Protein Bead Array for the Detection of HIV-1 Antibodies from Fresh Plasma and Dried-Blood-Spot Specimens

Simultaneous multianalyte immunoassays offer advantages, including reduction of technical operations, execution time, and overall costs. The suspension array technology (SAT), which uses microfluorospheres with flow cytometry, is a multianalyte immunoassay that requires small specimen volumes, such as those obtained by less-invasive blood sampling approaches such as heel or finger sticks, as well as those obtained in pediatric specimens. The dried-blood-spot (DBS) technology has become an important screening tool for clinical and epidemiologic testing (1)(2)(3). It is particularly convenient in rural, resource-limited settings where trained personnel and adequate facilities for blood collection, processing, transport, and storage may not be available. This approach, however, provides limited specimen volume (∼5–6 μL of serum/6-mm punch) (2). Fortunately, SAT is well suited to support such a format (4)(5). The range of applications for SAT includes antibody, oligonucleotide, peptide, and protein bead arrays (PBAs) (4)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17). PBAs have been used for the simultaneous detection of serum antibodies to infectious agents for systemic candidiasis (18), herpes viruses (19), measles, mumps (20), and HIV (21)(22). The successful detection of HIV-1 serum antibodies by a four-protein bead array was first reported by Scillian et al. (21), who used beads of various sizes and proteins noncovalently coupled to beads. Reagents are now available that permit covalent coupling of proteins to beads and simultaneous analysis of up to 100 color-coded bead sets in a high-throughput, 96-well plate format. In addition, a new generation of flow cytometers compatible with SAT has recently been introduced, e.g., the Luminex-100 System (Luminex Corp.) and the FACSArray (BD Biosciences). We have developed and evaluated a HIV-1 PBA for the rapid detection of antibodies to seven viral proteins from fresh plasma and DBS specimens. Seven recombinant HIV-1 proteins from the gag, pol, and envelope regions and a nonviral antigen control [bovine serum albumin (BSA)], all from commercial sources (see the Appendix in the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol50/issue7/), were covalently coupled to individual sets of color-coded microfluorospheres (Luminex Corp.). Diluted plasma and DBS specimens (in 50 μL) were incubated for 1 h with the mixture of microfluorospheres, to which seven or eight proteins had been coupled, in a filter-bottomed 96-well plate (Nunc). Bound antibodies were detected with biotinylated anti-human IgG conjugate (BD Biosciences; 1 h) and streptavidin-phycoerythrin conjugate (Sigma; 15 min). The beads were resuspended in 20 mL/L paraformaldehyde in phosphate-buffered saline and analyzed directly from the filter-bottomed, 96-well plate with a Luminex-100 System equipped with the Luminex Data Collector, Ver. 1.7, software. The median fluorescence intensity was used as the read-out. The HIV-1 PBA test was completed within 3 h. The interassay variation (CV) for antibody measurements for all seven proteins was 3–10% as evaluated by measuring a HIV-1 antibody plasma pool in duplicate in 13 consecutive assays. We tested the HIV-1 PBA against two commercial enzyme immunoassays (EIAs), the GS™ rLAV HIV-1 EIA (Bio-Rad) and the DETECT HIV EIA (Adaltis), and a Western blot (WB) assay (Calypte Biomedical Corp.), using fresh plasma from six HIV-1-infected individuals, obtained from the National Laboratory for HIV Reference Services (Health Canada). The results obtained with the PBA and the two commercial EIAs were similar (Table 1 ). Samples analyzed with the PBA were interpreted as reactive only when antibodies to proteins derived from at least two of the three gene regions were detected. In fresh plasma samples, the PBA detected antibodies to individual HIV-1 proteins more frequently than the WB (see Table 1 in the online Data Supplement). Antibodies to every viral protein were detected (endpoint titer ≥20), with the exception of p17 in sample 2. In contrast, the WB failed to detect antibodies to one or more proteins in one-half of the samples. Five of the six samples met the WB criteria for positive interpretation, i.e., they showed reactivity to at least two of the following three proteins: p24, gp41, and gp120/160. One sample (sample 3) showed antibody reactivity to p24 only (indeterminate reactivity) when tested with WB, whereas it was reactive for all seven HIV-1 proteins in the PBA (Table 1 in the online Data Supplement). Consistent with early seroconversion, blood samples taken from this patient at a later time were reactive when tested by WB (data not shown). The apparent increased sensitivity of the HIV-1 PBA for the detection of envelope antibodies would constitute a definite advantage over the WB for weakly reactive plasma samples. Antibody reactivity to BSA was detected in one of six fresh plasma samples (endpoint titer, 189; see Table 1 in the online Data Supplement). However, in any given individual, the antibody profile to BSA appeared to be independent of HIV antibody profile, as illustrated in the longitudinal testing of seroconverters with the HIV-1 PBA (Fig. 1A ). Comparison of HIV-1 antibody detection by the HIV-1 PBA, EIAs, and WB. Fresh plasma samples were tested with the HIV-1 PBA as serial 10-fold dilutions (1:20–1:200 000; see the Appendix in the online Data Supplement). Samples were considered positive when antibody reactivity to proteins derived from at least two of the three different gene regions, gag, pol, or envelope, were detected (see Table 1 in the online Data Supplement). IND, indeterminate; NT, not tested. Paper disks (6 mm) from DBS specimens were incubated overnight in elution buffer (see Table 1 in the online Data Supplement) before testing with the EIA and the PBA. For WB testing, the paper disk was incubated with the WB strip overnight in blotting buffer. Results of the HIV-1 PBA are expressed as the ratio of the median fluorescence intensity for the test samples to the cutoff, with values ≥1 considered reactive (see Table 1 in the online Data Supplement). Comparison of HIV-1 antibody detection by the HIV-1 PBA, EIAs, and WB. Fresh plasma samples were tested with the HIV-1 PBA as serial 10-fold dilutions (1:20–1:200 000; see the Appendix in the online Data Supplement). Samples were considered positive when antibody reactivity to proteins derived from at least two of the three different gene regions, gag, pol, or envelope, were detected (see Table 1 in the online Data Supplement). IND, indeterminate; NT, not tested. Paper disks (6 mm) from DBS specimens were incubated overnight in elution buffer (see Table 1 in the online Data Supplement) before testing with the EIA and the PBA. For WB testing, the paper disk was incubated with the WB strip overnight in blotting buffer. Results of the HIV-1 PBA are expressed as the ratio of the median fluorescence intensity for the test samples to the cutoff, with values ≥1 considered reactive (see Table 1 in the online Data Supplement). Antibody reactivity to HIV-1 proteins and BSA from three HIV-1 seroconversion panels. Plasma specimens were tested with the HIV-1 PBA (1:50 dilution; see the Appendix in the online Data Supplement). Results are shown as the ratio of the median fluorescence intensity of the test samples to the cutoff with values ≥1 considered positive. The cutoff for positivity is indicated by the dotted line in each panel. The shaded areas denote specimens reactive with two commercial EIAs (GS rLAV HIV-1 EIA and the DETECT HIV EIA). (A), profiles of antibody reactivity to HIV-1 envelope proteins and BSA from three HIV-1 seroconversion panels. •, gp41 antibodies; ▪, gp120 antibodies; ▾, gp160 antibodies. The dashed line corresponds to antibodies to BSA. (B), profiles of antibody reactivity to HIV-1 gag/pol proteins from three HIV-1 seroconversion panels. ○, p17 antibodies; ▿, p24 antibodies; □, p55 antibodies; ⋄, p66 antibodies. The three seroconversion panels tested were BCP6246 (a 21-specimen panel collected serially from a single plasma donor over a period of 82 days), PRB910 (a 7-specimen panel collected serially from a single plasma donor over a period of 40 days), and PRB923 (a 13-specimen panel collected serially from a single plasma donor over a period of 161 days). We evaluated the PBA for the detection of HIV-1 antibodies eluted from 15 DBS specimens (National Laboratory for HIV Reference Services, Health Canada). The HIV-1 PBA performed as well as the EIA and the WB for the determination of HIV-1-reactive DBS specimens obtained from HIV-1-infected individuals (Table 1 ). Five samples determined to be nonreactive with the EIA were also interpreted as nonreactive with the PBA. Antibodies to HIV-1 p24 were detected in two of the samples nonreactive with the PBA, but these samples were not tested with the WB (see Table 1 in the online Data Supplement). Antibodies to p17 and p55 proteins were detected more frequently with the PBA (p55, 10 of 10; p17, 8 of 10) than with the WB (p55, 7 of 10; p17, 7 of 10), but gp41 and gp120 antibodies were not detected more often (Table 1 in the online Data Supplement). In four DBS specimens, gp41 and/or gp120 antibodies were not detected with the PBA, whereas they were detected as weak bands on the WB. Overnight incubation of the paper disk with the WB strip might partly explain the discrepancy observed for the detection of envelope protein antibodies with the HIV-1 PBA. A shorter incubation period (1 h) was chosen for the PBA to evaluate its potential as a rapid test. Incubation of the microfluorospheres with the paper disk as the elution takes place or prolonging the incubation period with the eluate would likely improve the detection of the envelope antibodies. The HIV-1 PBA performed well for the early detection of HIV-1 antibodies in seroconverters (submitted for publication). In the present study, the PBA was compared with two EIAs for the early detection of HIV-1 antibodies in three seroconverters. Fig. 1 shows the HIV-1 antibody profiles (1:50 dilution) of the ratio of the median fluorescence intensity for the test samples to the cutoff for all seven viral proteins. The first reactive samples detected with the PBA in all three seroconverters were identical to those detected with the two EIAs (Fig. 1 , shaded areas). Antibodies to gp41 (Fig. 1A ) and p55 (Fig. 1B ) were often the first to be detected. As reported by others (23)(24), nonspecific antibody reactivity to the capsid protein p24 was observed in one donor (BCP6246). The PBA described here used only seven HIV-1 proteins. The inclusion of other HIV proteins from HIV-1/2 and various HIV-1 subtypes would be essential for widespread serology screening. HIV-specific peptides, as opposed to full-length recombinant proteins, could also potentially be used to resolve antibody cross-reactivity to the products of the GAG gene, which is encountered occasionally in uninfected individuals. Effort has been invested to improve the algorithm for confirming HIV-1/2 infection (25)(26)(27)(28). The currently used WB can produce false-negative results in cases of recent infection and false-positive results as a result of contaminating cellular antigens or cross-reactive antibodies (24)(29)(30)(31). Alternative immunoblot assays use recombinant HIV-1/2 proteins and peptides deposited as bands on a nitrocellulose membrane (25)(27)(28). Reactivity is estimated qualitatively from the intensities of the individual bands. This type of assay typically requires 20 μL of specimen per strip and 6–7 h for analysis. The use of recombinant proteins and peptides, free of cellular contaminants, has been shown to reduce the number of indeterminate results seen with WB. The HIV-1 PBA parallels the principle of these immunoblots in that it uses recombinant proteins or peptides. It potentially offers the same advantage of the recombinant immunoblot regarding the reduction of indeterminate results. However, the HIV PBA offers additional advantages compared with WB and recombinant immunoblots, such as consistency in the composition and concentration of viral proteins, numeric read-outs, smaller specimen volume requirements (≤5 μL), and less hands-on time (completion time ∼3 h). The HIV PBA offers the high throughput of EIA combined with the individual protein testing capacity of the WB. The assay can be tailored to the requirements of screening or confirmatory testing as well as to the needs of specific vaccine studies. Vaccine trials could benefit greatly from a PBA that differentiates protected from infected individuals. Because large-scale vaccine studies are likely to be conducted in settings where conditions may not permit easy handling of fresh blood specimens, the DBS technology constitutes an attractive alternative for specimen collection and the PBA appears to be well suited for this purpose. The HIV PBA offers substantial advantages over conventional immunoassays, but standardization, validation, and performance testing remain to be done before the assay can be used for clinical testing.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.