Home Search Abstracts Browse Sessions Program Committee E-mail Abstract Author View Session


Session 16 Oral Abstracts
Viral Genes and Cellular Co-Factors
Tuesday, 10 am - 12:30 pm
Presentation Time: 10:45 am
Room 2008


65
Cell-type-specific Targeting of HIV-1 Gag: Evidence of a Role for PIP2
A Ono*, and E O Freed
NCI-FCRDC, NIH, DHHS, Frederick, MD, USA

Background:  HIV-1 particle formation begins with the targeting of the viral structural protein Gag to the site of virus assembly. In most cell types, including HeLa and T cells, virus assembly takes place largely on the plasma membrane, whereas in macrophages virus particles are mainly formed in intracellular vesicles. A highly basic patch in the matrix domain of Gag is known to be a major determinant of Gag transport to the plasma membrane, but the molecular mechanism by which Gag is targeted to the appropriate subcellular location remains poorly understood. Interestingly, a variety of cellular proteins is known to be targeted to the plasma membrane through interactions between their basic domains and a plasma membrane lipid, phosphatidylinositol-(4,5)-bisphosphate (PIP2).

Methods:  To elucidate the mechanisms underlying the targeting of virus assembly, we examined virus release efficiency and Gag localization of wild type and mutant Gag proteins in various cell types by immunoprecipitation and confocal microscopy. To examine a possible involvement of PIP2 in targeting, we reduced cellular PIP2 levels by expressing phosphoinositide 5-phosphatase IV (5-ptase) in HIV-1-expressing HeLa cells and analyzed the effect on virus assembly by biochemical and microscopic methods.

Results:  In HeLa and T cells, matrix basic-domain mutants assemble in the multivesicular body (MVB) instead of at the plasma membrane. Interestingly, in primary macrophages, even wild type Gag is targeted to the MVB. But, despite the difference in their localization, wild type Gag particles are released from macrophages and HeLa cells with similar efficiencies. MVB targeting is still observed with Gag mutants lacking the L domain, which is known to interact with MVB-associated proteins. In HeLa cells, in which Gag is targeted directly to the plasma membrane, depleting PIP2 correlates with a 5- to 8-fold reduction in virus production. Strikingly, 5-ptase overexpression causes Gag to be mainly retargeted to the MVB.

Conclusions:  These results demonstrate the existence of 2 cell type-dependent pathways for Gag targeting. In macrophages, MVB targeting is the major physiological pathway for extracellular virus production, which likely utilizes exosomal release machinery. In cell types in which virus assembly occurs on the plasma membrane, PIP2 plays a role in directing Gag to the cell surface, and PIP2 depletion induces a switch from plasma membrane to MVB targeting. These findings provide novel insights into the early stages of HIV-1 assembly in virus-infected cells.

Keywords: virus assembly; Gag; phosphoinositide