/Daprt strains can be explained by the fact that adenosine, adenine, and inosine are funneled through hypoxanthine by adenosine phosphorylase, adenine deaminase, and nucleoside hydrolase enzymes, respectively (38), whereas guanosine and guanine are converted through nucleoside hydrolase (39) and guanine deaminase (40) activities to xanthine, the preferred XPRT substrate. Thus, a functional XPRT activity alone appears sufficient to enable L. donovani promastigotes to salvage adenine, guanine, hypoxanthine, and their corresponding ribonucleosides to the nucleotide level. Once a purine is salvaged, purine nucleotide interconversion enzymes then distribute the purine ring among the adenylate and guanylate nucleotide pools in order to meet the nucleotide requirements of the parasites. Although XPRT is sufficient for host purine acquisition, the ability of Dxprt mutants to grow in defined medium supplemented with most purines clearly demonstrates that the enzyme is not necessary for the salvage of all purines, i.e. all purines can be converted to the nucleotide level by XPRTindependent routes, except xanthine. The fact that Dxprt parasites can grow in guanosine and guanine was somewhat unexpected in view of radiolabel uptake data suggesting that guanine is mostly deaminated to xanthine prior to salvage. As

34410

Xanthine Phosphoribosyltransferase from L. donovani

the Dxprt line could still grow in guanosine- and guaninesupplemented medium, these growth experiments indicate that guanine deamination is sufficiently inefficient to permit adequate guanine phosphoribosylation through HGPRT. The presence of HGPRT, APRT, and adenosine kinase activities can account for the ability of Dxprt L. donovani promastigotes to proliferate in medium in which hypoxanthine, inosine, adenine, or adenosine is the only purine supplied. Clearly, it will be of great interest to attempt to introduce the Dxprt mutation into the pre-existing Dhgprt, Daprt, Dhgprt/Daprt L. donovani strains that have been constructed by targeted gene replacement (4), as this additional genetic complexity should permit a thorough genetic dissection of the purine salvage pathway in this parasite. The ability to create homozygous Dhgprt or Daprt null mutants by single targeted gene replacement followed by loss-of-heterozygosity (4) and the availability of a multiplicity of independent drug resistance markers for Leishmania (41) facilitates the construction of strains with multiple genetic lesions and makes this genetic analysis feasible. An increasing body of evidence has intimated that purine salvage is regulated in L. donovani and perhaps in other trypanosomatids (15, 42, 43). Data in this article demonstrate that XPRT activity is progressively up-regulated as a consequence of genetic impairments in purine salvage capability, i.e. there is a sequential increase in XPRT mRNA and XPRT protein and activity in Dhgprt and Dhgprt/Daprt parasites. Indeed the Dhgprt/Daprt strain in which XPRT expression is indispensable expresses an order of magnitude more XPRT transcript than the wild type line in which XPRT expression is nonessential. A parallel nutritional stress response is observed for HGPRT expression, as HGPRT mRNA and activity is augmented ;5-fold in parasites starved for purines (15). As Leishmania transcribe their genes as polycistronic transcripts (44), the proximal location of HGPRT and XPRT would intimate that they would be coordinately regulated by a common mechanism in response to conditions of nutritionally or genetically induced purine stress. The 39-nucleotidase/nuclease activity that presumably mobilizes host cell nucleic acids for parasite purine salvage is also elevated in a pronounced fashion in response to purine deprivation in L. donovani (43), as well as in the related mosquito parasite Crithidia fasciculata (42). How purine salvage is regulated in trypansomatids is completely unknown. The inability to generate purines de novo, the unique substrate specificity of XPRT, the fact that Dxprt parasites cannot utilize xanthine, and the up-regulation of XPRT in Dhgprt lines support the idea that HGPRT and XPRT are the principal routes for purine salvage in L. donovani and potential targets for the design of novel antiparasitic agents. One therapeutic paradigm would be a subversive substrate of either enzyme that is either not effectively recognized by the mammalian HGPRT, e.g. xanthine analogs, or not converted after phosphoribosylation to a cytotoxic nucleotide analog, e.g. pyrazolopyrimidine nucleobase analogs such as allopurinol (45). Another approach would be inhibitor development. However, the redundant nature of the purine pathway in L. donovani implies that this therapeutic strategy would require a combination of inhibitors specific for 6-oxopurine PRTs. However, the similar overall catalytic architecture thus far observed for all members of the HG(X)PRT family (46, 47) and the fact that HGPRT and XPRT accommodate structurally similar substrates and are presumed to catalyze phosphoribosylation through a common oxocarbonium intermediate (48) intimate that an inhibitor common to both enzymes, e.g. a high affinity transition state analog that binds selectively and irreversibly to the active sites of these PRTs, could be developed. The molecular and biochem-

ical reagents prerequisite for a rational and perhaps structurebased approach to the design and discovery of novel compounds that target HGPRT and XPRT, the central enzymes of purine salvage in L. donovani, are now in hand. REFERENCES 1. Grogl, M., Thomason, T. N., and Franke, E. D. (1992) Am. J. Trop. Med. Hyg. 47, 117–126 2. Berens, R. L., Krug, E. C., and Marr, J. J. (1995) in Biochemistry of Parasitic Organisms and Its Molecular Foundations (Marr, J. J. and Muller, M., eds) pp. 89 –117, Academic Press, New York 3. Tuttle, J. V., and Krenitsky, T. A. (1980) J. Biol. Chem. 255, 909 –916 4. Hwang, H.-Y., Gilberts, T., Jardim, A., Shih, S., and Ulman, B. (1996) J. Biol. Chem. 271, 30840 –30846 5. Hwang, H.-Y., and Ullman, B. (1997) J. Biol. Chem. 272, 19488 –19496 6. Jochimsen, B., Nygaard, P., and Vestergaard, T. (1975) Mol. Gen. Genet. 143, 85–91 7. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor NY 8. Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F., and Higgins, D. G. (1997) Nucleic Acids Res. 25, 4876 – 4882 9. Comeau, A. M., Miller, S. I., and Wirth, D. F. (1986) Mol. Biochem. Parasitol. 21, 161–169 10. Wilson, K., Hanson, S., Landfear, S. M., and Ullman, B. (1991) Nucleic Acids Res. 19, 5787 11. Miller, S. I., and Wirth, D. F. (1988) Mol. Cell. Biol. 8, 2597–2603 12. Cruz, A., Coburn, C. M., and Beverley, S. M. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 7170 –7174 13. Clayton, C., Adams, M., Almeida, R., Baltz, T., Barrett, M., Bastien, P., Belli, S., Beverley, S., Biteau, N., Blackwell, J., Blaineau, C., Boshart, M., Bringaud, F., Cross, G., C ruz, A., Degrave, W., Donelson, J., El-Sayed, N., Fu, G., Ersfeld, K., Gibson, W., Gull, K., Ivens, A., Kelly, J., Lawson, D., Lebowitz, J., Majiwa, P., Matthews, K., Melville, S., Merlin, G., Michels, P., Myler, P., Norrish, A., Opperdoes, F., Papadopoulou, B., Parsons, M., Seebeck, T., Smith, D., Stuart, S., Turner, M., Ullu, E., and Vanhamme, L. (1998) Mol. Biochem. Parasitol. 97, 221–224 14. Bradford, M. M. (1976) Anal. Biochem. 71, 248 –254 15. Allen, T. E., Hwang, H. Y., Jardim, A., Olafson, R. W., and Ullman, B. (1995) Mol. Biochem. Parasitol. 73, 133–143 16. Laemmli, U. K. (1970) Nature 227, 680 – 685 17. Towbin, H., Stahelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, 4350-4354 18. Langford, C. K., Ullman, B., and Landfear, S. M. (1992) Exp. Parasitol. 74, 360 –361 19. Allen, T. E., Hwang, H. Y., Jardim, A., and Ullman, B. (1995) Mol. Biochem. Parasitol. 74, 99 –103 20. Eads, J. C., Scapin, G., Xu, Y., Grubmeyer, C., and Sacchettini, J. C. (1994) Cell 78, 325–334 21. Jardim, A., and Ullman, B. (1997) J. Biol. Chem. 272, 8967– 8973 22. Hendrickson, N., Allen, T., and Ullman, B. (1993) Mol. Biochem. Parasitol. 59, 15–28 23. Fung, K., and Clayton, C. (1991) Mol. Biochem. Parasitol. 45, 261–264 24. Opperdoes, F. R., and Michels, P. A. (1993) Biochimie (Paris) 75, 231–234 25. Ablig, W., and Entian, K.-D. (1988) Gene (Amst.) 73, 141–152 26. Ullman, B., and Carter, D. (1997) Int. J. Parasitol. 27, 203–213 27. Mushegian, A. R., and Koonin, E. L. (1994) Protein Sci. 3, 1081–1088 28. Shih, S., Hwang, H. Y., Carter, D., Stenberg, P., and Ullman, B. (1998) J. Biol. Chem. 273, 1534 –1541 29. Giacomello, A., and Salerno, C. (1978) J. Biol. Chem. 253, 6038 – 6044 30. Mulligan, R. C., and Berg, P. (1981) Proc. Natl. Acad. Sci. U. S. A. 78, 2072–2076 31. Eells, J. T., and Spector, R. (1983) Neurochem. Res. 8,1451–1457 32. King, A., and Melton, D. W. (1987) Nucleic Acids Res. 15, 10469 –10481 33. Donald, R. G. K., Carter, D., Ullman, B., and Roos, D. S. (1996) J. Biol. Chem. 271, 14010 –14019 34. Beck, J. T., and Wang, C. C. (1993) Mol. Biochem. Parasitol. 60, 187–194 35. Wang, C. C., and Simashkevich, P. M. (1981) Proc. Natl. Acad. Sci. U. S. A. 78, 6618 – 6622 36. Sommer, J. M., Ma, H., and Wang, C. C. (1996) Mol. Biochem. Parasitol. 78, 185–193 37. Miller, R. L., Ramsey, G. A., Krenitsky, T. A., and Elion, G. B. (1972) Biochemistry 11, 4723– 4731 38. Lafon, S. W., Nelson, D. J., Berens, R. L., and Marr, J. J. (1982) Biochem. Pharmacol. 31, 231–238 39. Koszalka, G. W., and Krenitsky, T. A. (1979) J. Biol. Chem. 254, 8185– 8193 40. Hassan, H. F., and Coombs, G. H. (1985) Exp. Parasitol. 59, 139 –150 41. Freedman, D. J., and Beverley, S. M. (1993) Mol. Biochem. Parasitol. 62, 37– 44 42. Gottlieb, M. (1985) Science 227, 72–74 43. Sacchi, J. B., Jr., Campbell, T. A., and Gottlieb, M. (1990) Exp. Parasitol. 71, 158 –168 44. Lee, M. G., Atkinson, B. L., Giannini, S. H., and Van der Ploegh, L. H. (1988) Nucleic Acids Res. 16, 9567–9585 45. Marr, J. J. (1983) J. Cell. Biochem. 22, 187–196 46. Schumacher, M. A., Carter, D., Roos, D. S., Ullman, B., and Brennan, R. G. (1996) Nat. Struct. Biol. 3, 881– 887 47. Somoza, J. R., Chin, M. S., Focia, P. J., Wang, C. C., and Fletterick, R. J. (1996) Biochemistry 35, 7032–7040 48. Tao, W., Grubmeyer, C., and Blanchard, J. S. (1996) Biochemistry 35, 14 –21