Objective: We conducted this study to genetically characterize dual in
fection in individuals demonstrating a dual serological profile. Metho
ds: All subjects were first evaluated by immunoblot for antibody react
ivity to the major viral antigens for HIV-1 and HIV-2. Sera were judge
d to be dual-seropositive if they reacted with strong and equal intens
ity with the envelope antigens of both HIV-1 and HIV-2 and were confir
med with type-specific recombinant env peptides. We used nested polyme
rase chain reaction (PCR) to proviral gag and env sequence from periph
eral blood mononuclear cell amplify (PBMC) DNA from HIV-1- and HIV-2-i
nfected individuals. Positive amplification was detected after Souther
n blot hybridization. Results: Plasmid dilution and mixing showed equi
valent sensitivity of HIV-1 and HIV-2 primers that was not altered by
heterologous target sequences. The DNA PCR showed 100% sensitivity and
specificity for detection of monotypic HIV infection. Serologically d
efined HIV-dual reactives were evaluated by this assay, with 100% dete
ction in female sex workers (21 out of 21), but only 38.5% detection (
five out of 13) in hospitalized patients; all being HIV-1 positive onl
y. The lack of HIV-2 proviral signal was significantly correlated with
low CD4+ lymphocyte counts (P value = 0.04). Conclusion: The results
suggest that HIV dual infection may not be a static condition. Levels
of HIV-2 may decrease with disease progression or sequester in tissue
reservoirs; our results may also suggest that HIV-1 effectively overgr
ows HIV-2 in the dually exposed host individual.