Distribution of HIV-1 subtypes in female sex workers of Calcutta, India

Distribution of HIV-1 subtypes in female sex workers of Calcutta, India

Mandal, Dibyakanti

India is experiencing rapid and extensive spread of HIV and is reported to have entered the third stage of the epidemic in many urban areas. HIV infection has been reported from all the states in India’ and the majority of these have occurred through heterosexual contact. However, studies in the northeastern states have highlighted HIV infection among injecting drug users (IDUs), the route of transmission being through shared needles and syringes2. While a high prevalence of the infection in female sex workers of Maharastral3 and Tamil Nadu4 is well documented, a relatively low prevalence of HIV infection among sex workers5.6 in the eastern part of India has been reported. The important factors responsible for rapid spread of HIV in India might be frequent multiple partner sexual relationships, high incidence of sexually transmitted diseases and inadequate counselling services. An effective low cost vaccine is one of the major intervention priorities. The knowledge of viral variants circulating among the target population will be of help in designing such a vaccine. Sketchy information on the profile of the subtypes of HIV-1 prevalent in India has emerged as a result of a number of isolated attempts. Early studies from India, identified subtype C as the major circulating subtype among STD patients in Pune’, Mumbai8, truck drivers from Punjab9 and IDUs from Manipur10 The presence of other subtypes like All, B’z and Thai BI has also been reported. In a recent study, the samples isolated from HIV-1 seroconverters revealed the presence of recombinant strain with subtypes A and C13. However, there is very little information on the nature of HIV- I subtypes in eastern India. The present study aims to classify HIV-1 subtypes) cirulating among the female sex workers in Calcutta where HIV-1 seroprevalence is still relatively lows.e.

Material & Methods

Sample collection, testing and preservation: Blood samples were collected from female sex workers of different red-light areas of Calcutta during the period April 1998 to January 1999. Samples were tested using rapid spot test (Immunocomb HIV-1/2 Bi-spot, Orgenics, Israel), ELISA (Immunogenetics, Belgium) and line immunoassay (Inno- LIA HIV-1I/HIV-2 kit, Immunogenetics, Belgium) for HIV seropositivity. Peripheral blood mononuclear cells (PBMC) were separated from whole blood and stored at -70 deg C until use.

Peptide enzyme immuno assay: PEIA is a method based on the reactivity of the antibody present in the HIV infected sera to V3 loop peptide of HIV-1 envelope 14,11. Samples were analyzed using 8 synthetic 14 meric peptides representing the V3 loop of the envelope glycoprotein (gp120) of 8 subtypes A, B, C, D, E, F, Thai B and 0 (Table). A non-virally coded aspartic acid (D) was added to the start of each peptide yielding 15 meric peptide to improve the binding of the peptide to the plate. The peptides were synthesized using 9-flurenyl methoxycarbonyl (FMOC) chemistry in an automated synthesizer (Model 431A, Applied Biosystems, USA). The peptide antigens (0.5 mg/ml) were dissolved in 0.1 M bicarbonate buffer, pH 9.4 and adsorbed into microwell plates (Immunolon 2, Dynatech, USA) overnight at 4C and washed once with 10 mM phosphate buffered saline, pH 7.4 containing 0.05 per cent Tween 20 (Sigma, USA). Nonspecific binding was blocked with 150 gl per well milk buffer (5% non-fat dry milk and 0.3% Tween 20 in 10 mM phosphate buffered saline, pH 7.4) and incubated at 37C for 90 min. The diluted serum samples (100 gl at 1:400 dilution in milk buffer) were added to the antigen-coated wells and incubated at 37C for 60 min. The plates were washed and bound antibodies were detected with affinity purified, peroxidase conjugated anti-human IgG (H+L chain) and 3, 3′, 5, 5′ tetramethyl benzidine (TMB) hydrogen peroxide substrate (BioRad, USA). The reaction was stopped by IM H2S04 and the absorbance (OD) was recorded at 450 nm in Titretech version 4 spectrophotometer (Flow Laboratories, Switzerland). HIV-1 negative samples were included in each run of the assay. The cut-off value was determined [(mean of negative samples +3 SD)x 2]. The average cut-off was between 0.15 and 0.3 OD unit. A cut-off value of 0.3 was used throughout the study”.

The samples were considered reactive to a particular V3 peptide if the OD value was found to be greater than the cut-off value of 0.3. The highest OD value was taken as V3 ax . The per cent reactivity of each peptide was determined considering the V3 max as 100 per cent. The sample serotype was classified by those peptide(s) for which the sample reacted >! 90 per cent. The samples which reacted with multiple peptides (>!90%) were repeated at 1:1600 serum dilution and the serotype was determined by strongest reacting peptide. The samples, which reacted with multiple peptides (more than two peptides) even at higher dilution, were classified as non-typable.

Extraction of DNA from peripheral blood mononuclear cells : PBMC were separated from the whole blood by Ficoll-Hypaque gradient centrifugation”. PBMC were lysed with DNA extraction buffer consisted of 10 mM Tris-HCI, pH 8.0, 50 mM KCI, 2.5 MM MgC’2′ 0.5 per cent Tween-20, 0.5 per cent Nonidet P-40 (Sigma, USA) and incubated with proteinase K (Gibco-BRL, USA) for I h at 55’C. DNA was extracted by phenol/ chloroform (once with phenol, twice with 1:1 v/v phenol-chloroform, once with chloroform). DNA was then precipitated with 3M Na acetate, pH 7.0 and ethanol. DNA precipitate was washed, dried and dissolved in DEPC (diethyl pyrocarbonate) treated water

PCR amplification of HIV-1 envelope gene : C2-V3/ C2-VS region of HIV-1 envelope gene was amplified from PBMC DNA by nested PCR in thermal cycles (Gene Amp PCR system, 2400, Perkin Elmer, USA). Primers used for the first round PCR were EDS and ED 12. Sequences of the primers are


ED 12 5′ AGTGCTTCCTGCTGCTCCCAACCCAA 3′ (7822-7792). The nucleotide position of all the primers was based on the HIV-1-HXB2 genome (GeneBank accession number K03455).

0.5 – Ig of PBMC DNA was used as template for PCR amplification in the presence of 1.25 mM MgCl, 200 mu M dNTPs (Perkin Elmer, USA), 20 pmole of each primers and 2.5 units of Taq polymerase (Ampli Taq gold, Perkin Elmer, USA) in a total volume of 50 gl. PCR condition was 94’C 15 min; 3 cycles at 94 deg C I min, 50’C I min, 72’C 1 min; 35 cycles at 94 deg C 15 sec, 55’C 45 sec, 72C 1 min and final extention at 72 deg C for 5 min. Primers for second round PCR were ES7/ES8 or ED31/ED33.




(7359-7380) 5 Itl of the first product was used as template for second PCR. PCR condition was 94 deg C 15 min; 40 cycles at 94 deg C 15 sec, 55’C 45 sec, 72C I min and final extension at 72C for 5 min in the presence of 1.25 mM MgCI 20 pmole of each primers and 200 mM dNTPs and 2.5 U Taq polymerase. All the above primers were supplied with HMA kit obtained from the NIH AIDS Research and Reference Reagent Programme (NIH, Bethesda, Md. USA).

Heteroduplex mobility assay : HMA assay was conducted as described elsewhere”‘. C2 – V3/C2 V5 region of HIV-1 envelope gene of different reference strains (NIH AIDS Reference and Reagent programme) were amplified using the same sets of nested primers used to amplify HIV-1 DNA from PBMC. Amplified DNA fragments from reference strains were mixed separately with the amplicon obtained from sample DNA in the presence of annealing buffer (10 mM Tris-HCI, pH 8.3, 100 mM NaCl, 2mM EDTA). Mixtures of DNAs were heated at 94’C for 5 min and quickly chilled on ice. Heteroduplex formed was then analyzed on 5 per cent acrylamide gel. Mixtures of 700 base pair fragments amplified with primers ES7 and ES8 were electrophoresed for 2 h 30 min at 250 volts (]4x]6 vertical gel apparatus, ATTO, Japan) and 550 base pair fragments amplified with primers ED31 and ED33 were electrophoresed for 2 h at 250 volts. IX TBE (0.88 M Tris-borate, pH 8.0, 89 mM boric acid, 2 mM EDTA) was used as running buffer. Relative mobility of each sample was calculated by dividing the migration distance of the heteroduplex band by the migration distance of the corresponding homoduplex band. Mobility of the heteroduplexes is inversely proportional to the sequence divergence of two annealed strands. To obtain a standard curve representing mobility vs divergence relationship, we plotted the heteroduplex mobility (expressed in cm) against the percentage of divergence of the annealed PCR product from known HIV-1 sequences. Percentage of divergence was calculated on the basis of sequence alignment that discounted gaps due to nucleotide insertions or deletions. The best-fit curve was found with a linear function. Under our experimental condition with 550 bp DNA fragments and an electric field of 250 volt for 2 h the equation was

% Divergence = 37.4 – 36 (heteroduplex mobility)

The standard sequences used for HMA were Al(RW20), BI(BR20), B2(TH 14), CI(MA959), C2(ZM18), C3(IND868), C4(BR25), Dl(UG21) and E2(TH06).

Cloning and sequencing : PCR amplified fragments of HIV-1 envelope gene isolated from PBMC were cloned directly in TA cloning vector PCR 2.1 (Invitrogen, USA) which contains 37 overhang to enable direct ligation of PCR product having YA overhang produced by template independent polymerization of Taq polymerase. White colonies in the presence of IPTG/X-GaI (Bangalore-Genel, India) formed after transformation of the ligation mixture into Escherichia coli strain DH5(x F were selected. Release of 550 bp fragment with EcoRl from the DNAs isolated from the selected colonies indicated the insertion of HIV-1 env fragment. Such 4 clones -of each samples were sequenced using Sequenase version 2.0 kit (Unites States Biochemicals, USA) following manufacturer’s protocol with forward primer R124 (5′ AAATGGCAGTCTAGCAGAAG 3′) and reverse primer KK30 (5′ AATTTCTGGGTCCCCTCCTG 3′). Radioactive U_pl2 dATP (BRIT, India) was used as the labelling reagent. The reaction mixture were electrophoresed onto 8 per cent polyacrylamide gel in the presence of 6M urea in a vertical electrophoresis system (Gibco-BRL, USA).


Subtype distribution by PEIA : Of the blood samples collected from 973 female sex workers, 59 were found to be HIV-1 seropositive (6%). Fifty two samples were subjected to serotype analysis by PEIA. Majority of the samples (33/52, 60%) reacted with the peptide specific for subtype C; 7, 3 and I sample reacted with subtype D, A and E specific peptides respectively. Some samples (8/52, 16%io) reacted with more than one peptide thus remained non-typable. Interestingly, no samples reacted with subtype B or Thai B specific peptides.

HMA assay : To determine the HIV-1 subtype of HMA assay, the mobility of heteroduplexes formed between the 40 amplified DNA products from PBMC and reference strains Al (RW20), BI(BR20), B2 (TH14), C3 (IND868), Dl- (UG21) and E2 (TH06), were analyzed. Sample PCR product in the absence of any reference strains served as control. Heteroduplex formed between the unknown sample and the most closely related reference sequence migrated more, indicating the likely subtype of the sample. This assay determined that 38 samples belonged to HIV- I subtype C, -two samples remained nontypable. Most of the samples which reacted with D or E specific V3 peptides turned out to be subtype C by HMA. All 38 subtype C samples were further analyzed for classification of C subtype with four subtype C strains, viz., Cl (MA959, Malawi), C2 (ZM 18, Zambia), C3 (IND868, India) and C4 (BR25, Brazil). The major subclassified subtypes were C3 (26/38), followed by C2 (11/38), and Cl (1/38). No C4 subtype was detected with the samples tested so far. As majority of the samples showed homogeneity with subtype C (mostly C3) we measured the heteroduplex mobility of all the samples with respect to C3 subtype and plotted this against DNA divergence (Fig. 1). Less than 10 per cent DNA divergence was observed with the majority of samples. Two samples having DNA divergence of more than 15 per cent were designated as nontypable.

Sequence alignment and phylogenetic analysis : C2V3 region of HIV-1 envelope gene (325 bp) amplified from PBMC was cloned into the TA cloning vector. Four recombinants for each sample were sequenced with the primer and no variation in the sequences was observed. The sequences spanning the V3 loop were aligned with reference sequences of different geographical isolates. Multiple sequence alignment was done by pileup programme19. Phylogenetic analysis was performed by Phylip (version 3.2) package”. Evolutionary distances were measured following Kimura-two parameter distance matri X21 using DNADIST programme. Bootstrap analysis was performed for 100 replicates by SEQBOOT programme available in the Phylip package. In the representative tree of 11 I samples isolated from Calcutta, 8 samples clustered with subtype C (Fig. 2). However, on careful examination it was apparent that 3 samples (Ind. Cal 2, Ind. Cal 51, and Ind. Cal 20) were much closer to reference strain (IND 868) than the other 5 samples (Ind. Cal 462, Ind. Cal 250, Ind. Cal 554, Ind. Cal 145, and Ind. Cal 150). Three samples (Ind. Cal 684, Ind. Cal 242, and Ind., Cal 202) clustered outside C subtypes forming different operational taxonomic units (OTUs). The sequences of these 11 samples were submitted to EMBL nucleotide database and the accession numbers assigned were as follows:

AJ278375 (Ind. Cal2); AJ278381 (Ind. Ca)51); AJ278376 (Ind. Cal 20); AJ278380 (Ind. Ca1462);

AJ278379 (Ind. Ca125(If; AJ278382 (Ind. Ca1554); AJ278384 (Ind. Cal 145); AJ278383 (Ind. Call 50);

AJ278374 (Ind. Ca1684); AJ278378 (Ind.Cal242); and AJ278377 (Ind. Ca1202).

Amino acid sequence analysis : Amino acid .sequences predicted from the nucleotide sequences were compared and analyzed. Among the 35 amino acids within the V3 loop, the crown of the loop contained GPGQ motif found in HIV-1 strains of A, C, D and E subtypes (Fig. 3). All the samples had one, two or even more amino acid substitutions within this region. In the sample Ind. Cal 242, an unusual T (threonine) substitution instead of I (isoleucine) at position 24 was noticed near the GPGQ motif. In another sample Ind. Cal 554, G (Glycine) at position 34 was replaced by D (Aspartate). The potential glycosylation sites, aspargine(N)-X(any amino acid)-threonine(T)/ serine(S) within the C2-V3 region (amino acid position 24-73) were compared (Fig. 3). As in other subtype C sequences, the highly conserved site NCT (position 33, marked as 2) at the N-terminal boundary of the V3 loop (position 35-69) was missing in all these sequences. A few residues upstream, yet another conservative site, NQS (position 28, marked as 1) was found to be missing in two samples. The site, NNT (position 40, marked as 3) within the V3 loop was highly conserved among the Calcutta samples. Another site, NIS (position 70, marked as 4) at the C-terminal end of the V3 loop was found to be present in nine samples.


A major concern in the development of a broadly effective vaccine against HIV-1 infection has been the large amount of genetic diversity that characterizes this virus. Phylogenetic analysis of HIV-1 sequences world-wide has demonstrated the presence of at least eight genetic subtypes (A-H) as well as the highly divergent group 0 and N strains of HIV- 122-25. The HIV-1 V3 region of the envelope gene has been intensively investigated as a vaccine candidate, because it contains the principal neutralizing domain and epitopes for t-helper cells and plays a role in cytotoxic T-lymphocyte recognition, cell tropism and viral infectivity. Despite the genetic variation within the envelope gene, the hypervariable V3 region remains highly conserved within each subtype. The present study indicates that HIV-1 subtype C predominantly infects the female sex workers in Calcutta. Studies conducted. among different groups of individuals from Mumbai’, Pune7 and Tamil Nadu” also indicated that subtype C as the major subtype. The study at Pune7 is the only published report where an appreciable number of samples from a selected group (STD clinic attenders) was taken into consideration. In our study, we have selected a specific high risk group, female sex workers and used three methods, viz., PEIA, HMA and genomic analysis to determine the HIV-l subtypes. Majority of the HIV-1 positive samples (33/52) serotyped by PEIA was found to be of subtype C. PBMC DNA was directly amplified to avoid any bias in selecting viral variants that may grow more efficiently in culture C. Forty samples were amplified by nested PCR; the remaining samples could not be amplified. This might be due to the presence of very low copy of the viral genome. C2-V3 region amplified with ED31/ED33 primer pairs was used with 35 of the 40 samples. The remaining 5 samples did not produce good amplicon using the above primer pairs and ES7/ES8 primer pairs was used to amplify C2-VS region for these 5 samples. Subclassification of subtype C by HMA showed C3 (68%) and C2 (29%) as major subclasses present in this geographical region. In an identical study from Pune 7, a similar distribution of C3 and C2 subtypes (65 and 35% respectively) in the samples collected from both seropositives and seroconverters, has been reported. However, we could not detect any quasispecies complexity within the samples as reported earlier. Eleven samples which reacted with peptides other than C were found to be subtype C by HMA. This indicates that PEIA though rapid and easy cannot be used exclusively to determine the subtype. Eleven of the 40 amplified samples were selected randomly for DNA sequence analysis. Phylogenetic analysis based on DNA sequences showed that most of the samples clustered with reference C subtype and three samples (Ind. Cal 684, Ind. Cal 242 and Ind.Cal 202) clustered outside. Amino acid sequence analysis showed that there were unusual amino acid substitutions within or close to the V3 loop. However, the V3 apical tetrapeptide sequence Gly-Pro-Gly-Gln (GPGQ) is conserved in all the sequences. All these unusual amino acid substitutions may alter the pattern of immunogenicity and decrease the sensitivity of the PEIA serotyping. However, this decreased sensitivity was not reflected in HMA assay probably due to a very small number of nucleotide substitutions within the V3 loop. Comparison of potential N-linked glycosylation sites indicates the variations of the sites although the characteristic site is present in all the samples. Whether these changes in the glycosylation site have any effect on the immunogenicity, or infectivity is yet to be determined by expressing the full length envelope gene27–29in appropriate system.

The data presented here augment other studies and support the conclusion that the major subtype of HIV- 1 circulating in India belongs to C. We could not detect the presence of other subtypes as reported by other workers”‘z. In three samples, the partial envelope sequence phylogenetic ally distinct from others was observed. It would be interesting to sequence full-length envelope gene as well as the gag gene for these samples. However, this is the first report from this region of the country with an appreciable number of samples from a definite highrisk group and using all the methods available to determine HIV-1 subtypes. Although the sequencing of the amplified DNA fragments is still the gold standard for determining the HIV-1 subtypes, HMA can be used for rapid evaluation in order to have an idea regarding the trend of HIV-1 subtypes circulating within a particular region. It is clear from our study as well from those of others that HIV- in the Indian population no longer clusters into a tight monophyletic group but that several variants within the same subtype are present. Constant monitoring of the strains circulating in different high-risk groups might help in designing a candidate vaccine.


Authors thank the staff members of the STD/HIV Intervention Project, Calcutta for their co-operation in collecting samples and NIH AIDS Research and Reference Reagent Programme, USA for the supply of HMA subtyping kit version 2.0. The financial assistance from Department of Biotechnology, Govt. of India, New Delhi is acknowledged.

Copyright Indian Council of Medical Research Nov 2000

Provided by ProQuest Information and Learning Company. All rights Reserved.