Abstract: The purposes of this study were: (1) to describe the genetic variability of HIV strains found in Burkina Faso, (2) to characterize non-B HIV strains mutation profiles selected by ARVs and (3) to detect possible resistances induced by ARV drugs. From 30 October 2002 to 20 November 2003, 132 HIV1-positive patients taking Highly Active Antiretroviral Therapy (HAART) for more than one year in Bobo-Dioulasso and Ouagadougou were included. T-CD4+ lymphocytes count was done using Dynabeads technique while genotypic test and ARV-resistance tests were conducted using Pol sequencing that codes for reverse transcriptase reverse, integrase and protease. Due to undetectable viremia, 86 samples out of 132 could not be characterized. Whereas in the 46 others that had a viral load exceeding 1000 copies mL-1, the following HIV-1 subtypes were identified: CRF06 (54,55%); CRF02(38,63%); CRF01 (4,55%) and subtype A (2,27%). In addition, several mutations related to PI, NRTI and NNRTI resistance were isolated in 27 samples. This study found a huge genetic HIV-1 polymorphism in Burkina Faso. The level of acquired resistance to ARV after one year of treatment amounted 20.4%. These results clearly show that there is imperative need to set up an ARV resistance surveillance network in Burkina Faso to guide treatment strategies and follow the extension of the phenomenon in the country.
INTRODUCTION
Since the discovery of HIV-1 in 1983 (Vahlne, 2009; Miedema, 2008) and of HIV-2 in 1986 (Agrawal et al., 2010), the Human Immunodeficiency Viruses, the etiologic agents of the Acquired Immunodeficiency Syndrome (AIDS), kept propagating throughout the world, thus causing a pandemic with tragic consequences. HIV positive people who have resistance to antiretroviral drugs are often co-infected with HHV8 (Ilboudo et al., 2009), hepatitis B (Ilboudo et al., 2007), Toxoplasma gondii (Ouermi et al., 2009), pathogenic bacteria and intestinal parasites (Simpore et al., 2009). After 1987, the year when AZT (first antiretroviral) (De Clercq, 2009a; Zhang, 2010) marketing was authorized, many ARVs treating HIV infection were developed. They target various viral enzymes (reverse transcriptase, integrase and protease) and viral co-receptors. Today, these antiretroviral drugs belong to six classes: nucleoside/nucleotide and non-nucleoside reverse transcriptase analogues inhibitors, protease inhibitors, entry and fusion inhibitors and integrase inhibitors (De Clercq, 2009b). With the increasing availability of antiretroviral (ARV) drugs in developing countries, the emergence of ARV-resistant strains represents a major public health stake and its prevention constitutes one of the current treatment recommendations. In this context, the ARV-resistant mutated strains pose not only a direct problem in the management of the concerned patient but also an indirect challenge as they are likely to be transmitted to HIV-negative people. Viral resistance is related to mutations in the pol gene that codes for reverse transcriptase and protease, therefore modifying some amino-acids in these enzymes. Thus, their become insensitive to the concerned antiretroviral drugs. The first case of resistance to antiretroviral treatment was announced in 1989 (De Clercq, 2009a). In developing countries, social, economic and health conditions are factors conducive to a poor patient compliance and therefore to potential resistance emergence. Among the problems encountered, there are an anarchic circulation of molecules, problems in drug supply chain, lack of infrastructures, inadequate number of trained personnel, high cost of biological tests required for patient follow-up and the poor compliance of patients taking ARVs. It should be specified that the genetic diversity of HIV, particularly in Africa, is also a potentially important factor leading to natural resistance or emergence of resistance to some ARVs. Therefore, treatment efficacy could be influenced by HIV genetic diversity. Subsequently, HIV-2 and HIV-1 strains of group 0 are naturally resistant to Non-Nucleoside Analogue Reverse Transcriptase Inhibitors (NNRTI) (Wittkop et al., 2011). In group M, sub-type F strains are insensitive to NNRTI under trial (Almeida et al., 2009) and the sub-type G strains seem to be less sensitive to some protease inhibitors (IP) in vitro (Velazquez-Campoy et al., 2003). Moreover, mutations in nelfinavir resistance of sub-type G strains do not seem to be similar to those noted in sub-type B (Soares et al., 2010). A study carried out in Uganda suggests that sub-type D could more quickly develop a nevirapine resistance than sub-type A, in women receiving a single dose nevirapine to prevent mother-to-child HIV transmission (Eshleman et al., 2001). Besides, many minor mutations, in particular in protease gene, were discovered among treatment naive patients infected by non-B (Vergne et al., 2000; Simpore et al., 2007) HIV strains. These mutations are identified through genotypic tests. Therefore, characterization of HIV-1 strains is currently an irreplaceable test for patients facing therapeutic failures. The goal of this study was: (1) to describe the genetic variability of HIV strains found in Burkina Faso (2) to characterize non-B HIV strains mutation profiles selected by ARVs and (3) to detect possible resistances induced by ARV drugs.
MATERIALS AND METHODS
From October 30, 2002 to November 20, 2003 in Bobo-Dioulasso, Burkina Faso, we conducted a cross-sectional study. This study was done at the inception of the Highly Active Antiretroviral Theray (HAART) programme in Burkina Faso when less than 200 HIV patients were treated. We used Centre MURAZ Virology Laboratory in Bobo-Dioulasso and IRD (Institute of Research for Development) Virology Laboratory, a UNAIDS Collaborative Centre in Montpellier (France).
STUDY POPULATION AND PROCEDURES
The study population comprised HIV-1 patients under ARVs, regularly followed up in Bobo-Dioulasso at Internal Medicine Department, Sourô SANOU University Hospital) and Centre MURAZ and in Ouagadougou (Day Care Hospital, Yalgado OUEDRAOGO University Hospital). In addition to University Hospital, a small number of patients were followed up in various private and public health facilities in Ouagadougou: Saint Camille Medical Centre, Pissy District Hospital, Yentema Private Clinic, Office of the President Health Centre, Clinique de la Paix, Clinique Suka and the Burkinabe Association for Family Welfare (ABBEF) Health centre.
Inclusion criteria: The study involved patients who were taking ARV treatment for at least one year and have given their verbal consent.
Sample collection and processing: The 132 patients had venous blood samples collected in two EDTA tubes for medical tests.
Immunological testing: For the biological monitoring, the count of circulating TCD 4 lymphocytes was realized using Dynabeads technique in Centre MURAZ Microbiology-Immunology Laboratory.
Virological study: The Pol gene was sequenced at the IRD Virology Laboratory in Montpellier (France). For the sequencing, plasma and PBMC were separated by centrifugation on ficoll gradient. ARN was extracted from plasma using Qiamp viral RNA mini kit (QIAGEN France). Reverse transcription was realized using Expand Reverse Transcriptase enzyme (Boehringer Mannheim) and reverse IN3 primer (40 pmoles μL-1) 5’-TCTATBCCATCTAAAAATAGTACTTTCCTGA TTCC-3’. A pair of primers, G25REV (5’-GCAAGAGTTTTGGCTG AATGAG-3’) and IN3 (5’-TCTATBCCATCTAAAAAT AGTACTTTCCTGATTCC-3’) amplified a region of 2400 Pb which was used as matrix for the second amplification, with AV150 (5’-GTGGAAAGGAAGGAC ACCAAATGAAAG-3’) and polM4 primers (5’-CTATT AGCTGCCCCATCTACATA-3’). PCRs conditions (1st and 2nd rounds) were 5 min at 92°C (denaturation), followed by 35 cycles at 92°C for 20 sec (denaturation), 50°C for 30 sec (primers pairing), 72°C for 10 min (polymerization) with a final elongation lasting 10 min at 72°C. Amplification was verified on a BET-TBE-agarose gel (1%) at 120 V for 30 mn, with a molecular weight marker. Sequencing was realized using an automatic sequencer Applied Bio System 3100 with 16 capillaries. The Big Dye Terminator kit version 3.1 (Applied Biosystems) was used and the sequence reaction volume was 20 μL (2 μL of Tris-MgCl2 5X buffer, 1 μL of oligo at 3, 2 pmol μL-1, 2 μL of purified DNA 40 to 100 ng, 4 μL of premix and 11 μL of H2O). PCR conditions for sequencing were 25 cycles: 96°C for 20 sec for denaturation, 50°C for 30 sec for hybridation, 60°C for 4 min to reinforce Taq incorporation and a temperature drop to 4°C. The proteic sequences coding protease and RT were compared to a B consensus sequence for the detection of major and minor mutations using BETA TEST algorithm (http://hivdb.stanford.edu/). Pol gene phylogeny enables us to determine HIV-1 strains circulating in Burkina Faso.
Data analysis: Characteristics of the study population are described in number and percentages with their confidence intervals for categorical variables and in means and medians with their spread for continuous variables. The number and percentages of genetic polymorphism characterized in this population are reported overall. Resistance mutation profiles were interpreted with French National Agency for AIDS Research (ANRS) and Stanford algorithms. The number and percentage of patients with genotypic resistances detected are reported overall and according to treatment characteristics.
RESULTS AND DISCUSSION
Demographic and clinical characteristics of the study population: The study yielded the following results: the mean age of the 132 patients was 38.15 years±1.42 (15 to 64 years). Women represented 72% of patients with a 95% confidence interval spanned between 63.37 and 79.26. The mean of CD4 was 320 μL-1, range (25-1692). The distribution of the study population, according to Centers of Diseases Control and Prevention classification was as follows: stage A (18.18%), stage B (75.00%) and stage C (6.82%). The mean treatment duration for our HIV-infected patients was 14 months.
Treatment experiences: All patients received HAART in accordance with the first line protocol recommended in Burkina Faso. Hundred percent (100%) of the study population had received an NRTI; 73.5% had received an NNRTI; 28.0% had received a PI while 57.0% (12) of the patients were undergoing their 3rd line therapy and were more exposed to developing resistance mutations. Six patients who had interrupted their treatment presented resistance mutations. Forty patient who were in therapeutic failure had a CD4 rate <200 mm-3 after 14 months of treatment. First line treatment was maintained by 82 patients (62.12%): 2 NRTI plus 1 PI or 1 NNRTI. But the other patients (37.88%) had changed and gone through several lines of treatments. The main reasons for these changes in therapeutic lines were related either to poor treatment compliance, to HAART side effects or to therapeutic failures.
Pol gene characterization in study population: As shown in Fig. 1, the phylogenic analysis completed by bootscan analyses revealed genetic polymorphisms in the pol genes of HIV-1 strains found in Burkina Faso. Strains pol gene characterization (n = 44) showed an important genetic diversity of HIV-1 in Burkina Faso. The recombining CRF06 strain was predominant (54.55%). The recombining forms CRF02-AG (38.63%), CRF01-AE (4.55%) and sub-type A (2.27%) co-circulated in the study population.
Emergence of HIV-1 resistances to ARVs: Among the 132 patients, 86 (65.20%), who complied well with the treatment protocol, had undetectable viremia and therefore characterization of their HIV-1 strains was not possible. But 46 samples that had a viral load over 1000 copies mL-1 could be characterized and 27 other samples exhibited different mutation profiles associated with high rates of resistance to Protease Inhibitors (PI), to Nucleoside Reverse Transcriptase Inhibitor (NRTI) and to Non-Nucleoside Reverse Transcriptase Inhibitor (NNRTI). Figure 2 summarizes HIV-1 resistances to various ARVs detected in present study population.
Genotypic mutations associated with resistance to PI, NRTI and NNRTI: All sequenced samples have been analyzed at the level of possible sites of mutations leading to resistances in the sequence coding protease and reverse transcriptase. Table 1 summarizes the mutations found in the study population.
This study found a huge genetic HIV-1 polymorphism in Burkina Faso dominated by circulating recombinant forms CRF06-cpx (54.5%), CRF02-AG (38.6%) and CRF01-AE (4.6%) (Table 2). The level of acquired resistance to ARVs after one year of treatment amounted 20.4% (27/132). Present results agree with the literature which confirms the circulation of sub-type CRF06 in part of West Africa such Niger, Nigeria, Burkina Faso, Mali, Senegal and Côte d’Ivoire (Montavon et al., 2002; Nadembega et al., 2006; Simpore et al., 2007). The study population comprised a large majority of women (72%) which confirms the higher attendance of health centres by women as well as the feminization phenomenon of HIV. Thus, study presents a clear female predominance among PLWHIV/ AIDS (76.00%), among PLWHIV/ARV (66.23 %) and PLWHIV/AIDS/COTI (87.23%) (Montavon et al., 2002; Drabo et al., 2001).
| Fig. 1: | Radial phylogenic tree of nucleotide sequences of Pol gene (Reverse transcriptase, Protease) of reference strains of group M (in black) and of the 44 isolates of Burkina Faso strains from, 2002-2003 (in red). Boostraps (%) reflecting the strength of the branching are noted at the knots of the phylogenic tree |
| Fig. 2: | Percentage of any resistance detected according to ARV drug, Burkina Faso, 2002-2003. Anti-retrovirals used: AZT: Zidovudine; 3TC: Lamivudine; DDI: Didanosine; DDC: Zalcitabine; D4T: Stavudine; ABC: Abacavir; TDF: Ténofovir Disoproxyl Fumarate; NVP: Névirapine; EFV: Efavirenz; DLV: Delavirdine; IDV: Indinavir; SQV: Saquinavir; NFV: Nelfnavir; RTV: Ritonavir; APV: Amprenavir; ATV: Atazanavir; LPV: Lopinavir; INTR: Inhibiteurs Nucléosidiques de la Transcriptase Reverse; INNTI: Inhibiteur Non Nucléosidique de la Transcriptase Reverse; IP: Inhibiteur de la Protéase; ARV: Antiretroviral |
| Table 1: | Number of major and minor mutations inducing resistances to ARVs: PI, NRTI and NNRTI, Burkina Faso, 2002-2003 |
| Table 2: | Comparison of strains frequencies in present study and those found by Ouedraogo et al. (2003) (18) and Nadembega et al. (2006), all in Burkina Faso (19) |
| a→b: p = 0.977 (NS); a→c: p = 0.639 (NS); b→c : p = 0.624 (NS) d→-e: p = 0.635(NS); d→f: p = 0.919 (NS); e→f: p = 0.505 (NS) g→h: p = 0.949 (NS); g→i: p = 0.917 (NS); h→i : p = 0.549 (NS) NS: Non-Significant | |
In this sense, present findings confirm those of Pignatelli et al. (2006) and Drabo et al. (2001) who had also found a very high women’s attendance of health centres. The study population was relatively young with a mean age of 38.15 years±1.42 and extreme values ranging between 15 and 64 years. This mean age compares with data from a similar study conducted in Côte d’Ivoire (38±5) and Cameroon (Adje et al., 2001). The study population in our study as well as in a similar study conducted in Côte d’Ivoire reflects the population pattern in developing countries that show a mean age below 40 years (Toni et al., 2003). Thus, present findings overlap with UNAIDS findings that revealed that the 20-49 years age group is the largest one among people living with HIV/AIDS (Toni et al., 2003; Delfraissy, 2002; Sagna et al., 2010). After 14 months of treatment, CD4 median in the studied population was at 320 μL-1 [25-1692 CD4 μL-1]; which justifies the high rate of resistances (20.45%) found in our study population. However, the average rate of CD4 found in our study was higher than the 118 CD4 μL-1 rate found in Cameroon in a similar research with extreme values of 78-167 CD4 μL-1 (Adje et al., 2001). These results suggest that the Cameroonian patients were at a more advanced stage of AIDS at the time of the study. The characterization of our study population in clinical stages proposed by CDC shows that only 7% of our study, population was at the advanced C stage.
The recombining CRF06 strain was predominant (54.55%). However, the recombining CRF02-AG (38.63%), CRF01-AE (4.55%) forms and sub-type A (2.27%) also co-circulated in the study population. Present results tally with those found by Ouedraogo-Traore et al. (2003) and Nadembega et al. (2006) in studies carried out in Burkina Faso (Table 2). In all three studies, no statistically significant differences were found in strains frequencies: CRF06-cpx, CRF02-AG and A1. Among the 46 samples characterized, 27 (58.70%) persons presented resistances to the drugs currently used in triple therapy following a 14-month therapy. These data compare with findings of another study conducted in Abidjan in Côte d’Ivoire in which 57.4% of patients (n = 68) had acquired resistances after ARV (Adje et al., 2001). Nadembega et al. (2006) found 37.5% (6/16) resistances in patients under ARV. However, there is no statistically significant difference between the resistance rate we just found and Nadembèga’s findings (p = 0.143). Another study was conducted by Vergne et al. (2000) in Senegal, where HIV prevalence is very low, detected an 11.8% ARV resistance. Based on these Senegal data, it can be ascertained that the higher the number of HIV-infected people taking HAART, the higher the probability of isolating ARV resistant strains. In our study, twenty-three patients-16 of whom where in therapeutic failure, had a major ARVresistance. These major resistance mutations fell into three classes of ARV: M184V, T215Y, K103N, M46I, K20I, M36I, I84V observed in our study are the major standard mutations found in literature (Adje et al., 2001; Tebit et al., 2008; Varella et al., 2008; Djoko et al., 2011). The antiretroviral drugs helped in achieving a drastic decline in infection morbidity and mortality, turning it into a chronic infection. However, the prescriber and the patient should bear in mind that HIV infection remains potentially lethal because of the mutations induced by ARV and requires continuous long term treatment to obtain a virological and immunological control.
CONCLUSION
The results obtained from this study suggest that physicians should be more attentive to the monitoring of patients taking HAART, in order to avoid producing and disseminating numerous new resistant strains among their patients. Besides, we believe that HIV vaccine research for Burkina Faso cannot disregard the recombining strains we have just identified, namely: CRF06; CRF02-AG; CRF01-AE 4 and sub-type A 2.
ACKNOWLEDGMENTS
The authors are grateful to the staff of Centre MURAZ Bobo-Dioulasso, CHUSS Bobo-Dioulasso, Laboratoire de rétrovirologie IRD Montpellier, Faculté des sciences Université de Lomé/Togo, Centre Médical Saint Camille and CERBA/LABIOGENE, Université de Ouagadougou. They are deeply grateful to the ANRS (Agence nationale de recherches sur le Sida et les hépatites virales) for the financial support.