To the Editor: The provision of specific HIV diagnostic tests and the development of HIV vaccines and antiviral agents depends on understanding the global extent of the diversity of the genome. Sequences of the env and gag genes have been used to classify HIV-1 into a major (M) group, consisting of 10 subtypes designated A through J, and an outlying (O) group of sequences(1,2). The diversity of HIV-1 is probably greatest in Africa(3), and monitoring changes in subtype distribution there may predict the course of the pandemic. Fifteen blood samples were randomly selected from stored samples from a cohort of 684 HIV-1-infected adults from Nairobi, and the lymphocytes were separated. The samples were from six men and nine women; HIV acquisition was assumed to be by heterosexual intercourse, because there were no other known risk factors. Genomic DNA was isolated using a Nucleon I Nucleic Acid Extraction Kit (Scotlab Bioscience, Coatbridge, Strathclyde, U.K.). Portions of the HIV-1 env andgag genes were amplified by polymerase chain reaction methods, subtyped by heteroduplex mobility analysis (HMA), and sequenced as previously described(4-6). Of the 15 samples, 1(K13) failed to amplify with the env or gag primer. The subtyping results are shown in Table 1. The virus was first typed by HMA on Phast gels (Pharmacia, LKB Biotechnology AB, Uppsala, Sweden) using the algorithm previously described(6). A statistically significant result (p < 0.05, Krustal-Wallis test) was obtained by HMA of the 1200-bp amplicon from the env gene for six samples: K0, K2, K3, K6, K8, and K14. Another sample (K7), for which there was no amplification with the 1200-bp primers, gave a statistically significant subtype with the 700-bp env amplicon. Of the other seven samples, one (K13) was not amplified by any of the primers tried, including gag primers(4), and one (K11) produced multiple bands on the gel so that it was not possible to measure the heteroduplex mobility. Amplification and sequencing in thegag gene was used to subtype this sample(4). The remaining five samples (K1, K4, K5, K9, K10) did not give an easily interpretable band pattern when the HMA products were run out on Phast gels, but unequivocal results were obtained using Metaphor XR agarose gels (FMC BioProducts, Kent, U.K.). Sequencing was used to confirm the subtype of these latter samples (K1, K4, K5, K9, K10) and of K3 and K6, which were considered controls.TABLE 1: Subtyping of Kenyan HIV-1 samplesThis work also indicates that, for some samples, care must be taken if HMA alone is used to assign the subtype. The lack of dependability may reflect high sequence diversity, particularly of clade A viral genomes. Infection with more than one strain of HIV-1 and the presence of recombinant or mosaic genomes also may affect the resolution of HMA. In conclusion, we found 12 subtype A, 1 subtype C, and 1 subtype D, confirming the predominance of subtype A in Kenya(3, 7-9). Acknowledgements: This study was supported in part by a SEIMC-Russel-95 grant and by a grant (1995 BEA1300056) from CIRIT, Generalitat de Catalunya, to F. J. Belda. *F. J. Belda †C. Mwchari †M. Hawken *K. L. Barlow *J. P. Clewley *Virus Reference Division; Central Public Health Laboratory; London, United Kingdom;†Clinical Research Centre; Kenya Medical Research Institute; Nairobi, Kenya
Read full abstract