Cutting Out Spurious gp160 Bands from Human Immunodeficiency Virus Western Blot Results

The HIV Reference Laboratory of the Department of Pathology at the Singapore General Hospital serves both as a screening and a reference center for human immunodeficiency virus (HIV) serological testing. The HIV Blot 2.2 Western blot (Genelabs Diagnostics) is used to confirm HIV-reactive screening tests. The Western blot (WB) criteria for a positive result is the presence of at least one band from each of the Env, Gag, and Pol antigens, and a negative result is the absence of specific viral bands. Indeterminate results are reported when the viral bands do not meet the criteria for WB-negative or WB-positive results. The number of samples reported as indeterminate increased from 10.5% in 1996 to an average of 12.2% in 1997 to 1999. This increase was attributed to an unacceptable number of strips showing faint gp160 bands with or without other virus-specific bands. At the time that this increase was noted, the major change in the WB procedure was the conversion of the semiautomated Genelabs Autoblot machine from version 27 to version 36. Version 36 contains an altered design of the disposable plastic tray and processes 36 instead of 27 WB strips. Contamination of WBs with minute volumes of strongly positive serum may lead to faint Env bands even when the dilution is up to one million-fold (1). The gp160 bands, with or without other virus-specific bands, were suspected to be false positives due to carryover of positive sera from adjacent wells. Observation of the design of the tray showed that the troughs were separated by a ridge meant to prevent overflow of reagents from one trough to the next. However, a small ledge along the end of the ridge allowed access for solutions to flow across adjacent troughs. To demonstrate this, red stamp ink diluted 1:1 with distilled water was added to blotting buffer (BBi) as a marker in one trough, which was flanked by blotting buffer without ink (BB). The tray was subjected to the same overnight rocking process that the strips would undergo in a normal run. Throughout the incubation period, no overflow of colored blotting buffer was observed. However, as the tray was tilted forwards during the draining phase, overflow occurred. Figure ​Figure11 shows that after the draining procedure, BBi was present in the lower ends of up to four adjacent troughs containing BB. This implicated the design of the trough for the contamination of wells containing negative sera by adjacent positive sera. FIG. 1 The Genelabs Autoblot 36 disposable plastic tray in situ after the draining stage showing overflow across the ridges (5 to 8), resulting in carryover contamination in the adjacent troughs. Ridges 10 to 17 have been cut at the lower ends with a small flap ... In an attempt to interrupt trough-to-trough carryover, the tray was cut at the lower ends of the ridges and a small flap was raised to prevent overflow. No carryover contamination could be seen with the modified troughs (Fig. ​(Fig.11). This phenomenon was borne out by a routine WB assay run under similar conditions which resulted in spurious gp160 as well as p24 bands in up to four adjacent WB strips in troughs with unmodified ridges. There were no spurious bands on strips in modified troughs. The manufacturer has since modified the trays so that there is no ledge allowing access between the troughs. This observation may be useful for the design of all automated WB trays.