Original article Universal profiling of HIV-1 pol for genotypic study and resistance analysis across subtypes

Transcription

1 Antiviral Therapy 2011; 16: (doi: /IMP1892) Original article Universal profiling of HIV-1 pol for genotypic study and resistance analysis across subtypes Ting Nie 1, Mervi Detorio 1, Raymond F Schinazi 1 * 1 Center for AIDS Research, Department of Pediatrics, Laboratory of Biochemical Pharmacology, Emory University/VA Medical Center, Decatur, GA, USA *Corresponding author rschina@emory.edu Background: The increased use of anti-hiv-1 treatments in developing countries primarily infected by non-b subtypes necessitates development of novel tools to assess susceptibility and resistance. HIV-1 genomes are highly polymorphic and present challenges for the development of universal protocols capable of screening across subtypes. Currently available viral genotyping methods are useful for viral quantification, but are inadequate for sequence profiling or comprehensive mutation detection in the variable regions of HIV polymerase (pol). Methods: A novel set of universal primers within pol, with consensus among a variety of HIV-1 subtypes, was developed. One-round amplification was performed by one-step reverse transcription PCR on 79 samples from HIV-1 subtypes. Using a second set of primers, the amplified fragment was sequenced and assembled to produce a profile database per sample. Results: First-round amplification using universal primers generated a unique amplicon encompassing the major pol regions in all tested HIV-1 subtype samples. Sequence analysis of the amplified fragment not only confirmed the subtype of each HIV-1 isolate but also identified resistance mutations in the pol genes of HIV 1, including protease, reverse transcriptase, connection, RNase H, and integrase. Last, some of these primers were used to develop a viral load test using quantitative real time-pcr. Conclusions: A novel protocol was produced to effectively identify and simultaneously generate extensive sequence profiles of pol genes across HIV-1 subtypes. This protocol allows for expeditious and cost-effective mutation detection, genotypic evaluation and viral load determination in multiple HIV-1 subtypes. Introduction Three classes of HIV-1 exist: group M (major), group O (outlier) and group N (new). Group M, the most diverse group, is comprised of nine subtypes (A D, F H, J and K) and includes several circulating recombinant forms (CRFs) [1,2]. Geographic and demographic distribution amongst the subtypes is heterogeneous, with one or more subtypes predominating infection in particular geographic locations; for example, subtype C predominates in Southern and Eastern Africa, and China [3,4]. Due to high levels of migration and contact with persons from non-b endemic countries, the prevalence of non-b subtypes and infection is predicted to increase in the developed western world, including the United States [5]. Despite this trend, most researchbased efforts in North America and Europe have been concentrated on the regionally dominant subtype B. Recent reports demonstrated that HIV-1 subtype B accounts for only 12% of the estimated 40 million HIV-infected individuals worldwide, whereas subtypes C and A collectively account for greater than 70% of all new infections [6]. HIV-1 genomes are highly polymorphic and present challenges for the development of universal protocols capable of screening across subtypes. Viral genotyping was previously reported using a series of primers designed to amplify HIV-1 gag [7], which is primarily useful for viral quantification, yet inadequate for mutation detection in highly polymorphic reverse transcriptase (RT) regions. Recently, other regions in polymerase (pol) of HIV-1, for example, integrase (IN), were also utilized for developing viral load testing across subtypes [8,9]. These protocols are efficient for genotyping and quantification in clinical diagnosis, which are, however, not suitable for comprehensive assessment of drug resistance. A recent review indicates that roughly 60% of non-b infected persons who fail antiviral therapy 2011 International Medical Press (print) (online) 1267

2 T Nie et al. have genotypic evidence of drug resistance [2]. This resistance is due, in part, to natural polymorphisms in the pol gene: 10 15% sequence variation in protease (PR) and RT [10]. Another recent report demonstrated that naturally occurring polymorphisms in IN were closely associated with antiviral drug resistance among 73 subjects of subtype C [11]. Despite recent efforts to procure vaccine development initiatives for non-b subtypes [12], the antiviral chemotherapy, phylogenetic subtype identification and profiling, and drug resistance mutation database for non-b subtypes remains mostly understudied [13]. Independent of these shortcomings, recent additions to the HIV Drug Resistance Database [14], including protease inhibitor (PI), nucleoside reverse transcriptase inhibitor (NRTI), non-nucleoside reverse transcriptase inhibitor (NNRTI), and integrase inhibitor (INI)-related drug resistance mutations in 157 non-b infected persons, have provided a foundation for a more comprehensive understanding of resistance mutations conferred by multiple classes of antiretroviral agents across subtypes. Unfortunately, a significant detriment to analysing resistance across subtypes is the lack of a universal sequencing and direct detection protocol for HIV-1 RT mutations. The recent development of universal genotyping protocols have provided some advancement to this end [15], however, the increasing demand for a universal protocol for sequencing pol and profiling mutations across subtypes necessitates design of a novel approach. Recent reports indicate that other regions of pol, including connection domain, RNase H and IN, emerged as significant contributors to anti-hiv-1 drug resistance, in addition to RT [16 19]. Therefore, in order to amplify and sequence for drug resistance mutants or polymorphisms across subtypes, a set of novel universal primers within pol were designed, with consensus among HIV-1 subtypes. Successful one-round amplification was performed by one-step reverse transcription PCR in 79 HIV-1 subtype samples, to generate a unique amplicon with a size of approximately 2.6 kb. Using a second set of primers, the 2.6 kb amplicon encompassing the major pol regions was sequenced to produce a profile database. Together, this methodology allowed for the assessment of resistance mutations in the PR, RT, connection, RNase H and IN encoding regions of the HIV-1 genome. Lastly, some of these primers were used to develop a viral load assay using quantitative real time-pcr. Methods Primer design Universal primers (A, B, C and D) were specifically designed to target highly conserved regions of the pol gene in all HIV-1 subtypes. Additional primer sets (a1, a2, b1 and b2) were designed for sequencing. Two primer sets (O1f/O2r and O3f/O4r) were designed specifically for group O-based sequencing. The relative location of each primer is illustrated in Figure 1B and primer sequences are listed in Table 1. Bf and Br are the forward and reverse orientation of one primer, respectively; as are Cf and Cr, O1f and O2r, O3f and O4f. Isolation of RNA from isolates A total of 79 HIV-1 isolates from multiple subtypes and CRFs were obtained from the NIH AIDS Research and Reference Reagent Program (Germantown, MD, USA), including A (n=11), B (n=12), C (n=17), CRF01_AE (n=16), CRF02_AG (n=2), D (n=11), F (n=2) and G (n=4) of group M, group O (n=3) and N (n=1). 60 µl of total RNA was isolated from 300 µl viral supernatants using an EZ1 Advanced automated purification machine (Qiagen, Valencia, CA, USA). One-step reverse transcription PCR Using the universal primers A and D, one-step reverse transcription PCR was performed on 300 ng of extracted total RNA using a SuperScript III kit (Invitrogen, Carlsbad, CA, USA) in a 50 µl amplification PCR mixture (2 reaction buffer, 25 µl; total RNA sample, µl per 300 ng; Primer A (10 µm), 1 µl; Primer D (10 µm), 1 µl; SuperScript III RT/Platinum Taq mix, 1 µl; add dh 2 O up to 50 µl). The running conditions were as follows: 50 C, 30 min; 94 C, 5 min; 94 C, 30 sec; 49 C, 30 sec; 68 C, 2.5 min; 39 cycles; 68 C, 10 min. As a result, one approximately 2.6 kb DNA fragment was universally amplified across all the subtypes. All PCR products were purified using the QIAquick PCR Purification Kit (Qiagen). These reaction conditions also worked for the other universal primer sets, including Bf and Cr, Bf and D. After purification, all of the samples were sent to the DNA Sequencing Core at the University of Michigan (Ann Arbor, MI, USA) for sequencing using secondary sets of primers (forward primers: A, a1, a2, b1, Bf, b2f; reverse primers: Br, b2r, Cr; for group O, primers: O1f, O2r, O3f, O4r). Sequencing analysis for PR, RT, RNase H and IN All retrieved sequences were assembled and loaded into the Vector NTI Advance 11 software (Invitrogen). The cdna sequences were blasted using the HIV Drug Resistance Stanford Database for clade/subtype classification and mutation detection among the regions of PR, RT, connection domain, RNase H and IN. Quantitative real time-pcr The two universal primers, Cf and D, in the IN region generated one relatively small DNA fragment of 140 bp, a size suitable for developing quantitative real time-pcr across subtypes. The standard curve was generated using International Medical Press

3 Genotypic analysis of HIV-1 subtypes Figure 1. Universal and sequencing primers designed for genotyping and sequencing the pol of HIV-1 subtypes A A a1 a2 b1 Br/Bf b2r/b2f Cr/Cf D PR RT Connection RNase H IN B M (A) Diagram of the positions of the universal (A, B, C and D) and sequencing (a1, a2, b1 and b2) primers in the HIV-1 pol genome. (B) Electrophoresis of one approximately 2.6 kb amplicon using one-step reverse transcription PCR and universal primers A and D; shown are representative samples of subtypes A, B, C, D, F, G, groups O and N (lanes 1 8). f, forward; IN, integrase; M, molecular weight marker; PR, protease; r, reverse; RT, reverse transcriptase. Table 1. Sequences of HIV-1 pol universal primers and sequencing primers Primer Direction Type Sequence of primer (5 3 ) Gene A Forward Universal a CAGGAGCAGATGATACAG PR Bf Forward Universal a TGGACTGTCAATGATATACA RT Br Reverse Universal a TGTATATCATTGACAGTCCA RT Cf Forward Universal a ACAGTGCAGGGGAAAGAA IN Cr Reverse Universal a TTCTTTCCCCTGCACTGT IN D Reverse Universal a CCCTTCACCTTTCCAGAG IN a1 Forward Sequencing b ATAGGGGGAATTGGAGGTTTTAT PR a2 Forward Sequencing b AGGAATGGATGGCCCAAA PR b1 Forward Sequencing b GGGTTATGAACTCCATCCTGATAAATGGAC RT b2f Forward Sequencing b TGGAGAGCAATGGCTAGTGA RNaseH b2r Reverse Sequencing b TCACTAGCCATTGCTCTCCA RNaseH O1f Forward Sequencing c AACTAAAACCAGGAATGGATGG RT O2r Reverse Sequencing c CCTTGTTAGGCAATTGGATGG RT O3f Forward Sequencing c GCCTAACAAGGATGTGTGGACAG RNaseH O4r Reverse Sequencing c CATGATCTTCTTGTGCCTGGT RNaseH a Universal primers are designed for both PCR and sequencing. b Sequencing primers are designed only for sequencing. c Sequencing primers are designed specifically for group O. IN, integrase; PR, protease; RT, reverse transcriptase. a serial dilution of xxlai plasmid (containing the full genome of HIV-1, subtype B, length =9229 bp), a gift from John W Mellors (University of Pittsburgh, Pittsburgh, PA, USA). The quantitative real time-pcr was performed using the LightCycler RNA Master HybProbe Kit (Roche Applied Science, Indianapolis, IN, USA), with primers Cf and D as well as a synthetic 6-FAM-labeled probe (Applied Biosystems, Foster City, CA, USA), which was specifically designed against one IN region between primer Cf and primer D, and ng of total RNA extracts were used per reaction. The probe sequence was: 5 -(6-FAM)-TTC AAA ATT TTC GGG TTT ATT ACA GGG ACA GCA GAG A-(TAMRA)-3. The running conditions were as follows: 61 C, 20 min; 95 C, 30 sec; 95 C, 10 sec; 50 C, 20 sec; 72 C, 20 sec; 45 cycles; 40 C, 30 sec. Antiviral Therapy

4 T Nie et al. Table 2. Viral load determined in a subset of 79 HIV-1 isolates from multiple subtypes by quantitative real time- PCR using universal primers Cf and D Clade type ID name Copy number/ml sem A 93RW ,778 89,377 A RW , ,978 A UG ,987 27,679 B 92BR014 1,201,726 57,571 B BR ,676, ,247 B 92HT599 2,306, ,353 C 98T2017 3,461,224 64,320 C 93USNG31 1,241,942 41,530 C 98IN017 2,401, ,264 CRF01_AE 93TH051 1,077,422 59,453 CRF01_AE THA ,286, ,134 CRF01_AE THA , D 92UG001 1,408, ,881 D UG ,380, ,649 D UG ,231, ,903 F 93BR020 1,331,131 75,432 G G3 1,932, ,954 G JV1083 1,731, ,970 G R132 3,003,724 82,930 N YBF30 1, O BCF Negative control H 2 O 17 9 sem, the standard error of the mean copy number/ml. XL-TOPO cloning To clone the 2.6 kb amplicons of HIV-1 pol, the TOPO XL PCR Cloning kit (Invitrogen) was utilized. The DNA fragments were purified and incubated with the pcr-xl-topo vector for 5 min at room temperature, transformed and screened for positive clones. The clones were purified by QIAprep Spin MiniPrep kit (Invitrogen) and identified by PCR as positives. The positive clones were sequenced with 2 vector primers, M13 reverse and T7, as well as 4 HIV-1 pol primers (forward: a2, b1, reverse: Br, b2r). According to the TOPO XL PCR Cloning kit manual, the efficiency was improved by gel purifying the PCR products using the QIAquick Gel Extraction kit (Qiagen). Results Amplification of gene pol in HIV-1 subtypes using novel universal primers Novel universal primers (A, B, C and D) for HIV-1 subtypes were designed on homologous regions from a dozen sequences of subtypes and CRFs (Figure 1 and Table 1), with nearly 100% homology to sequences of group M pol and 94.4% to those of groups N and O. One-step reverse transcription PCR amplification and sequence analysis of the pol sequences using universal primers A and D produced a unique amplicon of roughly 2.6 kb in all of the subtypes examined (Figure 1B shows the PCR product from eight representative HIV-1 subtypes). These results confirmed that our primer set had the potential for amplification of all the available HIV-1 subtypes. The amplification reactions were performed successfully using universal primers A and D, as long as the total RNA from isolated samples was detectable (10 40 ng/µl for most RNA samples in our studies isolated by EZ1 machine, see Methods). However, adequate total RNA (300 ng per reaction) did not guarantee positive results due to variation in HIV-1 infection rate and growth. We then determined the minimum detection level of our universal primers A and D, which were capable of amplifying DNA samples at 1 10 pg of xxlai (equivalent to copies of the plasmid per reaction); relevantly, a few RNA samples of HIV-1 subtypes (from subtype A, B and C) previously undetected by PCR were quantified by the quantitative real time- PCR using primers Cf and D (as described in the next paragraph) at an average 8,000 10,000 copies per ml. Taken together, it is safe to conclude that our universal one-round amplification, that is, one-step reverse transcription PCR using primers A and D, can detect gene pol of HIV-1 more than 10 4 copies per ml (at least for group M). Primers B and C, also located in consensus regions of HIV-1, can generate shorter PCR products, some of which are suitable for development of quantitative real time-pcr across subtypes (for instance, primer Cf combined with D produces an 140 bp amplicon). The efficiency of primers Cf and D was evaluated by quantitative real time-pcr using dilutions of HIV-1 subtype B (xxlai) as a standard curve; the efficiency was calculated to be 2.0, which justifies the use of these universal primers for viral load analysis (Table 2). The subtype samples tested had viral loads with ranges of detection levels as low as 100 copies/ml, confirming that the sensitivity of our universal primer sets was at least comparable to other known methods [8]. Sequencing HIV-1 subtypes with primer walking Novel secondary primers designed with relatively less homology (approximately 88 90%) between HIV-1 subtypes were examined in sequencing the purified 2.6 kb amplicons, which were amplified using universal primers A and D (Figure 1B and Table 1). Using these internal primer sets, sequencing of the pol region was performed with primer walking; these pol sequence data were then analysed and assembled by Vector NTI. Subsequently, a BLAST of the HIV-1 Drug Resistance Database for subtype and mutation profiling was conducted. Sequencing of the PCR product not only confirmed the identity of each HIV-1 isolate, but also International Medical Press

5 Genotypic analysis of HIV-1 subtypes covered almost entire regions of pol genes. We therefore concluded that our protocol is suitable for extensive sequencing of gene pol, including regions from codon number 26 of PR to codon number 247 of IN (that is, PR, codons 26 99; RT/Connection/RNase H, codons 1 560; IN, codons 1 247) that segment confirmed to be 2640 bp in length and to comprise about 93% of the entire pol gene. To complete the pol sequencing, we successfully utilized a complementary set of primers A and D, that is, reverse A and forward D, and the RACE (rapid amplification of cdna ends) method to clone the missing segments of the pol sequence (PR codon no 1 25 and IN codon no , 66 amino acids in total) for further sequencing. Profiling HIV-1 subtypes for drug resistance and mutation Determining the mechanisms responsible for emergence of resistance mutations has culminated in a demand for a broad-based and comprehensive methodology to study mutations across multiple regions of the genome. To this end, mutations in connection domain, RNase H and IN, recently have become more of interest in definition of these mechanisms [17 19]. The main application of our novel protocol, irrespective of identification of HIV-1 subtypes, is to sequence the pol gene for profiling and analysing of resistance mutations in drugnaive or drug-treated subjects. As mentioned earlier, the pol genes of HIV-1 subtypes were sequenced and assembled with primer walking (Figure 1A). Using Vector NTI to analyse the sequencing data, we generated one HIV-1 genotypic profile for each sample. The pol genes of each subtype and CRF of HIV-1 obtained from the NIH AIDS Research and Reference Reagent Program, including A (n=11), B (n=12), C (n=17), CRF01_AE (n=16), CRF02_AG (n=2), D (n=11), F (n=2) and G (n=4) of group M, and group N (n=1), were then successfully sequenced and profiled, with the exception of samples from group O (n=3). Therefore, for group O sequencing, we designed two additional primer sets (Table 1), O1f/O2r and O3f/O4r, for sequencing the PCR product amplified by primers A and D on these group O isolates (n=3). Based on the HIV Drug Resistance Database, all drug resistance mutations were identified in the treatment-naive HIV-1 isolates (Table 3). A number of previously reported resistance mutations were detected and classified in Table 3; for instances, several INI mutations were identified, including V151I in one subtype A sample (92RW009) and subtype B samples (BR92019 and 92HT599); G163R in one subtype C sample (IN93905) and one subtype F sample (BZ126); E138D in one subtype C sample (98IN017); and M154I in one CRF01_AE sample (THA93060). The existence of the resistance mutations in group O samples was not totally surprising, as this group is known to be the most variable among all the subtypes [20]. Notably, besides the known resistance mutations listed in Table 3, a number of other polymorphisms existing among the pol genes, which could be natural variations or unknown mutations, were not listed. XL-TOPO cloning of HIV-1 pol and genotypic study In addition to the applications mentioned above, the extensive sequencing protocol for HIV-1 genotypic study was utilized for assessment of drug resistance among individual clones in the viral pools collected from β-d-3 -azido-2,3 -dideoxy-2,6-diaminopurine (AZD)-treated, HIV-infected peripheral blood mononuclear (PBM) cells [21]. Due to the length, the 2.6 kb PCR products amplified by primers A and D were successfully cloned using XL-TOPO cloning kit. Furthermore, all positive clones were sequenced with primers (forward: a2, b1; reverse: Br, b2) and two vector primers, T7 and M13 reverse. Using this method, mutations in viral pools collected from AZD-treated human PBM cells were determined (Table 4). As expected, no known PI or INI mutations were found in the PR and IN regions of these clones. Neither were any abnormal polymorphisms found in these regions. In the RT region, all samples had the mutation of T215Y, with either the single mutant T215Y or the double mutant M41L/T215Y occurring in 80% and 20% of the population, respectively. This analysis supports the use of this novel protocol for direct genotypic study on HIV-1 viral pools. Overall, this work represents a valuable protocol that provides a multitier assessment of viral sequences and resistance mutations across multiple regions of the HIV-1 genome, and which is broadly applicable across subtypes. Tandem use of these primers to quantify viral loads provides a secondary and broad-based application for these primers. Discussion A novel protocol was reported here for identification of HIV-1 subtypes and thorough analysis of pol gene and drug resistance mutations. The capacity for rapid profiling of HIV-1 sequences makes this protocol useful for efficient analysis of HIV-1 subtype demographics and drug resistance across populations around the world. This too provides a foundation for initiation of in-depth research designed to elucidate the effect of ongoing migration on the global distribution of HIV-1 subtypes. The existence of resistance mutations observed among group M samples might be due, in part, to random sequence variation (Table 3). However, for group Antiviral Therapy

6 T Nie et al. Table 3. Drug resistance mutation profiles of HIV-1 subtypes and CRFs in pol Clade type ID name PR RT IN A UG93089 None None None A UG94103 None None None A UG92035 None None None A UG93086 None None None A RW92024 None None None A ZA97009 None None None A UG92037 None None None A 93RW020 None None None A RW92008 None None None A UG92029 None None None A 92RW009 None None V151I B 96USHIPS7 A71V L74V None B THA93067 None None None B None None None B BZI67 None None None B QZ4589 None None None B None None None B None None None B 96SHIPS9 None None None B THA93074 None None None B 92BR014 None None None B BR92019 None D67N, T215D, K219Q V151I B 92HT599 None None V151I C IN98026 None None None C 97USNG30 None None None C TZ98013 None None None C ZA97003 None None None C None None None C ZA97012 None None None C None None None C 98CN009 None None None C IN98002 None None None C IN93905 None None G163R C 93IN101 None None None C BR98004 None D67N, K70R None C RW92006 None None None C None None None C 98T2017 None None None C 93USNG31 None T69S, K70R, M184V, K219Q None C 98IN017 None None E138D CRF01_AE THA93060 None None None CRF01_AE THA93064 None None M154I CRF01_AE THA93062 None None None CRF01_AE THA93069A None None None CRF01_AE THA None None None CRF01_AE THA92005 None None None CRF01_AE TH92020 None V90I None CRF01_AE CMV10 None None None CRF01_AE THA92001 None None None CRF01_AE THA92006 None None None CRF01_AE CMU02 None V106I None CRF01_AE THA92024 None None None CRF01_AE THA92003 None None None CRF01_AE 93TH051 None None None CRF01_AE THA93054 None None None CRFs, circulating recombinant forms; IN, integrase; PR, protease; RT, reverse transcriptase International Medical Press

7 Genotypic analysis of HIV-1 subtypes Table 3. Continued Clade type ID name PR RT IN CRF01_AE THA93053 None None None CRF02_AG 96USSN20 None D67N, T69D, K70R None CRF02_AG 96USSN54 None None None D UG94108 None None None D UG94105 None None None D UG94114 None M41L None D UG93082 None None None D UG9367 None None None D UG94117 None None None D UG9370 None None None D UG94118 None None None D 92UG001 None None None D UG92024 None None None D UG92005 None None None F BZ126 None None G163R F 93BR020 None None None G HH8793 None None None G G3 None None None G JV1083 None None None G R132 None None None N YBF30 None None None O BCF02 Q58E, A71V T67N, A98G, V106I, V179E, None Y181C, L210Y O BCF01 K43T, Q58E, A71V A98G, K103R, V179E, Y181C, None L210Y O MVP5180 K43T, Q58E, A71V A98G, K103R, V179E, L210Y None Table 4. Genotypic study assessing mutations in viral pools collected from AZD-treated, passage 55, HIV-infected PBM cells Clones PR (26 99) RT (1 560) IN (1 267) 1 None T215Y None 2 None T215Y None 3 None T215Y None 4 None M41L, T215Y None 5 None T215Y None 6 None T215Y None 7 None M41L, T215Y None 8 None T215Y None 9 None T215Y None 10 None T215Y None AZD, β-d-3 -azido-2,3 -dideoxy-2,6-diaminopurine; IN, integrase; PBM, peripheral blood mononuclear; PR, protease; RT, reverse transcriptase. O samples, in which major mutations have likely developed through HIV-1 evolution, the results are consistent with previous findings that antiviral resistance mutations, such as Y181C, are present in treatmentnaive samples of this group [4,22]. This underscores the relationship between existing resistance mutations and challenges in clinical application of antiretroviral therapy, which was often traditionally designed based on HIV-1 subtype B [2]. Therefore, this protocol is a valuable tool for scientists to evaluate and predict the effects of experimental or approved drugs on various HIV-1 subtypes. However, it should be noted that our protocol might not accommodate variables impacting subtype detection under non-optimal conditions. Many factors could contribute to detection failure, including sample preparation and isolation of total RNA. Moreover, ongoing evolution of the HIV-1 population will generate subtypes or CRFs beyond our detection. For analysis of the mechanisms responsible for emergence of resistance mutations, this protocol was applied Antiviral Therapy

8 T Nie et al. as an extended platform for screening of resistance mutations in HIV-1 pol (Table 4). One of the strengths of this protocol is that it is able to detect mutations or polymorphisms in multiple regions of pol simultaneously, for example, one subtype B sample (BR92019) with RT and IN resistance mutations (Table 3). Therefore, our sequence profiling of gene pol in the HIV-1 subjects may provide an infection history and resistance background useful for informing future treatment. In demographic investigation of HIV-1 populations and transmission, the sequence profiles of mutation and polymorphism may represent evolutionary footprints indicative of HIV-1 population migration, phylogenetic variation, and ongoing recombination. As mentioned earlier, our protocol covers approximately 93% of the HIV-1 pol gene, which is sufficient for most studies of HIV-1 resistance against drugs targeting RT, PR, RNase H or IN. For instance, the recently reported MK-2048 (an INI) resistance mutations G118K and E138K [23] are detectable within the coverage of our protocol. However, despite the utility of our protocol, it should be noted that resistance to HIV-1 drugs does not develop solely in the pol gene; one exception is that mutants in the gag gene are widely selected as resistance against PIs [24,25]. For viral load assay using quantitative real time- PCR, the efficiency of the primer set (Cf and D) was first determined on subtype B. It is noteworthy that because of sequence variation across HIV-1 subtypes, the samples from group N and O appear to have remarkably lower viral loads than those from group M (Table 2). Another great challenge for this assay is to design a consensus probe across subtypes. These caveats may reduce the applicability of our viral load assay in these two particular groups. Overall, the simplicity and efficacy of this study provides a novel and comprehensive tool to develop an assay of genotypic study and resistance analysis across subtypes to meet the growing global needs. Acknowledgements We would like to thank John Mellors for providing the plasmid xxlai and Tom North for his advice and review. We would also like to thank the NIH AIDS Research and Reference Reagent Program for all HIV-1 subtype samples. This work was supported in part by NIH grants 2P30-AI , 5R37-AI , 5R37-AI , RO1-AI , and the Department of Veterans Affairs (to RFS). Portions of this work were presented previously at the XVIII International HIV Drug Resistance Workshop: Basic Principles & Clinical Implications, 9 13 June, 2009, Fort Myers, FL, USA (Abstract 110); and the 10 th Annual Symposium of Antiviral Drug Resistance: Targets and Mechanisms, November, 2009, Richmond, VA, USA (Abstract 24). Disclosure statement The authors declare no competing interests. References 1. Martínez-Cajas JL, Pant-Pai N, Klein MB, Wainberg MA. Role of genetic diversity amongst HIV-1 non-b subtypes in drug resistance: a systematic review of virologic and biochemical evidence. AIDS Rev 2008; 10: Kantor R, Katzenstein D. Drug resistance in non-subtype B HIV-1. J Clin Virol 2004; 29: Katzenstein D. Diversity, drug resistance, and the epidemic of subtype C HIV-1 in Africa. J Infect Dis 2006; 194 Suppl 1:S45 S Spira S, Wainberg MA, Loemba H, Turner D, Brenner BG. Impact of clade diversity on HIV-1 virulence, antiretroviral drug sensitivity and drug resistance. J Antimicrob Chemother 2003; 51: Brennan CA, Stramer SL, Holzmayer V, et al. Identification of human immunodeficiency virus type 1 non-b subtypes and antiretroviral drug-resistant strains in United States blood donors. Transfusion 2009; 49: Osmanov S, Pattou C, Walker N, Schwardländer B, Esparza J. Estimated global distribution and regional spread of HIV-1 genetic subtypes in the year J Acquir Immune Defic Syndr 2002; 29: Lu W, Cao L, Ty L, Arlie M, Andrieu JM. Equivalent amplification of intrinsically variable nucleic acid sequences by multiple-primer-induced overlapping amplification assay: applications for universal detection and quantitation. Nat Med 1999; 5: Tang N, Huang S, Salituro J, et al. A RealTime HIV-1 viral load assay for automated quantitation of HIV-1 RNA in genetically diverse group M subtypes A-H, group O and group N samples. J Virol Methods 2007; 146: Van Laethem K, Schrooten Y, Covens K, et al. A genotypic assay for the amplification and sequencing of integrase from diverse HIV-1 group M subtypes. J Virol Methods 2008; 153: Kantor R, Katzenstein D. Polymorphism in HIV-1 nonsubtype B protease and reverse transcriptase and its potential impact on drug susceptibility and drug resistance evolution. AIDS Rev 2003; 5: Fish MQ, Hewer R, Wallis CL, Venter WD, Stevens WS, Papathanasopoulos MA. Natural polymorphisms of integrase among HIV type 1-infected South African patients. AIDS Res Hum Retroviruses 2010; 26: Nkolola JP, Essex M. Progress towards an HIV-1 subtype C vaccine. Vaccine 2006; 24: Wainberg MA. HIV-1 subtype distribution and the problem of drug resistance. AIDS 2004; 18 Suppl 3:S63 S Kantor R, Machekano R, Gonzales MJ, Dupnik K, Schapiro JM, Shafer RW. Human immunodeficiency virus reverse transcriptase and protease sequence database: an expanded data model integrating natural language text and sequence analysis programs. Nucleic Acids Res 2001; 29: Wittek M, Stürmer M, Doerr HW, Berger A. Molecular assays for monitoring HIV infection and antiretroviral therapy. Expert Rev Mol Diagn 2007; 7: Rhee SY, Liu TF, Kiuchi M, et al. Natural variation of HIV-1 group M integrase: implications for a new class of antiretroviral inhibitors. Retrovirology 2008; 5: Roquebert B, Marcelin AG. The involvement of HIV-1 RNAse H in resistance to nucleoside analogues. J Antimicrob Chemother 2008; 61: International Medical Press

9 Genotypic analysis of HIV-1 subtypes 18. Santos AF, Lengruber RB, Soares EA, et al. Conservation patterns of HIV-1 RT connection and RNase H domains: identification of new mutations in NRTI-treated patients. PLoS One 2008; 3:e Yap SH, Sheen CW, Fahey J, et al. N348I in the connection domain of HIV-1 reverse transcriptase confers zidovudine and nevirapine resistance. PLoS Med 2007; 4:e Tebit DM, Nankya I, Arts EJ, Gao Y. HIV diversity, recombination and disease progression: how does fitness fit into the puzzle? AIDS Rev 2007; 9: Zhang HW, Coats SJ, Bondada L, et al. Synthesis and evaluation of 3 -azido-2,3 -dideoxypurine nucleosides as inhibitors of human immunodeficiency virus. Bioorg Med Chem Lett 2010; 20: Accepted 14 March 2011; published online 14 September Leoz M, Depatureaux A, Vessière A, et al. Integrase polymorphism and HIV-1 group O diversity. AIDS 2008; 22: Bar-Magen T, Sloan RD, Donahue DA, et al. Identification of novel mutations responsible for resistance to MK-2048, a second-generation HIV-1 integrase inhibitor. J Virol 2010; 84: Yusa K, Harada S. Acquisition of multi-pi (protease inhibitor) resistance in HIV-1 in vivo and in vitro. Curr Pharm Des 2004; 10: Lambert-Niclot S, Flandre P, Malet I, et al. Impact of gag mutations on selection of darunavir resistance mutations in HIV-1 protease. J Antimicrob Chemother 2008; 62: Antiviral Therapy