(Received for publication, August 15, 1996, and in revised form, December 30, 1996)
From the Department of Biological Chemistry and Molecular
Pharmacology, Harvard Medical School, Boston, Massachusetts 02115 and
Department of Pharmacology and Program in Molecular
Medicine, University of Massachusetts Medical School,
Worcester, Massachusetts 01605
Trypanosomatid protozoans depend upon exogenous
sources of pteridines (pterins or folates) for growth. A broad spectrum
pteridine reductase (PTR1) was recently identified in Leishmania
major, whose sequence places it in the short chain alcohol
dehydrogenase protein family although its enzymatic activities resemble
dihydrofolate reductases. The properties of PTR1 suggested a role in
essential pteridine salvage as well as in antifolate resistance. To
prove this, we have characterized further the properties and relative roles of PTR1 and dihydrofolate reductase-thymidylate synthase in
Leishmania pteridine metabolism, using purified enzymes and knockout mutants. Recombinant L. major and Leishmania
tarentolae, and native L. major PTR1s, were tetramers
of 30-kDa subunits and showed similar catalytic properties with pterins
and folates (pH dependence, substrate inhibition with
H2pteridines). Unlike PTR1, dihydrofolate
reductase-thymidylate synthase showed weak activity with folate and no
activity with pterins. Correspondingly, studies of
ptr1 and dhfr-ts
mutants implicated only PTR1 in the ability of L.
major to grow on a wide array of pterins. PTR1 exhibited
2000-fold less sensitivity to inhibition by methotrexate than
dihydrofolate reductase-thymidylate synthase, suggesting several
mechanisms by which PTR1 may compromise antifolate inhibition in
wild-type Leishmania and lines bearing PTR1
amplifications. We incorporate these results into a comprehensive model of pteridine metabolism and discuss its implications in chemotherapy of this important human pathogen.
Leishmania are trypanosomatid protozoan parasites that infect millions of people worldwide (1). Leishmaniasis takes several forms, ranging from minor or severe disfiguring cutaneous lesions to the deadly visceral form, depending upon the species and immune status of the host. Vaccines against Leishmania are not yet available, and treatment currently relies on the antiquated pentavalent antimonial compounds. These drugs are often toxic, sometimes ineffective, and their mode of action remains unknown. A better understanding of novel biochemical pathways of this primitive eukaryotic parasite clearly would be helpful in the development of selective anti-Leishmania drugs. For example, although antifolates are a mainstay in the treatment of parasitic diseases such as malaria, they have not proven clinically effective against Leishmania (2, 3). This may reflect the fact that Leishmania and related trypanosomatids exhibit a number of unusual features in pteridine (pterin and folate) metabolism. Improved knowledge of this pathway would likely allow the development of antifolates effective against this important disease.
Leishmania and other trypanosomatids including Crithidia are unable to synthesize the pterin moiety from GTP and thus must acquire pteridines from the host by salvage mechanisms (2, 4-11). This feature led historically to an appreciation of the pterin requirement of eukaryotes, where pterins are now known to participate as essential cofactors in hydroxylations, ether-lipid cleavage, and NO synthase (12-15). However, the pathways involved in the salvage and metabolism of pterins, and their function in Leishmania, are only beginning to emerge (9, 10).
Recently, we identified a novel pteridine reductase
(PTR1)1 in Leishmania (10).
PTR1 (formerly hmtxr or ltdh)
was originally identified as the gene responsible for methotrexate
(MTX) resistance on the amplified H region in several species of
Leishmania (16, 17). Sequence comparisons placed the
predicted PTR1 protein in a large family of aldo-keto reductases and
short chain dehydrogenases, a family including both
dihydropteridine and sepiapterin reductases (16-19). The ability
of PTR1 to reduce pteridines such as biopterin and folate was
established by genetic and biochemical approaches in our laboratory
(10). First, ptr1 null mutants specifically
required H2- or H4biopterin for growth, a
requirement not satisfied by H2- or H4folate.
Second, partially purified recombinant PTR1 protein exhibited
NADPH-dependent reductase activity with biopterin and
folate and lesser activity with H2biopterin or
H2folate (10). These properties placed PTR1 in a position to play a key role in the salvage of oxidized pterins. Moreover, the
H2folate reductase activity of PTR1, when combined with its relative insensitivity to MTX inhibition (100 nM
versus 0.1 nM for DHFR-TS; Ref. 20), suggested
that PTR1 could compromise antifolate inhibition of
Leishmania (10).
Despite the homology of PTR1 to the short chain alcohol dehydrogenase superfamily (16-18), its enzymatic properties overlap those of many dihydrofolate reductases (DHFR), which is remarkable given their evolutionary divergence. The major role of DHFR is to convert H2folate to the biochemically active H4folate, a step needed for de novo synthesis of thymidylate, and in bacteria and higher eukaryotes, purine nucleotides (trypanosomatids are auxotrophic for purines). In Leishmania as well as all protozoans and plant species examined thus far, DHFR is part of a bifunctional polypeptide that also encodes thymidylate synthase (DHFR-TS; Refs. 21-23). Direct comparison of the enzymatic properties of PTR1 and DHFR-TS would help in the elucidation of the salvage and metabolism of pteridines in Leishmania. Additionally, such information could establish the suitability of PTR1 and/or DHFR-TS as targets for rational Leishmania chemotherapy.
Here we have purified both native and recombinant L. major
PTR1s as well as recombinant Leishmania tarentolae PTR1, and
have characterized their properties including Km,
Vmax, pH dependence, and inhibition by substrate
and MTX. Comparisons of the wild-type and ptr1
and dhfr-ts
knockout Leishmania
showed that the ability to grow in diverse pterins correlated with
their activity with PTR1 but not DHFR-TS, establishing PTR1 as the sole
mediator of oxidized pterin salvage. Comparisons of the properties of
PTR1 and DHFR-TS enzymes, and pteridine reductase activities in crude
Leishmania extracts (including those from
ptr1
and dhfr-ts
mutants), were used to establish the relative contribution of these
enzymes in pteridine metabolism. With this information, we have
developed a comprehensive model of the salvage and metabolism of
pteridines in Leishmania.
All lines of Leishmania
were derived from L. major strain LT252 clone CC-1 and
cultured in M199 medium containing 10% fetal bovine serum (24). In
this medium parasites grow as the promastigote form, which normally
resides extracellularly within the gut of the sand fly insect vector.
Null mutant Leishmania lacking DHFR-TS (dhfr-ts) or PTR1
(ptr1
) were created by targeted disruption of
both alleles of each gene (10, 25). The ptr1
mutant was grown with H2- or H4biopterin (2-4
µg/ml), and the dhfr-ts
mutant was grown
with 10 µg/ml thymidine. The lines
ptr1
/+PTR1 and
dhfr-ts
/+DHFR-TS represent the respective null
mutants transfected with plasmids pX63NEO-PTR1 (10) or pK300 (24) and
overexpress PTR1 and DHFR-TS, respectively (Ref. 10; this work). In
some experiments cells were grown in fdM199, which is M199 medium
lacking folate and thymidine and supplemented with 0.66% bovine serum
albumin (U. S. Biochemical Corp.) instead of serum. Pterin supplements were H4biopterin (RBI), 6-hydroxymethylpterin, pterin,
pteroic acid (Sigma), and a wide range of other pterins (Schircks
Laboratories, Jona, Switzerland or from S. Kaufman, National Institutes
of Health). H2neopterin was prepared from neopterin by
reduction with dithionite in the presence of ascorbate (26). Parasites
were enumerated using a Coulter Counter (Model Zf) at the time when
cultures grown in H4biopterin had reached late log
phase.
The initial steps of purification of recombinant L. major PTR1 have been described (10) and included expression in Escherichia coli using the pET-3a expression vector (27), induction, cellular lysis, and purification by ammonium sulfate precipitation and DEAE-cellulose chromatography. PTR1-containing fractions from the DEAE step were pooled and the buffer changed to 20 mM Mes, pH 6.0, by passage over PD10 columns of Sephadex G-25 (Pharmacia Biotech Inc.). Subsequent purification steps were carried out by fast protein liquid chromatography (Pharmacia). Protein was applied to an ion exchange Mono-S HR 5/5 column and eluted with a 20-min 0-0.2 M NaCl gradient at 1 ml/min. An ion exchange Mono-Q 5/5 column was also tested and found to give an equivalent purification. PTR1-containing fractions were combined, and the volume reduced to 1 ml using YM10 filters (Amicon). The concentrate was applied to a Superdex 200HR 10/30 column and eluted at a flow rate of 0.5 ml/min with 20 mM Mes, pH 6.0, containing 0.1 M NaCl. Recombinant PTR1 was purified 10-fold with overall yields of 80%.
The coding region for L. tarentolae PTR1 was
amplified by the polymerase chain reaction using Taq
polymerase, template DNA from the MG strain of L. tarentolae, and the primers SMB-8 (5-ggcagatcTCAGGCCCGGGTAAGGC) and SMB-9 (5
-cgcagatctcccatATGACGACTTCTCCGA; lowercase letters indicate bases not present in PTR1), with 25 amplification
cycles of 1 min at 94 °C, 1 min at 57 °C, and 2 min at 72 °C.
The expected fragment was obtained, digested with NdeI and
BglII, inserted into the pET-3a expression vector (Novagen),
and transformed into E. coli strain BL21(DE3)/pLysS (27).
The expression of L. tarentolae PTR1 was induced and the
enzyme purified as described for L. major.
Native PTR1 was purified from 7.5 × 1010
ptr1/+PTR1 L. major, in a manner similar
to that used for the recombinant enzyme except that the cells were
lysed by 3 cycles of freezing and thawing followed by sonication. The
lysate was centrifuged at 100,000 × g for 30 min, and
the supernatant was loaded onto a DEAE-cellulose column, eluted (10),
and further purified as described for the recombinant enzyme. Native
PTR1 was purified 200-fold and obtained in 72% yield. Purified PTR1
preparations were stored at
80 °C in the presence of 20% glycerol
and 20 mM
-mercaptoethanol.
The molecular weights of
nondenatured PTR1s were estimated on a Sephacryl S-200 column (120 × 0.8 cm) at a flow rate of 0.5 ml/min. Three different pH values were
tested using the following buffers: 20 mM Tris-HCl, pH 7.0, 20 mM NaPO4, pH 6.0, or 20 mM sodium acetate, pH 4.7, each containing 0.1 M NaCl.
Molecular mass markers were -amylase (200 kDa), alcohol
dehydrogenase (150 kDa), bovine serum albumin (66 kDa), carbonic
anhydrase (29 kDa), and cytochrome c (12.4 kDa). Fractions
were monitored at 280 nm and for PTR1 activity.
Spectrophotometric pteridine reductase
assays were performed at 30 °C in the presence of NADPH (usually 100 µM) and pteridines as indicated (10). The pH dependence
of PTR1 activity was determined using three overlapping buffers: 20 mM sodium acetate, pH 3.6-6.0, NaPO4, pH
5.5-7.5, or Tris-HCl, pH 7.0-8.0. Radiometric assays of folate and/or
H2folate reductase activities (28) were performed using 40 µM [3,4
,7,9-3H]folate (24.1 Ci/mmol,
Moravek Biochemicals), which was purified prior to use (29). To test
the nature of the product formed from reduction of biopterin or
H2biopterin by PTR1 or DHFR-TS, a coupled assay was used
(30) where the synthesis of H4biopterin is linked to the
hydroxylation of [4-3H]Phe (27 Ci/mmol, Amersham Corp.)
by mammalian phenylalanine hydroxylase (Sigma). After incubation for 30 min at 25 °C, the [3H]Tyr formed was iodinated, the
sample was passed over a Dowex 50 column, and the tritiated water was
quantified by scintillation counting.
The kinetic parameters Km and Vmax for the pteridine substrates were measured in a spectrophotometric assay with 100 µM NADPH as described previously (10). Extinction coefficients used for various pteridines were determined spectrophotometrically, and PTR1 activity was calculated based on the decrease in absorbance of both NADPH and the pteridine substrates. Kinetic data for oxidized pteridines were evaluated by fitting to the Michaelis-Menten equation by nonlinear regression (Hyper Version 1.02A; J.S. Eastby, Liverpool, UK). Both H2folate and H2biopterin showed substrate inhibition at concentrations above 5 and 10 µM, respectively, and for these, Km, Vmax, and Ki (for substrate) values were evaluated using graphical plots and the general equation for substrate inhibition (31). For inhibition studies, PTR1 was incubated with MTX and NADPH and the reaction initiated with the pteridine substrate (40 µM folate, 100 µM biopterin, 10 µM H2biopterin, or 5 µM H2folate). Inhibition was examined at several concentrations of enzyme, and the data were analyzed using a method for tight binding inhibitors to obtain Ki (32).
Purification and Assay for DHFR-TSRecombinant DHFR-TS from
L. major was purified from a
dhfr E. coli strain (33) bearing
the expression plasmid 02CLSA-4 (34). Cells were lysed by two cycles
through a French press (15,000 p.s.i.), and DHFR-TS was purified by
binding and elution from a MTX-Sepharose column (Sigma) (34, 35). The
eluate was concentrated using YM10 membrane filters (Amicon) and loaded
onto a Sephacryl S-200 column (120 × 0.8 cm). Electrophoretically
homogeneous enzyme was eluted with 50 mM Tris·HCl, 0.1 M NaCl at a flow rate of 0.5 ml/min, desalted over PD10
columns of Sephadex G-25 (Pharmacia), and stored at
80 °C in the
presence of 10% glycerol.
Polyclonal antiserum against PTR1 was elicited in New Zealand White rabbits using 200 µg of L. major PTR1 in Freund's complete adjuvant (Sigma) in the primary immunization. The rabbits were boosted 5 times with 100 µg PTR1 each in incomplete Freund's adjuvant at 3-week intervals, and serum was obtained after the last bleeding. For immunoblots, purified PTR1 and crude Leishmania extracts were separated on a 12.5% SDS-polyacrylamide gel (36) and electrophoretically transferred onto Millipore polyvinylidene difluoride membranes (37) using a semi-dry blot apparatus (Owl Scientific). Blots were incubated with antiserum to PTR1 (1:1000), and binding was detected using either horseradish peroxidase-conjugated goat anti-rabbit antibody (1:3000) and chemiluminescence (Amersham Corp.) or alkaline phosphatase-conjugated goat anti-rabbit antibody and developed with 5-bromo-4-chloro-3-indolyl phosphate and nitro blue tetrazolium.
Preparation of Crude Leishmania ExtractsLate logarithmic phase promastigotes were collected by centrifugation, washed twice with phosphate-buffered saline (138 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, and 1.8 mM KH2PO4), and resuspended (3 × 109/ml) in phosphate-buffered saline supplemented with 1 mM EDTA and a mixture of protease inhibitors suggested by Meek et al. (20). Cells were lysed by freeze thawing and sonication and the extracts clarified by centrifugation at 15,000 × g for 30 min.
Previously we reported upon the partial
purification of L. major PTR1, expressed in engineered
E. coli (10). Inclusion of two additional steps (ion
exchange and gel filtration chromatography) yielded preparations that
were electrophoretically homogeneous, even when the gel was overloaded
(Fig. 1A, lanes 3-5). We also overexpressed
and purified native PTR1 from L. major parasites and
recombinant L. tarentolae PTR1. The recombinant and native PTR1s behaved similarly during purification and exhibited similar mobilities upon SDS-polyacrylamide gel electrophoresis (Fig. 1A, lanes 5-7). The apparent subunit molecular masses were 30 kDa (10, 11, 16, 17).
Western blot analysis with a polyclonal antiserum to recombinant
L. major PTR1 detected a 30-kDa protein in wild-type
L. major extracts whose size was identical to that of
purified PTR1s (Fig. 1B, lanes 2-4). This protein was
absent in the ptr1 L. major
deletion mutant obtained previously by gene targeting (Fig. 1B,
lane 1) (10) and was expressed at approximately 100-fold higher
levels in the L.major line overexpressing PTR1 (Fig.
1B, lane 3; note that 100-fold less protein was loaded in
lane 3). In wild-type cells, PTR1 constituted about 0.01%
of the total cellular protein.
By gel filtration chromatography, the apparent molecular mass of PTR1 was estimated to be 116 and 117 kDa for the recombinant and native L. major enzymes, respectively (not shown). Similar values were obtained at pH values of 4.7, 6.0, and 7.0 (data not shown). We infer that PTR1 is a tetramer of identical 30-kDa subunits and that significant alterations in molecular shape are not associated with differences in the pH dependence of folate versus biopterin reduction (below).
Enzymatic Properties of PTR1Previous studies of partially purified PTR1 showed it to have two pH optima, one of about 4.7 for biopterin and H2biopterin and one of about 6.0 for folate and H2folate (10). Studies of the homogeneous L. major and purified L. tarentolae PTR1s have refined and extended these initial findings.
At the optimum pH for each substrate, PTR1 activity with oxidized
biopterin and folate exhibited standard Michaelis-Menten kinetics (Fig.
2). However, H2biopterin and
H2folate showed substrate inhibition at concentrations
above 10 and 5 µM, respectively (Fig. 2).
Vmax values with H2biopterin and
H2folate were derived from analyses that included
considerations of substrate inhibition (31) and yielded values that
were at least 50% that of the corresponding oxidized pteridines (Table
I). Previously, only substrate concentrations of 100 µM were tested (10), which led to a 3-4-fold
underestimate of the rate of reduction of H2pteridines by
PTR1. H2neopterin and H2sepiapterin also showed
substrate inhibition, whereas L- and
D-biopterin, L- and D-neopterin,
6-hydroxymethylpterin, L- and
D-monapterin, 6-formylpterin, and 6,7-dimethylpterin showed standard Michaelis-Menten kinetics (data not shown). This suggests that
substrate inhibition was a general feature of PTR1 activity, but only
with H2pteridines.
|
We then examined the pH dependence of PTR1 activity. With biopterin a
sharp peak of activity was observed at pH 4.7 (Fig. 3A). Activity with H2biopterin
was also optimal at pH 4.7, although the peak was somewhat less sharp
(Fig. 3B). A pH optimum of 4.7 was found for PTR1 activity
with every pterin tested (L- and D-biopterin, L- and D-neopterin, 6-hydroxymethylpterin,
L- and D-monapterin, 6-formylpterin, 6, 7-dimethylpterin, H2sepiapterin; data not shown). In
contrast, with folate maximal activity occurred at pH 6.0 (Fig. 3C), and with H2folate a broad pH optimum was
found, from about 5 to 7.5 (Fig. 3D). Thus, pH optima
criteria divide PTR1 substrates into pterins versus folates,
rather than by oxidation state as observed for substrate
inhibition.
Based on this information, we determined the kinetic properties for the recombinant and native L. major PTR1s, and L. tarentolae PTR1, at pH 4.7, 6.0, or 7.0 with biopterin, H2biopterin, folate, and H2folate (Table I). The properties of all three enzymes were very similar, showing first that the recombinant enzyme faithfully represented the native L. major enzyme and second that PTR1s from different species catalyze similar reactions.
For all PTR1s, the Km for NADPH was 9-15 µM, and this was insensitive to enzyme source, pH, and substrate (Table I). At optimum pH values, biopterin displayed the highest Km (10-12 µM); H2biopterin and H2folate were intermediate (3.4-8.5 µM), and folate had the lowest Km (1.9-2.6 µM; Table I). For H2biopterin and H2folate, substrate inhibition Ki values of 11-21 µM were obtained, 2-4-fold above the Km calculated for these substrates (Table I). In general these values were not strongly affected by pH. This suggests that the differences between pterins and folates, or oxidized and reduced pteridines, arise from factors involving interaction with the substrates themselves, rather than the assay conditions.
MTX was a potent inhibitor of the recombinant L. major and L. tarentolae PTR1s, with all pteridines and at different pH values (Table I). Using the method of Cha (32) to calculate the Ki for tight binding inhibitors at the optimum pH for each substrate, MTX inhibited PTR1 activity with biopterin most strongly (Ki = 30 nM), followed by H2biopterin (Ki = 60 nM), H2folate (Ki = 200 nM) and folate (Ki = 255 nM). For all substrates the Ki was higher at pH 7.0 than at optimal pH, showing an increase of 6-9-fold for biopterins and 3-4-fold for folates (Table I).
Products of Pteridine Reduction by PTR1We determined whether
the action of PTR1 on biopterin yielded the biologically active
H4biopterin by coupling this reaction to the
H4biopterin-dependent formation of tyrosine
(Tyr) by mammalian phenylalanine hydroxylase (30). In the
absence of phenylalanine hydroxylase or PTR1, little Tyr
formation was observed (Fig. 4A). Addition of
increasing amounts of PTR1 resulted in increasing Tyr synthesis, with
10 µg of PTR1 showing as much activity as 10 µM
H4biopterin (Fig. 4A; it should be noted that
the conditions of this assay, pH 6.8, are not optimal for PTR1
activity). Similar results were obtained when biopterin was replaced
with H2biopterin in the assay mixture (not shown). Thus,
PTR1 directs the synthesis of biologically active
H4biopterin, presumably the
(6R)-L-erythro-5,6,7,8-H4biopterin substrate of phenylalanine hydroxylase.
We next asked whether PTR1 activity generated H4folate. A radiometric assay was used where folate and H2folate but not H4folate were precipitated in the presence ZnSO4 (28). By these criteria, recombinant L. major PTR1 mediated the formation of H4folate from both folate (Fig. 4B) and H2folate (not shown). As expected, activity with folate was inhibited partially by 1.25 µM MTX, whereas bovine DHFR was completely inhibited (Fig. 4B). Thus, we conclude that PTR1 mediates the synthesis of H4pteridines (the biochemically active forms) starting from either oxidized or H2pteridines.
Comparison of PTR1 and DHFR-TS ActivitiesThe activities of PTR1 toward pterins and folates overlap those of DHFRs purified from various sources (Table I; Ref. 38). The activity of the Leishmania DHFR-TS enzyme with pterins or oxidized folate had not been reported, and we purified the L. major DHFR-TS from engineered E. coli (20, 34, 39). The recombinant enzyme prepared by these methods is known to exhibit the same properties as the native enzyme, when assayed with H2folate or for TS activity (20, 34, 39).
DHFR-TS activity was optimal at pH 5.0 with folate (Km = 4.1 ± 2.6 µM), and at pH 7.0 with H2folate (Fig. 3, E and F). In contrast, PTR1 activity was maximal at pH 6.0 with both substrates (Fig. 3, C and D). Relative to PTR1, DHFR-TS activity was 20-fold greater with H2folate and 100-fold less with folate (Table II). We were unable to detect biopterin or H2biopterin reduction by DHFR-TS in either spectrophotometric or coupled phenylalanine hydroxylase assays, at pH values from 4.7 to 7.4 (Table II; data not shown). Thus, Leishmania DHFR-TS has weak activity with folate and no detectable activity with pterin substrates.
|
To determine the relative contributions of PTR1 and DHFR-TS to the reduction of folates in L. major, we measured activities in crude cellular extracts. We were aided by the availability of targeted null mutants lacking either the PTR1 or DHFR-TS genes (10, 25), which permitted a genetic test of the contribution of each enzyme. Since nonspecific interference in crude extracts was high with the spectrophotometric assay, particularly at low pH, we used the radiometric assay with [3H]folates at substrate concentrations yielding highest activity (Table I and Fig. 2).
H2folate reduction was measured at pH 7, where both PTR1
and DHFR-TS exhibited high activity (Fig. 3, D and
F). Comparisons of the wild-type,
ptr1 (DHFR-TS only) and
dhfr-ts
(PTR1 only) lines showed that more
than 90% of cellular activity arose from DHFR-TS (Table
III). The predominance of DHFR-TS agrees with the
predicted relative contribution of these two enzymes, calculated from
estimates of the cellular levels of these two proteins and their
specific activities (Table II).
|
Folate reduction was measured at pH 5, 6, and 7; the data for pH 6 is
shown in Table III. Since the radiometric assay follows H4folate rather than H2folate formation (40),
it is relevant to note that the H2folate reductase
activities of both PTR1 and DHFR-TS were comparable to or greater than
that with folate (Fig. 3, C-F) and would thus
not be limiting. As with H2folate, most of the folate
activity could be assigned to DHFR-TS, as the
ptr1 mutant showed only an 18% reduction in
activity, comparable to the 11% activity remaining in the
dhfr-ts
mutant.
However, the predominance of DHFR-TS in folate reduction disagrees with
that deduced from estimates of the cellular levels of these two
proteins and their specific activities (Table II). We calculated that
the contribution of PTR1 should be 40-fold higher than that of DHFR-TS,
rather than 7-fold lower (Table III). This arises from discrepancies in
both DHFR-TS and PTR1 activities, which were observed to be about
11-fold higher and 24-fold lower than calculated, respectively (Table
III). To address this problem, we examined numerous different
preparations and experimental conditions (varying pH and folate
concentrations), verified that each assay was performed in the linear
range of crude extract addition, and confirmed that radiometric and
spectrophotometric assays yielded similar kinetic parameters with the
purified enzymes (data not shown). None of these variables
significantly altered the result shown in Table III. That the assay
used could detect high levels of PTR1 is shown by studies of the
ptr1/+PTR1 line, which shows a 200-fold
increase in activity with folate (Table III), and by addition of
purified PTR1 to the crude extracts, which yielded the expected
activity (not shown). Last, mechanistic studies of purified PTR1 and/or
DHFR-TS catalysis do not suggest an explanation for this observation
(20).2
Leishmania are able to utilize a wide range
of pteridines (8, 11), and we sought to establish whether PTR1,
DHFR-TS, or possibly some other pteridine reductase was responsible for salvage. We utilized a folate-deficient medium (fdM199) in these studies to determine the ability of different pteridines to support the
growth of wild-type or mutant L. major and compared these results with the relative activity of PTR1 with these substrates (Table
IV). In fdM199 medium, supplementation with an active
pteridine is required for growth, and this is not affected by provision of thymidine (which is required by the dhfr-ts
mutant).
|
Three different groups of oxidized pteridines emerged from these
studies (Table IV). "Good" pteridine nutrients (L- and
D-biopterin, 6-hydroxymethylpterin,
L-neopterin) sustained the growth of wild-type Leishmania and were good PTR1 substrates (>59% the
activity obtained with L-biopterin). PTR1 but not DHFR-TS
was essential for growth with these pterins, as the
ptr1 mutant failed to grow while the
dhfr-ts
mutant grew normally. "Poor"
pteridine nutrients (D-neopterin, L- and
D-monapterin, 6,7-dimethylpterin, 6-formylpterin) failed to
sustain growth of wild-type Leishmania but were able to
support growth of the PTR1 overproducer. These pteridines showed
reduced activity with PTR1, about 10-34% that of
L-biopterin (Table IV). DHFR-TS overproduction failed to
sustain growth with these nutrients, consistent with its lack of
activity with pterin substrates (Table II, 4). Last, "inactive"
pteridine nutrients (pterin, pteroic acid, xanthopterin,
isoxanthopterin, 6-carboxypterin, and 7-biopterin) were unable to
support growth of any Leishmania tested and,
correspondingly, were weak or inactive PTR1 substrates (0- 6% the
activity obtained with L-biopterin). Thus, the ability of
oxidized pterins to sustain growth of Leishmania was
correlated with their ability to serve as PTR1 substrates.
Several reduced pterins were also examined (Table IV). As expected,
H4biopterin supported growth in all lines. Remarkably, H2biopterin and H2neopterin also supported
growth of the ptr1 mutant. Previously, this
was attributed to the anticipated ability of DHFR-TS to reduce
H2biopterin; however, DHFR-TS lacks this activity (Table
II). Last, H2sepiapterin and H4-6-methylpterin behaved as good pteridine nutrients in that they supported wild-type growth but, unlike the other H2pteridines, failed to
support growth of the ptr1
mutant.
Catalytic Properties of PTR1We have purified and
determined the enzymatic properties of recombinant and native PTR1 from L. major and recombinant PTR1 from L. tarentolae.
These enzymes exhibited similar physical and catalytic properties,
indicating that PTR1 does not undergo Leishmania-specific
modification, and validating the use of the recombinant enzyme for more
detailed studies. All PTR1s displayed good activity with both pterins
(biopterin and others; Tables I, II, and IV) and folates (Tables I and II). However, there were significant variations in the catalytic properties among pteridine substrates, with PTR1 activity on pterins exhibiting a sharper, more acidic pH optimum relative to folates, and
H2pterins and H2folate both showing significant
substrate inhibition. The results also show that PTR1 is capable of
reducing oxidized pteridines completely to the tetrahydro form.
The properties of PTR1 may be compared with other well-known pteridine reductases, such as DHFR and dihydropteridine reductase (DHPR). Although DHPR shows sequence similarities placing it in the "short chain dehydrogenase family" with PTR1 (16-19), PTR1 is more closely related to other members of this family and does not exhibit activity with "quinonoid" H2biopterin (10). Conversely, DHPR does not exhibit activity with folates or H2pterins, other than those in the quinonoid form (19, 41).
DHFRs from various sources exhibit activity with both folates and pterins (42, 43) but, unlike PTR1, are much less active with folate than H2folate. Substrate inhibition has been observed previously with folate and H2folate with the Lactobacillus casei DHFR (44) and with a number of H2pterins with rat DHFR (42). The latter finding was attributed to either a lack of reducing agents in the assay mix or the presence of inhibitory pterins such as biopterin. However, biopterin would not inhibit PTR1 nor did reductants affect the activity (data not shown). Substrate inhibition is thus an intrinsic property of PTR1, perhaps arising from allosteric interactions of tetrameric PTR1, or mechanisms described previously with other proteins (31). Substrate inhibition is often considered non-physiological since, when present, it often occurs at high substrate levels. Current data suggest that the intracellular levels of folates and biopterin are 2-20 µM in Leishmania (7, 8, 29), but the activities of PTR1 and DHFR would be expected to keep the levels of H2pteridines low. When inhibited by the action of antifolates, H2pteridine levels could rise to a point where substrate inhibition could be significant.
Although the substrate specificities of PTR1 resemble those of DHFRs from other species, differences in catalytic mechanism relative to that of DHFRs were evident in the pH dependence of PTR1 activity. Typically DHFRs display weak activity with folate that is optimal around pH 5, whereas much higher activity is observed with H2folate with two pH optima around pH 5 and 7 (44-48). In contrast, PTR1 activity was comparable with folate and H2folate, with a pH optimum around 6 (Fig. 3).
Despite its shared evolutionary ancestry with the short chain dehydrogenase family which includes DHPR, the properties of PTR1 have converged on those of DHFR, albeit with important catalytic differences. A similar process may have occurred independently with the prokaryotic type II DHFRs, which lack sequence or structural homology to chromosomal DHFRs (49). How PTR1 independently attained its role as a novel pteridine reductase is an interesting question in the evolution of catalytic pathways. Currently, we are pursuing studies of the detailed catalytic mechanism and three-dimensional structure of PTR1 in our effort to shed light on this process.
A Comprehensive Model for Pteridine Metabolism in LeishmaniaOur findings have permitted us to develop a general
model for pteridine metabolism in Leishmania (Fig.
5), which provides a convenient framework for evaluating
current data and developing future studies. The evidence for this
model, and its implications to pteridine metabolism and
chemotherapeutic inhibition, is discussed below.
The Role of PTR1 in Pterin Salvage
We have tested and confirmed the proposal that PTR1 was responsible for salvage of oxidized pteridines (10) in several ways. First, the ability of L. major to grow on a wide range of oxidized pterins correlates well with their activity as PTR1 substrates (Table IV). Good PTR1 substrates support Leishmania growth, and poor substrates require elevated PTR1 levels to support growth. Notably, the most physiologically abundant pterins in mammals, neopterin and biopterin, are the best substrates for PTR1 activity and Leishmania growth, whereas insect pterins such as xanthopterin are inactive (Table IV). Our findings are also in good agreement with results presented previously for growth of Leishmania donovani (8) and growth and altered PTR1 expression in L. tarentolae (17, 50), suggesting that PTR1 plays the same role in all Leishmania species. Second, deletion of PTR1, but not DHFR-TS, resulted in loss of the ability to grow on oxidized pterins (Table IV). Third, DHFR-TS showed no activity with pterins such as biopterin (Table II) nor did overproduction of DHFR-TS alter the pterin growth profile of Leishmania (Table IV). Thus, PTR1 alone accounts for salvage of oxidized pterins in Leishmania.
The Relative Contributions of PTR1 and DHFR-TS to Pteridine MetabolismAlthough DHFR-TS plays no role in the reduction of pterins, PTR1 possesses significant activity with folates (Tables I and II). By studying the reduction of folate and H2folate in Leishmania crude extracts, from wild-type and lines lacking PTR1 or DHFR-TS, we were able to assess their relative contributions to pteridine metabolism. For H2folate, more than 90% of the activity arose from DHFR-TS, a finding supported by calculations based upon the levels of PTR1 and DHFR-TS protein and their respective specific activities (Table III). However, for folate discrepant results were obtained. Comparisons of the null mutants suggested that more than 80% of the activity was contributed by DHFR-TS, whereas we calculated that 98% of this activity should arise from PTR1. We were unable to reconcile this difference, despite extensive testing and variation of experimental conditions, and it may reflect the existence of other activities not yet accounted for in our studies (below). Minimally, genetic deletion studies establish the dependence of the cellular folate reductase activity upon the presence of either PTR1 or DHFR-TS. For this reason, Fig. 5 depicts DHFR-TS as the major path of folate reduction within Leishmania.
What Is Responsible for Reduction of H2biopterin?The ptr1 mutant
was shown to grow normally on H2biopterin alone (Table IV)
(10). Previously this was attributed to an expected H2biopterin activity of DHFR-TS; however, we showed here
that DHFR-TS lacks this activity (Table II). One explanation postulates the existence of an enzyme, "PTR2," possessing
H2biopterin but not biopterin reductase activity. An enzyme
exhibiting activity with both H2biopterin and
H2folate, but not biopterin and folate, has been described
previously in the related trypanosomatid Crithidia (51, 52),
and alternative pteridine reductases unrelated to either DHPR or DHFR
have been detected in E. coli (53). Thus far, we have not
been able to detect H2biopterin reductase in crude
preparations derived from ptr1
L. major (data not shown).
A number of studies have demonstrated that the trypanosomatid growth requirement for folate can be reduced or even eliminated by inclusion of pterins such as biopterin (5-11) (Table IV). Although growth studies can be compromised by the presence of trace contaminants, incorporation of radiolabeled biopterin into folates has been shown in L. donovani (9), suggesting the occurrence of a de novo synthetic pathway. In contrast, another study failed to find incorporation of radiolabeled para-aminobenzoic acid into folate in L. major (54), which would be expected assuming that folates are synthesized by the classic route of dihydropteroate synthase. Most dihydropteroate synthase inhibitors are ineffective in Leishmania (55-57), and the few that are active show an independent, non-folate based mode of action (6). Thus, the mechanism of pterin/folate interconversion is not specifically indicated in Fig. 5.
What Is the Role (If Any) of Biopterin in Leishmania?The role of biopterin in trypanosomatids is unknown. In other organisms, H4biopterin plays a key role in the hydroxylation of phenylalanine and tyrosine, cleavage of ether-linked lipids, and the biosynthesis of nitric oxide (13, 14, 58, 59). However, trypanosomatids lack phenylalanine hydroxylase activity (60), and recently we have shown that ether-linked lipid cleavage uses NADPH rather than H4biopterin as a cofactor (61). Thus, it is conceivable that Leishmania does not use H4biopterin directly.
However, H4biopterin has been demonstrated in the related
trypanosomatid Crithidia, and Crithidia and
Leishmania both possess DHPR activity, which in other
organisms is responsible for recycling the quinonoid
H2biopterin formed by enzymatic use of
H4biopterin (Fig. 5) (10, 62). Second, improvements in
defined media and methodology suggest that L. major is in
fact unable to grow in the presence of folate alone and that previous
results from our lab to the contrary reflect the occurrence of a pterin
breakdown product in most folate
preparations.3 Moreover, neither folate nor
H2folate can rescue the growth defect of
ptr1 Leishmania (10). Thus, pterins
are required for Leishmania growth independently of their
role in folate biosynthesis. Third, recently we have shown that PTR1
levels, by affecting the formation of reduced cellular biopterin,
affect the sensitivity of Leishmania to
oxidants.3 Cumulatively, these data point to an essential
role of H4biopterin in Leishmania
metabolism.
Amplification of the Leishmania PTR1 gene within the H region is often observed in MTX-resistant Leishmania (reviewed in Refs. 50 and 63). The data in this work now provide a clear rationale for this process. As an alternative H2folate reductase with 4000-fold less sensitivity to MTX than DHFR at physiological pH (500 versus 0.13 nM; Table I), PTR1 is poised to provide a metabolic "by-pass" of DHFR-TS inhibition (10). However, due to its weaker contribution (relative to DHFR-TS; Table III), PTR1 overexpression by gene amplification is apparently necessary to provide sufficient activity. Since the Ki for MTX inhibition is greater than 300 nM at pH 7 for all reactions performed by PTR1 (Table I), overexpression of any of these could also contribute to relieving inhibition of DHFR-TS, by increasing H2folate pools indirectly through increased utilization of biopterin or directly by reduction of folate (Fig. 5).
The sensitivity of Leishmania to antifolates is dramatically
affected (several orders of magnitude) by exogenous folate levels (6,
7, 10). For example, to show antifolate inhibition of the amastigote
stage infecting macrophages, a folate-free medium was required (64).
Modulation of antifolate inhibition also has been noted in the malaria
parasite Plasmodium falciparum (65). In contrast, mammalian
cells show relatively little effect and lack oxidized pteridine
reductase activity (51, 66). Under conditions where DHFR-TS is
inhibited, the ability of PTR1 in wild-type Leishmania to
synthesize reduced folates could play a significant role in the
modulation of MTX potency. Consistent with this,
ptr1 Leishmania show
hypersensitivity to MTX (10, 11).
Thus, for reasons both genetic and biochemical, future strategies oriented toward antifolate inhibition of Leishmania should include inhibition of PTR1. In this regard, we have identified an inhibitor which shows good potency against both DHFR-TS and PTR1 activities, as well as Leishmania promastigote and amastigote growth, in medium containing physiological folate levels.3 The principles established here promise to lead to improved chemotherapeutic inhibition of this important parasite, and in the future we hope to incorporate insights garnered from the three-dimensional structures of both DHFR-TS (34) and PTR1 in the search for clinically effective anti-parasite agents targeting this pathway.
In summary, improved understanding of the properties and roles of PTR1 and DHFR-TS in pteridine metabolism has permitted the establishment of a comprehensive model incorporating current knowledge of pteridine metabolism in Leishmania. This model provides a useful framework for formulating and testing new hypotheses of pterin metabolism and has led to an increased understanding of the question of antifolate inhibition and chemotherapy of Leishmania.
We thank Dr. Chen-Chen Kan of Agouron Pharmaceutical Inc. for advice, members of the Walsh lab at Harvard Medical School for helpful discussions on kinetic parameters of PTR1, and James Luba for pointing out the occurrence of substrate inhibition with H2pteridines. We also thank Alexandre Bello for assistance with Leishmania culture in defined media and heterologous expression of PTR1 and D. Dobson, L. Epstein, L. A. Garraway, F. Gueiros-Filho, D. Kwon, and J. Moore for critical comments on the manuscript.