Acta Tropica Use of Bacterial Surrogates As a Tool to Explore

Acta Tropica 149 (2015) 64–69

Contents lists available at ScienceDirect

Acta Tropica

journal homepage: www.elsevier.com/locate/actatropica

Use of bacterial surrogates as a tool to explore antimalarial drug interaction: Synergism between inhibitors of malarial dihydrofolate reductase and dihydropteroate synthase

Yuwadee Talawanich a, Sumalee Kamchonwongpaisan a, Worachart Sirawaraporn b, a, Yongyuth Yuthavong ∗

a National Centre for Genetic Engineering and Biotechnology, National Science and Technology Development Agency, 113 Thailand Science Park, Phahonyothin Road, Khlong Luang, Pathum Thani 12120, Thailand b Department of Biochemistry, Faculty of Science, Mahidol University, Rama VI Rd., Bangkok 10400, Thailand

a r t i c l e i n f o a b s t r a c t

Article history: Interaction between antimalarial drugs is important in determining the outcome of chemotherapy using Received 3 June 2014 drug combinations. Inhibitors of dihydrofolate reductase (DHFR) such as pyrimethamine and of dihy- Received in revised form 31 March 2015 dropteroate synthase (DHPS) such as sulfa drugs are known to have synergistic interactions. However, Accepted 13 May 2015 studies of the synergism are complicated by the fact that the malaria parasite can also salvage exogenous Available online 18 May 2015 folates, and the salvage may also be affected by the drugs. It is desirable to have a convenient system to study interaction of DHFR and DHPS inhibitors without such complications. Here, we describe the Keywords: use of Escherichia coli transformed with malarial DHFR and DHPS, while its own corresponding genes Bacterial surrogate Dihydrofolate reductase have been inactivated by optimal concentration of trimethoprim and genetic knockout, respectively, to Dihydropteroate synthase study the interaction of the inhibitors. Marked synergistic effects are observed for all combinations of Plasmodium falciparum pyrimethamine and sulfa inhibitors in the presence of trimethoprim. At 0.05 ␮M trimethoprim, sum of Antimalarial fractional inhibitory concentrations, �FIC of pyrimethamine with sulfadoxine, pyrimethamine with sul- Synergism fathiazole, pyrimethamine with sulfamethoxazole, and pyrimethamine with dapsone are in the range of 0.24–0.41. These results show synergism between inhibitors of the two enzymes even in the absence of folate transport and uptake. This bacterial surrogate system should be useful as a tool for assessing the interactions of drug combinations between the DHFR and DHPS inhibitors. © 2015 Elsevier B.V. All rights reserved.

1. Introduction 2005; Sibley and Price, 2012). Although the risk of adverse effects of sulfa drugs could complicate their deployment, the potential of With continuing global threat of malaria (WHO, 2012), there synergism between the DHFR and DHPS inhibitors is important, is an urgent need to search not only for new drugs, but also for not only for present therapy, but also for the understanding of the effective drug combinations. Combinations of inhibitors of Plas- nature of the synergism, which could form a basis in developing modium dihydrofolate reductase (DHFR) such as pyrimethamine new effective and safe combinations of antimalarial drugs. (PYR) and inhibitors of dihydropteroate synthase (DHPS) such as In Plasmodium falciparum, the synergism has been shown to be sulfa drugs known to have synergistic activities have long been dependent on the level of exogenous folate and p-aminobenzoate deployed in antimalarial chemotherapy (Bell, 2005; Wang et al., (pABA) and the ability of the parasite to salvage such exogenous 1999). However, the efﬁcacy of such combinations has been com- agents (Wang et al., 1999; Wang et al., 1997). Such dependence promised by emergence of resistance, primarily due to mutations affects the efﬁcacy of both inhibitors, and the degree of synergism of the genes coding for both DHFR and DHPS (Gregson and Plowe, between them. The inhibition of DHPS by sulfadoxine (SDX) can be bypassed by parasite uptake and utilization of exogenous folate, and PYR can also have an additional effect than inhibition of DHFR, namely, blocking of folate uptake. However, synergism between ∗ Corresponding author at. 113 Thailand Science Park, Phahonyothin Road, DHFR and DHPS inhibitors also exists in bacteria. Since bacteria like Khlong Nueng, Khlong Luang, Pathum Thani 12120, Thailand. Tel.: +66 2564 6698; Escherichia coli fax: +66 2564 7007. cannot utilize oxidized exogenous folates (Hussein E-mail address: [email protected] (Y. Yuthavong). et al., 1998; Noiriel et al., 2007), therefore complications due to such

http://dx.doi.org/10.1016/j.actatropica.2015.05.011 0001-706X/©Reproduced2015 fromElsevier ActaB.V. All Tropicarights reserved. 149: 64-69 (2015). Yuwadee Talawanich: Participant of the 23rd UM, 1995-1996.

232 233 Y. Talawanich et al. / Acta Tropica 149 (2015) 64–69 65 66 Y. Talawanich et al. / Acta Tropica 149 (2015) 64–69 salvage do not exist. Transformed bacteria carrying these enzymes supplemented with 0.4% arabinose, 100 ␮g/ml ampicillin, 34 ␮g/ml while their own enzymes have been knocked out genetically and/or chloramphenical, and 50 ␮g/ml kanamycin. Bacterial clones with chemically may be used as a tool to explore the synergism between positive growth complementation were selected for second trans- malarial DHFR and DHPS inhibitors. While this does not give com- formation by pBSDDRTS. The E. coli C600 folP::Kmr harboring plete information on the synergism of these inhibitors on treatment both pBSDPKDS and pBSDDRTS plasmids was grown on the same of Plasmodium, it can give useful basic information on drug inter- M9CA in the presence of 4 ␮M trimethoprim at 37 ◦C for 24 h to action at the enzyme level. In this paper, we present the results observe functional complementation of pBSDDRTS. on the interaction between PYR and different sulfa derivatives in E. coli transformed with P. falciparum DHFR and DHPS, while its 2.4. Measurement of IC50 values own corresponding enzymes have been inactivated. High degrees of synergism were observed with all tested combinations. The inhibitory activities of PYR and sulfa derivatives against bacterial cells expressing wild-type pfDHFR-TS and pfHPPK-DHPS 2. Materials and methods were monitored individually and in combinations using bacterial Fig. 1. Growth complementation of E. coli C600 folP::Kmr by P. falciparum DHPS and DHFR. The cells harboring different plasmids were streaked on M9CA agars in the r growth in liquid culture. All experiments were carried out at least in presence of thymidine or trimethoprim and incubated at 37 ◦C for 24 h. Streaks 1–4 are E. coli C600 folP::Km cells harboring pBAD18 + pBAD33 (streak 1), pBAD18 + pBSDDRTS (streak 2), pBSDPKDS + pBAD33 (streak 3) and pBSDPKDS + pBSDDRTS (streak 4). Plates A–C are LB agar plates supplemented with thymidine (A), M9CA plate (B) and M9CA 2.1. Materials duplicates and the concentration that yielded 50% inhibition (IC50- plate supplemented with 4 ␮M trimethoprim (C). value) was averaged from at least 3 independent experiments. 4-Aminophenyl sulfone (dapsone, DAP) was purchased from Briefly, an overnight culture in LB broth was diluted to A600 0.05 ∼ r Aldrich Chemical Company (USA). Pyrimethamine (PYR), trimetho- with liquid M9CA media containing 0.4% bacto casamino acid, 0.68% on sensitivity to DHFR inhibitors (Chusacultanachai et al., 2002) and and that they can rescue E. coli C600 folP::Km on thymidineless E. coli prim, sulfamonomethoxine (SMM), sulfathiazole (STZ), sul- Na2HPO4, 0.3% KH2PO4, 0.05% NaCl2, 0.1% NH4Cl, 2 mM MgSO4, for screening DHFR inhibitors (Bunyarataphan et al., 2006). An M9CA agar in the presence of trimethoprim. P. falciparum dhps fadimethoxine (SDM), sulfamethoxazole (SMX), and sulfamet- 0.1 mM CaCl2, and 0.4% glycerol supplemented with 0.4% arabinose, model with gene in place of the indigenous gene hazine (SMT) were from Sigma Chemical (USA). Sulfadoxine 100 ␮g/ml ampicillin, 34 ␮g/ml chloramphenical and 50 ␮g/ml was also constructed for analysis of sulfa resistance (Berglez et al., 3.2. Single drug susceptibility test (SDX) was a gift from Helm Mahaboon Ltd. (Thailand). Restric- kanamycin. To set up the inhibitor testing experiments, the diluted 2004). In order to study the antimalarial effect of the combined tion endonucleases and other DNA-modification enzymes were bacterial suspension (190 ␮l) was added the test inhibitor dissolved DHFR and DHPS inhibitors using a bacterial surrogate system, we In order to use E. coli C600 folP::Kmr expressing wild-type obtained from New England Biolabs and Promega. Plasmid DNA in DMSO (10 ␮l). The final concentration of DMSO in all reactions have mobilized both dhfr-ts and hppk-dhps genes of P. falciparum PfDHFR-TS and PfHPPK-DHPS system as a surrogate model for extraction kit, Gel purification kit and PCR purification kit were was kept constant at 0.5%. The concentrations of the inhibitors into the compatible expression pBAD plasmids and transformed the study of drug combination effect using trimethoprim to sup- r from Thermo Scientific (USA). All other chemicals, reagents and were varied from 0.195 ␮M to 1 mM. The culture was grown at them into E. coli C600 folP::Km of which the dhps gene was pre- press host DHFR, we first investigated trimethoprim effect on the media were from Sigma and Merck. pBAD18 and pBAD33 plas- 37 ◦C for 5 h with continuous agitation at 800 rpm in a microplate viously knocked out, while the host DHFR activity was inhibited inhibitory effect of a single antifolate drugs using liquid culture by trimethoprim, a specific bacterial DHFR inhibitor which also mids containing the arabinose PBAD promoter, (Guzman et al., 1995) incubator shaker (BioSan PST-60HL-4, Thermo-shaker). Bacterial condition. As summarized in Table 1, trimethoprim affects the IC50 inhibits PfDHFR but with less potency. r were used as vectors for expressing pfHPPK-DHPS and pfDHFR-TS, growth was measured at A600 using a microplate reader (Multi- of antifolates against E. coli C600 folP::Km expressing wild-type r Fig. 1 shows the results of folP complementation experiments. respectively. The DHPS-deficient E. coli C600 folP::Km (Fermer skan Ascent, Thermo Labsystems). The average A600 value from the PfDHFR-TS and PfHPPK-DHPS to different extent. In the absence E. coli C600 folP::Kmr cells transformed with plasmid carrying P. and Swedberg, 1997) was used as a host for the expression of control culture without inhibitor was used as 100% growth, and the of trimethoprim, PYR is inactive (IC50 > 50 ␮M) against the bacte- pfHPPK-DHPS and pfDHFR-TS. average reads of the tested culture were calculated by dividing the falciparum dhfr-ts gene (pBSDDRTS) and pBAD18 (streak 2), with rial surrogate model. At 0.05 ␮M trimethoprim, PYR is still inactive plasmid carrying P. falciparum hppk-dhps gene (pBSDPKDS) and absorbance by the absorbance of the control culture. The IC50-value (IC50 > 50 ␮M). At 4 ␮M trimethoprim, PYR becomes active with pBAD33 (streak 3) and with two plasmids carrying P. falciparum 2.2. Construction of expression plasmids was calculated using GraphPad Prism software version 5.0. IC50-value of 0.52 ␮M. This IC50-value of PYR is 6 times higher In some experiment, the diluted bacterial suspension was grown dhfr-ts and hppk-dhps genes, pBSDDRTS and pBSDPKDS (streak than the values reported for P. falciparum carrying the wild-type 4) and those with blank plasmids, pBAD18 and pBAD33 (streak The wild-type gene coding for pfHPPK-DHPS was amplified in 50 ml tubes as mixture of 2 ml cell suspension and 10 ␮l of the DHFR in vitro. Typically, the IC50 value of PYR against the wild- from pKOS-pfPPPK-DHPS (Kasekarn et al., 2004) using oligonu- test inhibitor for 8 h with continuous agitation at 200 rpm, 37 ◦C. 1) can grow normally on LB agar supplemented with thymidine type malaria parasite is less than 0.1 ␮M (Basco, 2003; Foote et al., (Fig. 1A). cleotide primers Nhe I-SDHPPK-f (5�-CGGCTAGCGGAGTGAA- Aliquots of cell suspension (200 ␮l) were taken out for measure- 1990; Khalil et al., 2003; Peterson et al., 1990; Zindrou et al., In the absence of thymidine supplementation, the bacterial cells ACGATGGAAACTATACAAGAACTAA-3�) and EcoR I-DHPS-r (5�- ment of turbidity using the microplate reader. 1996). A yeast surrogate model with replacement of the indigenous with blank plasmid (streak 1) and those carrying Pfdhfr-ts gene TCGAATTCCATGTTTGCACTTTCCTT -3�). The PCR reaction (50 ␮l) DHFR by P. falciparum wild-type (D6) enzyme was also reported to (streak 2) could not grow on M9CA agar, while those with plasmid containing 0.2 ␮g template DNA, 0.2 ␮M each of primers, 200 ␮M 2.5 Construction of isobolograms and measurement of FIC and give an IC50 value of PYR slightly lower than our bacterial model carrying Pf hppk-dhps gene could grow (streaks 3 and 4) (Fig. 1B). each of dNTPs and 1.5 U of Pfu DNA polymerase was initially FIC values (IC50 value = 0.35 ␮M)(Wooden et al., 1997). In our bacterial model, In the presence of trimethoprim, E. coli C600 folP::Kmr harbor- heated at 95 ◦C for 5 min, followed by 30 cycles of 95 ◦C for 30 s, trimethoprim is necessary for suppressing host DHFR which cannot ing only hppk-dhps gene could not grow on M9CA agar without 50 ◦C for 30 s, and 68 ◦C for 210 s. The resulting 2.1 kb of PCR Serial dilutions of each two combined inhibitors were mixed be inactivated by PYR. product was digested with Nhe I and EcoR I, and subsequently together in a checkerboard manner, i.e., the concentration of one thymidine supplementation since the endogenous DHFR activity Since it has been reported that trimethoprim also inhibits PfD- cloned into pBAD18 pre-digested with the same enzymes. The inhibitor was fixed while the concentrations of the other were was inhibited by trimethoprim (streak 3) (Fig. 1C). Only E. coli HFR with inhibition constant (K -value) of 10.3 nM (Sirichaiwat r i resulting recombinant plasmid, i.e., pBSDPKDS, was verified increased. The efficacy of the DHPS-DHFR combined inhibitors C600 folP::Km harboring plasmids for expressing PfDHFR-TS et al., 2004), which is only 3 times less effective than inhibiting and PfHPPK-DHPS could grow normally, even in the presence of by DNA sequencing. Likewise, the PfDHFR-TS expression plas- were monitored by determination of IC50 growth inhibition of the EcDHFR (Ki-value = 3.8 nM) (Zolli-Juran et al., 2003), the effect of mid was constructed using the same strategy; the wild-type bacterial cells as mentioned above. The fractional inhibition con- trimethoprim (streak 4). These confirm that both Pfhppk-dhps and trimethoprim on susceptibility of our bacterial surrogate model Pfdhfr-ts gene coding for PfDHFR-TS was amplified from pET17DHFR-TS centration (FIC) was calculated from the ratio of IC50 obtained expression plasmids produce functionally active enzymes to sulfa drugs was investigated. At high trimethoprim condition (Chitnumsub et al., 2004) using primers Sac I-SDDHFR-f (5�- upon using the combined inhibitors and IC50 obtained when single CGGAGCTCGGAGTGAAACGATGATGGAACAAGTCTGGGAGCTT-3�) inhibitor was used. The fractional Inhibition Concentration Index Table 1 r Individual IC50 and FIC values of PYR, sulfa drugs and dapsone against E. coli C600 folP::Km expressing pfDHFR-TS and pfHPPK-DHPS in the presence of trimethoprim to and Kpn I-DHFR-r (5�-GCCAGCGGTACCAATATTAAGCAGCCATATC- (FIC) for the combination of A and B is the sum of their individual suppress host DHFR. CATTGA-3�). The PCR reaction condition using Pfu DNA polymerase FIC values. The FIC values of the combination from at least three was 95 ◦C for 5 min, followed by 30 cycles of 95 ◦C for 30 s, 50 ◦C for individual experiments were used to calculate the mean values Inhibitor Control (no trimethoprim) 0.05 ␮M Trimethoprim 4 ␮M Trimethoprim 30 s, and 68 C for 240 s. The resulting PCR product was digested of FIC, which were then used to define synergism (FIC < 0.5), ◦ IC50 (␮M) IC50 (␮M) Estimated Mean FIC of IC50 (␮M) Mean FIC of PYR with Sac I and Kpn I, purified, and ligated with pBAD33 pre-digested additive (FIC = 1), and antagonism (FIC > 1.5) when the two PYR combination (f)* combination Sac Kpn inhibitors were combined. The interaction between two inhibitors with I and I. The plasmid obtained, i.e., pBSDDRTS, was Pyrimethamine (PYR) >50 >50 – 0.52 0.42 – ± verified for its sequence by DNA sequencing. was illustrated by isobologram plot of either IC50 or FIC values. Sulfadoxine (SDX) >500 325.1 65.2 0.39 < f < 0.41 21.10 1.85 0.08 0.06 ± ± ± Sulfathiazole (STZ) 45.5 16.7 15.9 8.0 0.30 < f < 0.34 0.50 0.19 0.08 0.01 ± ± ± ± r Sulfamethoxazole (SMX) 140.2 50.2 56.4 29.9 0.28< f < 0.31 2.20 0.73 ND** 2.3. Transformation of E. coli C600 folP::Km with Pf hppk-dhps 3. Results and discussion ± ± ± Dapsone (DAP) 185.9 87.6 53.6 28.8 0.24 < f < 0.28 1.31 0.20 0.08 0.02 and Pfdhfr-ts ± ± ± ± Sulfamonomethoxine (SMM) ND ND ND 0.63 0.29 0.07 0.02 ± ± 3.1. System verification by complementation assays Sulfadimethoxine (SDM) ND ND ND 3.91 0.55 ND ± The recombinant pBSDPKDS was transformed into E. coli Sulfamethazine (SMT) ND ND ND 3.87 1.67 ND r ± C600 folP::Km and was functionally verified by growth We have previously described the use of E. coli carrying Plas- * FIC-values estimated using FIC of sulfa for the low boundary and FIC of sulfa plus estimated FIC of PYR from IC50 – PYR alone of 50 ␮M as the high boundary. complementation on M9CA agar medium without thymidine, modium dhfr gene as a surrogate for testing the effect of mutations **Not determined.

234 235 Y. Talawanich et al. / Acta Tropica 149 (2015) 64–69 65 66 Y. Talawanich et al. / Acta Tropica 149 (2015) 64–69 salvage do not exist. Transformed bacteria carrying these enzymes supplemented with 0.4% arabinose, 100 ␮g/ml ampicillin, 34 ␮g/ml while their own enzymes have been knocked out genetically and/or chloramphenical, and 50 ␮g/ml kanamycin. Bacterial clones with chemically may be used as a tool to explore the synergism between positive growth complementation were selected for second trans- malarial DHFR and DHPS inhibitors. While this does not give com- formation by pBSDDRTS. The E. coli C600 folP::Kmr harboring plete information on the synergism of these inhibitors on treatment both pBSDPKDS and pBSDDRTS plasmids was grown on the same of Plasmodium, it can give useful basic information on drug inter- M9CA in the presence of 4 ␮M trimethoprim at 37 ◦C for 24 h to action at the enzyme level. In this paper, we present the results observe functional complementation of pBSDDRTS. on the interaction between PYR and different sulfa derivatives in E. coli transformed with P. falciparum DHFR and DHPS, while its 2.4. Measurement of IC50 values own corresponding enzymes have been inactivated. High degrees of synergism were observed with all tested combinations. The inhibitory activities of PYR and sulfa derivatives against bacterial cells expressing wild-type pfDHFR-TS and pfHPPK-DHPS 2. Materials and methods were monitored individually and in combinations using bacterial Fig. 1. Growth complementation of E. coli C600 folP::Kmr by P. falciparum DHPS and DHFR. The cells harboring different plasmids were streaked on M9CA agars in the r growth in liquid culture. All experiments were carried out at least in presence of thymidine or trimethoprim and incubated at 37 ◦C for 24 h. Streaks 1–4 are E. coli C600 folP::Km cells harboring pBAD18 + pBAD33 (streak 1), pBAD18 + pBSDDRTS (streak 2), pBSDPKDS + pBAD33 (streak 3) and pBSDPKDS + pBSDDRTS (streak 4). Plates A–C are LB agar plates supplemented with thymidine (A), M9CA plate (B) and M9CA 2.1. Materials duplicates and the concentration that yielded 50% inhibition (IC50- plate supplemented with 4 ␮M trimethoprim (C). value) was averaged from at least 3 independent experiments. 4-Aminophenyl sulfone (dapsone, DAP) was purchased from Briefly, an overnight culture in LB broth was diluted to A600 0.05 ∼ r Aldrich Chemical Company (USA). Pyrimethamine (PYR), trimetho- with liquid M9CA media containing 0.4% bacto casamino acid, 0.68% on sensitivity to DHFR inhibitors (Chusacultanachai et al., 2002) and and that they can rescue E. coli C600 folP::Km on thymidineless E. coli prim, sulfamonomethoxine (SMM), sulfathiazole (STZ), sul- Na2HPO4, 0.3% KH2PO4, 0.05% NaCl2, 0.1% NH4Cl, 2 mM MgSO4, for screening DHFR inhibitors (Bunyarataphan et al., 2006). An M9CA agar in the presence of trimethoprim. P. falciparum dhps fadimethoxine (SDM), sulfamethoxazole (SMX), and sulfamet- 0.1 mM CaCl2, and 0.4% glycerol supplemented with 0.4% arabinose, model with gene in place of the indigenous gene hazine (SMT) were from Sigma Chemical (USA). Sulfadoxine 100 ␮g/ml ampicillin, 34 ␮g/ml chloramphenical and 50 ␮g/ml was also constructed for analysis of sulfa resistance (Berglez et al., 3.2. Single drug susceptibility test (SDX) was a gift from Helm Mahaboon Ltd. (Thailand). Restric- kanamycin. To set up the inhibitor testing experiments, the diluted 2004). In order to study the antimalarial effect of the combined tion endonucleases and other DNA-modification enzymes were bacterial suspension (190 ␮l) was added the test inhibitor dissolved DHFR and DHPS inhibitors using a bacterial surrogate system, we In order to use E. coli C600 folP::Kmr expressing wild-type obtained from New England Biolabs and Promega. Plasmid DNA in DMSO (10 ␮l). The final concentration of DMSO in all reactions have mobilized both dhfr-ts and hppk-dhps genes of P. falciparum PfDHFR-TS and PfHPPK-DHPS system as a surrogate model for extraction kit, Gel purification kit and PCR purification kit were was kept constant at 0.5%. The concentrations of the inhibitors into the compatible expression pBAD plasmids and transformed the study of drug combination effect using trimethoprim to sup- r from Thermo Scientific (USA). All other chemicals, reagents and were varied from 0.195 ␮M to 1 mM. The culture was grown at them into E. coli C600 folP::Km of which the dhps gene was pre- press host DHFR, we first investigated trimethoprim effect on the media were from Sigma and Merck. pBAD18 and pBAD33 plas- 37 ◦C for 5 h with continuous agitation at 800 rpm in a microplate viously knocked out, while the host DHFR activity was inhibited inhibitory effect of a single antifolate drugs using liquid culture by trimethoprim, a specific bacterial DHFR inhibitor which also mids containing the arabinose PBAD promoter, (Guzman et al., 1995) incubator shaker (BioSan PST-60HL-4, Thermo-shaker). Bacterial condition. As summarized in Table 1, trimethoprim affects the IC50 inhibits PfDHFR but with less potency. r were used as vectors for expressing pfHPPK-DHPS and pfDHFR-TS, growth was measured at A600 using a microplate reader (Multi- of antifolates against E. coli C600 folP::Km expressing wild-type r Fig. 1 shows the results of folP complementation experiments. respectively. The DHPS-deficient E. coli C600 folP::Km (Fermer skan Ascent, Thermo Labsystems). The average A600 value from the PfDHFR-TS and PfHPPK-DHPS to different extent. In the absence E. coli C600 folP::Kmr cells transformed with plasmid carrying P. and Swedberg, 1997) was used as a host for the expression of control culture without inhibitor was used as 100% growth, and the of trimethoprim, PYR is inactive (IC50 > 50 ␮M) against the bacte- pfHPPK-DHPS and pfDHFR-TS. average reads of the tested culture were calculated by dividing the falciparum dhfr-ts gene (pBSDDRTS) and pBAD18 (streak 2), with rial surrogate model. At 0.05 ␮M trimethoprim, PYR is still inactive plasmid carrying P. falciparum hppk-dhps gene (pBSDPKDS) and absorbance by the absorbance of the control culture. The IC50-value (IC50 > 50 ␮M). At 4 ␮M trimethoprim, PYR becomes active with pBAD33 (streak 3) and with two plasmids carrying P. falciparum 2.2. Construction of expression plasmids was calculated using GraphPad Prism software version 5.0. IC50-value of 0.52 ␮M. This IC50-value of PYR is 6 times higher In some experiment, the diluted bacterial suspension was grown dhfr-ts and hppk-dhps genes, pBSDDRTS and pBSDPKDS (streak than the values reported for P. falciparum carrying the wild-type 4) and those with blank plasmids, pBAD18 and pBAD33 (streak The wild-type gene coding for pfHPPK-DHPS was amplified in 50 ml tubes as mixture of 2 ml cell suspension and 10 ␮l of the DHFR in vitro. Typically, the IC50 value of PYR against the wild- from pKOS-pfPPPK-DHPS (Kasekarn et al., 2004) using oligonu- test inhibitor for 8 h with continuous agitation at 200 rpm, 37 ◦C. 1) can grow normally on LB agar supplemented with thymidine type malaria parasite is less than 0.1 ␮M (Basco, 2003; Foote et al., (Fig. 1A). cleotide primers Nhe I-SDHPPK-f (5�-CGGCTAGCGGAGTGAA- Aliquots of cell suspension (200 ␮l) were taken out for measure- 1990; Khalil et al., 2003; Peterson et al., 1990; Zindrou et al., In the absence of thymidine supplementation, the bacterial cells ACGATGGAAACTATACAAGAACTAA-3�) and EcoR I-DHPS-r (5�- ment of turbidity using the microplate reader. 1996). A yeast surrogate model with replacement of the indigenous with blank plasmid (streak 1) and those carrying Pfdhfr-ts gene TCGAATTCCATGTTTGCACTTTCCTT -3�). The PCR reaction (50 ␮l) DHFR by P. falciparum wild-type (D6) enzyme was also reported to (streak 2) could not grow on M9CA agar, while those with plasmid containing 0.2 ␮g template DNA, 0.2 ␮M each of primers, 200 ␮M 2.5 Construction of isobolograms and measurement of FIC and give an IC50 value of PYR slightly lower than our bacterial model carrying Pf hppk-dhps gene could grow (streaks 3 and 4) (Fig. 1B). each of dNTPs and 1.5 U of Pfu DNA polymerase was initially FIC values (IC50 value = 0.35 ␮M)(Wooden et al., 1997). In our bacterial model, In the presence of trimethoprim, E. coli C600 folP::Kmr harbor- heated at 95 ◦C for 5 min, followed by 30 cycles of 95 ◦C for 30 s, trimethoprim is necessary for suppressing host DHFR which cannot ing only hppk-dhps gene could not grow on M9CA agar without 50 ◦C for 30 s, and 68 ◦C for 210 s. The resulting 2.1 kb of PCR Serial dilutions of each two combined inhibitors were mixed be inactivated by PYR. product was digested with Nhe I and EcoR I, and subsequently together in a checkerboard manner, i.e., the concentration of one thymidine supplementation since the endogenous DHFR activity Since it has been reported that trimethoprim also inhibits PfD- cloned into pBAD18 pre-digested with the same enzymes. The inhibitor was fixed while the concentrations of the other were was inhibited by trimethoprim (streak 3) (Fig. 1C). Only E. coli HFR with inhibition constant (K -value) of 10.3 nM (Sirichaiwat r i resulting recombinant plasmid, i.e., pBSDPKDS, was verified increased. The efficacy of the DHPS-DHFR combined inhibitors C600 folP::Km harboring plasmids for expressing PfDHFR-TS et al., 2004), which is only 3 times less effective than inhibiting and PfHPPK-DHPS could grow normally, even in the presence of by DNA sequencing. Likewise, the PfDHFR-TS expression plas- were monitored by determination of IC50 growth inhibition of the EcDHFR (Ki-value = 3.8 nM) (Zolli-Juran et al., 2003), the effect of mid was constructed using the same strategy; the wild-type bacterial cells as mentioned above. The fractional inhibition con- trimethoprim (streak 4). These confirm that both Pfhppk-dhps and trimethoprim on susceptibility of our bacterial surrogate model Pfdhfr-ts gene coding for PfDHFR-TS was amplified from pET17DHFR-TS centration (FIC) was calculated from the ratio of IC50 obtained expression plasmids produce functionally active enzymes to sulfa drugs was investigated. At high trimethoprim condition (Chitnumsub et al., 2004) using primers Sac I-SDDHFR-f (5�- upon using the combined inhibitors and IC50 obtained when single CGGAGCTCGGAGTGAAACGATGATGGAACAAGTCTGGGAGCTT-3�) inhibitor was used. The fractional Inhibition Concentration Index Table 1 r Individual IC50 and FIC values of PYR, sulfa drugs and dapsone against E. coli C600 folP::Km expressing pfDHFR-TS and pfHPPK-DHPS in the presence of trimethoprim to and Kpn I-DHFR-r (5�-GCCAGCGGTACCAATATTAAGCAGCCATATC- (FIC) for the combination of A and B is the sum of their individual suppress host DHFR. CATTGA-3�). The PCR reaction condition using Pfu DNA polymerase FIC values. The FIC values of the combination from at least three was 95 ◦C for 5 min, followed by 30 cycles of 95 ◦C for 30 s, 50 ◦C for individual experiments were used to calculate the mean values Inhibitor Control (no trimethoprim) 0.05 ␮M Trimethoprim 4 ␮M Trimethoprim 30 s, and 68 C for 240 s. The resulting PCR product was digested of FIC, which were then used to define synergism (FIC < 0.5), ◦ IC50 (␮M) IC50 (␮M) Estimated Mean FIC of IC50 (␮M) Mean FIC of PYR with Sac I and Kpn I, purified, and ligated with pBAD33 pre-digested additive (FIC = 1), and antagonism (FIC > 1.5) when the two PYR combination (f)* combination Sac Kpn inhibitors were combined. The interaction between two inhibitors with I and I. The plasmid obtained, i.e., pBSDDRTS, was Pyrimethamine (PYR) >50 >50 – 0.52 0.42 – ± verified for its sequence by DNA sequencing. was illustrated by isobologram plot of either IC50 or FIC values. Sulfadoxine (SDX) >500 325.1 65.2 0.39 < f < 0.41 21.10 1.85 0.08 0.06 ± ± ± Sulfathiazole (STZ) 45.5 16.7 15.9 8.0 0.30 < f < 0.34 0.50 0.19 0.08 0.01 ± ± ± ± r Sulfamethoxazole (SMX) 140.2 50.2 56.4 29.9 0.28< f < 0.31 2.20 0.73 ND** 2.3. Transformation of E. coli C600 folP::Km with Pf hppk-dhps 3. Results and discussion ± ± ± Dapsone (DAP) 185.9 87.6 53.6 28.8 0.24 < f < 0.28 1.31 0.20 0.08 0.02 and Pfdhfr-ts ± ± ± ± Sulfamonomethoxine (SMM) ND ND ND 0.63 0.29 0.07 0.02 ± ± 3.1. System verification by complementation assays Sulfadimethoxine (SDM) ND ND ND 3.91 0.55 ND ± The recombinant pBSDPKDS was transformed into E. coli Sulfamethazine (SMT) ND ND ND 3.87 1.67 ND r ± C600 folP::Km and was functionally verified by growth We have previously described the use of E. coli carrying Plas- * FIC-values estimated using FIC of sulfa for the low boundary and FIC of sulfa plus estimated FIC of PYR from IC50 – PYR alone of 50 ␮M as the high boundary. complementation on M9CA agar medium without thymidine, modium dhfr gene as a surrogate for testing the effect of mutations **Not determined.

234 235 Y. Talawanich et al. / Acta Tropica 149 (2015) 64–69 67 68 Y. Talawanich et al. / Acta Tropica 149 (2015) 64–69

(4 ␮M), the IC50 values for six sulfa drugs and DAP range from 0.5 ␮M for STZ to 21.1 ␮M for SDX (Table 1). These IC50 values are lower than those reported earlier where E. coli C600 folP::Kmr expressing only wild-type PfHPPK-DHPS was employed and no trimethoprim added. In one report, our IC50-value of SDX is about 10-fold lower (IC50 = 71 ␮g/ml or 230 ␮M) (Berglez et al., 2004) and in another report, our IC50-values of SDX, STZ, DAP, SDM and SMX are about 1.6 to 7 fold lower than the reported values (Kasekarn − − et al., 2004). In the condition of low trimethoprim (0.05 ␮M), the IC50 values of sulfa drugs increase to 15.9 ␮M for STZ to 325 ␮M for SDX. These values are slightly higher than the reported values; the IC50-value of SDX is 325 ␮M in this model comparing with 230 and 150 ␮M in the two reports. These suggest that the condition with low concentration of trimethoprim, which yields comparable IC50- sulfa results, is optimal for determining antifolates susceptibility in this bacterial model.

3.3. Drug combination assay

We then investigated the effect of trimethoprim on combi- Fig. 2. Isobolograms illustrating the synergistic interactions between PYR and STZ nation assay of sulfathiazole (STZ) pyrimethamine (PYR). Fig. 2 using bacterial surrogate system with various concentrations of trimethoprim. shows the isobolograms of STZ and PYR in the absence and presence of trimethoprim. We found that in all conditions, the system shows synergistic effect of PYR and STZ even in the absence of

a) b) 1 1

0.8 0.8

0.6 0.6

Fig. 4. Isobolograms showing marked synergistic interaction between PYR and sulfa drug in the presence of 4 ␮M trimethoprim (a) PYR-SDX, (b) PYR-STZ, (c) PYR-SMM, (d) FIC of SDX of FIC 0.4 FIC of STZ 0.4 PYR-DAP.

trimethoprim. The cells are more sensitive to PYR and STZ when the folate, ranging from 0.25 ␮M in its absence to an estimate of 100 ␮M 0.2 0.2 concentration of trimethoprim is increased, making the isobolo- in the presence of 45 nM folate (Wang et al., 1999). The degree of grams shift toward the left. At 0.25 ␮M and 4 ␮M trimethoprim, synergism is markedly reduced in the absence of exogenous folate. 0 the large synergy is observed. The synergistic effect of PYR and STZ The high degree of synergism in the presence of exogenous folate 0 0 0.2 0.4 0.6 0.8 1 in the absence of trimethoprim has made it unnecessary to add indicates that PYR may be acting, not only as DHFR inhibitor, but 0 0.2 0.4 0.6 0.8 1 trimethoprim to the system. Nevertheless, we found that adding also as inhibitor of folate uptake and/or utilization (Wang et al., FIC of PYR FIC of P YR low concentration of trimethoprim to partially suppress host DHFR 1999). It was furthermore shown that PYR and SDX do not act syn- c) d) of this bacterial surrogate system has eased the IC50 determination ergistically on P. falciparum DHFR, hence ruling out the possibility of the sulfa drugs in drug combination assay. that the synergism observed is due to simultaneous inhibition of the 1 1 Figs. 3 and 4 show isobolograms illustrating the interactions DHFR by the two inhibitors (Chulay et al., 1984). With P. chabaudi between PYR and the chosen sulfa inhibitors such as SDX, STZ, SMM, DHFR, it was shown that PYR and SDX have a slight synergistic 0.8 0.8 SMX and DAP using the bacterial surrogate system in the presence inhibitory effect, far lower than the synergism shown by the drugs of 0.05 ␮M and 4 ␮M trimethoprim, respectively. All the combina- on the parasite in vivo (Sirawaraporn and Yuthavong, 1986). How- tions show synergism. At 4 ␮M trimethoprim, the values of sum ever, even though the synergism may not depend on simultaneous 0.6 0.6 of fractional inhibitory concentrations ( FIC) of all combinations DHFR inhibition, it may still be due to interactive effect of the two DAP SMX f f are in the range of 0.07 - 0.08 (Table 1). Since IC50-values of PYR drugs acting at their corresponding target enzymes. In addition, alone could not be determined under the condition with 0.05 ␮M the pterin-sulfa adduct produced from DHPS reaction should also FIC o FIC FIC o FIC 0.4 0.4 trimethoprim, it is estimated that FIC values of all combinations be taken into consideration since it was found to exert antimalarial are in the range of 0.24–0.41 (Table 1). These FIC values obtained activity in vitro (Mberu et al., 2002). The same inhibitory effect by from low trimethoprim experiments are close to the values pre- sulfonamide-adduct was also observed in yeast (Patel et al., 2003) 0.2 0.2 viously reported for different Plasmodia; i.e., a value for FIC of but not in E. coli (Roland et al., 1979). However, the target of these 0.25 was reported for PYR–SDX combination against Plasmodium adducts has not been identiﬁed although the adduct may interfere 0 0 chabaudi in vivo (Sirawaraporn and Yuthavong, 1986). This model with the downstream enzymes in folate biosynthesis. The inter- 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 with low trimethoprim should therefore be optimal for investigat- pretation of synergism in malaria parasites is also complicated by FIC of PYR FIC of PYR ing PYR-sulfa drug combination. the presence of folate salvage which is not inhibited by PYR or The IC50 value for SDX against P. falciparum lines which can sal- sulfa drugs (Wang et al., 2007). It is therefore important to study Fig. 3. Isobolograms showing synergistic interaction between PYR and sulfa drug in the presence of 0.05 ␮M trimethoprim (a) PYR-SDX, (b) PYR-STZ, (c) PYR-SMX, (d) vage folate was shown to depend critically on the level of exogenous potential synergism between the DHPS and DHFR inhibitors, in the PYR-DAP.

236 237 Y. Talawanich et al. / Acta Tropica 149 (2015) 64–69 67 68 Y. Talawanich et al. / Acta Tropica 149 (2015) 64–69

3.3. Drug combination assay

0.8 0.8

0.6 0.6

236 237 Y. Talawanich et al. / Acta Tropica 149 (2015) 64–69 69 absence of folate salvage. Our E. coli bacterial surrogate system, Guzman, L.M., Belin, D., Carson, M.J., Beckwith, J., 1995. Tight regulation, which lacks such salvage, yields synergistic effect of PYR and sulfa, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177, 4121–4130. suggesting that indeed a direct inhibition of DHFR and DHPS is Hussein, M.J., Green, J.M., Nichols, B.P., 1998. Characterization of mutations that accounted, at least in part, for the observed synergism. This system allow p-aminobenzoyl-glutamate utilization by Escherichia coli. J. Bacteriol. may yield crucial information for drug development, even though 180, 6260–6268. Kasekarn, W., Sirawaraporn, R., Chahomchuen, T., Cowman, A.F., Sirawaraporn, W., the results need to be treated with caution due to differences from 2004. Molecular characterization of bifunctional hydroxymethyldihydropterin P. falciparum. pyrophosphokinase-dihydropteroate synthase from Plasmodium falciparum. The inhibition data by PYR and sulfa drugs reported thus far Mol. Biochem. Parasitol. 137, 43–53. show dependence on the condition and the strain of parasite used, Khalil, I., Ronn, A.M., Alifrangis, M., Gabar, H.A., Satti, G.M., Bygbjerg, I.C., 2003. Dihydrofolate reductase and dihydropteroate synthase genotypes associated reflecting the complexity of the synergistic interactions. Our sys- with in vitro resistance of Plasmodium falciparum to pyrimethamine, tem, utilizing a bacterial surrogate in place of the parasite, should trimethoprim, sulfadoxine, and sulfamethoxazole. Am. J. Trop. Med. Hyg. 68, be useful as a preliminary screening step to test the interac- 586–589. Mberu, E.K., Nzila, A.M., Nduati, E., Ross, A., Monks, S.M., Kokwaro, G.O., Watkins, tion of the malarial DHPS and DHFR inhibitors in the absence of W.M., Hopkins Sibley, C., 2002. Plasmodium falciparum: in vitro activity of folate salvage. The role of folate salvage has not been considered sulfadoxine and dapsone in field isolates from Kenya: point mutations in in this system, but could be addressed, for example, by further dihydropteroate synthase may not be the only determinants in sulfa resistance Plasmodium falciparum: in vitro activity of sulfadoxine and dapsone in field modifying the surrogate system to include incorporation of Plas- isolates from Kenya: point mutations in dihydropteroate synthase may not be modium folate transporters (Salcedo-Sora et al., 2011; Wang et al., the only determinants in sulfa resistance. Exp. Parasitol. 101, 90–96. 2007). Noiriel, A., Naponelli, V., Gregory 3rd, J.F., Hanson, A.D., 2007. Pterin and folate salvage plants and Escherichia coli lack capacity to reduce oxidized pterins. Plant physiol. 143, 1101–1109. Acknowledgements Patel, O., Satchell, J., Baell, J., Fernley, R., Coloe, P., Macreadie, I., 2003. Inhibition studies of sulfonamide-containing folate analogs in yeast. Microbial. Drug Resistance (Larchmont N.Y.) 9, 139–146. This work was supported by grants (to Y.Y.) from European Peterson, D.S., Milhous, W.K., Wellems, T.E., 1990. Molecular basis of differential Union (INCO-Dev) and Medicines for Malaria Ventures (MMV). S.K. resistance to cycloguanil and pyrimethamine in Plasmodium falciparum was supported by Howard Hughes Medical Institute and NSTDA’s malaria. Proc. Natl. Acad. Sci. U. S. A. 87, 3018–3022. Roland, S., Ferone, R., Harvey, R.J., Styles, V.L., Morrison, R.W., 1979. The Cluster Management Program. characteristics and significance of sulfonamides as substrates for Escherichia coli dihydropteroate synthase. J. Biol. Chem. 254, 10337–10345. References Salcedo-Sora, J.E., Ochong, E., Beveridge, S., Johnson, D., Nzila, A., Biagini, G.A., Stocks, P.A., O’Neill, P.M., Krishna, S., Bray, P.G., Ward, S.A., 2011. The molecular basis of folate salvage in Plasmodium falciparum: characterization of two Basco, L.K., 2003. Molecular epidemiology of malaria in CameroonXVI. folate transporters. J. Biol. Chem. 286, 44659–44668. Longitudinal surveillance of in vitro pyrimethamine resistance in Plasmodium Sibley, C.H., Price, R.N., 2012. Monitoring antimalarial drug resistance: applying falciparum. Am. J. Trop. Med. Hyg. 69, 174–178. lessons learned from the past in a fast-moving present. Int. J. Parasitol. Drugs Bell, A., 2005. Antimalarial drug synergism and antagonism: mechanistic and Drug Res. 2, 126–133. clinical significance. FEMS Microbiol. Lett. 253, 171–184. Sirawaraporn, W., Yuthavong, Y., 1986. Potentiating effect of pyrimethamine and Berglez, J., Iliades, P., Sirawaraporn, W., Coloe, P., Macreadie, I., 2004. Analysis in sulfadoxine against dihydrofolate reductase from pyrimethamine-sensitive Escherichia coli of Plasmodium falciparum dihydropteroate synthase (DHPS) and pyrimethamine-resistant Plasmodium chabaudi. Antimicrob. Agents alleles implicated in resistance to sulfadoxine. Int. J. Parasitol. 34, Chemother. 29, 899–905. 95–100. Sirichaiwat, C., Intaraudom, C., Kamchonwongpaisan, S., Vanichtanankul, J., Bunyarataphan, S., Leartsakulpanich, U., Taweechai, S., Tarnchompoo, B., Thebtaranonth, Y., Yuthavong, Y., 2004. Target guided synthesis of Kamchonwongpaisan, S., Yuthavong, Y., 2006. Evaluation of the activities of 5-benzyl-2,4-diamonopyrimidines: their antimalarial activities and binding pyrimethamine analogs against Plasmodium vivax and Plasmodium falciparum affinities to wild type and mutant dihydrofolate reductases from Plasmodium dihydrofolate reductase-thymidylate synthase using in vitro enzyme falciparum. J. Med. Chem. 47, 345–354. inhibition and bacterial complementation assays. Antimicrob. Agents Wang, P., Brobey, R.K., Horii, T., Sims, P.F., Hyde, J.E., 1999. Utilization of exogenous Chemother. 50, 3631–3637. folate in the human malaria parasite Plasmodium falciparum and its critical role Chitnumsub, P., Yuvaniyama, J., Vanichtanankul, J., Kamchonwongpaisan, S., in antifolate drug synergy. Mol. Microbiol. 32, 1254–1262. Walkinshaw, M.D., Yuthavong, Y., 2004. Characterization, crystallization and Wang, P., Read, M., Sims, P.F., Hyde, J.E., 1997. Sulfadoxine resistance in the human preliminary X-ray analysis of bifunctional dihydrofolate malaria parasite Plasmodium falciparum is determined by mutations in reductase-thymidylate synthase from Plasmodium falciparum. Acta Crystallogr. dihydropteroate synthetase and an additional factor associated with folate D Biol. Crystallogr. 60, 780–783. utilization. Mol. Microbiol. 23, 979–986. Chulay, J.D., Watkins, W.M., Sixsmith, D.G., 1984. Synergistic antimalarial activity Wang, P., Wang, Q., Sims, P.F., Hyde, J.E., 2007. Characterisation of exogenous folate of pyrimethamine and sulfadoxine against Plasmodium falciparum in vitro. Am. transport in Plasmodium falciparum. Mol. Biochem. Parasitol. 154, 40–51. J. Trop. Med. Hyg. 33, 325–330. WHO, 2012. World Malaria Report. WHO, Geneva. Chusacultanachai, S., Thiensathit, P., Tarnchompoo, B., Sirawaraporn, W., Wooden, J.M., Hartwell, L.H., Vasquez, B., Sibley, C.H., 1997. Analysis in yeast of Yuthavong, Y., 2002. Novel antifolate resistant mutations of Plasmodium antimalaria drugs that target the dihydrofolate reductase of Plasmodium falciparum dihydrofolate reductase selected in Escherichia coli. Mol. Biochem. falciparum. Mol. Biochem. Parasitol. 85, 25–40. Parasitol. 120, 61–72. Zindrou, S., Nguyen, P.D., Nguyen, D.S., Skold, O., Swedberg, G., 1996. Plasmodium Fermer, C., Swedberg, G., 1997. Adaptation to sulfonamide resistance in Neisseria falciparum: mutation pattern in the dihydrofolate reductase-thymidylate meningitidis may have required compensatory changes to retain enzyme synthase genes of Vietnamese isolates, a novel mutation, and coexistence of function: kinetic analysis of dihydropteroate synthases from N. meningitidis two clones in a Thai patient Plasmodium falciparum: mutation pattern in the expressed in a knockout mutant of Escherichia coli. J. Bacteriol. 179, 831–837. dihydrofolate reductase-thymidylate synthase genes of Vietnamese isolates, a Foote, S.J., Galatis, D., Cowman, A.F., 1990. Amino acids in the dihydrofolate novel mutation, and coexistence of two clones in a Thai patient. Exp. Parasitol. reductase-thymidylate synthase gene of Plasmodium falciparum involved in 84, 56–64. cycloguanil resistance differ from those involved in pyrimethamine resistance. Zolli-Juran, M., Cechetto, J.D., Hartlen, R., Daigle, D.M., Brown, E.D., 2003. High Proc. Natl. Acad. Sci. U. S. A. 87, 3014–3017. throughput screening identifies novel inhibitors of Escherichia coli Gregson, A., Plowe, C.V., 2005. Mechanisms of resistance of malaria parasites to dihydrofolate reductase that are competitive with dihydrofolate. Bioorg. Med. antifolates. Pharmacol. Rev. 57, 117–145. Chem. Lett. 13, 2493–2496.

238 239