CYP2B6 poor metaboliser alleles involved in efavirenz and nevirapine metabolism: CYP2B6*9 and CYP2B6*18 distribution in HIV-exposed subjects from Dschang, Western Cameroon

Infection, Genetics and Evolution 35 (2015) 122–126

Contents lists available at ScienceDirect

Infection, Genetics and Evolution journal homepage: www.elsevier.com/locate/meegid

CYP2B6 poor metaboliser alleles involved in efavirenz and nevirapine metabolism: CYP2B6*9 and CYP2B6*18 distribution in HIV-exposed subjects from Dschang, Western Cameroon Giacomo Maria Paganotti a,b,c,1,⇑, Gianluca Russo c,1, Martin Sanou Sobze d, George Bouting Mayaka e, Charles Waithaka Muthoga a,b, Leabaneng Tawe a,b, Axel Martinelli f, Rita Romano c, Vincenzo Vullo c a

University of Botswana-University of Pennsylvania Partnership, Gaborone, Botswana Medical Education Partnership Initiative (MEPI) Laboratory, Gaborone, Botswana c Department of Public Health and Infectious Diseases, ‘‘Sapienza” University of Rome, Rome, Italy d Biomedical Sciences Department, Faculty of Sciences, University of Dschang, Dschang, Cameroon e District Hospital of Dschang, Dschang, Cameroon f Wellcome Trust Genome Campus, Hinxton, UK b

a r t i c l e

i n f o

Article history: Received 13 May 2015 Received in revised form 30 July 2015 Accepted 3 August 2015 Available online 4 August 2015 Keywords: CYP2B6 Efavirenz Nevirapine Pharmacogenetics Poor metabolisers Cameroon

a b s t r a c t The prescription of patients’ tailored anti-infectious treatments is the ultimate goal of pharmacogenetics/ genomics applied to antimicrobial treatments, providing a basis for personalized medicine. Despite the efforts to screen Africans for alleles underlying defective metabolism for a panel of different drugs, still more research is necessary to clarify the interplay between host genetic variation and treatments’ response. HIV is a major infectious disease in sub-Saharan African countries, and the main prescribed anti-HIV combination therapy includes efavirenz (EFV) or nevirapine (NVP). The two drugs are both mainly metabolised by cytochrome P450 2B6 liver enzyme (CYP2B6). Defective variants of CYP2B6 gene, leading to higher drug exposure with subsequent possible side effects and low compliance, are well known. However, little is known about CYP2B6 alleles in Cameroon where only one study was done on this subject. The main objective of the present work is to assess, in a subset of HIV-exposed subjects from Dschang in West Cameroon, the prevalence of two SNPs in the CYP2B6 gene: 516G>T (rs3745274) and 983T>C (rs28399499), both associated to a defective EFV and NVP metabolism. We analyzed 168 DNA samples collected during two cross-sectional surveys performed in Dschang, West Cameroon. In the population studied the observed allele frequencies of 516G>T and 983T>C were 44.35% (95%CI, 36.84– 51.86%) and 12.80% (95%CI, 7.75–17.85%), respectively. Moreover, concerning the CYP2B6 expected phenotypes, 28.57% of the population showed a poor metaboliser phenotype, while 27.38% and 44.05% showed an extensive (wild-type) and an intermediate metaboliser phenotype, respectively. Here we found that an important fraction of the subjects is carrying EFV/NVP poor metaboliser alleles. Our findings could help to improve the knowledge about the previewed efficacy of anti-HIV drug therapy in Cameroon. Finally, we designed a new method of detection for the 983T>C genetic variation that can be applied in resource-limited laboratories. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction Advances in pharmacogenetics over the past decades have resulted in a deeper understanding of the aetiology of variability in drug metabolism and response, including drugs used in the antiretroviral therapy (ART) against HIV infection. ART is based

⇑ Corresponding author at: P.O. Box AC157 ACH, Gaborone, Botswana. 1

E-mail address: [email protected] (G.M. Paganotti). These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.meegid.2015.08.003 1567-1348/Ó 2015 Elsevier B.V. All rights reserved.

on a long-life combination of antiretroviral drugs (ARVs) that have reduced significantly the morbidity and mortality burden of HIV infection worldwide. The vast majority of HIV-infected patients on ART in resource-limited countries (RLCs) take a nonnucleoside retro-transcriptase inhibitor (NNRTI) containing regimen as recommended first-line therapy (WHO, 2013). In Cameroon the HIV-prevalence is 4.3% and in the Western Region is 2.8%. In the Western Region only 38% of the population have a known HIV status (CNLS/UNAIDS, 2014). Moreover, in the whole Cameroon 49% of HIV-infected patients in need are taking

G.M. Paganotti et al. / Infection, Genetics and Evolution 35 (2015) 122–126

ART, and 39% of those taking ART are lost to follow-up after 12 months (CNLS/UNAIDS, 2014). According to data provided by the Dschang District Hospital, during 2010–2011, 436 HIVinfected adults started the ART (67% NNRTI-based), and 19% of them were lost to follow-up after 12 months. Presently, response to ART is often limited by the occurrence of toxicity or by the emergence of drug resistance. Since the beginning of ART roll-out in RLCs, according to mathematical models, the prevalence of ART resistance would increase as ART availability increases (Baggaley et al., 2006), as confirmed also by WHO surveys (WHO, 2012) and by a systematic review of the literature (Stadeli and Richman, 2013). This phenomenon, together with the increase of patients on ART lost to follow-up (Grimsrud et al., 2014), represents a serious risk in a public health point of view for HIV programs efficacy in RLCs. Moreover, genetic variation in drug metabolising enzymes may predispose to adverse event to drugs, or reduce drugs efficacy as well by influencing drug plasma exposure, this being potentially more frequent in African populations where most of the human genetic variability is present (Campbell and Tishkoff, 2008). In the field of ART, the genetic variability could be associated to different metabolic phenotypes, including poor metaboliser (PM) phenotypes that could cause high drug exposure and possibly toxicity. Moreover, ARVs with a prolonged half-life (i.e. NNRTIs) prescribed to PM subjects with low ART adherence, may result in exposure to the slowly metabolised drug alone – functional monotherapy – which may cause emergence of resistance, and treatment failure (Ribaudo et al., 2006). Furthermore, according to WHO, HIV-resistance is increasing in RLCs, and in Africa it is principally related to NNRTI drugs (WHO, 2012). NNRTIs are metabolised by cytochrome P450 (CYP) and also UDP-glucuronosyl-transferase (UGT) enzymes. In particular, efavirenz (EFV) and nevirapine (NVP) are hydroxylated mainly by CYP2B6 liver enzyme, with a minor contribution of CYP2A6 and UGT2B7 for EFV, and of CYP3A4 for NVP. Then the hydroxylated metabolites of both drugs undergo glucuronidation (Riska et al., 1999; Ward et al., 2003). Furthermore, it seems that SNPs of the nuclear receptor CAR (constitutive androstane receptor) may play some role in the metabolism of EFV (Wyen et al., 2011), possibly because its expression correlate to CYP2B6 expression in liver (Wortham et al., 2007). For both drugs other cytochromes participate at varied levels on their metabolic pathway, however, many studies have shown a major role of CYP2B6 and its genetic polymorphisms in the variations of the plasma concentration and clearance rate of both drugs and their possible associated toxicity. A huge literature links EFV and NVP metabolism to the activity of CYP2B6 enzyme and the modulation in pharmacokinetic parameters to different factors including genetic polymorphisms at CYP2B6 gene (on human chromosome 19). For example, it has been clearly demonstrated that poor metaboliser phenotypes of NNRTI are more prone to have higher than expected plasma concentration of the drug (Wang et al., 2006; Penzak et al., 2007; Saitoh et al., 2007; Rotger et al., 2007; Wyen et al., 2008; Bertrand et al., 2012; Sarfo et al., 2014), lower clearance (Saitoh et al., 2007; Haas et al., 2004; Lehr et al., 2011), increased half-life (Ribaudo et al., 2006), higher number of side effects episodes and then lower compliance (Haas et al., 2004; Lubomirov et al., 2011; Yuan et al., 2011; Mukonzo et al., 2013), leading to a possible treatment failure (Ribaudo et al., 2006); however, data in literature are conflicting and mostly on non-African HIV-infected patients (Ribaudo et al., 2010; Frasco et al., 2012; Haas et al., 2014; Lee et al., 2014). Although PM patients may have an increased risk to select viral drug resistance, possibly because of a reduced compliance linked to higher risk of toxicity, to our knowledge no studies have been performed in order to evaluate the association between viral resistance and CYP2B6 metaboliser phenotypes.

123

The reduced rate of EFV and NVP metabolism is mostly linked to the presence of two SNPs in the CYP2B6 gene: 516G>T (rs3745274) and 983T>C (rs28399499), causing an aberrant liver mRNA splicing (Hofmann et al., 2008) and the disruption of the alpha helix structure in the protein, respectively (Klein et al., 2005). The SNP 983T>C affects the metabolism of both EFV and NVP; the SNP 516G>T influences mainly the EFV metabolism, while data on its impact on NVP metabolism are less conclusive (Chen et al., 2010; Lehr et al., 2011). The two SNPs define two different poor metaboliser alleles (CYP2B6*9 and CYP2B6*18, for 516 and 983 SNPs, respectively), but they are also included in different haplotype patterns together with other SNPs, defining other CYP2B6 alleles (http://www.cypalleles.ki.se), the most studied being CYP2B6*6 (for 516) and CYP2B6*16 (for 983). In literature, the frequency of the 516G>T in African populations ranges between 20% (Ikediobi et al., 2011) and 70% (Mehlotra et al., 2006), while the frequency of the 983T>C ranges between 1.6% (Mehlotra et al., 2006) and 17% (Ikediobi et al., 2011). There is only one single report about the frequency of polymorphisms in CYP2B6 gene from Cameroon (Swart et al., 2012), where the authors report the allele frequency of 516G>T being 37% in 69 subjects from Yaoundé (Central Cameroon), while no information is available about the frequency of 983T>C alone or in combination with other SNPs. In the present study we analyze the presence and genotype composition of the two functional polymorphisms on the CYP2B6 gene (516G>T and 983T>C) associated with the defective metabolism of EVF and NVP in a sample of HIV-exposed individuals from Dschang, Western Region of Cameroon. Furthermore, we report a new touchdown PCR-RFLP method for detection of 983T>C genetic variation, that could be used as alternative to real-time PCR or sequencing approaches in resource-limited laboratories.

2. Material and methods The blood samples were collected during two cross-sectional surveys performed at the District Hospital of Dschang, West Cameroon. The surveys were carried out with the purpose of evaluating the prevention of mother-to-child transmission of HIV (2003–2006) (Turriziani et al., 2008), and the malaria impact on HIV-infected adults (2012). We enrolled 168 subjects (100 unrelated HIV negative children born from HIV positive mothers, and 68 unrelated HIV-infected adults) all belonging to the Bamileké ethnic group that is one of the Bantu-speaking related ethnic groups in Africa. Signed informed consent for multiple genetic and epidemiological surveys was obtained from all adults and from all children’s parents/carers. This study was conducted in accordance with the guidelines of the Helsinki Declaration of 2000, with the approval of the ethics committee of the ‘‘Sapienza” University of Rome and the Comité National d’Ethique et de la Recherche en Sciences Humaines (CNERSH) in Cameroon. We extracted human DNA from sera samples of the subjects using Chelex-100 resin (Bio-Rad). CYP2B6*9 (defined by the presence of the SNP rs3745274, 516G>T) detection was carried out using a PCR-RFLP technique according to the protocol of Lavandera et al. (2011) with minor modifications. For CYP2B6*18 (defined by the presence of the SNP rs28399499, 983T>C) detection, we develop a new touchdown PCR-RFLP essay. We designed two primers that amplify a 215 bp fragment of the CYP2B6 gene (forward primer: 50 -GCCAGGCTCATCTTGAACTC-30 ; reverse primer: 50 -ATTTTGGCTCGGTCATGAAG-30 ). For touchdown PCR cycling, after 4 min of denaturation at 95 °C, the PCR mixtures were subjected to the following conditions: 30 s at 95 °C, 1 min at 67 °C and 1 min at 72 °C for 11 cycles decreasing the annealing temperature by 0.5 °C every cycle. Then, 30 s at 95 °C, 30 s at 62 ° C and 30 s at 72 °C for additional 30 cycles with a delayed last step

124

G.M. Paganotti et al. / Infection, Genetics and Evolution 35 (2015) 122–126

of 10 min at 72 °C. The PCR product was then incubated with BsmAI restriction enzyme that cuts the mutant allele only (C) in two fragments of 151 bp and 64 bp; undigested products represent the wild-type allele (T). The digested fragments were visualized on a 3% agarose gel. Controls for human genotyping were utilized after sequencing the different genotypes. In addition, all the heterozygous TC were repeated to avoid the genotyping error. We assigned the metabolic profile according to Haas et al. (2009). Chi-square analyzis and F-statistics were applied to evaluate Hardy–Weinberg equilibrium. Arlequin v3.5 (Excoffier and Lischer, 2010) was used to test for possible linkage disequilibrium (LD) between the two loci. The input consisted of diploid genotypic DNA data with unknown gametic phase and assuming co-dominance. The Expectation–Ma ximization (EM) algorithm to test for LD was run for 20,000 permutations and 3 initial conditions, based on recommended criteria. Haplotype reconstruction was performed using Arlequin 3.5 and the PHASE 2.1.1 software developed by Stephens et al. (2001). For Arlequin, the EM algorithm was run on the same input file used for the LD test at the haplotype level, with 50 starting points and 1000 iterations. Other parameters were set to default. The Empirical Linear Bayes (ELB) algorithm (with default settings) was used to generate haplotype counts. For PHASE, the algorithm was run for 100 iterations, with a burn-in of 100 and a thinning interval of 1. The ‘‘-MS” modeling option was used which provides more confident calls and is better suited for small data sets. The algorithm was rerun on the same data set 5 times using a different random seed each time, as recommended by the authors.

51.86%). The allele frequency for the mutant allele C at position 983 was 12.80% (95%CI, 7.75–17.85%). The genotypes for 516G>T polymorphism where in Hardy–Weinberg equilibrium [v2 = 3.5 (df = 1), P = n.s.; FIS (Wright’s F) = 0.14, P = n.s.], while it was no possible to calculate it for the 983T>C because the homozygous CC genotype was not found. The test to determine the presence of LD in Arlequin produced an exact P-value of 0.00649 ± 0.00056 and a v2 = 7.55710 (df = 1, P = 0.00598). Thus, the test indicates the statistically significant (P < 0.05) presence of LD between the two loci. Each round of haplotype reconstruction with PHASE produce virtually identical results, underscoring the robustness of the results (see Supplementary materials). Only 9 individuals had alternative haplotypes. The most likely haplotype combination (TC and GT, where the first nucleotide refers to the 516 polymorphism and the second to the 983) had an average probability of 0.737. A similar result was obtained when reconstructing haplotypes with Arlequin and haplotype counts were identical for both PHASE and Arlequin (Table 1). Haplotype frequencies were estimated using Arlequin and also calculated manually from the given Haplotype counts (Table 1). The differences between the two calculations are minimal and underscore the strong prevalence of the GT and TT haplotypes in the samples. Concerning the CYP2B6 expected phenotypes (Table 2), 48 subjects (28.57%) showed a poor metaboliser phenotype. Overall, the majority of the subjects (n = 122; 72.62%) had a defective metabolic phenotype (intermediate and poor metaboliser).

3. Results

To our knowledge, the present study is the largest genotyping work on CYP2B6 performed in subjects from Cameroon. The only previous known study reporting the CYP2B6 516G>T frequency in Cameroon was performed on 69 subjects from Yaoundé (Central Region) only defined as Bantu-speaking, without further ethnic specification (Swart et al., 2012). There is no published study on CYP2B6 983T>C from Cameroon. We used an approach focusing on defective metabolism of EFV and NVP driven by the two pure poor metaboliser alleles (CYP2B6*9 and *18). Very few studies have followed this approach based on the composite CYP2B6 516/983 genotype (Haas et al., 2009). For example, the most studied CYP2B6 allele, *6, is defined by 516G>T and 785A>G polymorphisms (http://www.cypalleles.ki.se), the first being a defective variation (Hofmann et al., 2008), while the second is linked to an increased EFV (Wang et al., 2006) and NVP (Bertrand et al., 2012) metabolism. Thus, the resulting CYP2B6*6 allele is mostly a poor metaboliser allele because of a reduced enzyme expression (linked to 516G>T variant). For that reason, many reports are focused only on the detection of 516G>T as marker for CYP2B6*6 (Ikediobi et al., 2011; Nyakutira et al., 2008; Mukonzo et al., 2009; Habtewold et al., 2011; Ngaimisi et al., 2013), which is not a rigorous criterion. In fact, some studies have shown that a proportion of carriers of

We obtained successful DNA amplification for both SNP from 168 subjects. The genotypes for the two polymorphisms were distributed as follows: for the 516G>T we accounted 58 GG genotypes (34.5%), 71 GT genotypes (42.3%) and 39 TT genotypes (23.2%); for the 983T>C we accounted 125 TT genotypes (74.4%) and 43 TC genotypes, while no CC genotype was found. The allele frequency for the mutant allele T at position 516 was 44.35% (95%CI, 36.84– Table 1 Reconstructed haplotype counts and frequencies as determined by PHASE and Arlequin softwares. Haplotype

GT

TT

GC

TC

All

Total Frequency raw (%) Frequency Arlequin (%)

175 52.08 51.4

118 35.12 35.81

12 3.57 4.26

31 9.23 8.53

336 100 100

Haplotype reconstruction by PHASE determined GT and TC as the most likely haplotypes in the case of genotype GT/TC. The same was the case in Arlequin and both gave the same counts. Haplotype frequencies in Arlequin were estimated using the EM algorithm, while the raw frequency was calculated from the haplotype counts.

4. Discussion

Table 2 Combined CYP2B6 516G>T and 983T>C expected metabolic phenotypes. CYP2B6 516G>T/983T>C

CYP2B6 genotypes Total By phenotype (%)

CYP2B6 phenotypes Extensive metaboliser (EM)1

Intermediate metaboliser (IM)2

Poor metaboliser (PM)3

GG/TT 46 46 (27.38)

GT/TT 62

GT/TC 9

GG/TC 12 74 (44.05)

Combined CYP2B6 expected phenotypes were assigned as follows: 1 EM (extensive metabolisers): no variant allele at either position 516 wor 983 (wild type). 2 IM (intermediate metabolisers): 1 variant allele at either position at 516 or 983, but not both. 3 PM (poor metabolisers): 2 variant alleles at either 516 or 983, and both; or 3 variant alleles.

TT/TT 17 48 (28.57)

TT/TC 22

All 168 168 (100)

G.M. Paganotti et al. / Infection, Genetics and Evolution 35 (2015) 122–126

516GG (wild-type) genotype had a risk of sub-therapeutic EFV plasma concentration (Rodriguez-Novoa et al., 2005; ter Heine et al., 2008); the possible explanation could be due to the presence of 785A>G polymorphism, not assessed in both studies. Furthermore, also the CYP2B6*16 allele contains the 785A>G polymorphism (together with 983T>C) and the high EFV plasma concentration of carriers of *16 allele (Wang et al., 2006) is mostly due to the presence of the 983T>C polymorphism. Our data show a slightly higher allele frequency for the 516G>T variant allele compared with the study of Swart et al. (2012) (44% vs 37%). This could be due to a smaller sample size of the first assessment (Swart et al., 2012), or to a possible different ethnicity component between the two studies. The absence of the CYP2B6 983CC genotype was already known from the literature (Wyen et al., 2008; Haas et al., 2009). Moreover, in the dataset at dbSNP related to 983T>C polymorphism (http://www.ncbi.nlm.nih.gov/ projects/SNP/snp_Ref.cgi?rs=28399499) is clearly reported that most of the African population screened do not show any CC genotype except the Youruba from Nigeria with a frequency of 1.8% and the Maasai from Kenya with a frequency of 0.7%. Thus, our results are in line with other findings. The most relevant result from this study is that almost 3 out of 4 subjects (72.62%) analyzed are carrying defective CYP2B6 alleles leading to defective metabolism (intermediate or poor metabolisers), with the intermediate metabolisers possibly having an intermediate PK and therefore an intermediate chance of toxicity. Almost 1 out of 3 subjects (28.57%) has a PM CYP2B6 phenotype. Moreover, 22 (13.09%) subjects show a genotype pattern with 3 variant alleles at position 516 and 983 (516TT/983TC), leading possibly to a further additional reduction of enzyme activity. Thus, they do have very low predictive chance to metabolise well EFV and, to less extent also NVP (Stadeli and Richman, 2013; Micheli et al., 2013), that are the most prescribed ARVs drug in RLCs (WHO, 2013). This phenomenon raises the possibility of increased risk of ARVs adverse events, followed by lower ART compliance and then higher selective pressure for drug resistance selection. Thus, the possible increase of HIV resistance related to genetic defective ARVs metabolism, together with an high rate (39%) of HIV-infected patient on ART lost to follow-up (CNLS/UNAIDS, 2014), risk to severely affect the public health approach of ART rollout program in Cameroon. The implication of our results on a possible dose adjustment of NNRTIs, leaded by patients’ CYP2B6 genotype, in order to maximize ART efficacy and minimizing the risk of drug resistance selection, needs further and larger studies evaluating also drug-drug interaction with concomitant treatments (i.e. anti-tuberculosis therapy) (Kwara et al., 2010; Mukonzo et al., 2014). Furthermore, LD analyzis gave a positive result for the two loci. Haplotype reconstruction revealed the predominance of the GT and TT haplotypes in the samples. This could indicate a tendency for the alleles in locus 516 to segregate preferentially with the wild-type allele T on locus 983 and not with the mutant allele C, possibly because of its deleterious effects on the metabolism of xenobiotics (Nebert et al., 1989; Gonzalez and Nebert, 1990; Danielson, 2002). Our study has some limitations. The sample size is relatively small and related to only one ethnic group living in Cameroon. Moreover, data on EFV or NVP plasma exposure, as well as virologic data (viral load, genetic viral analyzis) on all the HIV-infected participants were not available. 5. Conclusions In conclusion, the present study is the largest genotyping work on CYP2B6 516G>T (*9) and the first on 983T>C (*18) from Cameroon. Our results provide a new insight in order to improve the efficacy of the ART in Cameroon. Furthermore, we set a new method for detection of 983T>C genetic variation that could be

125

used as alternative to real-time PCR or sequencing approaches in resource-limited laboratories. Acknowledgements We are particularly grateful to the subjects enrolled in the study, the technical staff for their support and health care service directors for their collaboration throughout the investigation. We also thank Dr. Gaseitsiwe S. from Botswana-Harvard Partnership for sequencing of the reference genotypes for 983T>C polymorphism and Dr Verra F. for a critical revision of the manuscript. This work has been funded by the Faculty of Medicine of the ‘‘Sapienza” University of Rome, Italy. The molecular work has been performed at the Medical Education Partnership Initiative (MEPI) laboratory, Faculty of Medicine, University of Botswana, Gaborone, Botswana. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.meegid.2015.08. 003. References Baggaley, R.F., Garnett, G.P., Ferguson, N.M., 2006. Modelling the impact of antiretroviral use in resource poor settings. PLoS Med. 3 (4), e124. Bertrand, J., Chou, M., Richardson, D.M., Verstuyft, C., Leger, P.D., Mentré, F., Taburet, A.M., Taburet, D.W., 2012. ANRS 12154 study group. Multiple genetic variants predict steady-state nevirapine clearance in HIV-infected Cambodians. Pharmacogenet. Genomics 22 (12), 868–876. Campbell, M.C., Tishkoff, S.A., 2008. African genetic diversity: implications for human demographic history, modern human origins, and complex disease mapping. Ann. Rev. Genomics Hum. Genet. 9, 403–433. Chen, J., Sun, J., Ma, Q., Yao, Y., Wang, Z., Zhang, L., Li, L., Sun, F., Lu, H., 2010. CYP2B6 polymorphism and nonnucleoside reverse transcriptase inhibitor plasma concentrations in Chinese HIV-infected patients. Ther. Drug Monit. 32 (5), 573–578. CNLS (Comité National de Lutte contre le SIDA)/UNAIDS (United Nation against AIDS), Yaoundé, Cameroon: Global Aids Response Progress (GARP), 2014. (accessed 7 November 2014). Danielson, P.B., 2002. The cytochrome P450 superfamily: biochemistry, evolution and drug metabolism in humans. Curr. Drug Metab. 3 (6), 561–597. Excoffier, L., Lischer, H.E., 2010. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol. Ecol. Resour. 10 (3), 564–567. Frasco, M.A., Mack, W.J., Van Den Berg, D., Aouizerat, B.E., Anastos, K., Cohen, M., De Hovitz, J., Golub, E.T., Greenblatt, R.M., Liu, C., Conti, D.V., Pearce, C.L., 2012. Underlying genetic structure impacts the association between CYP2B6 polymorphisms and response to efavirenz and nevirapine. AIDS 26 (16), 2097–2106. Gonzalez, F.J., Nebert, D.W., 1990. Evolution of the P450 gene superfamily: animalplant ‘warfare’, molecular drive and human genetic differences in drug oxidation. Trends Genet. 6 (6), 182–186. Grimsrud, A., Balkan, S., Casas, E.C., Lujan, J., Van Cutsem, G., Poulet, E., Myer, L., Pujades-Rodriguez, M., 2014. Outcomes of antiretroviral therapy over a 10-year period of expansion: a multicohort analysis of African and Asian HIV programs. J. Acquir. Immune Defic. Syndr. 67 (2), e55–e66. Habtewold, A., Amogne, W., Makonnen, E., Yimer, G., Riedel, K.D., Ueda, N., Worku, A., Haefeli, W.E., Lindquist, L., Aderaye, G., Burhenne, J., Aklillu, E., 2011. Longterm effect of efavirenz autoinduction on plasma/peripheral blood mononuclear cell drug exposure and CD4 count is influenced by UGT2B7 and CYP2B6 genotypes among HIV patients. J. Antimicrob. Chemother. 66 (10), 2350–2361. Haas, D.W., Ribaudo, H.J., Kim, R.B., Tierney, C., Wilkinson, G.R., Gulick, R.M., Clifford, D.B., Hulgan, T., Marzolini, C., Acosta, E., 2004. Pharmacogenetics of efavirenz and central nervous system side effects: an adult AIDS clinical trials group study. AIDS 18 (18), 2391–2400. Haas, D.W., Gebretsadik, T., Mayo, G., Menon, U.N., Acosta, E.P., Shintani, A., Floyd, M., Stein, C.M., Wilkinson, G.R., 2009. Associations between CYP2B6 polymorphisms and pharmacokinetics after a single dose of nevirapine or efavirenz in African Americans. J. Infect. Dis. 199 (6), 872–880. Haas, D.W., Severe, P., Jean Juste, M.A., Pape, J.W., Fitzgerald, D.W., 2014. Functional CYP2B6 variants and virologic response to an efavirenz-containing regimen in Port-au-Prince, Haiti. In: J. Antimicrob. Chemother. 69 (8), 2187–2190.

126

G.M. Paganotti et al. / Infection, Genetics and Evolution 35 (2015) 122–126

Hofmann, M.H., Blievernicht, J.K., Klein, K., Saussele, T., Schaeffeler, E., Schwab, M., Zanger, U.M., 2008. Aberrant splicing caused by single nucleotide polymorphism c.516G>T [Q172H], a marker of CYP2B6*6, is responsible for decreased expression and activity of CYP2B6 in liver. J. Pharmacol. Exp. Ther. 325 (1), 284–292. Ikediobi, O., Aouizerat, B., Xiao, Y., Gandhi, M., Gebhardt, S., Warnich, L., 2011. Analysis of pharmacogenetic traits in two distinct South African populations. Hum. Genomics 5 (4), 265–282. Klein, K., Lang, T., Saussele, T., Barbosa-Sicard, E., Schunck, W.H., Eichelbaum, M., Schwab, M., Zanger, U.M., 2005. Genetic variability of CYP2B6 in populations of African and Asian origin: allele frequencies, novel functional variants, and possible implications for anti-HIV therapy with efavirenz. Pharmacogenet. Genomics 15 (12), 861–873. Kwara, A., Ramachandran, G., Swaminathan, S., 2010. Dose adjustment of the nonnucleoside reverse transcriptase inhibitors during concurrent rifampicincontaining tuberculosis therapy: one size does not fit all. Expert Opin. Drug Metab. Toxicol. 6 (1), 55–68. Lavandera, J.V., Parera, V.E., Rossetti, M.V., Batlle, A., Buzaleh, A.M., 2011. Identification of CYP3A5 and CYP2B6 polymorphisms in porphyria cutanea tarda associated to human immunodeficiency virus. J. Clin. Exp. Dermatol. Res. S2, 006. Lee, K.Y., Lin, S.W., Sun, H.Y., Kuo, C.H., Tsai, M.S., Wu, B.R., Tang, S.Y., Liu, W.C., Chang, S.Y., Hung, C.C., 2014. Therapeutic drug monitoring and pharmacogenetic study of HIV-infected ethnic Chinese receiving efavirenzcontaining antiretroviral therapy with or without rifampicin-based antituberculous therapy. PLoS One 9 (2), e88497. Lehr, T., Yuan, J., Hall, D., Zimdahl-Gelling, H., Schaefer, H.G., Staab, A., MacGregor, T. R., Jayadev, S., 2011. Integration of absorption, distribution, metabolism, and elimination genotyping data into a population pharmacokinetic analysis of nevirapine. Pharmacogenet. Genomics 21 (11), 721–730. Lubomirov, R., Colombo, S., di Iulio, J., Ledergerber, B., Martinez, R., Cavassini, M., Hirschel, B., Bernasconi, E., Elzi, L., Vernazza, P., Furrer, H., Günthard, H.F., Telenti, A., 2011. Swiss HIV cohort study. Association of pharmacogenetic markers with premature discontinuation of first-line anti-HIV therapy: an observational cohort study. In: J. Infect. Dis. 203 (2), 246–257. Mehlotra, R.K., Ziats, M.N., Bockarie, M.J., Zimmerman, P.A., 2006. Prevalence of CYP2B6 alleles in malaria-endemic populations of West Africa and Papua New Guinea. Eur. J. Clin. Pharmacol. 62 (4), 267–275. Micheli, J.E., Chinn, L.W., Shugarts, S.B., Patel, A., Martin, J.N., Bangsberg, D.R., Kroetz, D.L., 2013. Measuring the overall genetic component of nevirapine pharmacokinetics and the role of selected polymorphisms: towards addressing the missing heritability in pharmacogenetic phenotypes? Pharmacogenet. Genomics 23 (11), 591–596. Mukonzo, J.K., Röshammar, D., Waako, P., Andersson, M., Fukasawa, T., Milani, L., Svensson, J.O., Ogwal-Okeng, J., Gustafsson, L.L., Aklillu, E., 2009. A novel polymorphism in ABCB1 gene, CYP2B6*6 and sex predict single-dose efavirenz population pharmacokinetics in Ugandans. Br. J. Clin. Pharmacol. 68 (5), 690– 699. Mukonzo, J.K., Okwera, A., Nakasujja, N., Luzze, H., Sebuwufu, D., Ogwal-Okeng, J., Waako, P., Gustafsson, L.L., Aklillu, E., 2013. Influence of efavirenz pharmacokinetics and pharmacogenetics on neuropsychological disorders in Ugandan HIV-positive patients with or without tuberculosis: a prospective cohort study. BMC Infect. Dis. 13, 261. Mukonzo, J.K., Owen, J.S., Ogwal-Okeng, J., Kuteesa, R.B., Nanzigu, S., Sewankambo, N., Thabane, L., Gustafsson, L.L., Ross, C., Aklillu, E., 2014. Pharmacogeneticbased efavirenz dose modification: suggestions for an African population and the different CYP2B6 genotypes. PLoS One 9 (1), e86919. Nebert, D.W., Nelson, D.R., Feyereisen, R., 1989. Evolution of the cytochrome P450 genes. Xenobiotica 19 (10), 1149–1160. Ngaimisi, E., Habtewold, A., Minzi, O., Makonnen, E., Mugusi, S., Amogne, W., Yimer, G., Riedel, K.D., Janabi, M., Aderaye, G., Mugusi, F., Bertilsson, L., Aklillu, E., Burhenne, J., 2013. Importance of ethnicity, CYP2B6 and ABCB1 genotype for efavirenz pharmacokinetics and treatment outcomes: a parallel-group prospective cohort study in two Sub-Saharan Africa populations. PLoS One 8 (7), e67946. Nyakutira, C., Roshammar, D., Chigutsa, E., Chonzi, P., Ashton, M., Nhachi, C., Masimirembwa, C., 2008. High prevalence of the CYP2B6 516G>T(*6) variant and effect on the population pharmacokinetics of efavirenza in HIV/AIDS outpatients in Zimbabwe. Eur. J. Clin. Pharmacol. 64 (4), 357–365. Penzak, S.R., Kabuye, G., Mugyenyi, P., Mbamanya, F., Natarajan, V., Alfaro, R.M., Kityo, C., Formentini, E., Masur, H., 2007. Cytochrome P450 2B6 (CYP2B6) G516T influences nevirapine plasma concentrations in HIV-infected patients in Uganda. HIV Med. 8 (2), 86–91. Ribaudo, H.J., Haas, D.W., Tierney, C., Kim, R.B., Wilkinson, G.R., Gulick, R.M., Clifford, D.B., Marzolini, C., Fletcher, C.V., Tashima, K.T., Kuritzkes, D.R., Acosta, E.P., 2006. Pharmacogenetics of plasma efavirenz exposure after treatment discontinuation: an adult AIDS clinical trials group study. Clin. Infect. Dis. 42 (3), 401–407. Ribaudo, H.J., Liu, H., Schwab, M., Schaeffeler, E., Eichelbaum, M., Motsinger-Reif, A. A., Ritchie, M.D., Zanger, U.M., Acosta, E.P., Morse, G.D., Gulick, R.M., Robbins, G.

K., Clifford, D., Haas, D.W., 2010. Effect of CYP2B6, ABCB1, and CYP3A5 polymorphisms on efavirenz pharmacokinetics and treatment response: an AIDS clinical trials group study. J. Infect. Dis. 202 (5), 717–722. Riska, P., Lamson, M., MacGregor, T., Sabo, J., Hattox, S., Pav, J., Keirns, J., 1999. Disposition and biotransformation of the antiretroviral drug nevirapine in humans. Drug Metab. Dispos. 27 (8), 895–901. Rodriguez-Novoa, S., Barreiro, P., Rendón, A., Jiménez-Nacher, I., González-Lahoz, J., Soriano, V., 2005. Influence of 516G>T polymorphisms at the gene encoding the CYP450-2B6 isoenzyme on efavirenz plasma concentrations in HIV-infected subjects. Clin. Infect. Dis. 40 (9), 1358–1361. Rotger, M., Tegude, H., Colombo, S., Cavassini, M., Furrer, H., Décosterd, L., Blievernicht, J., Saussele, T., Günthard, H.F., Schwab, M., Eichelbaum, M., Telenti, A., Zanger, U.M., 2007. Predictive value of known and novel alleles of CYP2B6 for efavirenz plasma concentrations in HIV-infected individuals. Clin. Pharmacol. Ther. 81 (4), 557–566. Saitoh, A., Sarles, E., Capparelli, E., Aweeka, F., Kovacs, A., Burchett, S.K., Wiznia, A., Nachman, S., Fenton, T., Spector, S.A., 2007. CYP2B6 genetic variants are associated with nevirapine pharmacokinetics and clinical response in HIV-1infected children. AIDS 21 (16), 2191–2199. Sarfo, F.S., Zhang, Y., Egan, D., Tetteh, L.A., Phillips, R., Bedu-Addo, G., Sarfo, M.A., Khoo, S., Owen, A., Chadwick, D.R., 2014. Pharmacogenetic associations with plasma efavirenz concentrations and clinical correlates in a retrospective cohort of Ghanaian HIV-infected patients. J. Antimicrob. Chemother. 69 (2), 491–499. Stadeli, K.M., Richman, D.D., 2013. Rates of emergence of HIV drug resistance in resource-limited settings: a systematic review. Antivir. Ther. 18 (1), 115–123. Stephens, M., Smith, N., Donnelly, P., 2001. A new statistical method for haplotype reconstruction from population data. Am. J. Hum. Genet. 68 (4), 978–989. Swart, M., Skelton, M., Wonkam, A., Kannemeyer, L., Chin’ombe, N., Dandara, C., 2012. CYP1A2, CYP2A6, CYP2B6, CYP3A4 and CYP3A5 polymorphisms in two bantu-speaking populations from Cameroon and South Africa: implications for global pharmacogenetics. Curr. Pharmacogenomics Person Med. 10 (1), 43–53. ter Heine, R., Scherpbier, H.J., Crommentuyn, K.M., Bekker, V., Beijnen, J.H., Kuijpers, T.W., Huitema, A.D., 2008. A pharmacokinetic and pharmacogenetic study of efavirenz in children: dosing guidelines can result in subtherapeutic concentrations. Antivir. Ther. 13 (6), 779–787. Turriziani, O., Russo, G., Lichtner, M., Strano, A., Tsague, G., Maida, P., Vullo, V., Antonelli, G., 2008. Study of the genotypic resistant pattern in HIV-infected women and children from rural west-Cameroon. In: AIDS Res. Hum. Retroviruses 24 (6), 781–785. Wang, J., Sönnerborg, A., Rane, A., Josephson, F., Lundgren, S., Ståhle, L., IngelmanSundberg, M., 2006. Identification of a novel specific CYP2B6 allele in Africans causing impaired metabolism of the HIV drug efavirenz. Pharmacogenet. Genomics 16 (3), 191–198. Ward, B.A., Gorski, J.C., Jones, D.R., Hall, S.D., Flockhart, D.A., Desta, Z., 2003. The cytochrome P450 2B6 (CYP2B6) is the main catalyst of efavirenz primary and secondary metabolism: implication for HIV/AIDS therapy and utility of efavirenz as a substrate marker of CYP2B6 catalytic activity. J. Pharmacol. Exp. Ther. 306 (1), 287–300. WHO: the HIV drug resistance report, 2012. (accessed 17 Oct 2014). WHO: consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection: recommendations for a public health approach, 2013. (accessed 17 Oct 2014). Wortham, M., Czerwinski, M., He, L., Parkinson, A., Wan, Y.J., 2007. Expression of constitutive androstane receptor, hepatic nuclear factor 4 alpha, and P450 oxidoreductase genes determines interindividual variability in basal expression and activity of a broad scope of xenobiotic metabolism genes in the human liver. Drug Metab. Dispos. 35 (9), 1700–1710. Wyen, C., Hendra, H., Vogel, M., Hoffmann, C., Knechten, H., Brockmeyer, N.H., Bogner, J.R., Rockstroh, J., Esser, S., Jaeger, H., Harrer, T., Mauss, S., van Lunzen, J., Skoetz, N., Jetter, A., Groneuer, C., Fätkenheuer, G., Khoo, S.H., Egan, D., Back, D.J., Owen, A., 2008. German competence network for HIV/AIDS. Impact of CYP2B6 983T>C polymorphism on non-nucleoside reverse transcriptase inhibitor plasma concentrations in HIV-infected patients. In: J. Antimicrob. Chemother. 61 (4), 914–918. Wyen, C., Hendra, H., Siccardi, M., Platten, M., Jaeger, H., Harrer, T., Esser, S., Bogner, J.R., Brockmeyer, N.H., Bieniek, B., Rockstroh, J., Hoffmann, C., Stoehr, A., Michalik, C., Dlugay, V., Jetter, A., Knechten, H., Klinker, H., Skaletz-Rorowski, A., Fätkenheuer, G., Egan, D., Back, D.J., Owen, A., 2011. German Competence Network for HIV/AIDS Coordinators. Cytochrome P450 2B6 (CYP2B6) and constitutive androstane receptor (CAR) polymorphisms are associated with early discontinuation of efavirenz-containing regimens. J. Antimicrob. Chemother. 66 (9), 2092–2098. http://dx.doi.org/10.1093/jac/dkr272. Yuan, J., Guo, S., Hall, D., Cammett, A.M., Jayadev, S., Distel, M., Storfer, S., Huang, Z., Mootsikapun, P., Ruxrungtham, K., Podzamczer, D., Haas, D.W., 2011. Nevirapine Toxicogenomics Study Team. Toxicogenomics of nevirapineassociated cutaneous and hepatic adverse events among populations of African, Asian, and European descent. In: AIDS 25 (10), 1271–1280.