HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 46, 36369-36379, November 17, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/46/36369    most recent M004036200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Yamage, M. Articles by Dwyer, D. M. Search for Related Content PubMed PubMed Citation Articles by Yamage, M. Articles by Dwyer, D. M. Social Bookmarking                      What's this?

Molecular Characterization of a Hyperinducible, Surface Membrane-anchored, Class I Nuclease of a Trypanosomatid Parasite^*

Mat Yamage, Alain Debrabant^§, and Dennis M. Dwyer^¶

From the Cell Biology Section, Laboratory of Parasitic Diseases, Division of Intramural Research, NIAID, National Institutes of Health, Bethesda, Maryland 20892-0425

Received for publication, May 11, 2000, and in revised form, August 11, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The 3'-nucleotidase/nuclease (3'-NT/NU) is a surface enzyme unique to trypanosomatid parasites. These organisms lack the pathway for de novo purine biosynthesis and thus are entirely dependent upon their hosts to supply this nutrient for their survival, growth, and multiplication. The 3'-NT/NU is involved in the salvage of preformed purines via the hydrolysis of either 3'-nucleotides or nucleic acids. In Crithidia luciliae, this enzyme is highly inducible. For example, in these organisms purine starvation triggers an ~1000-fold up-expression of 3'-NT/NU activity. In the present study, we cloned and characterized a gene encoding this intriguing enzyme from C. luciliae (Cl). Sequence analysis showed that the Cl 3'-NT/NU deduced protein possessed five regions, which we defined here as being characteristic of members of the class I nuclease family. Further, we demonstrated that the Cl 3'-NT/NU-expressed protein possessed both 3'-nucleotidase and nuclease activities. Moreover, we showed that the dramatic up-expression of 3'-NT/NU activity in response to purine starvation of C. luciliae was concomitant with the ~100-fold elevation in steady-state mRNA specific for this gene. Finally, results of our nuclear run-on analyses demonstrated that such up-regulation in 3'-NT/NU enzyme activity was mediated at the posttranscriptional level.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Trypanosomatid protozoa are a group of primitive parasitic organisms, which are of importance because many of them cause serious or fatal diseases in humans and domestic animals worldwide (54). Although purines are essential for the growth, multiplication and survival of these organisms, they are incapable of synthesizing the purine ring de novo. Thus, they are totally dependent upon their hosts to provide an exogenous source of preformed purines (1). In that regard, these purine auxotrophic organisms possess extracellular salvage mechanisms to acquire these essential nutrients from their host environments (2). One such trypanosomatid salvage enzyme is the externally oriented, bifunctional, surface membrane 3'-nucleotidase/nuclease (3'-NT/NU),¹ which is capable of generating free nucleosides via the hydrolysis of either 3'-nucleotides or nucleic acids (3, 4). Such 3'-NT/NU activities have been reported from a variety of different trypanosomatids (4, 5). Among these organisms, however, only the 3'-NT/NU of Crithidia luciliae was shown to be dramatically up-expressed by ~1000-fold in response to purine starvation conditions (6). Further, the C. luciliae 3'-NT/NU was shown to possess some biochemical and kinetic properties similar to those of the class I, single-strand-specific, nucleases of fungi and germinating plant seedlings (3, 7).

In light of these intriguing biochemical and physiological properties, experiments were designed in the current report to delineate the structure, function, and expression of the gene encoding this highly inducible enzyme of C. luciliae.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Reagents-- Reagents used in this study were purchased from Sigma unless otherwise specified.

Parasites-- Cultures of C. luciliae (ATCC strain no. 30285) (8) were grown at 26 °C in medium M199 (Life Technologies, Inc.), pH 6.8, as described previously (9) and supplemented here with 10% (v/v) heat-inactivated fetal bovine serum (Life Technologies, Inc.). A clone (MY₁₀₁) was isolated by limiting dilution from such cultures and used for all subsequent experiments. For testing the effects of adenosine depletion on C. luciliae, replete cells were grown and maintained in chemically defined medium, RPMI 1640+ (9) with 100 µM adenosine, and "adenosine-starved" cells were incubated in such medium without adenosine for various periods up to 72 h. In all experiments, log-phase cell cultures (2-4 × 10⁷/ml) were harvested by centrifugation as described previously (10). Both the cells and the resulting cell-free culture supernatants were processed for various purposes as described below. For some experiments, C. luciliae were treated with 0.5 µg/ml tunicamycin as described previously (11, 12).

Nomenclature-- The designations used in this report for genes, proteins, and plasmids follow the genetic nomenclature for Trypanosoma and Leishmania as outlined by Clayton et al. (13).

Oligonucleotide Primers, PCR, and Probe Preparation-- Degenerate oligonucleotide primers were designed based on the amino acid sequences conserved among the Leishmania donovani 3'-nucleotidase/nuclease (5), Aspergillus oryzae S1 nuclease (14), and Penicillium citrinum P1 nuclease (15) following a codon usage of C. luciliae RNA polymerase II (16). Such primers were synthesized by -cyanoethylphosphoramidite chemistry using an Expedite nucleic acid synthesis system (PE Applied Biosystems, Foster City, CA). Primer 1 (P1) (forward, 5'-GG(T CG)G A(CT) AT(C T)CA (AG)C CIC T(GC T)CA-3') and primer 2 (P2) (reverse, 5'-(TC)T T(CG A)GC IAG ICG GTA (GCA) CC-3') were used in PCR amplification of C. luciliae cDNA with Taq polymerase (Roche Molecular Biochemicals). Conditions for amplification were 94 °C for 30 s, 53 °C for 1 min, 72 °C for 2 min (36 cycles), and 74 °C for 10 min. The 436-nt product of this reaction was ligated to pCRII by TA cloning (Invitrogen, Carlsbad, CA), and the resulting plasmid was designated pCl-436. Both strands of this plasmid were sequenced and found to have high homology with that of the L. donovani 3'-NT/NU gene (5). The pCl-436 insert was labeled with digoxigenin-dUTP (Roche Molecular Biochemicals) by PCR according to manufacturer's instructions. The resulting digoxigenin-labeled probe (DIG-436) was used for screening the C. luciliae cosmid library and for other hybridization studies.

Cosmid Library Construction and Screening-- Genomic DNA was isolated from 10⁹ washed C. luciliae using the GNome DNA isolation kit (Bio101, Vista, CA). A C. luciliae genomic DNA library was constructed from EcoRV (Roche Molecular Biochemicals)-restricted genomic DNA, blunt end-ligated to a SuperCos I vector, phage-packaged and adsorbed onto host Escherichia coli XL1 Blue MR all according to manufacturer's instructions (Stratagene, La Jolla, CA). Over 3000 colonies from this library were screened by hybridization with the DIG-436 probe at high stringency (hybridization and washing in 0.1× SSC and 0.1% SDS at 65 °C). Several positive clones were obtained. A 2.8-kb NcoI fragment from one of these cosmid clones (CosMY1) was subcloned into PCRScript (Stratagene), and the resulting plasmid (pCl-3) was used for sequence analysis.

Nucleotide Sequencing and Analyses-- Plasmid DNA was sequenced using the fluorescent dideoxy-chain terminator method of cycle sequencing (Dye Terminator Cycle Sequencing Ready Reaction kit, PerkinElmer Life Sciences) on a model 373A Applied Biosystems automated DNA sequencer.

Sequence data from both strands were analyzed using the Wisconsin Package, version 10.0, Genetic Computer Group (GCG) software package running on the National Institutes of Health Unix System and Sequencher 3.0 software (Gene Codes Corp., Ann Arbor, MI). Protein multiple sequence alignments were done using the ClustalW program (17).

Southern Blot Analysis and Pulsed-field Gel Electrophoresis-- For Southern blotting, C. luciliae gDNA was digested with various restriction endonucleases (AccI, AvaI, BglI, and SalI, Roche Molecular Biochemicals), separated by 1% agarose gel electrophoresis, transferred under vacuum onto nylon membranes (HyBond-N, Amersham Pharmacia Biotech) and cross-linked by UV irradiation. DIG-labeled PCR products were generated using the pCl-3 insert above as template and the appropriate oligonucleotide primers specific to the 5'-flanking (DIG-300, which spans nt 980 to nt 664) and 3'-flanking (DIG-420, which spans nt +864 to nt +1291) regions of the Cl 3'-NT/NU gene, respectively. Membranes were processed for hybridization at high stringency with either the DIG-300- or DIG-420-labeled probes according to the Genius system users' guide (Roche Molecular Biochemicals), and the hybridized fragments were visualized using the Genius detection system (Roche Molecular Biochemicals).

Agarose plugs containing ~1 × 10⁸ washed C. luciliae or C. fasciculata were prepared for PFGE according to previously described methods (18). Chromosomes were separated by PFGE in a Bio-Rad DRII PFGE apparatus using conditions described by Joshi et al. (19). Gels were stained with ethidium bromide, blotted onto nylon membranes (Hybond-N, Amersham Pharmacia Biotech), and processed for Southern blot hybridization with the DIG-436 probe as above.

cDNA Analyses-- cDNA analyses were done in order to characterize the 5'-splice acceptor site and 3'-polyadenylation site of the Cl 3'-NT/NU gene. Since C. luciliae up-regulate their expression of 3'-nucleotidase activity in response to purine starvation (6), mRNA was isolated from such cells using Poly(A) Pure (Ambion, Austin, TX). cDNA was synthesized by reverse transcription of poly(A)⁺ RNA obtained from cells starved for adenosine for 24 h using SuperScript II (Life Technologies, Inc.) and an oligo(dT)₁₄ adaptor (5'-AGA AGA CGT AGG TTG ACT GCT GCA GTT TTT TTT TTT TTT-3'). Following RNase H (Roche Molecular Biochemicals) treatment, the resulting cDNA was used as template for PCR. In such reactions, the 5' and 3' ends of the Cl 3'-NT/NU cDNA were generated using the RACE method (20). For 5'-RACE (i.e. splice acceptor site), forward primer was 5'-AAC GCT ATA TAT AAG TAT CAG TTT C-3' and reverse was 5'-GAG AAC TCG TTG GCG TTG T-3'. For 3'-RACE (i.e. polyadenylation site), forward primer was 5'-ACC TAC GCC TCT ACG CTG A-3' and adaptor was 5'-AGA AGA CGT AGG TTG ACT G-3'. PCRs were carried out using conditions as above, and the resulting fragments were also cloned into pCRII (Invitrogen) as above. Subsequent sequencing of these fragments verified them as a part of the C. luciliae 3'-nucleotidase/nuclease gene.

Full-length Cl 3'-NT/NU cDNA was generated by XL-PCR (Roche Molecular Biochemicals) using the 5'-RACE forward and 3'-RACE adapter primers and the cDNA synthesized above. The resulting 2.2-kb PCR product was ligated into PCRScript (Stratagene). Subsequently, this cloned insert was sequenced. Those results showed that this cDNA clone possessed sequences identical to those obtained from the 5'- and 3'-RACE products, as well as the Cl 3'-NT/NU ORF present in the pCl-3 genomic clone, above.

Homologous Episomal Expression of a Truncated 3'-Nucleotidase/Nuclease-- A truncated construct of the Cl 3'-NT/NU gene was generated by PCR using pCl-3 as template and the appropriate oligonucleotide primers. The resulting fragment contained the Cl 3'-NT/NU 5'-UTR (nt 105 to nt 1) and a portion of the Cl 3'-NT/NU ORF (nt +1 to nt +1005). The resulting ~1.1-kb PCR fragment was digested with SpeI and ligated into the SpeI site of the [pKSNEO] leishmanial expression vector (21). The resulting plasmid [pKSNEO-Cl 3'-nt/nu^C] and the [pKSNEO] control plasmid were transfected into C. luciliae cells by electroporation essentially as described by Descoteaux et al. (22). Following an overnight recovery in medium M199+ (9) with 10% fetal bovine serum at 26 °C, these cells were selected for their growth under increasing concentrations of Geneticin (G-418, Life Technologies, Inc.) up to 200 µg/ml. In some experiments, C. luciliae were treated with 0.5 µg/ml tunicamycin as described previously (11, 12). Lysates and cell-free culture supernatants from these transfectants were used in various experiments.

3'-Nucleotidase and Nuclease Enzyme Assays-- 3'-Nucleotidase enzyme activity was measured in cell lysates and in cell-free culture supernatants in test tube assays, using adenosine 3'-monophosphate (3'-AMP) as substrate, as described previously (7). One unit of enzyme activity is defined as 1 nmol of P_i released from the 3'-AMP/min per ml of cell-free culture supernatant or per mg of cell lysate protein. Protein concentrations were determined using the bicinchoninic acid method (BCA, Pierce). Similar samples of C. luciliae were also assayed for total nuclease activity using poly(A) as substrate as described previously (23).

SDS-PAGE, in Situ Enzyme Activity Gels, and Western Blotting-- Cell-free culture supernatants and cell lysates of C. luciliae were separated by SDS-PAGE. These gels were processed for either in situ staining for 3'-nucleotidase activity (24, 7) or in situ staining for nuclease activity (25). Similar SDS-PAGE gels were processed for Western blot analysis with various rabbit antisera or their preimmune sera as described previously (5). One of these was a rabbit antiserum (no. 1336, anti-Ld3'-NT/NU-specific) generated against a single internal peptide of the L. donovani 3'-nucleotidase/nuclease (amino acid residues Glu²⁰¹-Tyr²²⁶) (5). A second antiserum (no. 1398) was generated in a New Zealand White rabbit against an E. coli-expressed Ld 3'-NT/NU protein, which lacked both the N-terminal 25-aa signal peptide and the C-terminal 44 aa residues of the Ld 3'-NT/NU. Preimmune sera from these rabbits were used as controls.

Northern Blot Analyses-- A DIG-labeled 1000-nt PCR product (DIG-1000), which spans nt 1-1005 of the Cl 3'-NT/NU ORF, was generated using the pCl-3 insert above as template and the appropriate oligonucleotide primers. An identical ³²P-labeled probe (³²P-1000) was similarly generated. These probes were used in Northern blot analyses to quantitate the steady-state levels of specific mRNA present in adenosine-replete and adenosine-starved cells. Total RNAs were isolated from both adenosine-replete and adenosine-starved cells using STAT60 (Tel-Test, Inc., Friendswood, TX). Equivalent amounts (20 µg) of these RNAs were separated by agarose gel electrophoresis (26), transferred onto nylon membranes (HyBond-N, Amersham Pharmacia Biotech), and cross-linked by UV irradiation. Blots were hybridized at high stringency with either the DIG-1000 probe using Genius detection system (Roche Molecular Biochemicals) as above or with the ³²P-1000 probe as described previously (26). Images obtained with the DIG-labeled probe were analyzed using the National Institutes of Health Image densitometry software package (available via the National Institutes of Health web site) and those obtained with the ³²P-1000 labeled probe were quantified by phosphorimaging analyses using a PhosphorImager (model SI, Molecular Dynamics, Sunnyvale, CA) driven by a Scanner Softwaremanager package (Molecular Dynamics). The above blots were subsequently rehybridized with either a 440-nt DIG- or ³²P-labeled fragment of a cloned constitutively expressed C. luciliae -tubulin gene ORF.²

Analysis of Nascent RNAs-- Nuclear run-on experiments were done with adenosine-replete and adenosine-starved cells to ascertain the rates at which they transcribed the Cl 3'-NT/NU gene. Nuclei were isolated from C. luciliae maintained for 24 h in either adenosine-replete or adenosine-starved medium using previously described methods (27). Run-on transcriptions with [-³²P]rUTP (Amersham Pharmacia Biotech) were performed as described previously (28) and similar experiments were also done using digoxigenin-UTP (Roche Molecular Biochemicals). Slot blots were prepared on HyBond-N (Amersham Pharmacia Biotech) membranes containing equivalent amounts of either a 436-bp internal DNA fragment (nt +446 to nt +881) of the Cl 3'-NT/NU gene, a similar sized (440 bp) internal fragment of the C. luciliae -tubulin gene above, or a control linearized pBluescript plasmid (Stratagene). Such blots were hybridized with either the ³²P-labeled or DIG-labeled nascent mRNAs, and the resulting signals were captured and analyzed as above.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Identification of the Cl 3'-NT/NU Gene-- A 436-bp DNA fragment was obtained by RT-PCR using cDNA (derived from cells starved for adenosine for 12 h) as a template and a pair of the degenerate oligonucleotide primers: P1 and P2 (see "Experimental Procedures"). This fragment was cloned, sequenced, DIG-labeled, and used as a probe (DIG-436) to screen a C. luciliae gDNA cosmid library. Of the several clones identified from such screening, one cosmid clone (Cos MY1) was chosen for further analysis. Following restriction digestion of Cos MY1, a 2.8-kb NcoI fragment was isolated and subcloned into PCRScript and the resulting plasmid, pCl-3, was used for sequence analysis. Both strands of the cloned NcoI fragment were sequenced. Results of sequence analysis showed that the pCl-3 plasmid contained a complete ORF of 1,134 bp (Cl 3'-NT/NU, Fig. 1). This ORF showed high nt sequence identity (69%) to the L. donovani 3'-nucleotidase/nuclease gene (5). Moreover, sequence analysis revealed that G or C was the preferred base used at the third codon position in the Cl 3'-NT/NU. Out of 378 codons including TAG stop in this ORF, nearly 90% contained a C or G at the third nucleotide position (186 end with C and 154 end with G). Further, the overall composition of the Cl 3'-NT/NU is very GC-rich (65.0%).

View larger version (66K): [in this window] [in a new window]   Fig. 1.   Nucleotide and deduced amino acid sequences of the C. luciliae 3'-nucleotidase/nuclease (Cl 3'-NT/NU). Nucleotide numbers are shown at the left; those constituting the 5'-UTR and 5'-flanking region of this gene are designated (). The ORF (nt+1 to nt+1134) is shown in uppercase letters, and the 5'- and 3'-flanking regions are shown in lowercase letters. The TAG stop codon of the ORF is denoted AMB. The P1 and P2 arrows designate the oligonucleotide primers used to generate the 436-bp PCR amplification product. The putative splice acceptor site ag (nt 106/107) in the 5'-UTR and the putative polyadenylation aataaa motifs in the 3'-UTR are shown in bold type. Putative CAAT and TATA boxes are underlined in the 5'-flanking region. The deduced amino acid sequence is shown in italics, and residues are numbered on the right. The 25-aa predicted signal peptide sequence is denoted with a dashed underline. The 25-aa putative trans-membrane-spanning anchor region (Gly336-Leu360) is underlined in bold. The C-terminal 17-aa residues constitute a putative cytoplasmic tail. Asterisks indicate the two predicted N-glycosylation sites (Asn240 and Asn294).

The relationship between the pCl-3 gDNA clone, a full-length cDNA Cl 3'-NT/NU clone and four DIG-labeled probes (i.e. DIG-300, -420, -436, and -1000) generated from these clones is shown diagrammatically in Fig. 2A. The combined results of restriction digestion of these two clones are summarized as a restriction map in Fig. 2B. It is of importance to note that, with only a few exceptions (29), trypanosomatid protozoa generally do not possess introns within their ORFs (30). Further, all pre-mRNAs in these organisms are joined to a 39-nt conserved spliced leader at their 5' end by trans-splicing (31, 32). To further characterize the organization of the Cl 3'-NT/NU, we compared the sequence of our genomic pCl-3 clone with that of the full-length cDNA clone. Results of these analyses revealed that the first in-frame start codon, ATG, was preceded by 105 bp of 5'-UTR (Figs. 1 and 2B). The PyAG spliced-leader acceptor site was mapped to 106 bp from the start codon (Figs. 1 and 2B). As shown in Fig. 1, regions upstream of this site contained the sequences: CCTTGAC (220 to 214), TGTTGAC (156 to 150), and TTTCAAC (125 to 119), which conform to the PyNPyPyPuAPy consensus (33). In the latter, nt A is the potential target for lariat formation during the pre-mRNA trans-splicing events (34). Further, short segments typical of canonical CAAT and TATA boxes, which are present in higher eukaryotes, but remain to be defined in trypanosomatid protozoa, were also found in the 5'-UTR of the Cl 3'-NT/NU gene at 421 and 384 bp, respectively (Fig. 1). Similar analyses of the region downstream from the 3' end of the Cl 3'-NT/NU ORF revealed three putative polyadenylation signals (AATAAA) at nt 2,062, 2,131, and 2,135 (Figs. 1 and 2B). These signals could direct a poly(A) tail to be added downstream of one of these sites.

View larger version (34K): [in this window] [in a new window]   Fig. 2.   Characterization of the Cl 3'-NT/NU gene and deduced protein. A, diagrammatic representations of the pCl-3 gDNA clone (2.8-kb NcoI fragment, solid bar) and the full-length cDNA Cl 3'-NT/NU clone (bold dashed bar) used for sequencing. The spatial relationships of the digoxigenin-labeled oligonucleotide probes (DIG-300, -420, -436, and -1000; dashed bars) generated from these clones are shown. B, map of the Cl 3'-NT/NU locus. Restriction endonuclease sites (Ac, AccI; Av, AvaI; Bg, BglI; Na, NaeI; Nc, NcoI; Sa, SalI; Sm, SmaI) and the start (ATG) and stop (TAG) codons of the ORF (hatched box) are shown. The spliced leader acceptor site (downward arrow, nt 105) in the 5'-UTR and the three putative polyadenylation sites (upward arrows) in the 3'-UTR regions of this gene are shown. C and D, Southern blots of C. luciliae gDNA digested with AccI, AvaI, BglI, and SalI and hybridized with the DIG-300 probe (C) or the DIG-420 probe (D). Molecular size markers (kb) are shown on the right of each figure. E, Crithidia chromosomal DNA separated by PFGE and stained with ethidium bromide: lane M, yeast chromosome size markers; lane 1, C. luciliae; lane 2, C. fasciculata; lanes 3 and 4, Southern blots of these gels hybridized with the DIG-436 probe. F, schematic representation of the Cl 3'-NT/NU deduced protein. Amino acid residues are denoted by the numeric scale; the putative 25-aa N-terminal signal peptide (hatched area, SP) and predicted 25-aa trans-membrane domain (hatched area, TM) are shown. Arrowheads mark the two potential N-linked glycosylation sites (Asp240 and Asp294). The widely hatched areas (I-V) represent blocks of aa residues conserved among proteins related to the Cl 3'-NT/NU. G, Kyte-Doolittle hydropathy plot of the Cl 3'-NT/NU deduced protein. The overall hydrophobicity of both the putative N-terminal signal peptide and C-terminal trans-membrane anchor domains of this protein are shown. The C-terminal 17-aa residue cytoplasmic tail region represents the most hydrophilic portion of this deduced protein.

Characterization of the Cl 3'-NT/NU Locus and Chromosomal Localization-- To assess the structure and copy number of the Cl 3'-NT/NU, genomic DNA was digested with several restriction endonucleases and subjected to Southern hybridization with the DIG-labeled probes shown in Fig. 2A. These enzymes were chosen because they possessed at least one predicted restriction site within the Cl 3'-NT/NU ORF (Fig. 2B). Results of Southern hybridizations obtained with the DIG-300 probe (i.e. the 5'-flanking region of pCl-3) and with the DIG-420 probe (i.e. the 3'-flanking region of pCl-3) are shown in Fig. 2 (C and D), respectively. The cumulative results of these analyses indicated that at least three copies of the Cl 3'-NT/NU gene are present within the diploid genome of C. luciliae.

To identify the physical location(s) of the Cl 3'-NT/NU genes, chromosomal DNA from C. luciliae and C. fasciculata was separated by pulsed-field gel electrophoresis and stained with ethidium bromide (Fig. 2E, lanes 1 and 2). These gels were blotted onto nylon membranes and hybridized with the Cl 3'-NT/NU DIG-436 probe above. Results of these assays demonstrated that the Cl 3'-NT/NU genes were all apparently localized to a single >1.6-megabase chromosome in C. luciliae (Fig. 2E, lane 3) and to a similarly sized chromosome in Crithidia fasciculata (Fig. 2E, lane 4). The latter is in agreement with a previous observations which demonstrated that C. fasciculata possess 3'-nucleotidase and nuclease activities (35). Taken together, these results indicate that the 3'-NT/NU gene is conserved among these closely related species of trypanosomatid parasites.

Characterization of the Cl 3'-NT/NU Deduced Protein-- The Cl 3'-NT/NU open reading frame of 1,134 bp encodes a deduced protein of 377 amino acids with a calculated mass of 40,575 Da (Fig. 1). This deduced protein is acidic (~16% acidic aa residues) and has a predicted isoelectric point (pI) of 5.63, which is in agreement with the pI of ~5.8 as determined previously by two-dimensional electrophoresis of the native C. luciliae enzyme (7). Hydropathy analysis using the Kyte and Doolittle algorithm (46) revealed that the Cl 3'-NT/NU deduced protein possessed a hydrophobic domain at its N terminus (Fig. 2G). Based on the von Heijne algorithm (36), the N-terminal 25 aa residues (Met¹-Ala²⁵) represent a putative signal peptide sequence (Figs. 1 and 2F). Cleavage between aa residue 25 and 26, presumably in the endoplasmic reticulum, would result in a mature protein with Trp²⁶ as the N-terminal residue and a calculated molecular mass of 37,968 Da.

GAP analysis (37) demonstrated that the Cl 3'-NT/NU deduced protein had a 69% identity to that of the L. donovani 3'-nucleotidase/nuclease (GenBank^TM accession no. L35078) (5). In addition, results of BLAST (38) analyses showed that the Cl 3'-NT/NU also had significant aa homologies with other class I nucleases (39) and related proteins with putative nuclease functions from a variety of sources including: the endonuclease S1 homolog of Mesorhizobium loti (GenBank^TM AF049243) (40), senescence-associated protein 6 (SA6) of Hemerocallis hybrid cultivar (GenBank^TM AF082031), bifunctional nucleases (nucZe1 and nucZe2) of Zinnia elegans (GenBank^TM U90265 and U90266), the ZEN1 endonuclease of Zinnia elegans (GenBank^TM AB003131) (41), the bfn1 bifunctional nuclease of Arabidopsis thaliana (GenBank^TM U90264), BEN1 nuclease of Hordeum vulgare (NBRF Protein Data Base T04401) (41), the P1 nuclease of P. citrinum (Swiss-Prot P24289) (15), which is identical to endonuclease PA3 of Penicillium sp. (NBRF Protein Data Base JE0408) (42), and the S1 nuclease of A. oryzae (NBRF Protein Data Base JX0180, Refs. 14 and 43). Alignment of the Cl 3'-NT/NU sequence with the above proteins was done using the ClustalW multiple sequence alignment program (17), and the results are shown in Fig. 3. Results of such analyses revealed that this group of proteins contained five highly conserved blocks of aa residues designated here as domain I (Trp-X₃-Gly-His-X₃-(Ala-Cys)-X-Ile-Ala-Gln), domain II (Trp-Ala-Asp-X₂-(Arg-Lys)-X_6,8-Ser-X₂- His-Phe-Ile-X-Thr-Pro), domain III (Leu-X₂-Leu-X-His-Phe-(Met-Thr-Val-Ile)-Gly-Asp-Ile-His-Gln-Pro-Leu-His), domain IV (Asp-X-Gly-Gly-Asn-X₃-Val-X_n-(Lys-Phe-His-Thr)-X₂-Leu-His-X₂-Trp-Asp), and domain V (Glu-X₃-Leu-(Ala-Pro-Val)-X_4,5-Tyr-X-Gly-X_n-Gly-X-Thr-Leu-X₃-Tyr-X₆-(Ile-Val)-X₃-(Arg-Gln)-(Ile-Val-Leu)-X₂-Gly-Gly-X-Arg-Leu-Ala-X₂-Leu-Asn). The location of these domains within the Cl 3'-NT/NU deduced protein is shown in Figs. 2F and 3. It is noteworthy that four out of the five conserved domains contained one or more His residues that could be involved in the binding of divalent metal co-factors. Further, the nine conserved residues, underlined in domains I-IV above, have been shown to be responsible for the coordinate binding of zinc ions by the P1 nuclease of P. citrinum, the prototype member of this class of enzymes (44). These observations are of relevance as the native C. luciliae 3'-NT/NU was shown previously to require Zn²⁺ as a co-factor (7). Similar observations pertain to the native L. donovani 3'-NT/NU (45).

View larger version (112K): [in this window] [in a new window]   Fig. 3.   Alignment of the deduced aa sequence of the C. luciliae 3'-nucleotidase/nuclease (Cl 3'-NT/NU, GenBankTM AF140355) with related proteins from L. donovani (3'-nucleotidase/nuclease (Ld 3'-NT/NU, GenBankTM L35078); M. loti (S1 nuclease homolog: M.loti-S1 nuc, GenBankTM AF049243); H. hybrid cultivar (senescence-associated protein 6: H.h cult-SA6, GenBankTM AF082031); Z. elegans (bifunctional nuclease nucZe1: Z.eleg-bif.nuc, GenBankTM U90265); Z. elegans (ZEN1 endonuclease: Z.eleg-endonuc, GenBankTM AB003131); A. thaliana (bfn1 bifunctional nuclease: A.thal-bif.nuc1, GenBankTM U90264); H. vulgare (BEN1 nuclease: H.vulga-endonuc, GenBankTM D83178); P. citrinum (P1 nuclease: P.cit.Nuc P1, Swiss-Prot P24289); and A. oryzae (S1 nuclease: A.oryz-Nuc S1, NBRF Protein Data Base JX0180). This alignment was constructed using the ClustalW program. Gaps introduced to maximize alignments are indicated by dashes. Identical aa are shown in reverse-shaded boxes. Conserved aa substitutions are marked in shaded areas. Numbers reflect the sequence of aa residues within the C. luciliae 3'-NT/NU deduced protein. The domains conserved among these proteins are numbered I-V and underlined with a shaded bar.

Some members of the class I nuclease family above have been shown to possess two intrachain disulfide bridges (e.g. Cys⁸⁰-Cys⁸⁵ and Cys⁷²-Cys²¹⁶ in the P1 and S1 nuclease), which are necessary for the function of these enzymes (14, 15). The absence of such disulfide bridges from both the Cl 3'-NT/NU and the Ld 3'-NT/NU might account for the known resistance of these two enzymes to inactivation by disulfide reducing reagents (i.e. dithiothreitol, cysteine, and 2-mercaptoethanol; Refs. 7 and 24).

Analysis of the Cl 3'-NT/NU deduced protein sequence using the MOTIF program (available via the worldwide web, from the Institute for Chemical Research, Kyoto University, Kyoto, Japan) revealed that it contained eight potential sites for casein kinase II phosphorylation, three potential sites for protein kinase C phosphorylation and eight potential sites for N-myristoylation. In addition, the Cl 3'-NT/NU deduced protein contained two potential sites for N-linked glycosylation, i.e. Asn²⁴⁰ and Asn²⁹⁴ (Figs. 1 and 2F). This is in agreement with previous observations, which demonstrated that the native C. luciliae 3'-NT/NU was a N-linked glycoprotein with an apparent molecular mass of ~43 kDa as determined by SDS-PAGE (7). Presumably N-linked glycosylation could account for the difference between the calculated mass of the Cl 3'-NT/NU deduced protein and that of the native protein. Results of hydropathy analysis (46) suggested that the 25-aa hydrophobic region (Gly³³⁶-Leu³⁶⁰) in the C terminus of the Cl 3'-NT/NU deduced protein could function as a trans-membrane domain anchoring this protein in the parasite surface membrane (Fig. 2, G and F). This would result in the 17-aa hydrophilic C terminus of the deduced protein being exposed to the cytosol. These observations are in agreement with the known cell surface membrane localization of this enzyme (6, 23).

Episomal Expression of the Cl 3'-NT/NU and Characterization of the Expressed Protein-- To demonstrate that the Cl 3'-NT/NU gene encoded a functional 3'-nucleotidase/nuclease, a truncated Cl 3'-NT/NU gene was ligated it into the [pKSNEO] leishmanial expression vector (21). This construct contained a portion of the 5'-UTR (from nt 105 to nt 1) and a portion of the Cl 3'-NT/NU ORF (nt +1 to nt +1005). Thus, this construct lacked the C-terminal aa region (Gly³³⁶-Val³⁷⁷) encoding both the hydrophobic putative trans-membrane domain (Gly³³⁶-Leu³⁶⁰) and the hydrophilic cytoplasmic tail (His³⁶¹-Val³⁷⁷) of the Cl 3'-NT/NU. Both this plasmid construct [pKSNEO-Cl 3'-nt/nu^C] and the [pKSNEO] control plasmid were transfected into C. luciliae, and these cells were selected for growth under increasing G-418 concentrations up to 200 µg/ml. In order to induce their endogenous 3'-NT/NU activities for visualization in gels (below), log-phase cells were harvested, washed, resuspended in adenosine-starved medium, and incubated at 26 °C for 12 h prior to assay. Such cells were harvested, washed, lysed, and subjected to SDS-PAGE. These gels were stained in situ for 3'-nucleotidase and nuclease activities, respectively. Results of these assays showed that parasites transfected with the control [pKSNEO] plasmid contained only a single 43-kDa band of 3'-nucleotidase activity corresponding to the endogenous surface membrane enzyme activity (Fig. 4A, lane 1). Parasites transfected with the [pKSNEO- Cl 3'-nt/nu^C] plasmid showed a similar 43-kDa band of endogenous 3'-nucleotidase activity. In addition, they also showed a 38-kDa band of enzyme activity corresponding to the truncated Cl 3'-NT/NU-expressed protein (Cl 3'-nt/nu^C) encoded by the [pKSNEO-Cl 3'-nt/nu^C] plasmid (Fig. 4A, lane 2). The difference in apparent molecular mass between the native 43-kDa 3'-NT/NU and the 38-kDa expressed enzyme is due to the absence of the 45 C-terminal amino acid residues in the latter (i.e. reflecting a calculated reduction in molecular mass of 4,287 daltons). In parallel, results obtained from similar gels stained for nuclease activity showed that lysates of both control and [pKSNEO-Cl 3'-nt/nu^C] transfectants contained a 43-kDa band of nuclease activity corresponding to the endogenous Cl 3'-NT/NU enzyme activity (Fig. 4B, lanes 1 and 2). Moreover, lysates of parasites transfected with the [pKSNEO-Cl 3'-nt/nu^C] plasmid also showed a 38-kDa band of nuclease activity corresponding to the Cl 3'-nt/nu^C-expressed protein (Fig. 4B, lane 2). These results demonstrated that the Cl 3'-nt/nu^C construct encoded the bifunctional enzymatic activities of the native Cl 3'-NT/NU. Further, they showed that the C-terminal putative trans-membrane domain and hydrophilic tail region of the Cl 3'-NT/NU were not required for either the 3'-nucleotidase or nuclease activities of this protein. Moreover, it was hypothesized that episomal expression of Cl 3'-nt/nu^C would result in the production of a soluble (non-membrane-bound) Cl 3'-nt/nu^C being released/secreted from transfected parasites into their culture supernatants. To address the latter, C. luciliae transfected with control [pKSNEO] and [pKSNEO-Cl 3'-nt/nu^C] plasmids were grown to log phase in replete medium and their culture supernatants were harvested and assayed for both 3'-nucleotidase and nuclease activities. Results of spectrophotometric assays showed that only parasites transfected with the [pKSNEO-Cl 3'-nt/nu^C] plasmid released 3'-nucleotidase activity into their culture supernatant (data not shown). Similarly, only [pKSNEO-Cl 3'-nt/nu^C] transfectants were found to release nuclease activity into their culture supernatants (data not shown). Aliquots of such culture supernatants were separated by SDS-PAGE and stained in situ for 3'-nucleotidase and nuclease activities, respectively. Results of such assays showed that culture supernatants from C. luciliae transfected with the [pKSNEO] control plasmid had no detectable 3'-nucleotidase or nuclease activity (Fig. 4, A and B, lanes 3). In contrast, culture supernatants from cells transfected with the [pKSNEO-Cl 3'-nt/nu^C] plasmid contained a 38-kDa band of both 3'-nucleotidase (Fig. 4A, lane 4) and nuclease activity (Fig. 4B, lane 4). Release of the soluble Cl 3'-nt/nu^C-expressed protein into the culture supernatants of these transfectants presumably occurs via default into the parasite secretory pathway. The latter results confirmed that the C-terminal region of the native Cl 3'-NT/NU functions as a membrane anchor domain, but it is not necessary for either the nucleotidase or nuclease activities of this enzyme. Further, results of enzyme assays with culture supernatants from [pKSNEO-Cl 3'-nt/nu^C] transfectants demonstrated that the soluble expressed Cl 3'-nt/nu^C had K_m of 0.36 mM with 3'-AMP as substrate (data not shown). The latter is in close agreement with that reported previously (0.4 mM) for the native C. luciliae 3'-nucleotidase (7). Taken together, these results demonstrated for the first time that the Cl 3'-NT/NU gene encodes a bifunctional protein, which has both 3'-nucleotidase and nuclease activities.

View larger version (50K): [in this window] [in a new window]   Fig. 4.   SDS-PAGE gels stained in situ for 3'-nucleotidase (A) and nuclease (B) activities using 3'-AMP and poly(A) as substrates, respectively. Cell lysates (CL) of C. luciliae transfected with the control [pKSNEO] plasmid (C), or with the [pKSNEO-Cl 3'-nt/nuC] plasmid (T), are shown in lanes 1 and 2, culture supernatants (SN) from each of these cell types (C or T) are shown in lanes 3 and 4, respectively. Arrows indicate the 43-kDa endogenous and 38-kDa Cl 3'-nt/nuC-expressed 3'-nucleotidase/nuclease activities. Molecular masses in kDa of protein standards are shown at the left of this figure.

To determine whether N-linked glycosylation contributed to either the apparent molecular mass or the enzymatic activities of the Cl 3'-NT/NU, C. luciliae transfected with either control [pKSNEO] or [pKSNEO-Cl 3'-nt/nu^C] plasmids were subjected to treatment with tunicamycin. Lysates of untreated and tunicamycin-treated cells were separated by SDS-PAGE, and these gels were stained in situ for 3'-nucleotidase and nuclease activities. Results of such assays showed that in tunicamycin-treated cells both the endogenous 43-kDa Cl 3'-NT/NU and the 38-kDa Cl 3'-nt/nu^C-expressed enzymes were reduced in apparent molecular mass by ~4 kDa.; however, these proteins retained both their 3'-nucleotidase and nuclease activities, respectively (data not shown). Similar in situ gel activity assays were done with culture supernatants from untreated and tunicamycin-treated [pKSNEO-Cl 3'-nt/nu^C]-transfected C. luciliae. Results of such assays showed that the soluble Cl 3'-nt/nu^C-expressed enzyme released/secreted by these tunicamycin treated cells was reduced by ~4 kDa as compared with untreated controls and that it retained its 3'-nucleotidase activity (Fig. 5A). Identical results were obtained with these samples in gels stained for nuclease activity (data not shown). Cumulatively, these results demonstrate that N-linked glycosylation contributes to the molecular mass of the C. luciliae 3'-nucleotidase/nuclease, but it is not required per se for the activities of this bifunctional enzyme.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   A, SDS-PAGE gel showing the 3'-nucleotidase activities present in culture supernatants from untreated () and tunicamycin-treated (+)[pKSNEO-Cl 3'-nt/nuC]-transfected C. luciliae. Arrows designate the 38-kDa glycosylated and 34-kDa unglycosylated Cl 3'-nt/nuC enzymes secreted/released by these cells. B and C, Western blots of culture supernatants from C. luciliae transfected with the control [pKS NEO] vector (V), the [pKSNEO-Cl 3'-nt/nuC] plasmid (Cl), or a plasmid[pKS NEO Ld3'-nt/nus] encoding the Ld 3'-nt/nus protein of L. donovani (Ld). Blots were probed with either a rabbit antibody (1398) against an E. coli-expressed Ld 3'-nt/nus protein (B) or a rabbit antibody (1336) against a single internal peptide of the L. donovani 3'-NT/NU (C). Arrows designate the 38-kDa expressed proteins recognized by these antibodies.

Having shown that the C. luciliae 3'-NT/NU gene encoded a protein that had both 3'-nucleotidase and nuclease activities, it was of interest to ascertain whether it possessed any conserved antigenic epitopes with that of a recently characterized homologous enzyme from L. donovani. To that end, culture supernatants from C. luciliae transfected with [pKS NEOPT3'], [pKSNEO-Cl 3'-nt/nu^C] or a plasmid [pKS NEO Ld3'-nt/nu^s] encoding a truncated L. donovani secreted 3'-nucleotidase/nuclease (Ld 3'-nt/nu^s)³ were separated by SDS-PAGE, transblotted onto nylon membranes, and used for Western blot assays. Such blots were probed with several different antibodies. These included a rabbit antibody generated against an E. coli expressed Ld 3'-nt/nu^s protein (rabbit no. 1398), a rabbit anti-peptide antibody generated against a single internal 26-aa peptide (shown in Fig. 3 as Ld3'-NT/NU residues Glu²⁰²-Tyr²²⁷) of the L. donovani 3'-NT/NU (rabbit no. 1336) (5), and preimmune sera from these rabbits. None of the preimmune sera showed reactivity in Western blots with culture supernatants from any of the C. luciliae transfectants (data not shown). Similarly, neither rabbit 1398 nor rabbit 1336 showed any reactivity with culture supernatants from C. luciliae transfected with the control [pKS NEO] vector alone (lane V in Fig. 5 (B and C), respectively). However, both rabbit 1398 and 1336 reacted with a single 38-kDa protein present in the culture supernatants of C. luciliae transfected with the plasmid [pKS NEO Ld3'-nt/nus] encoding the Ld 3'-nt/nu^s protein (lane Ld in Fig. 5 (B and C), respectively). Identical results were obtained in blots with these antibodies and culture supernatants from L. donovani transfected with the homologous [pKS NEO Ld3'-nt/nu^s] construct (data not shown). These results demonstrated that C. luciliae could express a heterologous gene product which still possessed antigenic reactivity with both the anti-Ld 3'-nt/nu^s-expressed protein (rabbit 1398) and anti-Ld 3'-NT/NU peptide (rabbit 1336) specific antibodies. In contrast, the 38-kDa soluble/secreted Cl 3'-nt/nu^C protein expressed by C. luciliae [pKSNEO-Cl 3'-nt/nu^C] transfectants was only recognized in Western blots by rabbit 1398 (Fig. 5B, lane Cl) and not by rabbit 1336 (Fig. 5C, lane Cl). The results obtained with the No.1398 antibody demonstrated that the soluble Cl 3'-nt/nu^C-expressed protein possessed at least some antigenic epitopes in common with those of the heterologous Ld 3'-NT/NU protein. Presumably some of those epitopes must reflect linear sequences of aa identity between these two proteins (cf. Fig. 3). Conversely, the low level of conserved residues from aa 202 to aa 227 between the Ld 3'-NT/NU and the Cl 3'-NT/NU (cf. Fig. 3) accounts for the lack of reactivity of anti-Ld 3'-NT/NU peptide (1336) antibody with the Cl 3'-nt/nu^C protein.

Up-expression of Cl 3'-NT/NU Enzyme Activity and mRNA during Purine Starvation-- Previously, it was shown that purine starvation of C. luciliae resulted in a significant increase in their expression of 3'-nucleotidase enzyme activity (6). To ascertain whether Cl 3'-NT/NU mRNA levels were also altered in cells starved for purines, Northern blot analyses were performed using total RNA isolated from log-phase C. luciliae maintained under either replete or adenosine-starved conditions for 24 h (see below). Results from these blots showed that our Cl 3'-NT/NU DIG-labeled gene-specific probe (DIG-1000) hybridized to a single ~2.3-kb mRNA from both adenosine-replete and adenosine-starved cells (Fig. 6A, lanes 1 and 2). However, the signal obtained from adenosine-starved cells was significantly stronger than that from adenosine-replete cells, suggesting that they possessed higher levels of Cl 3'-NT/NU mRNA. To address that observation, time-course experiments were done to correlate the changes in Cl 3'-NT/NU-specific enzyme activity with Cl 3'-NT/NU-specific mRNA levels in cells starved for purines for various periods. To that end, C. luciliae cells were grown to log phase (~2 × 10⁷cells/ml) in chemically defined replete medium, harvested by centrifugation, washed once in such medium lacking adenosine by centrifugation, and resuspended to the same cell density in the latter medium. Such cells were incubated at 26 °C in medium lacking adenosine for various periods from T₀ up to 72 h (T₇₂). Control cell populations were handled identically except that they were maintained in replete medium throughout the course of the experiments. The 3'-nucleotidase activity of replete or adenosine-starved cells was quantitated colorimetrically using 3'-AMP as substrate. Results of these assays showed that cells maintained from 24 to 72 h in adenosine-starved medium expressed highly elevated levels (>1000-fold) of 3'-nucleotidase-specific activity compared with T₀ cells (Fig. 6B, top panel) or those continuously maintained under replete conditions (data not shown). Aliquots of the above cell samples were also separated by SDS-PAGE and stained in situ for 3'-nucleotidase activity. Results of those assays showed that cells maintained for 24 h or more under adenosine-starved conditions expressed a single strong 43-kDa band of 3'-nucleotidase activity compared with adenosine-starved cells at T₀ (Fig. 6B, lower panels) or those maintained for similar periods under replete conditions (data not shown). Results identical or very similar to those shown in Fig. 6B were also obtained from multiple other independent experiments. Thus, the currents results showed that C. luciliae expressed up to a 1000-fold higher levels of 3'-nucleotidase-specific activity in response to adenosine depletion and are in agreement with those reported previously by Neubert and Gottlieb (7).

View larger version (23K): [in this window] [in a new window]   Fig. 6.   Analyses of Cl 3'-NT/NU enzyme activity and mRNA during purine starvation. A, Northern blot showing hybridization of the DIG-labeled Cl 3'-NT/NU gene-specific (DIG-1000) probe with total RNA (20 µg) isolated from C. luciliae maintained under adenosine-replete or adenosine-starved conditions for 24 h (lanes 1 and 2, respectively). Arrow indicates the 2.3-kb Cl 3'-NT/NU-specific mRNA present in each cell type. Molecular mass markers are shown in kb at the left. B, analyses of Cl 3'-nucleotidase (3'-NT) activity in lysates of cells starved for adenosine for 0, 24, 48, and 72 h, respectively. Top panel shows the 3'-NT-specific activities (nmol/min/mg protein) in these samples. Results shown reflect the mean values obtained from triplicate samples from three independent experiments using 3'-AMP as substrate. Vertical bars indicate the standard error of each mean. Bottom panel shows the 43-kDa band of 3'-NT activity (arrow) present in these cell samples as revealed by in situ staining of SDS-PAGE gels. C, Northern blots showing hybridization of the DIG-labeled Cl 3'-NT/NU (DIG-1000) probe (middle panel, 3'-NT) with total RNA (20 µg) isolated from C. luciliae maintained under adenosine-starved conditions from 0 to 72 h. Arrow shows the single 2.3-kb specific Cl 3'-NT/NU message present in these samples. Lower panel shows the hybridization of the above RNA samples with a DIG-labeled -tubulin probe (Tubulin). Arrow shows the single 2.0-kb specific Cl -tubulin message present in these samples. Upper panel, quantitation of the Northern blot images obtained with the DIG-labeled probes. Values shown reflect the -fold increase in signal intensity between those obtained from adenosine-starved cells at 24, 48, and 72 h, respectively and that obtained from adenosine-starved cells at T0. Solid bars represent the signals obtained using the 3'-NT probe (middle panel) and the open bars signals obtained with the tubulin probe (lower panel).

Total RNA was also isolated from the replete or adenosine-starved cell samples above and used for Northern blot analyses. Equivalent amounts of total RNA (20 µg) from both cell types were separated in 1% agarose gels, blotted onto nylon membranes, and hybridized with the Cl 3'-NT/NU DIG-labeled gene-specific probe (DIG-1000). These Northern blot assays showed that cells continuously maintained (up to 72 h) under replete conditions characteristically had only a very low level signal for the single ~2.3-kb steady-state Cl 3'-NT/NU mRNA. This observation similarly applies to the results obtained with this probe and adenosine-starved cells at T₀ (Fig. 6C, 3'-NT probe, T₀). In contrast, cells maintained in adenosine-starved medium for 24 or more hours contained significantly higher levels of this single 2.3-kb 3'-NT/NU-specific message than adenosine-starved cells at T₀ (cf. Fig. 6C, 3'-NT probe, at 24, 48, and 72 h). As a control in these experiments, duplicate blots of the above RNAs were also hybridized with a DIG-labeled Cl -tubulin-specific gene probe. Results of those assays showed that both replete (data not shown) and adenosine-starved cells constitutively expressed very similar levels of a single ~2-kb -tubulin-specific mRNA over the time course of these experiments (Fig. 6C, -tubulin probe, 0-72 h). Northern blot images obtained with these DIG-labeled probes were quantitated using the National Institutes of Health Image software package (available from the National Institutes of Health web site). Results of such quantitation demonstrated that cells starved for adenosine for 24 or more hours expressed ~100-fold higher levels of steady-state Cl 3'-NT/NU mRNA than adenosine-starved cells at T₀ (Fig. 6C, top panel, solid bars) or those maintained under replete conditions throughout the experiment (data not shown). In contrast, over the time course of these experiments, adenosine starvation had no significant effect on the expression of -tubulin-specific mRNA by these cells (Fig. 6C, top panel, open bars). Results of these assays were confirmed using quantitative phosphorimaging analyses of identical blots hybridized with ³²P-labeled Cl 3'-NT/NU (³²P-1000) and ³²P-labeled Cl -tubulin probes. Taken together, results of these experiments demonstrated that the Cl 3'-NT/NU mRNA was specifically up-regulated in response to adenosine starvation and was concomitant with the up-expression of 3'-NT/NU enzyme-specific activity.

The differences observed in steady-state levels of Cl 3'-NT/NU mRNA between adenosine-replete and adenosine-starved cells might indicate their differential rates of transcription of this gene or reflect their differential post-transcriptional regulation of this message. To address this, nuclear run-on experiments were performed using DIG-labeled nascent mRNA from both cell types. These labeled mRNAs were hybridized with slot blots containing equivalent amounts of either a 436-bp internal DNA fragment of the Cl 3'-NT/NU gene (see Fig. 2A), a similarly sized 440-bp internal fragment of the Cl -tubulin ORF or linearized pBluescript plasmid. Images obtained from these assays were quantitated using the National Institutes of Health Image software package as above. Results of these analyses showed that both replete cells and adenosine-starved cells (i.e. starved for 24 h) contained very similar levels of nascent Cl 3'-NT/NU mRNA, indicating that their rates of transcription for this gene were virtually identical (Fig. 7A). Similar results were obtained regarding the rates of transcription for the tubulin gene in these replete and adenosine-starved cells (Fig. 7B). Virtually no hybridization to the pBluescript plasmid control was observed with the labeled mRNAs in these experiments (data not shown). Results very similar to those above were also obtained from parallel experiments using ³²P-labeling and quantitative phosphorimaging analyses.

View larger version (15K): [in this window] [in a new window]   Fig. 7.   Comparison of nascent messenger RNA transcripts from replete and adenosine-starved C. luciliae. A, slot blots of a 436-bp internal fragment of the Cl 3'-NT/NU ORF (see Fig. 2A) hybridized with DIG-labeled nascent mRNAs from adenosine replete cells (left panel) and adenosine-starved cells for 24 h (right panel). B, slot blots of a 440-bp internal fragment of the Cl -tubulin ORF hybridized with DIG-labeled nascent mRNAs as above. Signals from these hybridizations were quantified using National Institutes of Health Image, and the values shown are in arbitrary density units.

Cumulatively, results of these assays demonstrated that both adenosine-replete and adenosine-starved cells transcribed the Cl 3'-NT/NU gene at virtually the same rate. Thus, the accumulation of Cl 3'-NT/NU mRNA in adenosine-starved cells must reflect the posttranscriptional events that regulate the expression of this gene product.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Previously, it was reported that the 3'-nucleotidase/nuclease enzyme activity of C. luciliae was significantly up-expressed in response to purine starvation conditions (6). Based on its biochemical and kinetic properties, this bifunctional, surface membrane-bound enzyme was shown to share properties similar to those of the class I, single-strand-specific nucleases of fungi and germinating plant seedlings (3, 7). Results of our multiple sequence alignment analyses identified five regions of aa sequence, which appear to be highly conserved among members of the class I nuclease family (47, 48). Further, our results demonstrated that the deduced protein of the C. luciliae 3'-NT/NU gene reported here possessed all five of these conserved aa domains. Such conservation in primary structure strongly suggests that the C. luciliae 3'-NT/NU is a member of this family of enzymes. Included among these conserved residues are histidines, which in some members of this family have been shown to be involved in the binding of divalent metal ions (e.g. Zn²⁺) co-factors (47, 48). This is of relevance as the native C. luciliae 3'-NT/NU requires Zn²⁺ as a co-factor to retain its enzymatic activities (7). The latter also applies to the native L. donovani 3'-NT/NU (45). Taken together, these data reinforce the kinship in functions previously observed among the fungal and plant class I nucleases and the trypanosomatid 3'-nucleotidase/nucleases (5, 6, 7).

In contrast to other members of the class I nuclease family (47, 48), the C. luciliae 3'-nucleotidase/nuclease has been shown to be an externally-disposed, surface membrane-anchored protein (6, 23). In order to demonstrate that our Cl 3'-NT/NU gene actually encoded a functional enzyme and that it contained an anchoring domain, a truncated construct of it (Cl 3'-nt/nu^C)was episomally expressed in C. luciliae. The latter construct contained the putative N-terminal signal peptide but lacked both the putative C-terminal hydrophobic trans-membrane anchor domain and the hydrophilic cytoplasmic tail. This homologous episomal expression system was used to ensure that processing of the expressed protein would be equivalent to that of the native endogenous cell surface homolog. Results of these experiments showed that such transfectants released/secreted into their culture supernatants a soluble, highly active 38-kDa Cl 3'-nt/nu^C protein, which had both 3'-nucleotidase and nuclease activities. The latter demonstrated that our Cl 3'-NT/NU gene in fact encoded an active parasite enzyme. Further, these results showed that, although the putative trans-membrane domain was required for anchoring the native Cl 3'-NT/NU in the surface membrane of these parasites, it was not necessary for either of the bifunctional activities of this enzyme. Like most other members of the class I nuclease family, the native C. luciliae 3'-NT/NU is an N-linked glycoprotein with an apparent molecular mass of ~43 kDa as determined by SDS-PAGE (7). In agreement with this, results of our tunicamycin experiments demonstrated that N-linked glycosylation contributed ~4 kDa to the apparent molecular mass of the native Cl 3'-NT/NU protein. However, we showed that such N-linked glycosylation was not required per se for either the 3'-nucleotidase or the nuclease activities of this bifunctional enzyme.

Trypanosomatids are a group of primitive protozoan parasites, which are incapable of de novo purine biosynthesis. Thus, these purine auxotrophs require extracellular salvage mechanisms to acquire this essential nutrient from their host environments (2). One such trypanosomatid salvage enzyme is the externally oriented, surface membrane 3'-nucleotidase/nuclease, which is capable of generating free nucleosides via the hydrolysis of either 3'-nucleotides or nucleic acids (3, 4). Such enzyme activities have been reported from a variety of different trypanosomatids (4, 5). Thus, among the members of the class I nuclease family that we examined, our FASTA analysis showed that the closest structural ortholog to the C. luciliae 3'-NT/NU was that from a distantly related trypanosomatid parasite, L. donovani. The latter observation is in agreement with our Western blot analyses, which demonstrated that both of these parasite enzymes shared some common antigenic epitopes presumably reflecting conservation in their primary aa structures. In addition, we identified two nucleotide sequences recently deposited in the data bases, one from Trypanosoma brucei (GenBank^TM AC013353) and the other from Leishmania pifanoi (GenBank^TM AF057351) whose deduced peptides contained four out of the five conserved aa domains identified in the current report as characteristic of the class I nuclease family. Taken together, these observations suggest that the 3'-nucleotidase/nuclease has been structurally and functionally conserved among divergent members of the trypanosomatid family.

We showed, in the current study, that in response to adenosine starvation C. luciliae expressed up to a 1000-fold higher levels of 3'-nucleotidase activity than cells maintained under adenosine replete conditions. These observations verify those previously reported by Neubert and Gottlieb (7). Further, we identified and characterized a gene (Cl 3'-NT/NU) that encodes this unique parasite enzyme and showed that its specific steady-state mRNA was elevated by ~100 fold in adenosine-starved cells. Such a significant increase in the Cl 3'-NT/NU-specific message level most likely accounts for the dramatic elevation in 3'-nucleotidase activity observed in these cells. Since steady-state levels of mRNA are determined by a balance between their rate of synthesis and their rate of decay (49-51), our results suggested that either adenosine-starved cells transcribed the Cl 3'-NT/NU gene at a faster rate than adenosine-replete cells or that their transcribed message was more stable. To address this issue, nuclear run-on experiments were done with adenosine-replete and adenosine-starved cells to determine their rates of Cl 3'-NT/NU mRNA synthesis. Results of those experiments demonstrated that the specific Cl 3'-NT/NU message was transcribed at virtually the same rate by both adenosine-starved and adenosine-replete cells. Similar results were obtained from these cells with regard to their rates of transcription for a constitutively expressed gene, i.e. tubulin. Taken together, these observations indicate that the transcribed Cl 3'-NT/NU message must be more stable in adenosine-starved cells than in adenosine-replete cells. Such stability could lead to its accumulation and account for the ~100-fold difference observed in steady state mRNA levels between these two cell types. Thus, these experiments indicated that the remarkable up-expression of the Cl 3'-NT/NU is most likely modulated at the post-transcriptional level. Although such post-transcriptional regulation is unusual for higher eukaryotes, it has been shown to be the predominant mode for regulating gene expression in these primitive trypanosomatid protozoans (28, 52, 53).

In summary, in this report, we showed that in response to starvation for a single essential nutrient (i.e. adenosine), the primitive trypanosomatid protozoan, C. luciliae dramatically up-regulated its expression of a gene which encodes the surface membrane enzyme involved in the acquisition of such nutrients. Further, this unparalleled up-expression in enzyme activity appeared to be mediated by post-transcriptional events presumably involving the stability of its specific mRNA. Both the events that trigger this response and the precise mechanisms responsible for such post-transcriptional regulation of this Cl 3'-NT/NU gene remain to be explored experimentally in this primitive organism. Reagents generated in the current report should provide valuble tools for such studies. In addition, based on our observations, it also seems probable that the Cl 3'-NT/NU gene locus might be usefully exploited for the expression of foreign genes in a highly inducible fashion.

	ACKNOWLEDGEMENTS

We thank Dr. Michael Gottlieb of the Parasitology and International Programs Branch, NIAID, National Institutes of Health for encouragement and valuable discussions during the course of these studies. We also thank Jennifer Uyeda for assistance in the initial phases of this work and Dr. Alison M. Shakarian (Laboratory of Parasitic Diseases, NIAID, National Institutes of Health) for advice in designing oligonucleotide primers and her insightful critiques of these studies. In addition, we thank Drs. Hugues Charest (Laboratory of Parasitic Diseases, NIAID, National Institutes of Health) and Greg Matlashewski (Institute of Parasitology, McGill University) for providing us with the [pKSNEO] expression vector.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The 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 GenBank^TM/EMBL Data Bank with accession number(s) AF140355.

National Institutes of Health visiting fellow supported by postdoctoral fellowships from the Fogarty International Center and the NIAID, National Institutes of Health.

^§ National Institutes of Health visiting associate supported by postdoctoral fellowships from the Fogarty International Center and the NIAID, National Institutes of Health. Present address: Laboratory of Parasitic Biology and Biochemistry, Center for Biologics Evaluation and Research, Food and Drug Administration, Bethesda, MD 20892.

^¶ To whom correspondence should be addressed. Tel.: 301-496-5969; Fax: 301-402-2201; E-mail: ddwyer@niaid.nih.gov.

Published, JBC Papers in Press, August 16, 2000, DOI 10.1074/jbc.M004036200

² M. Yamage, unpublished data.

³ A. Debrabant, E. Ghedin, and D. M. Dwyer, (2000) J. Biol. Chem. 275, 16366-16372.

	ABBREVIATIONS

The abbreviations used are: NT/NU, nucleotidase/nuclease; aa, amino acid(s); PAGE, polyacrylamide gel electrophoresis; bp, base pair(s); PCR, polymerase chain reaction; Py, pyrimidine; Pu, purine; ORF, open reading frame; PFGE, pulsed field gel electrophoresis; UTR, untranslated region; nt, nucleotide(s); kb, kilobase pair(s); RACE, rapid amplification of cDNA ends; DIG, digoxigenin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES