Manuscript

Selection of Resistance to Pyrimethamine and Atovaquone in a Mouse Model of

Suci Nuralithaa, Josephine E. Siregara, Jan Verhoefb and Sangkot Marzukia

a
Eijkman Institute for Molecular Biology, Jakarta, Indonesia
b
Eijkman-Winkler Institute of Medical and Clinical Microbiology, University Medical Center,
Utrecht, Netherlands

INTRODUCTION

Antimalarial drug resistance has emerged as one of the greatest challenges facing malaria control today.
The evolutionary selection of malaria parasites within an individual host is important in the emergence
of drug resistance, but is still not well understood [1]. The probability of de novo resistance mutation
selection during the initial phase of treatment depends on several external factors; the number of
parasites exposed to the drug, the concentrations of drug to which these parasites are exposed, the
pharmacokinetic and pharmacodynamic properties of the antimalarial drugs, the degree of resistance
that results from the genetic changes, the level of host defence (non-specific and specific immunity).

In experimental models, drug-resistance mutations can be selected without mosquito passage (i.e.
without meiotic recombination) by exposure of large numbers of malaria parasites (either in vitro, in
animals, orin the pastin volunteers) to sub-therapeutic drug concentrations (Peter, 1987).

The genetic events which confer antimalarial drug resistance to be independent of the drug. These can
be mutations in or changes in the copy number of genes relating to the drug’s target (e.g. Pfdhfr) or
pumps which affect intraparasitic concentration of the drug (e.g. PfMDR).

There is important to define the standard pattern of resistance selection process in general according to
ability that selection method applied could be reproduce and precise.

The highest rates of emergence of resistance in population are for pyrimethamine and atovaquone
(Peters 1987; Looareesuwan et al. 1996)
This study examined the selection of drug resistance to two antimalarial agents, pyrimethamine and
atovaquone, in a mouse malaria model. Pyrimethamine is an antimalarial whose biochemical
mechanism of action as inhibitor of dihydrofolate reductase is well understood; resistance to this drug is
associated with single genetic lesions in the nuclear dhfr gene [1]. Atovaquone's mechanism of action as
inhibitor of mitochondrial respiratory complex III) is similarly well established; resistance is associated
with single genetic lesions in the mitochondrial cytb gene [2].

We also employed the pure wild type clone of Plasmodium berghei to defeat preexisting mutant cells
accumulated during previous growth and passage.

Materials and Methods:

1.Malarial parasite

Plasmodium berghei ANKA strains (from LUMC, Leiden) maintained by serial blood passage and
inoculated intraperitoneally into 10-12 week old BALB/c mice at aprox 10 6 parasitized red blood
cells /mouse.

2. Determination of parasitemia blood level

Peripheral blood smears were prepared onced per day from tail vein bleeds. The thin films were fixed in
methanol (3 min) and then stained with Giemsa 10% (Merck). Parasitemia level was determined under
100x oil immersion light microscopy for estimation of the proportion of infected erythrocytes.

3.Determination of effective Dose (ED50)

Determination of effective dose (ED50) was done especially only for pyrimethamine. Atovaquone ED 50 is
0.001 reffer to our report before [Josephine et al] . An inoculum of 10 6 parasitized erythrocytes was
prepared by dilution of blood harvested in PBS pH 7.2. with the 1 ml dissposible syringe, 100 µl of
solution injected to the experimental mice. At the parasitemia level 1-5%, the P. berghei-infected mice
were treated by intraperitoneally injection of pyrimethamine with different dose, 0.003 to 1 mg kg -1 BW
and no therapy for control. Drug given daily for 3 consecutive days and still monitored until the day 5 th.

The dose response of parasite growth data was analyzed by logarithmic regression function using the
Sigma-plot computer program (SPSS incorporation, USA) to determine the ED 50.

4. Antimalarial drugs

Atovaquone was kindly provided by Dr. Marry Pudney of the Wellcome Research Laboratories, UK. The
stock solution was dissolved in dimethyl sulfoxide and stored at -20C. Pyrimethamine was purchased
from Merck Company, USA. One hundred times dose of stock solution was dissolved in sterilized pure
water with 1 % glacial acetic acid and stored at -20C.
5. In vivo selection of P. berghei resistant clones to incomplete treatment of antimalarial drugs

The parasites were treated daily by intraperitoneal injection of pyrimethamine or atovaquone. The
adequate dose of each drug divided into therapeutic dose and ten times ED50 dose. Plasmodium
berghei was inoculated intraperitoneally into BALB/c mice with approx 10 6 parasitized red blood
cells/mouse. Treatment was initiated at parasitemia level of 3-5% and was continued until it was
reduced to 0.2-0.4%, following recovery in level of parasitemia by the absence of the drug. This
incomplete regime was repeated for several cycles until resistance was observed, indicated by level of
parasitemia still increased in the treatment period. The stability of resistant phenotypes was established
by passage to new mice and challenge with initial dose of the antimalaria.

6. Isolation of the P. berghei wild type pure clone

A pure wild type strain of parasite were obtained by serial limiting dilution until only 1-10 parasitized red
blood cells ready to inoculated into BALB/c mouse. After parasite growth at level 3-5%, the similar
incomplete treatment method above was started. This selection resistant process only to therapeutic
dose of atovaquone or pyrimethamine.

7. Isolation of parasite DNA

Approximately 50µl of tail-blood mouse-infected parasites mix with heparin were obtained and kept in
1.5 ml Eppendorf tube. Parasites DNA was isolated from the tubes using saponin (Sigma-Aldrich, St.
Louis, MO, USA) and chelex-100 (Sigma-aldrich) technique (Wooden et al., 1993). Eppendorf tube
centrifugated by Sorvall MC 12 V at 4.000 g for 5 minutes and discard the supernatant. Washing with 1
ml phosphate buffered saline (PBS) pH 7.2 and centrifugated. This washed blood dissolved by cold 0.5%
saponin in PBS and homogenized. After incubated in ice for 5 minutes, centrifugated at 4.000g for 5
minutes. Wash the blood three times with 1 ml PBS and centrifugated at 4.000 for 5 minutes. Put 200µl
of the 20% v/v chelex-100 solution in ddH 2O (pH 9.2) into pellet and boiled for 8 minutes. Centrifugated
at 12.000 g for 10 minutes and supernatant that contain the DNA saved in the new eppendorf tube. DNA
was used immediately for PCR process or stored at -20C.

8. Polymerase Chain Reaction (PCR)

The total DNA obtained was directly amplified using 50µl of the PCR mixture: 5 µl of template, 5µl of 10x
PCR buffer (NEB Thermo), and 10 mM of dNTPs, 20 pmol of each primer, 0.25 µl of Tag DNA Polymerase.
To amplify the mitochondrial cyt b gene of P. berghei used the sense primer F1
CCTTTAGGGTATGATACAGC and antisense primer R1 GTTTGCTTGGGAGCTGTAATC (QO2 domain) and F2
TGCCTAGACGTATTCCTGAT and R2 TGATGTATCATACCCTAAAG (QO1 domain). The thermocycle condition
was: 34 cycles of denaturation at 94C for 15 sec, annealing at 55C (for first paired primer) and 52C
(for second paired primer) for 15 sec and extension at 72C for 2 min, except for the first cycle of which
denaturation is prolonged for 5 min. To complete the run-off product, final extension was perfomed at
72C for 5 min.
To amplify the dhfr gene of P. berghei used the PfF 99-316. The PCR was run in 29 cycles of denaturation
at 94C for 30 sec, annealing at 50C for 30 sec and extension at 65C for 2 min, except for the first cycle
of which denaturation is prolonged for 5 min. To complete the run-off product, further extension was
perfomed at 65C for 5 min

9. Sequencing

Results and Discussion

Features of the experimental system employed in the present study are illustrated in Fig. 1, which shows
cycles of incomplete treatment of P. beghei infected mice with adequate doses of the antimalaria
atovaquone, but interupted every time the decrease of parasitemia reached level of approx. 10% of that
at the start of the treatment, to allow recovery of parasitemia before repeat treatment. In this early
experiment, stable resistant phenotype was already observed during the second treatment cycle.
Similar early experiments with pyrimethamine also suggested early development of resistance.

To establish the generality and reproducibility of this phenomena, we repeated the experiments with
larger numbers of mice (Table 1). For pyrimethamine, the treatment time was found to be #-# days,
whereas the parasitemia recovery time varied between #-# days. Similarly, the treatment and the
parasitemia recovery times for atovaquone were found to vary between # to # days and # to # days,
respectively. The average number of treatment cycles for stable resistant phenotype to develop was
found to be #+# (mean+SD; n=#) for pyrimethamine. For atovaquone it was slightly faster (#+#).

What are the mutational changes underlying the resistance? Fig. 2 shows that Dfhr mutations detected
include S108N, the most common in Plasmodium falciparum. Signicantly, all atovaquone resistant
mutations found following treatment with the therapeutic dose of atovaquone (Y268N,Y268C, L271V+
K272R) are located in the Qo2 domain of the cytochrome b gene, while those associated the lower dose
of 10xED50 include M133I in the Qo1 domain.

Is the rapid appearance of antimalarial resistant mutants an anomaly, due unusually high number of
preexisting mutant cells accumulated during previous growth and passages of the laboratory P. berghei
strain? To ensure that this is not the case, we have cloned the P. berghei strain, and examined the
appearance of pyrimethamine and atovaquone resistance on the various clones developed. For both
antimalarial, the results (Table 2) confirm the generality of our finding that cycles of incomplete
treatment with pyrimethamine and atovaquone lead to rapid development of resistance to the
antimalarials in mice.
Conclusions:

Incomplete treatment of P. berghei infection in mice lead to rapid selection of resistance mutants
against pyrimethamine and atovaquone.

Pyrimethamine resistant mutation S108N in the dhfr gene identified in P. berghei is similar to that in P.
palcifarum; other dhfr gene mutations are expected.

Atovaquone mutations affecting the QO1 or QO2 domain of the cytochrome b observed suggest dose
dependent selection of mutations.

Figure and Table Legends:

Fig.1. Development resistance to atovaquone following incomplete treatment cycles with the antimalarial. (a)
resistant phenotype was observed in the second cycle of treatment, following # days of parasitemia recovery after
the # days first treatment cycle. (b) the resistance persisted following passage of the parasite to a new mouse to
test the stability of the resistant phenotype.

Fig. 2. Mutations underlying resistance to atovaquone affect the Qo1 or Qo2 domains of the cytochrome b protein.

Table 1. Rapid development of resistance to pyrimethamine and atovaquone following cycles of incomplete
treatment of P. berghei infected mice.

Table 2. Clones of the P. berghei laboratory strain also showed rapid development of resistance following cycles of
incomplete treatment of infected mice.