Abstract
The surface of HIV-1, like that of other retroviruses, is studied with virally encoded glycoproteins which appear ultrastructurally as electron-dense spikes or knobs. The glycoprotein that forms the spike structure, gp120, is non-covalently bound to the transmembrane glycoprotein gp41. Mature HIV-1 virions do not have as many spikes as the genetically related retroviruses HIV-2 and SIV. gp120 is lost from HIV-1 during viral morphogenesis and after incubation of the virus with the soluble form of cellular receptor CD4. In this study we used ultrastructural cytochemistry and morphometry to quantitate the distribution of envelope glycoprotein spikes on budding and mature HIV-1 virions and to look for alternatives to the laborious and somewhat subjective spike-counting technique for envelope spike analysis on HIV-1. HIV-1, strain HTLV-IIIB, was examined after staining of envelope glycoproteins with either tannic acid, immunogold staining for gp120 (gp120-immunogold), or lectin-gold staining with concanavalin A for mannose residues (ConA-HRP-gold) and frequency distributions of spikes or gold particles per micron HIV-1 membrane generated. Envelope spikes were normally distributed on membranes of budding and mature HIV-1. However, the density of spikes per micron viral membrane on mature HIV-1 virions was approximately 50% of that observed on budding virions. ConA-HRP-gold and gp120-immunogold did not efficiently label budding virions. The shape of the frequency distribution for ConA-HRP-gold particles on mature virions was similar to that for envelope spikes and could be used to quantitate envelope glycoproteins on HIV-1. In addition, ConA-HRP-gold staining was able to detect the loss of envelope proteins after treatment of virus with soluble CD4. gp120-immunogold labeling was patchy and many virions were unlabeled. ConA-HRP-gold staining proved to be a rapid, reliable, and easily quantifiable method for estimation of envelope glycoprotein density on mature HIV-1. However, the loss of spike structures throughout the life cycle of HIV-1 can effectively be determined only by direct spike counting.