In Vitro Activities of 2,4-Diaminoquinazoline and 2,4-Diaminopteridine Derivatives against Plasmodium falciparum

The activities of 28 6-substituted 2,4-diaminoquinazolines,2,4-diamino-5,6,7,8-tetrahydroquinazolines, and 2,4-diaminopteridinesagainst Plasmodium falciparum were tested. The 50% inhibitoryconcentrations (IC₅₀s) of six compounds were <50 nM, andthe most potent compound was 2,4-diamino-5-chloro-6-[N-(2,5-dimethoxybenzyl)amino]quinazoline(compound 1), with an IC₅₀ of 9 nM. The activity of compound1 was potentiated by the dihydropteroate synthase inhibitordapsone, an indication that these compounds are inhibitors ofdihydrofolate reductase. Further studies are warranted to assessthe therapeutic potential of this combination in vivo.

INTRODUCTION

Top

Introduction

Inhibition of the synthesis of folate derivatives has been exploitedfor the development of drugs with activities against Plasmodiumfalciparum infection. The enzyme target of antifolates is dihydrofolatereductase (DHFR), which catalyzes the reduction of dihydrofolateto tetrahydrofolate. The latter is an essential cofactor inthe synthesis of thymidine, purines, and methionine (10). Inhibitorsof DHFR, such as pyrimethamine and chlorcycloguanil, are usedin combination with inhibitors of dihydropteroate synthase (DHPS),a second key enzyme of folate metabolism in P. falciparum. Anumber of sulfa drugs (inhibitors of DHPS), including sulfadoxineand dapsone (DDS), are known to potentiate the activities ofDHFR inhibitors or to act synergistically with them. One combinationalready used for a number of years for the mass treatment ofmalaria in Africa is pyrimethamine-sulfadoxine (SP). However,the therapeutic value of this combination has been significantlydiminished recently as a result of drug resistance (6-9, 11,12, 15). A new antimalarial combination consisting of chlorproguanil(whose active metabolite is chlorcycloguanil) and DDS has beendeveloped as an alternative to pyrimethamine-sulfadoxine (4,7, 16). Because of the intrinsic ability of P. falciparum toquickly develop different mechanisms of drug resistance, neweragents are urgently needed.

As part of an effort to discover potent new antifolates thatpotentially may be drugs with activities against multidrug-resistantstrains of P. falciparum, we have assessed the activities ofa library of 2,4-diaminoquinazolines, 2,4-diamino-5,6,7,8-tetrahydroquinazolines,and 2,4-diaminopteridines against the highly resistant V1S strainof P. falciparum. The compounds were chosen on the basis ofthe findings presented in a recent paper (5) documenting theiractivities against Saccharomyces cerevisiae yeast cells whoseDHFR genes were replaced with the P. falciparum DHFR gene.

MATERIALS AND METHODS

Top

Materials and Methods

The compounds were synthesized at the Dana-Farber Cancer Institute,Boston, Mass., by procedures described elsewhere (5) and arelisted by name in Table 1. Pyrimethamine and DDS were purchasedfrom Sigma Aldrich Co., St. Louis, Mo. Chlorcycloguanil wasa gift from AstraZeneca, Cheshire, United Kingdom.

View this table:
[in this window]
[in a new window]
TABLE 1. In vitro activities of 2,4-diaminoquinazoline, 2,4-diamino-5,6,7,8-tetrahydroquinazoline, 2,4-diaminothieno[2,3-d]pyrimidine, and 2,4-diaminopteridine derivatives against multidrug-resistant isolate P. falciparum V1S^a

The highly pyrimethamine-resistant V1S isolate that was usedin the assays has four mutations in its DHFR gene: Asn-51-Ile(a mutation from asparagine to isoleucine at codon 51), Cys-59-Arg,Ser-108-Asn, and Ile-164-Le. This DHFR genotype is associatedwith a substantial decrease in the affinity of binding of pyrimethamineto the enzyme, explaining why parasites that carry these mutationsare highly resistant to pyrimethamine (13). In vitro analysesof growth inhibition were performed in RPMI 1640 medium (GIBCO)containing physiological concentrations of para-aminobenzoicacid (0.5 µg/liter) and folic acid (10 µg/liter),10% (vol/vol) normal human serum, 25 mM bicarbonate, and 25mM HEPES buffer. Antimalarial activities in the presence ofvarious concentrations of each test compound were determinedby the radioisotope incorporation method (14). Results wereexpressed as the drug concentration required for 50% inhibitionof [³H]hypoxanthine incorporation into parasite nucleic acid(IC₅₀) by regression analysis of the dose-response curve.

Synergy was analyzed by calculating the sum of the minimum fractionalinhibitory concentrations (sFICs) (1). Synergy is demonstratedwhen the sFIC is <0.5. An sFIC >4.0 denotes an antagonisticeffect, and an sFIC between 0.5 and 4 indicates either a nonsynergisticor a nonantagonistic interaction.

RESULTS AND DISCUSSION

Top

Results and Discussion

The in vitro activities of the test compounds against pyrimethamine-resistantisolate V1S are shown in Table 1. Quinazoline compounds 1, 2,and 3 and tetrahydroquinazoline compounds 8 and 9 were the mostactive, with IC₅₀s of <50 nM; compound 1 had the lowest IC₅₀(ca. 9 nM). The potency of compound 1 in our assay was greaterthan those of pyrimethamine (IC₅₀, 4,200 nM) and chlorcycloguanil(IC₅₀, 81 nM) and, to our pleasant surprise, approached thatof WR99210 (IC₅₀, 3 nM), the most potent antifolate describedto date against P. falciparum. Interestingly, compound 2, whichdiffered structurally from compound 1 only by the fact thatit is a 3,4,5-trimethoxy derivative as opposed to a 2,5-dimethoxyderivative, showed a ca. threefold loss in potency relativeto that of compound 1. The N-methyl derivative (compound 3)was equipotent with compound 2, whereas the 2-bromo-3,4,5-trimethoxyanalogue (compound 4) was sevenfold less potent than compound2, suggesting greater tolerance for substitution on the bridgenitrogen than at the ortho position of the benzyl group. Elongationof the bridge by one carbon, as in compound 5, led to 157-foldless potency relative to that of compound 1. Three of the tetrahydroquinazolineanalogues (compounds 8, 9, and 10) were reasonably active, withIC₅₀s <50 nM. However, the thienopyrimidine and pteridineanalogues proved to be very weak inhibitors, with IC₅₀s >1,000nM in the majority of cases, and thus, they were clearly ofless interest than the quinazolines.

Although the compounds in Table 1 were not tested for theireffects on mammalian cells as part of this study, it may benoted that compounds 1 and 2 had IC₅₀s of 85 ± 8.0 and22 ± 4.0 nM, respectively, when they were tested in vitroat the Dana-Farber Cancer Institute against CCRF-CEM human leukemiclymphoblasts grown for 72 h in standard RPMI 1640 medium supplementedwith 10% fetal bovine serum (unpublished results). Thus, whilecompound 1 was found to be more potent than compound 2 againstP. falciparum in the present work, the opposite appears to bethe case with regard to human cells, presumably reflecting subtlespecies-specific differences in the three-dimensional structureof the active site of DHFR in P. falciparum versus that in humans.However, bearing in mind that the antimalarial assays were basedon [³H]purine (from hypoxanthine) incorporation into nucleicacids, whereas the assays of activities against human cellswere based on cell growth, this conclusion would have to beverified by directly comparing the activities of these compoundsagainst purified enzymes.

It has been known for more than 50 years that the combinationof a DHPS inhibitor and a DHFR inhibitor can synergisticallyblock de novo folate synthesis in P. falciparum and the othermicroorganisms in which this pathway is essential for growth(2, 3). Pyrimethamine-sulfadoxine and chlorproguanil-DDS areexamples of drug combinations that take advantage of this effect.Because a number of the quinazolines tested in this study hadpreviously been found to inhibit the P. falciparum DHFR geneexpressed in yeast (5), we postulated that these dicyclic compounds,too, would likewise act synergistically with DHPS inhibitorsin retarding the growth of intact P. falciparum organisms inculture. Accordingly, the most potent compound in Table 1, compound1, was tested in culture in the presence of various concentrationsof DDS. The results are presented in Table 2. The IC₅₀s of compound1 and DDS alone were 9 and 184,300 nM, respectively. In thepresence of 9,200, 6,100, and 4,600 nM DDS, compound 1 IC₅₀swere reduced to 0.08, 0.09, and 0.12 nM, respectively; sFICswere between 0.037 and 0.061, a clear indication that DDS actsin synergy with compound 1. In comparison, we have includeddata on the activity of chlorcycloguanil, a well-establishedDHFR inhibitor, in combination with DDS (Table 2). DDS increasedthe activity of chlorcycloguanil; however, the range of chlorcycloguanil-DDSsFICs was higher (0.26 to 0.38) than that for compound 1-DDS,an indication that the latter combination is more synergistic.All this information supports our hypothesis that compound 1and, presumably, the other active compounds in Table 1 are inhibitorsof DHFR.

View this table:
[in this window]
[in a new window]
TABLE 2. In vitro activities of the combinations of compound 1 and-DDS, and chlorcyloguanil-DDS against P. falciparum V1S

In summary, this paper reports that several 6-substituted 2,4-diaminoquinazolinesand 2,4-diamino-5,6,7,8-tetrahydroquinazolines are potent inhibitorsof the growth of the highly pyrimethamine-resistant strain V1S,with IC₅₀s of <50 nM. One compound, compound 1 (IC₅₀, 9 nM),was more potent against this strain than pyrimethamine by 2orders of magnitude and was nearly as potent as WR99210. Theseresults point to compound 1 as a promising lead for furtherstructure-activity optimization with the goal of designing newdrugs of the quinazoline family, and further studies are warrantedto assess the therapeutic potential of the combination compound1-DDS in vivo, in an animal model, for the treatment of pyrimethamine-resistanthuman malaria.

ACKNOWLEDGMENTS

We thank the director of Kenya Medical Research Institute forpermission to publish these data.

This work was supported by the Wellcome Trust of Great Britain(grants 056769 and 062372) and the U.S. National Institutesof Health (NIH Fogarty International grant TW 01186). E.N.,A.N., E.M., and K.M. are grateful to the Wellcome Trust forpersonal support.

FOOTNOTES

* Corresponding author. Mailing address: Kenya Medical Research Institute (KEMRI)/Wellcome Trust Collaborative Research Program, Wellcome Trust Research Laboratories, P.O. Box 43640 GPO, 00100 Nairobi, Kenya. Phone: 254-2-2710672. Fax: 254-2-2711673. E-mail: anzila{at}wtnairobi.mimcom.net.

REFERENCES

Top