Characterisation and sequence analysis of a carbamate kinase gene from the diplomonad Hexamita inflata.

Hexamita inflata can derive energy from the degradation of arginine via the arginine dihydrolase pathway. Carbamate kinase catalyses the third enzymatic step of the pathway synthesising ATP from the catabolism of carbamyl phosphate. This study reports the identification and characterisation of a carbamate kinase gene from this free-living diplomonad, together with measurements of carbamate kinase enzyme activity in cell-free extracts and a preliminary analysis of the carbamate kinase mRNA by reverse-transcription polymerase chain reaction. Analysis of the carbamate kinase gene revealed the use of non-canonical codons for glutamine. Phylogenetic studies showed a consistent close relationship between carbamate kinase sequences of H. inflata and Giardia intestinalis.

Trichomonas vaginalis: characterization, expression, and phylogenetic analysis of a carbamate kinase gene sequence.

The gene encoding carbamate kinase (CBK, ATP:carbamate phosphotransferase, EC 2.7.2.2) from Trichomonas vaginalis has been sequenced and its expression in this protozoon has been verified using reverse-transcription polymerase chain reaction. The codon usage and percentage nucleotide composition in the coding and noncoding regions are consistent with other genes isolated from this parasite. Phylogenetic analysis of this gene has suggested possible speciation events that are congruent with other studies, with suggestions of lateral gene transfer. The gene was expressed in Escherichia coli using the pQE-30 expression system, and the recombinant protein was purified using affinity chromatography. The expression of the recombinant protein was identified via Western blotting and matrix-assisted laser desorption ionization mass spectrometry of tryptic peptides. Preliminary kinetic assays have revealed that the recombinant enzyme has a K(m) similar to that of the native enzyme and size-exclusion chromatography has shown that the enzyme is active as the homodimer.

Characterisation and expression of the carbamate kinase gene from Giardia intestinalis.

The arginine dihydrolase pathway in Giardia intestinalis produces energy via the carbamate kinase (CBK, ATP:carbamate phosphotransferase, EC 2.7.2.2) reaction. Characterisation of the CBK gene from the Portland 1 strain indicated that it is located on either chromosome 3 or 4, does not appear to contain introns and is expressed in both the trophozoite and early cyst stages. Heterologous expression of CBK in Escherichia coli, using the pQE-30 expression system (QIAGEN), enabled a one-step purification of the recombinant enzyme via affinity chromatography. The expressed protein was identified by enzyme assay and mass spectrometry. The native and recombinant forms of the enzyme have similar physical properties and the recombinant enzyme appears to be active as the homodimer.

Trichomonas vaginalis: expression and characterisation of recombinant S-adenosylhomocysteinase.

The gene encoding S-adenosylhomocysteinase activity (S-adenosylhomocysteine hydrolase, SAHH; EC 3.3.1.1) in Trichomonas vaginalis has been expressed in Escherichia coli to facilitate the characterisation of the enzyme. Expression of this gene using the pQE-30 (6xHis N-terminal tag) expression system (QIAGEN) has enabled the one-step purification of 6 mg of active recombinant enzyme from a 100-ml bacterial culture by affinity chromatography using a nickel-NTA matrix. The recombinant enzyme has a molecular weight of approximately 56,000 and identification of tryptic peptides by matrix-assisted laser desorption ionisation (MALDI) mass spectrometry has shown that the purified recombinant protein is identical in primary structure to the predicted sequence. The presence of the N-terminal 6xHis tag in the recombinant enzyme did not appear to affect its kinetic and other properties, which are similar to those exhibited by the "native" enzyme present in cell-free extracts of T. vaginalis. These properties include a similar apparent Km for adenosine (20-25 microM for the recombinant and 5-10 microM for the native enzymes, respectively) and similar inhibition/inactivation patterns exhibited by adenosine analogues such as arabinosyl adenine (ara-A).

Molecular characterisation of adenosylhomocysteinase from Trichomonas vaginalis.

The enzyme S-adenosylhomocysteine hydrolase (SAHH) has been identified as a potential target for chemotherapy in protozoan parasites including Trichomonas vaginalis. To investigate this area of trichomonad metabolism in more detail, we have isolated and characterised a gene which encodes this activity from the WAA38 strain of this parasite. The gene was isolated by probing a Bg/II genomic mini-library with a fragment of the gene generated by thermal cycling using degenerate oligonucleotide primers. A 5.9-kb Bg/II clone was isolated and has been partially sequenced to reveal a 1458-bp open reading frame which encodes a 486-residue polypeptide (computed molecular mass of 53.4 kDa). The deduced amino acid sequence showed a high degree of sequence similarity to the hydrolases from other species, but was most similar to the enzyme from photosynthetic organisms. The trichomonal sahh gene also contains two "insertion sequences', one of which appears to be unique to this parasite while the second has previously been found only in photosynthetic organisms and in Plasmodium falciparum. Characterisation of the sahh mRNA from T. vaginalis confirmed that both of these insertion sequences (encoding 9 and 37 amino acid residues, respectively) are expressed in the protein product. The sahh mRNA is similar to those characterised from other protozoa in having a short, 12-bp untranslated 5'-leader sequence but the leader sequence does not conform well with the consensus sequence of the other mRNAs. Finally, Southern blots and sequence differences between genomic and cDNA clones indicate that there are multiple copies of the sahh gene in T. vaginalis.

Sequences upstream and downstream from the glutamine-dependent carbamoyl phosphate synthetase-encoding gene from the protozoan Babesia bovis.

In the protozoan parasite, Babesia bovis, the glutamine-dependent carbamoyl phosphate synthetase-encoding gene (CPSII) contains contiguous amidotransferase- and synthetase-encoding sequences. Unlike the organisation in most eukaryotes, the gene is not fused with other genes encoding enzymes of pyrimidine biosynthesis de novo. The nucleotide sequences immediately upstream and downstream from the gene contain motifs which may be involved in regulating its expression.

An organelle-like small subunit ribosomal RNA gene from Babesia bovis: nucleotide sequence, secondary structure of the transcript and preliminary phylogenetic analysis.

Investigations aimed at identifying the mitochondrial genome of Babesia bovis using the polymerase chain reaction (PCR) have established the existence of an organelle-like small subunit ribosomal RNA (SSU rRNA) gene in the parasite. The sequence, compiled from three main PCR products, was 1448 bp in length (including the primer regions), had a 73% A+T content and showed significant similarity (68% sequence identity) to the "organellar" SSU rRNA gene from Plasmodium falciparum. The proposed secondary structure of the transcript showed several features which were consistent with a eubacterial origin for the organelle-like gene. The presence of putative binding sites for streptomycin and tetracycline also supported an "organellar" location for the gene and suggested that the SSU rRNA transcript is functional in protein synthesis because tetracycline has anti-babesial activity. Phylogenetic analyses based on the conserved regions of the SSU-like rRNA genes from a wide variety of organisms showed only a weak association of the babesial sequence with its mitochondrial homologues and an even weaker association with the corresponding genes of plastid origin. The origin of this organelle-like gene in B. bovis therefore remains unresolved, as is the case for its homologue from P. falciparum.

The toxicity of antifolates in Babesia bovis.

A variety of anti-folate compounds have been tested for their ability to inhibit the growth of Babesia bovis as measured by the incorporation of [3H]hypoxanthine into the parasite's nucleic acids. Inhibitors of folate synthesis (including 7-methylguanosine and several sulpha drugs) were without effect but several structural analogues of folate were toxic. The most potent folate analogues were the lipophilic compounds piritrexim and trimetrexate, each causing 50% inhibition of [3H]hypoxanthine incorporation (IC50) at a concentration of 2.9 nM; other classical anti-folates such as pyrimethamine, methotrexate and trimethoprim were at least 100-fold less effective with IC50 values of 1.2, 0.29 and 0.50 microM, respectively. From these results we conclude that B. bovis does not synthesize folate de novo under cell culture conditions. However, the toxic effects of piritrexim and trimetrexate suggest that dihydrofolate reductase (DHFR) activity is essential for the parasite, most probably because of the role of this enzyme in the synthesis of thymidine nucleotides via thymidylate synthase.

Identification of a Babesia bovis gene with homology to the small subunit ribosomal RNA gene from the 35-kilobase circular DNA of Plasmodium falciparum.

Isolation of a 552-base pair (bp) fragment of a putative extrachromosomal small subunit ribosomal RNA (SSrRNA) gene from Babesia bovis was achieved using the polymerase chain reaction (PCR) followed by cloning and sequencing of the PCR product. The sequences of the oligonucleotide primers used for the PCR were derived from selected known sequences in the organellar SSrRNA gene which is encoded within the 35-kilobase (kb) circular DNA from Plasmodium falciparum. Comparison of the sequence of the 552-bp fragment from B. bovis with gene sequences from other organisms showed 71% identity with the organellar SSrRNA gene from P. falciparum and up to 65% identity with the plastid SSrRNA gene sequences from various other organisms. We conclude that the 552-bp fragment amplified by PCR from B. bovis is possibly derived from an organellar genome of this parasite.

Deoxyadenosine toxicity in an adenosine deaminase-inhibited human CCRF-CEM T-lymphoblastoid cell line causes cell swelling.

The human T-lymphoblastoid cell line CCRF-CEM, pre-treated with 2'-deoxycoformycin, was used as a model for adenosine deaminase deficiency to investigate how 2'-deoxyadenosine exerts its cytotoxic effects. Incubation of these cells with 1 microM or 5 microM deoxyadenosine for 24 and 48 h caused an increase of up to 50% in their modal cell volume as measured by a Coulter Size Distribution Analyzer and this increase in cell volume was accompanied by an increase in their fragility and deformability. The swelling of cells was concomitant with the phosphorylation of deoxyadenosine and its intracellular accumulation as dATP. There was no evidence of osmotic imbalance or of inhibition of the Na+/K(+)-dependent ATPase activity as the intracellular concentrations (and the intracellular:extracellular ratios) of Na+, K+ and Ca2+ were essentially unchanged. Cytochalasin B (20 microM) also caused lymphoblasts to swell over a 6-h period and its effect on cell size was similar to that of either 1 microM or 5 microM deoxyadenosine over 24 or 48 h. Longer time-courses of incubation with cytochalasin B caused severe toxicity leading to the death and lysis of a significant proportion of the cells. Other drugs, such as colchicine, vincristine and vinblastine that are known to affect various components of the cytoskeleton also caused swelling of cells in a concentration- and time-dependent manner but there was no evidence that these effects were additive or synergistic with those of deoxyadenosine. Inhibition of DNA synthesis, either directly by aphidicolin or indirectly by hydroxyurea, was less cytotoxic than the effect caused by deoxyadenosine. We conclude that one of the toxic effects resulting from the excessive phosphorylation of deoxyadenosine and its accumulation as dATP in human T-lymphoblasts is not dependent on inhibition of DNA synthesis but may be caused by the disruption of the cytoskeleton in these cells.

Mitochondrial function in Babesia bovis.

A variety of anti-mitochondrial drugs that had previously been found to inhibit the growth of the malarial parasite Plasmodium falciparum were tested on Babesia bovis in vitro. Several of these drugs were found to be non-toxic towards B. bovis. However, those drugs that were found to inhibit babesial growth included compounds (shown in parentheses) that have the following putative mitochondrial targets in the parasite: ATP synthetase complex (rhodamine 123, oligomycin, Janus Green); ATP-ADP translocase (bongkrekic acid); electron transport (rotenone, n-heptyl-4-hydroxyquinoline-N-oxide (HQNO), antimycin A); ubiquinone (CoQ) function (BW58C, menoctone); protein synthesis (tetracycline); and the proton pump (CCCP). We have also investigated the effects of some of these drugs on pyrimidine biosynthesis de novo by following the incorporation of [14C]bicarbonate into pyrimidine nucleotides and into the pyrimidine moieties of nucleic acids. The ubiquinone analogues BW58C and menoctone inhibited this pathway in the nM-microM range of concentrations. Inhibitors of electron transport (antimycin A and oligomycin) and an uncoupler (CCCP) were also effective inhibitors of pyrimidine biosynthesis de novo. We conclude that B. bovis has a functional mitochondrion that contributes significantly to pyrimidine biosynthesis de novo and to the overall energy metabolism of the parasite.

Routine screening for potential babesicides using cultures of Babesia bovis.

A procedure for screening of potential anti-malarial agents, based on the incorporation of [3H]hypoxanthine by the parasite, was adapted for the testing of anti-metabolites against Babesia bovis (Lismore and Samford isolates) cultured in vitro. A close correlation was found between [3H]hypoxanthine incorporation in a standard assay and percentage of parasitized cells as determined by microscopic examination. The concentrations of compounds causing 50% inhibition of [3H]hypoxanthine incorporation (ID50 values) for the established babesicides, Imidocarb and Amicarbalide, were determined to be 3 ng ml-1 (8.6 nM) and 5-10 ng ml-1 (17-34 nM), respectively. A variety of other anti-metabolites were tested in the system. ID50 values for some of the more effective compounds were tubercidin (75 nM), tetracycline (25 microM), menoctone (100 nM) and TN 108, a di-Mannich base derived from 4-(7'-trifluoromethyl-quinolin-4'-ylamino)phenol (0.13 microM). No significant differences between results with the two isolates of B. bovis were observed.

Purine salvage and metabolism in Babesia bovis.

Studies of the incorporation of radio-labelled purine precursors into the erythrocytic forms of Babesia bovis under tissue-culture conditions have confirmed the presence in the parasite of enzymatic activities responsible for the salvage of preformed purines. The results also revealed that the parasite was capable of a variety of nucleotide interconversions, such that exogenous hypoxanthine and adenosine were incorporated into both adenine and guanine nucleotides followed by the incorporation of these nucleotides into the adenine and guanine moieties of RNA and DNA. No evidence was found for salvage of preformed pyrimidines. Evidence was also obtained for the insertion of a parasite-specific nucleoside/nucleobase transporter into the membrane of the bovine (host) red cell. Thus, whereas normal (non-parasitised) bovine red cells are essentially incapable of transporting nucleosides across their membranes, the invasion of these cells by B. bovis introduces a transporter that can be inhibited by classic nucleoside transport inhibitors.

Adenosine kinase from human erythrocytes: kinetic studies and characterization of adenosine binding sites.

The reaction catalyzed by adenosine kinase purified from human erythrocytes proceeds via a classical ordered sequential mechanism in which adenosine is the first substrate to bind to and AMP is the last product to dissociate from the enzyme. However, the interpretation of the steady-state kinetic data is complicated by the finding that while AMP acts as a classical product inhibitor at concentrations greater than 5 mM, at lower concentrations AMP can act as an apparent activator of the enzyme under certain conditions. This apparent activation by AMP is proposed to be due to AMP allowing the enzyme mechanism to proceed via an alternative reaction pathway that avoids substrate inhibition by adenosine. Quantitative studies of the protection of the enzyme afforded by adenosine against both spontaneous and 5,5'-dithiobis(2-nitrobenzoic acid)-mediated oxidation of thiol groups yielded "protection" constants (equivalent to enzyme-adenosine dissociation constant) of 12.8 microM and 12.6 microM, respectively, values that are more than an order of magnitude greater than the dissociation constant (Kia = 0.53 microM) for the "catalytic" enzyme-adenosine complex. These results suggest that adenosine kinase has at least two adenosine binding sites, one at the catalytic center and another quite distinct site at which binding of adenosine protects the reactive thiol group(s). This "protection" site appears to be separate from the nucleoside triphosphate binding site, and it also appears to be the site that is responsible for the substrate inhibition caused by adenosine.

The biochemical and clinical consequences of 2'-deoxycoformycin in T cell chronic lymphocytic leukaemia.

The mechanisms for cell toxicity with adenosine deaminase inhibition by 2'-deoxycoformycin (dCF) in non replicating lymphoid cells include S-adenosylhomocysteine (SAH) hydrolase inactivation and reduction of cellular ATP content. These postulates were explored in a patient with T-CLL receiving dCF with a resultant fall in peripheral blood lymphocytes from 740 X 10(9)/1 to 90 X 10(9)/1 over 15 d. In red cells there was complete inhibition of adenosine deaminase and SAH hydrolase activities, progressive deoxyadenosine triphosphate (dATP) accumulation and ATP depletion but no significant alteration in adenosine monophosphate (AMP) deaminase activity or distribution in purine intermediates from radioactive adenosine. In T-CLL lymphocytes, there was incomplete lymphoid SAH hydrolase inactivation, reduced AMP deaminase activity and progressive dATP accumulation. The limited decrease in lymphocyte ATP content was related more to dCF administration than dATP accumulation, nor accompanied by significant changes in the distribution of purine intermediates from adenosine. These findings suggest that ATP depletion with dCF therapy does not reflect AMP deaminase activity modulation nor is of critical importance for cell toxicity. The exact role for elevated cellular dATP content and SAH hydrolase inactivation in this toxicity remains to be established.

Mechanism of deoxyadenosine-induced catabolism of adenine ribonucleotides in adenosine deaminase-inhibited human T lymphoblastoid cells.

Loss of ATP accompanying accumulation of dATP has recently been reported to occur in the erythrocytes and lymphoblasts of patients with T lymphocytic leukemia during treatment with deoxycoformycin, an inhibitor of adenosine deaminase (adenosine aminohydrolase, EC 3.5.4.4) that causes the accumulation of deoxyadenosine. We have studied the mechanisms responsible for adenine ribonucleotide depletion in cultured human CEM T lymphoblastoid cells treated with deoxycoformycin and deoxyadenosine. Accumulation of dATP was accompanied by depletion of total soluble adenine ribonucleotides without change in the adenylate energy charge, by the route ATP --> AMP --> IMP --> inosine --> hypoxanthine; conversion of IMP to AMP and de novo purine synthesis were inhibited in these cells. ATP degradation did not occur in a mutant of CEM that was incapable of phosphorylating deoxyadenosine, or in a B cell line with very limited ability to accumulate dATP. We found that dATP and ATP were both able to stimulate markedly the deamination of AMP by lymphoblast AMP deaminase; dAMP was a poor substrate for this enzyme (K(m) = 2.4 mM, vs. 0.4 mM for AMP). Similarly, dATP as well as ATP caused marked activation of IMP dephosphorylation by a lymphoblast cytoplasmic nucleotidase. Inhibition of intracellular AMP deaminase with coformycin prevented degradation of adenine ribonucleotides without affecting dATP accumulation. We propose that ATP-dependent phosphorylation of deoxyadenosine generates ADP and AMP. Simultaneously, dATP accumulation stimulates deamination of AMP, but not dAMP, and the dephosphorylation of IMP to inosine. Coupling of AMP degradation to ATP utilization in deoxyadenosine phosphorylation maintains the adenylate energy charge despite net depletion of cellular ATP.

Uridine and inosine: pressor nucleosides from man, rat and dog.

Short-acting pressor compounds isolated from rat kidney, brain, heart and spleen have been identified as inosine and uridine by gas chromatography, mass spectrometry, high pressure liquid chromatography and analysis of ultraviolet spectra. Inosine was further identified by nuclear magnetic resonance. These compounds have also been found in kidneys from hypertensive man, rejected renal transplants and dog and beef kidney. Tissues were extracted by acid ethanol extraction followed by gel filtration and high voltage paper electrophoresis. Compounds found to be pressor in the anaesthetised rat resisted proteolytic enzymes, boiling for 10 min, extremes of pH and incubation with plasma from the source species for up to 20 min. The pressor effects of bolus injections of active gel filtration fractions, uridine and inosine were short-lived with a maximum effect at 5-6 s. Intravenous (I.V.) infusions of extracts gave a sustained pressor response without a concurrent change in heart rate. The effect on blood pressure was not accompanied by increased heart rate and persisted when the pressor effects of angiotensin II, noradrenaline and 5-hydroxy-tryptamine were blocked pharmacologically.