Diversity in APOBEC3 and CCR5 host genes and HIV-1 in a South African population

Abstract

Introduction Human Immunodeficiency Virus (HIV-1) continues to be a global public health concern, even though Antiretroviral drugs (ARV), especially Highly Active Antiretroviral Therapy (HAART) has significantly reduced morbidity and mortality due to AIDS globally in developed and developing countries. However, there is still a great need to explore every avenue for new therapeutic interventions due to the limitations and side effects of HAART. Potential major breakthroughs for future therapeutic development were the discoveries more than 10 years ago of the role of HIV-1 co-receptors and anti-viral activities of host restriction factors such as APOBEC3G protein, which is a member of the DNA cytosine deaminase family. Entry of HIV in to the host cell is through the attachment of the viral envelope glycoprotein to the CD4 receptor, and subsequent interaction, mainly with either CCR5 or CXCR4 co-receptors. Inhibitors, such as Maraviroc, which binds to CCR5 inhibiting entry of CCR5 utilizing viruses (R5 viruses), is currently reserved for salvage therapy in many countries including South Africa. In the course of HIV infection, CXCR4 utilizing viruses (X4 viruses) may emerge and outgrow R5 viruses, and potentially limit the effectiveness of Maraviroc. Several host cell APOBEC3 genes (A3D, A3F, A3G and A3H) have been shown to restrict HIV, and the HIV viral infectivity factor (Vif) protein serves to antagonize the action of APOBEC3 proteins, promoting viral replication. The CCR5 co-receptor and the HIV Env V3 loop have also been documented as playing roles in HIV-1 disease progression. The interplay between host and viral genes still needs widespread attention, given that disease outcomes of HIV depend on many factors, including host cell genetics. Since the discovery of these genes and their role in HIV replication, many studies have been conducted that show their association with viral polymorphism. The polymorphisms found in host cell genes can have significant effects on viral replication, transmission and fitness and can also contribute to the overall diversity in HIV-1 populations. It is hypothesized that there are significant polymorphisms in HIV-1 and cellular genes that may differ among different populations. Population-based studies have tried to establish a relationship between host factors such as APOBEC3 and CCR5 polymorphism and the rate of disease progression, but most studies have focused on Caucasian populations. In vi contrast, little information is available for the effects of variation in these genes in African populations such as South Africa, where the HIV epidemic has expanded at an alarming rate. Although several population studies have focused on African Americans, these do not give us a complete picture of the potential variation in Africans, though the studies can be a good guide on which to base additional studies. A more comprehensive analysis involving different African populations will likely provide a better understanding of the mechanisms underlying host-pathogen interactions, especially in view of the fact that African Americans are primarily infected with HIV subtype B, which is rarely seen in Africa. Methodology This study characterized the genetic variability of the APOBEC3 D, F, G and H genes as well as the HIV-1 vif, in an ethnically diverse HIV-1 infected South African cohort using Next Generation Sequencing (NGS). In addition, polymorphism in CCR5 was analyzed in conjunction with an analysis of the V3 loop sequences in HIV-1 from this cohort. Genomic DNA was extracted from peripheral blood mononuclear cells (PBMCs) of 192 HIV-1 infected drug-experienced individuals who presented for routine care at the HIV/AIDS Prevention Group Wellness Clinic (HAPG) in Bela-Bela, Donald Fraser Hope Clinic (DFHC) in Vhufhuli and in local clinics in the Vhembe district of Limpopo Province, South Africa. Next generation sequencing custom based (Tn5 tagmentation and amplicon based) protocols to prepare libraries for host and HIV-1 genes were developed and validated with commercially available library preparation kits. The Tn5 tagmentation methods were used for longer DNA fragments and the custom amplicon based methods were used mainly for the shorter DNA fragments. To determine the variability of the APOBEC3 and CCR5 host genes, gene-specific primers were designed to amplify complete 12.16 kb A3D, 13.31 kb A3F, 10.74 kb A3G, 6.8 kb A3H and 1.3 kb CCR5 genes targeting the regions containing the exons. Libraries for the resulting amplicons were prepared using Tn5 transposase tagmentation methods and sequenced on an NGS Illumina MiSeq platforms generating millions of reads with good read coverage for variant calling. Single nucleotide polymorphisms (SNPs) and indels were determined, verified in dbSNPs and compared to SNPs in other populations reported in the 1000 Genome Phase III and HapMap. A Chi-square goodness-of-fit was used to verify if whether observed genotype frequencies were in agreement with the Hardy-Weinberg Equilibrium. Haplotypes and Linkage disequilibrium were inferred to determine SNP association. vii The HIV-1 vif and env V3 loop genes were also sequenced to determine their degree of variability of these genes and to infer co-receptor usage in the South African population. Gene-specific primers were designed to amplify the 579 bp Vif region and 440 bp containing the 105 bp V3 loop. Sequencing libraries from the resulting amplicons were prepared using either the Tn5 transposase or custom-based library preparation methods and sequenced on either an Illumina MiSeq or a MiniSeq platform generating millions of reads with good read coverage for variant calling. Phylogenetic analysis was done to determine the relatedness of the sequences. Major and minor variants were determined for HIV-1 and env V3 loop quasispecies was analysed for co-receptor usage; in an effort to draw inferences for the subsequent utility of Maraviroc as salvage therapy in South Africa. Results and Discussion Next generation library preparation; Tn5 tagmentation based and custom amplicon based protocols to sequence host and HIV genes were successfully developed and used to sequence and characterize variability in host cell APOBEC3D, F, G H, CCR5 and the HIV-1 vif gene and the V3 loop region of the env gene. The HIV-1 env V3 loop sequences generated (and quasispecies analyzed) were used to infer co-receptor usage in treatment-experienced individuals; in an effort to draw inferences for the subsequent utility of Maraviroc as salvage therapy in South Africa. Quality V3 loop sequences were obtained from 72 patients, with 5 years (range: 0-16) median duration on treatment. Subtypes A1, B and C viruses were identified at frequencies of 4% (3/72), 4% (3/72) and 92% (66/72) respectively. Fifty four percent (39/72) of patients were predicted to exclusively harbor R5 viral quasispecies; and 21% (15/72) to exclusively harbor X4 viral quasispecies. Twenty five percent of patients (18/72) were predicted to harbor a dual/mixture of R5X4 quasispecies. Of these 18 patients, about 28% (5/18) were predicted to harbor the R5+X4, a mixture with a majority R5 and minority X4 viruses, while about 72% (13/18) were predicted to harbor the R5X4+ a mixture with a majority X4 and minority R5 viruses. The proportion of all patients who harboured X4 viruses either exclusively or dual/mixture was 46% (33/72). Thirty-five percent (23/66) of the patients who were of HIV-1 subtype C were predicted to harbor X4 viruses (χ2=3.58; p=0.058), and 57% of these (13/23) were predicted to harbor X4 viruses exclusively. CD4+ cell count less than 350 cell/μl was associated with the presence of X4 viruses (χ2=4.99; p=0.008). The effectiveness of Maraviroc as a component in salvage therapy may be compromised for a significant number of chronically infected patients harboring CXCR4 utilizing viii viruses in the study cohort. Although from the current study a subset of patients harboring CCR5 utilizing viruses may benefit from Maraviroc, characterizing and identifying if variation in CCR5 are located at Maraviroc binding sites was of importance to investigate. The following variants; P35P, S75S, Y89Y, A335V and Y339F and their varying frequencies were detected in the CCR5 gene. The A335V variant was detected at a higher frequency of 17.4% (29/167). The G265S variant is reported for the first time in this study at 0.6% (1/167) frequency. The SNPs detected were in strong LD (D’= 1, R2 = 0.0) with minor deviation from the Hardy-Weinberg Equilibrium. These variants were not located at the binding motif of Maraviroc. The variants A335V and Y339F were detected at a higher frequency in this study than previously reported in South Africa. Variability in APOBEC3 host cell genes was also characterized in our study cohort. The following APOBEC3 variants compared to the GRCh37 consensus sequence were detected: R97C, R248K and T316T in A3D; R48P, A78V, A108S, S118S, R143R, I87L, Q87L, V231I, E245E, S229S, Y307C and S327S in A3F; S60S, H186R, R256H, Q275E and G363R in A3G and N15Δ, G105R, K140E, K121D, E178D in A3H. Minor allele frequency variants (MAF<5%); L221L, T238I, C224Y and C320Y in A3D; I87L, P97L and S229S in A3F; R256H, A109A, F119F and L371L in A3G, which are frequent in the European population, were also detected. In addition, novel R6K, L221R and T238I variants in A3D and I117I in A3F were detected. Most of the SNPs were in strong LD with minor deviation from the Hardy-Weinberg Equilibrium. Four, six, four, and three haplotypes were identified for A3D, A3F, A3G, and A3H respectively. In general, polymorphism in A3D, 3F, 3G and 3H were higher in our South African cohort than previously reported among other African, European and Asian populations. The APOBEC3 antagonist HIV-1 vif gene was also sequenced to determine the level of diversity in a South African population and also correlated with APOBEC3 variation. Functional Vif without frameshift mutation was observed in all samples except in 4 samples. The functional domain and motifs, such as Zn binding motifs, proline-rich domain, human casein kinase, and the N and C-terminal CBF interaction site were highly conserved. APOBEC binding motifs and the nuclear localization signal were less conserved in the South African HIV-1 Vif. APOBEC3 H variation strongly correlates with Vif variation. All the Vif sequences were subtype C, except one sample, which was identified as an A1/C recombinant. The vif gene in a South African population was under purifying selection, with the dS= 0.2581 and dN= 0.0684 and the dN/dS value = 0.265. There is a high genetic diversity in the South African vif gene, which may ix influence the neutralization and restriction of APOBEC genes. Conclusions In conclusion, the protocols developed in this study can be applied to amplify and sequence any host and HIV-1 genes of interest allowing much deeper and more sensitive profiling of host gene and HIV-1 genetic diversity. Our findings show that a highly significant number of chronically HIV-1 subtype C infected patients in Maraviroc-free treatment harbor CXCR4 utilizing viruses. The data is useful in the consideration of whether to include entry antagonists such as Maraviroc in alternative forms of treatment for patients failing second line treatment regimen in the study setting. The determination of co-receptor usage prior to initiation of therapy consisting of Maraviroc is suggested. Variation in the CCR5 coding region were observed at higher frequencies compare to other studies conducted in South African populations at different locations. This data may suggest that different populations in South Africa have different SNP frequencies. All the polymorphisms identified in the study were not located at the Maraviroc binding motif, therefore the subset of patient infected by R5 viruses may benefit from this drug. We have shown that significant APOBEC3 variation exists among an ethnically diverse population of South Africa by providing extensive data for 4 different A3 genes that are known to restrict HIV infection, but have only been sparsely studied in African populations. This study provides a baseline for future studies that would functionally characterize SNPs identified in this population, in order to understand the role of novel and/or low frequency variants observed. Ex vivo and in vivo studies will increase our understanding of how these variants might have cumulatively impacted the epidemic in Northern South Africa. This study also shows that there is a high level of HIV-1 Vif diversity in the study area. This diversity may impact the expression and packaging of Vif proteins, and the infectivity of HIV. In addition, a significant correlation was observed between HIV-1 Vif variation and APOBEC3 H haplotypes.