Background

In the past decades, international travel has increased and more than 125 million international travellers visit malaria endemic countries every year [1]. This has turned the attention of scientific and public health authorities to the problem of imported and introduced malaria. The leading cause of imported malaria is Plasmodium falciparum [2–6] and antibody reaction to the parasite circumsporozoite antigen among international travellers has shown a high prevalence of inoculation [7]. Infection is especially relevant in sub-Saharan Africa travellers [7].

Accordingly, the number of imported malaria cases has increased. There are approximately 30,000 cases of imported malaria notified per year in non-endemic countries with important morbidity and mortality [2–6, 8]. Both these indices can be lowered by adequate prophylaxis. Indeed, most cases of malaria are due to non-compliance or inadequate prophylaxis [4, 5, 9]. It is, therefore, imperative to find a chemoprophylactic antimalarial with a good tolerability and safety profile as well as an attractive prophylactic regimen.

The fixed-dose combination of 250 mg atovaquone and 100 mg proguanil per tablet (Malarone®) has recently been introduced for treatment and prophylaxis of P. falciparum malaria. Based on its similarity to ubiquinol, atovaquone acts on P. falciparum cytochrome b (pfcytb), affecting the parasite mitochondrial respiratory chain and collapsing the mitochondria membrane potential [10–12]. This leads to apoptosis [12]. In addition, the combination also blocks dihydroorotate dehydrogenase, inhibiting pyrimidine synthesis [11, 13]. Proguanil enhances the ability of atovaquone to collapse the mitochondrial membrane by a mechanism still unexplained [14]. The combination effectively inhibits the development of the liver and blood stages of P. falciparum [15–17].

As recently reviewed atovaquone-proguanil is highly effective as prophylaxis for up to six months [18, 19] with an improved tolerability profile, a more convenient prophylactic regimen and better compliance.

It is now recommended as a prophylactic alternative to mefloquine or doxycycline in chloroquine-resistant areas [20]. Nevertheless, in many countries it is considered the first line antimalarial for prophylaxis in areas of chloroquine resistance and for imported malaria treatment (including emergency treatment) [21, 22].

There have been some reports of atovaquone-proguanil treatment failure in travellers, associated to pfcytb gene mutations, particularly at codon 268, namely T802A and A803C [23–27]. Epidemiologic surveillance of emerging resistance to this combination can be done by screening for these mutations.

The presence of atovaquone-proguanil in vivo resistance conferring mutations in pfcytb gene in Luanda, Angola, was investigated. This was done in order to estimate the prevalence of these mutations in this region and make recommendations on prescribing this antimalarial as prophylaxis for travellers.

Methods

Infected blood samples were obtained from children younger than 12 years of age hospitalized at Luanda Pediatric Hospital, Angola, during the years of 2003/2004. Both the hospital and IHMT Ethics Committee approved all the applied protocols.

An estimated prevalence of mutation T802A and A803C of 0.77 and 0.96%, respectively, was considered for population size calculation [27]. For an expected prevalence of 0.77%, with a significance level of 99%, the size of the studied population should be at least 213 children (StatCal, EpiInfoVersion 6, CDC, Atlanta). Samples were individually spotted on Whatman no 4 filter paper, after microscopic confirmation of P. falciparum infection. Chelex DNA extraction was done according to established protocols [28]. A modified semi-nested PCR-RFLP method was carried out to search for polymorphisms in pfcytb (T802A and A803C) [24, 29]. Primer sequences are presented in Table 1. All amplifications were performed in an I-Cycler-IQ, BIO-RAD, thermocycler. The first amplification using the "OD/OR" primers was designed to produce a 716 bp fragment in the following conditions: 92°C (three minutes) (1st segment – 1 cycle); 92°C (30 seconds), 57°C (30 seconds), 72°C (one minute) (2nd segment – 9 cycles); 92°C (30 seconds), 52°C (30 seconds) 72°C (one minute) (3td segment – 34 cycles); 72°C (three minutes) (4th segment – 1 cycle). This amplicon contained all the single point polymorphisms. Reagent concentrations were: 1 mM MgCl2 (Fermentas, Lithuania), 10 μM primers (MWG-Biotech AG, Germany), 1 mM dNTPs (Fermentas, Lithuania), 120 U Taq polymerase (Fermentas, Lithuania), 500 ng DNA template from field samples. Outer PCR product was further diluted (1:50) and 1 μL of this solution was used for the second amplification.

Table 1 Primer sequences used for the amplification of pfcytb codon 268.
Full size table

In the second amplification (nested or semi-nested amplification), three different pairs of primers were used, namely OD/WtR; 802D/802R; 803D/OR to distinguish between the three polymorphisms. The following conditions were used for amplication: 92°C (three minutes) (1st segment – 1 cycle); 92°C (30 seconds), 65°C (30 seconds), 72°C (one minute) (2nd segment – 9 cycles); 92°C (30 seconds), 60°C (30 seconds) 72°C (one minute) (3td segment – 34 cycles); 72°C (three minutes) (4th segment – 1 cycle). Reagent concentrations were the same as for the first amplification. The products of the second amplification were confirmed by electrophoresis in ethidium bromide-stained 3% agarose gel.

For RFLP analysis, 4 μL of each nested product was used, with 0.1 U of each enzyme and specific buffer in a 20 μL final volume, incubated overnight at 37°C. Table 2 summarizes primer pairs and restriction enzymes used for the identification of each polymorphism.

Table 2 Primer pairs and restriction enzymes used for the identification of each pfcytb codon 268 polymorphism.
Full size table

The laboratory-established clone K1 of P. falciparum (pfcytb wild type) was used as amplification reaction intrinsic control. Pfmdr1 gene from the 3D7 clone and serca gene from Plasmodium chabaudi have a cutting site for Dra-I and Cai-I, respectively. They were used as restriction reaction positive controls.

Results

Two hundred and forty nine blood samples were collected. DNA was successfully amplified from 224. Figure 1 illustrates the result of restriction of each of the second amplification products with the corresponding enzyme.

Figure 1
figure 1

Restriction digests for detection of pfcytb codon 268 mutations on field samples. A. 600 bp amplification product with the primer pair for wild type (TAT) detection, digested by Mph1103-I; lane 1 and 7 – non-digested PCR product; lane 2 – K1 wild type; lanes 3–6, 8 field isolates. MWM – 100 bp molecular weight marker. B. 146 bp amplification product with the primer pair for mutation T802A (AAT) detection, digested by Dra-I; lane 1 and 7 – non-digested PCR product; lane 2 – K1 wild type; lanes 3–6 field isolates. MWM – 100 bp molecular weight marker. C.173 bp amplification product with the primer pair for mutation A803C (TCT) detection, digested by Cai-I; lane 1 and 7 – non-digested PCR product; lane 2 – K1 wild type; lanes 3–6 field isolates. MWM – 100 bp molecular weight marker.

Full size image

Products of the amplification with primers for the identification of wild type pfcytb were restricted with Mph1103-I. Clone K1 and all field samples were cut. None of the field samples amplified with primers for identification of T802A (AAT) and A803C (TCT) mutations was cut with either Dra-I or Cai-I, respectively. According to the enzyme restriction pattern, all samples were identified as pfcytb codon 268 wild type.

Discussion

In the studied population, no mutations associated to atovaquone-proguanil treatment failure were found. Prevalence of these mutations in Luanda was estimated to be less than 0.77% with a 99% significance level. Therefore, Malarone® can be recommended for use by travellers to this region with expected high efficacy. This represents an improvement in view of the known prevalence of resistance against other currently used prophylactic antimalarials.

This was the first work focusing on atovaquone-proguanil treatment failure associated mutations ever done in this area. It gives baseline information of the prevalence of these mutations in the region. After a more extensive atovaquone-proguanil introduction in the market, future studies will be important to monitor these numbers. Screening for the same mutations in Guinea Bissau, Zanzibar and Ghana [29, 30] were also negative. Limited use of this antimalarial in endemic countries can explain the low prevalence of resistance-associated mutations. Restricted use by travellers will probably not increase pressure selection of mutants.

However, other cases of treatment failure, not associated to point mutations, have been described [27]. Study of prevalence of mutations in field samples, especially in treatment failure cases and travellers using the drug for prophylaxis, will help understanding the importance of these mutations in resistance. It is, therefore, imperative to continue surveillance.