We investigated the presence of HIV or SIV in 70 HIV EIA-reactive, WB-indeterminate serum specimens collected in 1984 and 1986 from employees of Mama Yemo Hospital in Kinshasa, Zaire (currently the Democratic Republic of Congo) (11). WB bands were identified by using an HIV-1/2 WB assay (version 2.2, Genelabs Diagnostics, Singapore) and classified as indeterminate according to criteria for interpreting HIV-1 (http://www.cdc.gov/mmwr/preview/mmwrhtml/00001431.htm). Briefly, a positive WB result indicates reactivity to at least p24 and 1 of the 3 env (gp41, gp120, or gp160) proteins with banding intensity at least as strong as that seen in the weak-positive control; a negative WB result indicates no reactivity and an indeterminate WB result indicates reactivity to at least 1 protein.

Figure

Figure. . A) Distribution of HIV-1 Western blot–indeterminate patterns among 69 serum specimens from Kinshasa, Zaire, reactive by enzyme immunoassay. B) Phylogenetic classification of HIV-1 protease sequence ZA30972 (GenBank accession no. AY562558)...

Of the 70 WB-indeterminate serum specimens, 69 showed a broad range of HIV-1 WB-indeterminate band patterns, and 1 had an HIV-2 WB-indeterminate pattern with weak reactivity against gp36, p68, and gp80 proteins. Analysis of HIV-1 WB-indeterminate patterns provided several important observations. First, most of the specimens (41/69, 59%) showed reactivity to multiple viral proteins (Figure), whereas the remaining specimens (28/69, 41%) demonstrated reactivity against single HIV-1 proteins, including p24 (13/69, 18.8%), p17 (n = 8, 11.6%), gp160w (w = weak) (n = 4, 5.8%), and p51, p66, and gp41w (n = 1 each; 1.4% each). Second, in all but 3 specimens the WB-indeterminate patterns with reactivity to multiple viral proteins combined proteins of p24 or p17 gag or both with others, giving the following 21 profiles: p17/p24 (14/69, 20.3%); p24/p66 and p17/p24/p66 (n = 3 each, 4.3% each); p24/p51, p17/p24/p51, and p17/p24/gp41w (n = 2 each, 2.9% each); p17/p66, p17/gp160, p24/gp160, p24/p51/gp41w, p17/p51/p66, p17/p31/gp41w, p24/p51/p55, p24/p51/p66, p17/p24/p51/p66, p17/p24/p51/p55, p17/p24/p55/p66, p24/p31/p51/p55, p31/p66, p51/gp41w, and p31/p66/gp41w (n = 1 each, 1.4% each). Overall, this analysis indicated reactivity to 8 HIV-1 proteins that occurred as single bands or multiple combinations. Whereas reactivity to p24 might represent either early infection with HIV or reaction to nonspecific antigens (8), reactivity to other HIV-1 proteins (p17, p31, p51, p55, p66, gp41, or gp160) could reflect at least exposure to HIV-1, HIV-2, or SIV variants as well as cross-reactivity with yet unidentified HIV- or SIV-like strains (3,10).

To determine whether HIV- and SIV-like RNA could be detected in the 70 WB-indeterminate specimens, we attempted reverse transcription–polymerase chain reaction (RT-PCR) amplification of HIV-1, HIV-2, and SIV. However, because of a small volume (<300 mL) of some serum specimens, a limited number of primers was used. We selected 4 sets of primers for 3 highly conserved gene regions, including env-gp41, pol protease, and pol integrase, which we previously developed and extensively tested for the efficient PCR amplification of field specimens. Briefly, gp41 primers allow the amplification of HIV-1 groups M (subtypes A–K and U [unclassifiable]) N, and O and SIVcpz with an efficiency of 95% (13); HIV-1 type specific protease primers efficiently (>95%) detect group M viruses (subtypes A–K and U [unclassifiable]), and allow successful amplification of some PCR–gp41 negative specimens. HIV-2 type-specific protease primers allow the amplification of HIV-2 at least subtypes A and B (13), and HIV-2/SIV integrase primers amplify HIV-2 and at least 5 major SIV lineages including SIVcpz, SIVsm, SIVagm (African green monkey), SIVmndBK12 (mandrill), and SIVsyk (Sykes monkey) (14). The integrase primers should also amplify other SIV lineages including SIVcol, SIVLhoest/SIVsun, SIVgsn, SIVmus, SIVmon, SIVtal, and SIVdeb (3). This prediction is based on the fact that genetic diversity of these sequences within the 3´-end of primer regions was the same as in strains previously used for testing (15), which was clearly visible through DNA alignment of the integrase primers with all SIV genomic integrase sequences deposited in GenBank. Positive controls for PCR amplification included HIV-1 Mn, HIV-2 ROD, and SIVsm. Using this approach, we amplified a 297-bp HIV-1 protease (PR) gene from only 1 (1.4%) of the 70 serum specimens. The remaining 69 specimens were PCR negative for all the primers tested. Phylogenetic analysis of the PR gene sequence showed a close phylogenetic relationship with the HIV-1 MAL strain (Figure), a recombinant virus that contains portions of subtypes A and D and an unclassifiable region that was identified in 1985 in Zaire (16).

The PR gene sequence was identified in the serum of the person with reactivity to only the p17 gag band by WB, which suggests that this person was recently infected and that antibodies to all the HIV-1 antigens had not yet developed. No information was available on demography, risk, or clinical status of this person. In the remaining 69 WB-indeterminate specimens, we could not rule out exposure to SIV from nonhuman primates, from handling infected animal meat before consuming it, or from keeping monkeys as pets. Although the limited molecular scope of this study and the quality of the old serum specimens may not be adequate for molecular confirmation of such viruses, our findings of complex HIV WB-indeterminate patterns with reactivity to multiple viral proteins in serum specimens from persons living in the Democratic Republic of Congo provide comprehensive insights into HIV WB-indeterminate sera in the mid-1980s.