Natural Polymorphisms of integrase Among HIV Type 1-Infected South African Patients

Abstract

An HIV-1 subtype C specific assay was established for integrase genotyping from 51 integrase inhibitor-naive patient plasma samples and 22 antiretroviral drug-naive primary viral isolates from South Africa. Seventy-one of the 73 samples were classified as HIV-1 subtype C and two samples were unique AC and CG recombinants in integrase. Amino acid sequence analysis revealed there were no primary mutations (Y143R/C/H, Q148H/R/K, and N155H/S) associated with reduced susceptibility to the integrase inhibitors raltegravir and elvitegravir. However, one sample had the T97A mutation, three samples had the E157Q and V165I mutations, and the majority of samples contained the polymorphic mutation V72I. The expected finding of no major integrase mutations conferring resistance to integrase inhibitors suggests that this new antiretroviral drug class will be effective in our region where HIV-1 subtype C predominates. However, the impact of E157Q and other naturally occurring polymorphisms warrants further phenotypic investigation.

T he HIV-1 integrase enzyme (32 kDa, 288 amino acids) is responsible for the irreversible integration of the newly synthesized viral cDNA into the host cell chromosomal DNA, an essential step in the HIV-1 replication cycle.¹ Integrase is composed of three functional domains: the N-terminal domain (amino acids 1–50), which coordinates zinc binding, the catalytic core domain (amino acids 51–212), which contains three catalytic residues forming a DDE motif, and the C-terminal domain (amino acids 213–288), which is involved in host DNA binding and integrase oligomerization.²

Integrase functions in cells within the context of high-molecular-weight preintegration complexes that are formed in the cytoplasm after reverse transcription of the viral RNA. HIV-1 integration is subdivided into two steps with distinct kinetics and requirements. In the cytoplasm, a dinucleotide adjacent to a phylogenetically conserved CA is hydrolytically cleaved off the 3′ end of each strand of the linear viral cDNA by integrase in the preintegration complex (3′-processing).^3

–7 After nuclear transport of the preintegration complex, two phosphodiester bonds on each strand of the chromosomal DNA are targeted by the 3′-hydroxyls at the ends of the viral DNA, resulting in its insertion within the host genome (strand transfer).^3

–7 Removal of the two unpaired nucleotides on the 5′ viral DNA ends and gap filling of the interfaces between the viral and host genomic DNA are then done using host DNA repair machinery via mechanisms that are not yet fully understood. Completion of integration results in a fully functional provirus, which can then be used to initiate viral DNA transcription. There is no known human homologue to integrase, thus making it an ideal antiretroviral drug target.

Raltegravir (Isentress, Merck & Co., Inc., Whitehouse Station, NJ), the first integrase inhibitor to be approved by the U.S. Food and Drug Administration (October 2007), has successfully been used in combination with other antiretroviral drugs for treatment-experienced HIV-1 patients with multidrug-resistant strains. Raltegravir blocks viral replication by inhibiting DNA strand transfer.⁸ However, drug resistance to integrase inhibitors has been shown to occur both in vivo and in vitro, and is the consequence of mutations that are selected in the viral integrase gene targeted by the strand transfer antiretroviral drugs such as raltegravir and elvitegravir (Gilead Sciences, Foster City, CA). To date, there are at least 30 substitutions that have been specifically associated with the development of resistance to raltegravir and/or elvitegravir (http://hivdb.stanford.edu). However, the main mutational pathways associated with raltegravir and elvitegravir resistance include signature mutations at integrase positions 143, 148, and 155.

Access to antiretroviral drugs has escalated dramatically in the sub-Saharan African region in the past decade. Regimens in this region almost universally place a thymidine analogue with lamivudine and a nonnucleoside reverse transcriptase inhibitor as first line therapy, with a protease inhibitor-based regimen for those failing first line therapies. However, these regimens carry significant toxicities, requiring drug substitutions. In addition, an increasing number of patients require alternative regimens as they fail first and second line regimens due to nonadherence, toxicity, or the development of drug resistance. Thus, the effectiveness of integrase inhibitors such as raltegravir needs to be investigated as alternative antiretroviral drug treatment options are required. Several groups have published genotypic assays for the amplification and sequencing of integrase from diverse HIV-1 group M subtypes,^9
–11 however, we opted to establish a subtype C-specific integrase genotyping assay. This study describes the baseline mutations in integrase in HIV-1 subtype C (and unique recombinant) infected integrase inhibitor-naive individuals in South Africa.

A total of 80 samples from patients attending the Charlotte Maxeke Johannesburg Hospital were available for purposes of this study. These included 58 randomly selected patient plasma samples sent for routine viral load monitoring during 2009, and 22 well-characterized primary HIV-1 isolates from antiretroviral drug-naive AIDS patients isolated during 2005.¹² The patients viral loads ranged from 1500 to >3 million RNA copies/ml, and included antiretroviral drug-naive and treatment-experienced patients. Ethical clearance for the study was obtained for Research on Human Subjects (Medical) at the University of the Witwatersrand (clearance number M090575). Viral RNA was extracted from the 58 patient plasma samples using the automated Roche MagNa Pure LC analyzer and the MagNa Pure LC Total Nucleic Acid Isolation kit (Roche, Germany), according to the manufacturer's instructions. Proviral DNA from the 22 viral isolates was extracted from p24 antigen-positive cultures with a High Pure PCR Template Preparation kit (Roche, GmbH, Germany) according to the manufacturer's instructions. The full-length integrase gene was reverse transcriptase polymerase chain reaction (RT-PCR) amplified into cDNA from the extracted viral RNA samples using the reverse primer SQ9RC(3) (5′-TTGGATGTCTGCTTTCATAATGAT-3′)¹³ and the Expand Reverse Transcriptase (RT) kit (Roche Diagnostics, Germany), as per the manufacturer's instructions. Amplification from synthesized cDNA or extracted proviral DNA was performed using primers SQ7F(2) (5′-AGGAGCAGAAACTTTCTATGTAGATGG-3′)¹³ and SQ9RC(3) with the Expand High Fidelity^PLUSkit (Roche, Germany), as per the manufacturer's instructions. PCR amplicons (approximately 1.3 kb in length) were purified using the High Pure PCR Product Purification kit (Roche), as per the manufacturer's instructions.

Primers spanning the full-length integrase (∼880 bp) were used to sequence the PCR products in both directions (population-based sequencing). These included QFIN1F (5′- ATGGGTACCAGCACACAAAG-3′), QFIN2F (5′-TAGCAGGAAGATGGCCAGTC-3′), QFIN1R (5′-CTCTCCTTGAAATATACATATGGTGC-3′), and QFIN2R (5′-CTACTACTCCTTGACTTTGGGG-3′). Purified amplicons were sequenced with the ABI Big Dye terminator system (v3.1) according to the manufacturer's instructions on the ABI Prism 3730 and 3100-Avant Genetic Analyzer (Applied Biosystems, Foster City, CA). Sequence data were edited using the Sequencing Analysis V3.3 program (Applied Biosystems), and the complete integrase sequences were assembled and manually edited using Sequencher V4.7 (Genecodes, Ann Arbor, MI). For subtyping, a multiple alignment of the full length integrase region with references from HIV-1 subtypes A to K, CRF01_AE, and CRF02_AG (http://hiv-web.lanl.gov) was generated in Clustal X. Aligned sequences were converted to MEGA V3.0 format and used in phylogenetic and molecular evolutionary analyses. A phylogeny reconstruction of each integrase gene was performed by neighbor-joining using the Kimura two-parameter distance matrix. The stability of the nodes was assessed by bootstrap analysis (1000 replicates) and bootstrap values >70% were considered significant. Nucleotide sequences were converted to amino acid sequences and compared to the 2004 consensus C sequence available from the Los Alamos website (http://hiv-web.lanl.gov). The amino acid sequence alignment was extensively analyzed for the presence of primary mutations, nonpolymorphic and polymorphic mutations associated with resistance to known integrase inhibitors (http://hivdb.stanford.edu).

RT-PCR and PCR amplification was successful for 91.3% of all samples, and included 51 of the 58 patient plasma samples (not dependent on viral load) and all 22 proviral DNA samples, respectively. All 73 of the amplified samples were successfully sequenced. The integrase amplification and sequencing success rates are similar to those reported previously.^9
–11 Nucleotide sequences of the 73 newly characterized full-length integrase genes ranged from 867 to 876 base pairs, with intact open reading frames confirming the presence of functional integrase genes in all patients and primary virus isolates. The integrase intrasubtype nucleotide diversity for all 73 newly characterized samples varied between 2.2% and 7.9%, similar to what has been reported previously.¹⁴ Phylogenetic analysis of the full-length integrase nucleotide sequences, together with viruses from the major subtypes, revealed that 71 sequences clustered within HIV-1 subtype C with a bootstrap value of 100%, whereas two samples (09ZAP35 and 09ZAP52) showed evidence of recombination (Fig. 1). These two sequences were further analyzed using the RIP 3.0 Program (http://hiv-web.lanl.gov), which showed that samples 09ZAP35 and 09ZAP52 were unique AC and CG recombinants in integrase (results not shown). Since subtype A predominates in East, Central, and West Africa, and subtype G predominates in Central and West Africa, it is possible that these patients originated from these countries and a recombination event occurred in South Africa.

FIG. 1.

Phylogenetic relationships of the 73 newly characterized full-length integrase sequences with HIV-1 subtype reference sequences from the Los Alamos database (http://hiv-web.lanl.gov). The unique recombinant sequences are boxed. Phylogenetic trees were constructed from nucleotide sequence alignments, using neighbor joining and the Kimura two-parameter distance matrix. The reliability of the branching order was estimated from 1000 bootstrap replicates, and only bootstrap values of 70% or higher are shown.

An alignment of the 73 integrase amino acid sequences was extensively analyzed and compared, and a distribution of the variants is shown in Fig. 2. The proteins varied from 288 to 292 amino acids in length, primarily due to the presence of an unusual tetrapeptide insertion (QNME) in the C terminus of the protein in 23 patient and primary virus isolate samples. This unusual insertion has been noted previously.^15,16 The C terminus has been shown to be important for integrase dimerization and DNA binding. This is crucial for proper function and positioning of the catalytic domain close to its target site. This finding warrants further investigation to determine the impact of the insertion on the function of the integrase enzyme.

FIG. 2.

Distribution of variants among the 73 newly characterized full-length integrase sequences. The 2002 consensus subtype C sequence (http://hiv-web.lanl.gov) is shown at the top of each 30 amino acid sequence section. Numbers below the consensus sequence indicate the number of annotated sequences (out of 73) containing an unambiguous amino acid at that position. Beneath that are the amino acid variants with superscript numbers indicating the amount of times they occur. Integrase functional domains and important residues are highlighted. The N terminal domain is highlighted in yellow, the catalytic core domain in white, and the C terminal domain in blue; the catalytic triad (DDE) residues are in red and relevant integrase inhibitor resistance mutations are in green.

The sequences were analyzed for the presence of integrase mutations associated with reduced in vitro susceptibility to raltegravir (E92Q, F121Y, E138AK, G140AS, Y143RCH, S147G, Q148HRK, N155HS) and elvitegravir (T66IAK, E92Q, F121Y, E138AK, G140AS, S147G, Q148HRK, S153Y, N155HS, R263K) (Stanford Database, updated 15 September, 2009). None of the samples contained the mutations associated with primary resistance (Y143RCH, Q148HRK, and N155HS).

The E157Q mutation was observed among 4.1% (n = 3) of patient samples, including two subtype C patient samples (09ZAP34 and 09ZAP53) and one recombinant (09ZAP52) patient sample (Fig. 2). Until recently the E157Q mutation was listed among the mutations associated with reduced in vitro susceptibility to raltegravir and elvitegravir (http://hivdb.stanford.edu), but is now described as a minimally polymorphic mutation that is selected in vitro by elvitegravir and rarely in vivo by raltegravir, having a minimal, if any effect on integrase inhibitor susceptibility. Interestingly, among several other studies characterizing a total of 501 subtype C integrase sequences,^9,11,14,17 the E157Q mutation was reported in only one sample (1.4%).¹¹

The isoleucine at position 72 (I72) was identified at a high frequency among the subtype C samples and the 05ZAP52 recombinant (67.1%; n = 49), whereas V72 and T72 were found in 28.7% (n = 21, including the 05ZAP35 recombinant) and 1.4% (n = 1) of the patients, respectively, confirming their status as a naturally occurring subtype C polymorphism. This confirms the findings by Rhee et al. ¹⁴ who characterized naturally occurring variation in previously published HIV-1 group M integrase sequences, and found that 390 of 432 (90.3%) integrase-inhibitor-naive subtype C samples analyzed contained isoleucine at position 72, and only 9.4% and 0.2% contained a valine and threonine, respectively. The polymorphic mutation, T97A, associated with integrase inhibitors was observed in one primary virus sample (05ZAFV27), whereas K156N, another accessory integrase inhibitor resistance mutation, was found in patient 09ZAP34. The uncommon polymorphism K156N was not reported in subtype C, but occurs in 9% of subtype B and 5% of subtype D sequences.¹⁴ By contrast, T97A was observed in subtypes A, B, C, D, and CRF01_AE and CRF02_AG, albeit at low frequencies.¹⁴

Because the antiretroviral drug history of the patient samples analyzed in 2009 was unknown, we could not address the influence of prior antiretroviral therapy on integrase polymorphisms. Ceccherini-Silberstein et al. ¹⁸ reported several integrase polymorphisms, for example, M154I, V165I, and M185L, that were positively associated with specific reverse transcriptase mutations (F227L and T215Y) in treated patients. In our study, the V165I polymorphism was noted in three samples, 09ZAP3, 09ZAP54 (unknown drug history), and 05ZAFV29 (antiretroviral drug naive). This highlights the need for ongoing surveillance of integrase mutations as well as HIV-1 subtype C-specific phenotypic studies to elucidate the relevance of the above-mentioned polymorphisms (as well as naturally occurring polymorphisms) in the absence of the signature integrase mutations on susceptibility/resistance to integrase inhibitors.

In conclusion, none of the three previously reported major mutations (Y143R/C/H, Q148H, and N155H) associated with resistance to integrase inhibitors was observed in our cohort, indicating that integrase inhibitors may be used as a treatment option in South Africa.

Sequence Data

The 73 integrase sequences were submitted to GenBank using Sequin V5.35 (http://www.ncbi.nlm.nih.gov), and are available under the accession numbers GQ872465 to GQ872538.

Footnotes

Acknowledgments

We are grateful for research funding from AuTEK Biomed (Mintek and Harmony Gold Mining Company) and the Department of Molecular Medicine and Haematology.

Author Disclosure Statement

No competing financial interests exist.

References

Engelman

, Mizuuchi

, Craigie

. HIV-1 DNA integration: Mechanisms of viral DNA cleavage and DNA strand transfer. Cell, 1991; 67:1211–1221.

Van Maele

, Busschots

, Vandekerckhove

, Christ

, Debyser

. Cellular co-factors of HIV-1 integration. Trends Biochem Sci, 2006; 31:98–105.

LaFemina

, Schneider

, Robbins

et al. Requirement of active human immunodeficiency virus type 1 integrase enzyme for productive infection of human T-lymphoid cells. J Virol, 1992; 66:7414–7419.

Sakai

, Kawamura

, Sakuragi

et al. Integration is essential for efficient gene expression of human immunodeficiency virus type 1. J Virol, 1993; 67:1169–1174.

Miller

, Farnet

, Bushman

. Human immunodeficiency virus type 1 preintegration complexes: Studies of organization and composition. J Virol, 1997; 71:5382–5390.

Van Maele

, Debyser

. HIV-1 integration: An interplay between HIV-1 integrase, cellular and viral proteins. AIDS Rev, 2005; 7:26–43.

Yoder

, Bushman

. Repair of gaps in retroviral DNA integration intermediates. J Virol, 2000; 74:11191–11200.

Pommier

, Johnson

, Marchand

. Integrase inhibitors to treat HIV/AIDS. Nat Rev Drug Discov, 2005; 4:236–248.

Van Laethem

, Schrooten

, Covens

et al. A genotypic assay for the amplification and sequencing of integrase from diverse HIV-1 group M subtypes. J Virol Methods, 2008; 153:176–181.

10.

Van Baelen

, Rondelez

, van Eygen

et al. A combined genotypic and phenotypic human immunodeficiency virus type 1 recombinant virus assay for the reverse transcriptase and integrase genes. J Virol Methods, 2009; 161:231–239.

11.

Eshleman

, Hudelson

, Smith

et al. Analysis of pol integrase sequences in diverse HIV type 1 strains using a prototype genotyping assay. AIDS Res Hum Retroviruses, 2009; 25:343–345.

12.

Connell

, Michler

, Capovilla

et al. Emergence of X4 usage amongst HIV-1 subtype C: Evidence for an evolving epidemic in South Africa. AIDS, 2008; 22:896–899.

13.

Rousseau

, Birditt

, McKaya

et al. Large-scale amplification, cloning and sequencing of near full-length HIV-1 subtype C genomes. J Virol Methods, 2006; 136:118–125.

14.

Rhee

, Liu

, Kiuchi

et al. Natural variation of HIV-1 group M integrase: Implications for a new class of antiretroviral inhibitors. Retrovirology, 2008; 5:74.

15.

Hunt

, Papathanasopolous

, Gray

, Tiemessen

. Characterisation of near-full length genome sequences of three South African human immunodeficiency virus type 1 subtype C isolates. Virus Genes, 2003; 26:49–56.

16.

Tzitzivacos

, Tiemessen

, Stevens

, Papathanasopoulos

. Viral genetic determinants of nonprogressive HIV-1 subtype C infection in antiretroviral drug naïve children. AIDS Res Hum Retroviruses, 2009; 25:1141–1148.

17.

Passaes

, Guimaraes

, Fernandez

SLC

et al. Lack of primary mutations associated with integrase inhibitors among HIV-1 subtypes B, C, and F circulating in Brazil. J Acquir Immune Defic Syndr, 2009; 51:7–12.

18.

Ceccherini-Silberstein

, Malet

, Fabeni

et al. Specific mutations related to resistance to HIV-1 integrase inhibitors are associated with reverse transcriptase mutations in HAART-treated patients. Antivir Ther, 2007; 12:S6.