HIV-1 pol Gene Polymorphism and Antiretroviral Resistance Mutations in Drug-Naive Pregnant Women in Yaoundé, Cameroon

To the Editor: A recent demographic and health survey was performed in Cameroon in 2004 that showed an HIV-1 seroprevalence rate of 5.5% in the general population.1 In Yaoundé, the capital city of Cameroon, HIV-1 prevalence is as high as 10% among pregnant women attending antenatal centers.2 Since the year 2000, a public health pilot program for the prevention of mother-to-child transmission started providing a single dose of the nonnucleoside reverse transcriptase (RT) inhibitor, nevirapine (NVP), to pregnant women who were diagnosed as HIV-1 positive in a maternity ward in Yaoundé.2 This program is now scaled to at least 60% of health districts of the country.3 One of the main portals of antiretroviral (ARV) entry in Cameroon is via the HIV-1 prevention of mother-to-child transmission program. To date, no information is available on ARV resistance in this sentinel population. The present study aimed to investigate whether genetic polymorphisms were present in RT and protease (PR) of HIV-1 genes at positions affecting the susceptibility to ARV and to evaluate baseline drug resistance among HIV-1-infected pregnant women, naive to NVP in Yaoundé, Cameroon, where HIV-1 non-B subtypes predominate. After informed consent was obtained, HIV-1-positive pregnant women who were willing to participate in the study were included and interviewed for ARV history. Blood samples (2 × 5 mL) were drawn by venous puncture on EDTA-treated tubes. CD4+ T-lymphocyte enumeration was performed by flow cytometry with a FACSCount (Becton-Dickinson, San Jose, CA) and plasma viral load was quantitated by bDNA (Bayer Diagnostics, Paris, France). Viral RNA for the genotypic resistance study was extracted from 1 mL of plasma by using the Qiagen viral RNA extraction mini kit (Qiagen, Inc., Chatsworth, CA) as recommended by the manufacturer. The RT gene (codons 1-255) and PR gene (codons 1-99) were amplified by using the primers MJ3/MJ4 and Prot1FW/Prot1RV,4-6 respectively. These first polymerase chain reaction (PCR) were followed by nested PCR with primers A35/NE1 for RT and Prot2FW/Prot2Rev 4,7 for PR. PCR products were sequenced on both strands by Genome Express (Meylan, France) using the BigDye Terminator v3.1 Cycle Sequencing Kit on a 3730xl DNA analyzer (Applied Biosystems, Foster City, CA). Prot2FW/Prot2Rev and A35/NE1 primer sets were used for the PR and RT, respectively. Sequences were manually corrected by using the Ceq2000 (Beckman Coulter) software using the HIV-1 HXB2 (GenBank Acc. Number: K03455) sequence as template. Genotypic drug resistance interpretation was performed according to the latest version (September 2005) of the French Agence Nationale de Recherche contre le Sida et les Hépatites algorithm (http://www.hivfrenchresistance.org/). Between 2001 and 2004, a total of 110 HIV-1-positive pregnant women consented to participate in the study. They were all ARV therapy-naive before the index pregnancy. One hundred ten amino acid sequences were obtained in the PR gene and 109 in the RT gene. In this report, we considered only samples with complete sequences for both strands in the PR and RT genes. Overall, 96 ARV therapy-naive pregnant women whose HIV-1 were full-length sequenced in both PR and RT genes were analyzed. Age at inclusion was 27 years on average (range, 16-41 years). HIV-1 status was known for an average of 7 months (range, 3-12 months). The mean T-lymphocyte count was 365 (range, 9-962 cells/μL) and viral load was, on average, 4.8 log10 RNA copies/mL (range, 1.7-5.7 log10 RNA copies/mL). Phylogenetic analysis of the 96 PR and RT sequences, together with Basic Local Alignment Search Tool search, revealed a high genetic diversity in both genes. Almost all known HIV-1 subtypes were found, with CRF02-AG representing, however, 63 of 96 (65.6%) sequences in PR and 56 of 96 (58.3%) in RT genes. Other circulating recombinant forms were marginally found: CRF11 (6.2% in PR and 5.2% in RT), CRF14 (5.2% in PR), and other subtypes represent 14 of 96 sequences in PR and 22 of 96 sequences in RT. Complex recombinant forms were encountered more often in the RT genes, where they represent up to 26% (25/96). Sequence analysis for genotypic drug resistance in the RT gene revealed no major mutations associated with HIV-1 ARV resistance. One sample (M4587N, HIV-1/F2 Beckman-Coulter, Nyon, Switzerland) presented the L210W and T215S amino acid mutations. These mutations confer a possible resistance to the nucleoside RT inhibitors, zidovudine and stavudine, and to the nucleotide analog RT inhibitor, tenofovir. In the PR gene, all of the 96 sequences analyzed harbored 2 mutations or more at positions associated with a lower susceptibility to PR inhibitors in subtype B HIV-1 strains (Fig. 1). More than 87% of the sequences harbored 5 to 9 mutations. Minor amino acid substitutions K20I, M36I, R41K, and H69K were found at 79%, 99%, 86%, and 95% frequencies, respectively. One sample (M3988N) showed the mutation N88S, conferring resistance to the PR inhibitor nelfinavir. Overall, 1 (M3988N, 1.04%) of 96 samples we analyzed in the RT and PR presented a major HIV-1 drug resistance mutation in the PR gene (N88S) that confers resistance to the PR inhibitor, nelfinavir. In the present study, we have shown that HIV-1 genetic diversity in both PR and RT genes is very high in viruses circulating in seropositive pregnant women in Yaoundé, confirming previous findings from the same country but targeting a mixed population of men and women.8,9 Genotypic drug resistance analysis of the pol gene uncovered no major drug resistance mutation but one (M3988N) in both RT and PR, despite the great genetic diversity observed. This lack of major drug resistance mutation is consistent with an ARV-naive population, in an area like Cameroon, where mass anti-HIV-1 drugs are of relatively recent introduction. In the PR gene, the high genetic diversity observed seems to be correlated to a high degree of polymorphism at positions known to be associated with drug resistance in subtype B HIV-1. This observation was repeatedly reported in various studies on samples obtained from Africa and, to a lesser extent, from southeast Asia.10,11 The implication of these polymorphisms at key amino acid positions in the PR gene remains a concern for long-course anti-PR treatment of HIV-1-infected patients in southern countries. In the course of the present study, a small group of 10 women who already experienced NVP in a previous pregnancy was screened as well. Analysis of corresponding samples did not reveal either the presence of NVP-related mutations or related to other ARV. It should be mentioned, however, that the time between the 2 pregnancies was always greater than 20 months for the 10 women. Hence, single-dose NVP-associated mutations fade up in the plasma by approximately 18 months after drug application, as previously reported.12,13 These findings also emphasize the urgent need for genotypic drug resistance interpretation algorithms adapted to nonsubtype B HIV-1 strains. Our present data indicate that less than 1% of HIV-1-positive, ARV-naive pregnant women in Yaoundé carried viruses with major genotypic drug resistance mutations. It would be very informative to repeat such a study to evaluate the incidence of HIV-1 drug resistance mutations among ARV-naive and ARV-experienced pregnant women in Cameroon and the impact of a greater availability of ARV to other subpopulations.FIGURE 1: HIV-1 PR gene polymorphism at position associated with drug resistance in non-B subtypes. HIV-1 PR gene (codons 1-99) were amplified and sequenced, and the corresponding amino acid sequence deduced. The figure shows the frequency of amino acids at indicated positions, different from the reference sequence of HIV-1 subtype B (HXB2). Black bars represent any amino acid different from that of HXB2. Dashed bars represent resistance-associated mutations if present in a subtype B HIV-1. The black arrow indicates the unique major drug resistance mutation found in our sample.Aurélia Vessière, MSc* Eric Nerrienet, PhD*† Anfumbom Kfutwah, MSc* Elisabeth Menu, PhD‡ Mathurin Tejiokem, MD* Patricia Pinson-Recordon, PhD‡ Françoise Barré-Sinoussi, PhD‡ Hervé Fleury, MD§ Ahidjo Ayouba, PhD*‡ *Centre Pasteur du Cameroun BP 1274 Yaoundé, Cameroun †Laboratoire VIH/Hépatites Institut Pasteur du Cambodge BP 983 Phnom Penh Kingdom of Cambodia ‡Institut Pasteur de Paris Unité Régulation des Infections Rétrovirales, Paris, France and §Laboratoire de Virologie EA2968 Université Victor Segalen Bordeaux, France