Background
Despite a clear need, an effective anti-malarial vaccine that offers a high level of protection against the disease has not yet become available. Chemotherapy is still the major tool in the fight against malaria. However, the rapid rise in drug-resistant malaria is a major factor compromising the use of current anti-malarial drugs. New drug candidates can be found either through random screening [
1] or from target-based drug development [
2]. In the latter approach, the major goal is to elucidate and characterize new drug targets against which inhibitor molecules can be designed and evaluated. This method can take advantage of the available
Plasmodium genome database and what is known about the metabolic processes of these parasites. The folate pathway is attractive for chemotherapeutic targeting, as it plays a crucial role in 1-C metabolism and purine biosynthesis [
3]. Several enzymes in this pathway such as dihydropteroate synthase (DHPS) and dihydrofolate reductase (DHFR) are validated targets for the clinical treatment of malaria infection. Nevertheless, there are other enzymes in the pathway that have received less attention which should be investigated, as they may prove to be more effective targets for new anti-folate development.
Serine hydroxymethyltransferase (SHMT; EC. 2.1.2.1) is a pyridoxal-5-phosphate (PLP) dependent enzyme and belongs to a member of the α-elimination and replacement reaction class [
4]. SHMT catalyses the conversion of L-serine and tetrahydrofolate (THF) to glycine and 5, 10-methylenetetrahydrofolate (5,10-CH
2-THF) [
5]. In addition to its role in dTMP synthesis, this reaction involves the cycling of folate derivatives required for various anabolic and catabolic reactions. The enzyme has been characterized from various organisms including
Plasmodium faciparum and
P. vivax[
6,
7]. The expression of the
Plasmodium SHMT gene is noticeably increased during late trophozoite to schizont stages when high levels of folate and nucleotides are needed for cell multiplication process, emphasizing the indispensable role of this enzyme [
8]. Unlike the SHMTs of other eukaryotes that are tetrameric enzymes [
9,
10],
Plasmodium SHMTs are dimers [
6,
7]. Furthermore, in contrast to other mammalian enzymes,
Plasmodium SHMTs can bind and use D-serine as a substrate [
6,
7]. Interestingly, the Food and Drug Administration (FDA) recently approved a new anti-folate drug, pemetrexed, for the treatment of cancer which inhibits several enzymes in the folate pathway including SHMT [
11]. Considering the central metabolic role of SHMT in the malarial parasite, it is likely to be a molecular target suitable for anti-malarial development [
6,
7,
12‐
14]. Therefore, further investigation into the mechanism of
Plasmodium SHMTs inhibition is of interest such that the possibility of developing specific inhibitors against the enzyme can be explored.
As the first step in developing a convenient method for obtaining a higher yield of SHMT, the study demonstrates that the use of an auto-induction system significantly improves the production of the recombinant
Plasmodium SHMTs in
Escherichia coli. A convenient spectrophotometric enzyme activity assay which does not require radioactive substrates or anaerobic conditions was developed, based on coupling the reactions of
Plasmodium SHMT with
E. coli 5,10-methylenetetrahydrofolate dehydrogenase (MTHFD). Inhibition of
Plasmodium SHMTs was investigated using anti-folate compounds previously synthesized as inhibitors against
Plasmodium DHFR [
15‐
17]. In addition, inhibition of
Plasmodium SHMTs by the amino acid analogue, thiosemicarbazide was explored. Results obtained from this study should be useful for the future rational design of new inhibitors of
Plasmodium SHMTs.
Methods
Chemicals and reagents
All chemicals used in the study were analytical grade. L-serine, NADPH, NADP+, PLP, polyethyleneimine (PEI) solution (50% w/v), D-glucose, N-Z-amine AS (casein enzymatic hydrolysate), thiosemicarbazide, and α-lactose were purchased from Sigma-Aldrich (St Louis, MO, USA). [6R,S] THF, [6 S] THF, and [6R] 5,10-CH2-THF were obtained from Merck Eprova AG (Schaffhausen Switzerland). D-cycloserine, dithiothreitol (DTT) and yeast extract were from Bio-Science Inc. (Allentown, PA, USA). Isopropyl thio-β-D-galactoside (IPTG) was purchased from Fermentas Life Sciences (Glen Burnie, MD, USA). All chromatographic media were purchased from GE Healthcare Biosciences (Uppsala, Sweden). N-(2-hydroxyethyl) piperazine-N’-(2-ethane-sulfonic acid) (HEPES) was purchased from Research Organics (Cleveland, OH, USA). Escherichia coli BL21 (DE3) (Novagen, Madison, WI, USA) was used as the host strain for protein expression.
Protein expression and purification
Two expression media types, LB-IPTG and auto-induction media were used to express the recombinant
Plasmodium SHMTs in an
E. coli system. Protein expression of Pf- and PvSHMT using LB-IPTG media was performed according to previous reports [
6,
7]. The auto-induction media used was modified from the standard formula previously described [
18]. Briefly, a starter culture was grown at 37°C overnight in ZYP-0.8G media (1% w/v N-Z-amine AS, 0.5% w/v yeast extract, 62.5 mM (NH
4)
2SO
4, 125 mM KH
2PO
4, 125 mM Na
2HPO
4, 1 mM MgSO
4, and 0.8% w/v D-glucose) supplemented with 50 μg/ml ampicillin. The starter culture (0.5% v/v) was inoculated in ZYP-5052 media (1% w/v N-Z-amine AS, 0.5% w/v yeast extract, 0.5% w/v glycerol, 0.2% w/v α-lactose, and 0.05% w/v glucose) containing 50 μg/ml ampicillin, and the culture was vigorously shaken at 37°C until the OD
600 reached ~1.0 (6–7 hours). The temperature was lowered to 16°C, and the cells were incubated at this temperature for 16–18 hours before they were harvested. Protein purification was carried out according to the procedures previously described [
6,
7], except that only a Ni-Sepharose column was used for PfSHMT purification. For long-term storage at −80°C, the purified PvSHMT was kept in 50 mM HEPES, pH 7 containing 0.5 mM EDTA and 1 mM DTT (Buffer A), and PfSHMT was kept in Buffer A with 10% v/v glycerol added (Buffer B). Unless otherwise indicated, biochemical studies of Pf- and PvSHMT were performed in Buffer A.
The expression and purification of
E. coli MTHFD was performed as described in [
19] with some modifications. Briefly, BL21DE3 carrying pET22b(+)::FolD was grown at 37°C until OD
600 reached 1.2, at which IPTG was added to 0.4 mM. Cells were cultured until OD
600 reached 5 before harvesting. Cell pellet was re-suspended in 50 mM potassium phosphate buffer pH 6.5, 1 mM DTT, 1 mM EDTA and 0.1 mM PMSF, and lysed by ultrasonication (Sonic Vibra cell
TM; model VCX750). MTHFD was precipitated using 0-30% ammonium sulfate and the protein precipitation was dissolved in 50 mM potassium phosphate buffer pH 6.5, 1 mM DTT, 0.3 mM EDTA (buffer C). The dissolved protein was dialyzed against buffer C and loaded onto a DEAE-column previously equilibrated with the same buffer. Proteins were eluted with a linear gradient of 0–300 mM NaCl in buffer C. The activity of MTHFD was determined spectrophotometrically by monitoring the increase in absorbance at 375 nm due to the formation of NADPH by the oxidation of 5,10-CH
2-THF. The purified MTHFD stored at −80°C was stable for at least three months.
Protein quantitation
The concentration of proteins was determined by the Bradford method [
20] using the standard dye reagent (Bio-Rad Life Science, CA, USA). The protein concentration was calculated from a standard curve using bovine serum albumin as a protein standard. Alternatively, protein concentrations were determined according to the enzyme UV-visible absorption using absorption coefficient values at 420 (5,400 M
-1 cm
-1), 422 nm (6,370 M
-1 cm
-1), and 280 nm (14,690 M
-1 cm
-1) for PfSHMT, PvSHMT, and MTHFD respectively [
6,
7]. The MTHFD absorption coefficient was calculated based on the primary amino acid sequence [
21].
SHMT activity assay
To monitor Plasmodium SHMT activity during enzyme preparation, the SHMT reaction was coupled with a MTHFD reaction (SHMT-MTHFD) and performed under regular aerobic conditions in Buffer A. A typical assay reaction contained 5 μM MTHFD, 2 mM L-serine, 0.4 mM THF, 0.25 mM NADP+, and SHMT in a final volume of 1 mL at 25°C. Progression of the reaction was monitored by an increase in absorbance at 375 nm. Measurement of steady-state kinetic parameters of Plasmodium SHMTs was performed using the MTHFD coupled assay with a rapid-mixing apparatus (SFA-20, TgK Scientific, Bradford-on-Avon, UK) connected to a double-beam spectrophotometer (SHIMADZU 2501 PC, Shimadzu corp., Kyoto, Japan). To prolong the stability of THF, a stock solution of THF was prepared in an anaerobic glove box. The apparent Michaelis constant (Kmapp) for THF was determined by fixing the concentration of L-serine at 2 mM and varying the concentration of THF between 0.025-0.4 mM. A similar set-up was used in determining Kmapp for L-serine, except that the concentration of THF was fixed at 0.4 mM and the concentrations of L-serine were varied between 0.05-1.6 mM. All concentrations indicated were final concentrations after mixing.
Inhibitor screening for Plasmodium SHMTs
Inhibition of SHMT was studied by measuring the initial rates of the reaction using the SHMT-MTHFD coupling system, as described in “SHMT activity assay” of the Methods section, in the presence of inhibitors. Inhibitors used in this study were anti-folates (2,4-diaminopyrimidine) and amino acid analogues (D-serine, D-alanine, D-threonine, L-allo-threonine, D-cycloserine and thiosemicarbazide). Stock solutions of anti-folates were prepared in absolute dimethyl sulfoxide (DMSO) and amino acid analogues were prepared in Buffer A. The final concentrations used for anti-folates were 0.05-0.5 mM, depending on the solubility of each compound. The final concentration for the amino acid analogues was 1 mM. The efficacy of the inhibitors is presented as % inhibition, which is a relative percentage of enzyme activity compared to the reaction in the absence of the inhibitor.
Kinetics of Plasmodium SHMT inactivation by thiosemicarbazide
Inactivation of Pf- and PvSHMT by thiosemicarbazide was investigated by monitoring the residual SHMT activity upon incubation of the enzyme with various thiosemicarbazide concentrations at various incubation times using a rapid-mixing apparatus connected to a double-beam spectrophotometer. One syringe of the rapid-mixing apparatus contained 1 μM Pf- or PvSHMT, 5 μM MTHFD and various thiosemicarbazide concentrations (0.03-1 mM). Another syringe contained 2 mM L-serine, 0.4 mM THF and 0.25 mM NADP+. All reactions were performed in Buffer A at 25°C and the reaction was initiated by mixing the solutions from both syringes. Time-dependent inactivation was performed by varying the incubation time (5–30 min) of enzyme with thiosemicarbazide in the first syringe before mixing with the solution in the second syringe.
The inactivation reaction appeared to follow first-order kinetics since a plot of ln
V/
V0versus time was linear.
V and
V0 represent initial velocities of the reaction in the presence and the absence of inhibitor, respectively. An observed rate constant (
kobs) at each thiosemicarbazide concentration was determined from a slope of the plot of ln
V/
V0versus incubation time. A rate constant for the inactivation step (
kinact) and the equilibrium dissociation constant for binding of the inhibitor (
KI) were calculated from Equation
1, where [I] is the concentration of the inhibitor, using non-linear algorithms found in KaleidaGraph software (Synergy Software, Reading, PA, USA).
(1)
Analysis of product from the inactivation of PvSHMT by thiosemicarbazide
The product that resulted from the inactivation of PvSHMT by thiosemicarbazide was analysed by UV-visible absorption, retention time analysis after HPLC separation, and molecular mass determination by LC-MS. PvSHMT with OD422 ~ 0.4 AU (62.79 μM) was incubated with 10 mM thiosemicarbazide for 50 min in Buffer A at 25 °C, and the absorption spectrum change was recorded. The enzyme was de-natured by adding SDS (final concentration of 1% w/v). The de-natured enzyme was separated from small molecular weight compounds by a Centricon device with a 10 kDa molecular weight cut-off membrane (Millipore, Carrigtwohill, Co. Cork, Ireland), and the spectrum of the filtrate was recorded.
The filtrate from ultrafiltration of the PvSHMT-thiosemicarbazide mixture was subjected to reverse phase HPLC chromatography (Polaris 3 C8-A, 50 x 4.6 mm; Agilent Technologies, Inc. Santa Clara, CA, USA). The column was pre-equilibrated with 25 mM sodium formate pH 4.3 and was eluted using the same buffer at a flow rate of 1 mL min-1. The eluted compounds were detected by UV-visible absorption.
Additionally, the filtrate was analysed by LC-MS (Bruker AXS Inc., Madison, WI, USA) to separate small molecules using a Polaris 3 C8-A column pre-equilibrated with 25 mM ammonium formate pH 6.5 at a flow rate of 0.5 mL min-1 at 25°C. Eluents were analysed for their masses using a linear ion trap MS equipped with an electrospray ionization (ESI) source. The parental and fragmented mass profiles were analysed. All buffers used were pre-filtered through a 0.45 mm membrane (Millipore, Carrigtwohill, Co, Cork, Ireland).
Similar experiments as described above were applied for free PLP (OD388 ~ 0.1 AU) in the presence of 10 mM thiosemicarbazide.
Fluorescence changes of Plasmodium SHMTs upon binding of amino acids
Changes in the fluorescence properties of Pf- and PvSHMT upon binding of amino acids and folate analogues were monitored using a spectrofluorophotometer (SHIMADZU RF5301 PC, Shimadzu corp., Kyoto, Japan) at 25°C. The emission and excitation monochromator slits were set at 5 nm, the light source was from xenon lamp (150 W), and the scanning rate was set at medium speed. The concentrations of free PLP, Pf- and PvSHMT were ~ 23 μM (PLP; OD388 ~ 0.12, PfSHMT; OD420 ~ 0.12, and PvSHMT; OD422 ~ 0.15). Free PLP, Pf- and PvSHMT were excited at the wavelengths 388, 420 and 422 nm, respectively. L-serine, D-serine, L-alanine, or glycine was added to the protein or PLP in Buffer A (at the above concentrations) to give a final amino acid concentration of 10 mM, except for folinic acid, which was added to a final concentration of 1 mM. The binding of folinic acid to Plasmodium SHMTs was performed in the absence and presence of 10 mM glycine. For the measurement performed in the presence of both ligands, the enzyme was incubated with glycine for 5 min prior to the addition of folinic acid.
Acknowledgements
This work was supported by grants from the Cluster Program and Management Office, National Science and Technology Development Agency (P-00-20029) to UL and PC, the Thailand Research Fund (BRG5480001) and Faculty of Science, Mahidol University to PC, and the Cluster Program and Management Office for Discovery based Development Grant (CPMO-DD/P-10-11274) to UL. We gratefully acknowledge Medicines for Malaria Venture (MMV) for the use of TV-P-0-113. KS is a recipient of a scholarship from Thailand Graduate Institute of Science and Technology (TGIST). We thank Dr Martino di Salvo (Università di Roma) and Merck Epova AG (Schaffhausen Switzerland) for providing the plasmid for MTHFD expression and high quality folate compounds, respectively. We also thank Dr Bongkoch Tarnchompoo, Dr Thichakorn Jittawuttipoka, and Dr Somchart Maenpuen for valuable discussion.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
KS performed the study and drafted the manuscript. CT and TV provided anti-folates. YY discussed and commented on the manuscript. PC and UL conceived of the study and drafted the manuscript. All authors read and approved the final manuscript.