Specific Roles for Lipids in Virus Fusion and Exit Examples from the Alphaviruses - Fusion of Biological Membranes and Related Problems - page 436

membranes, which make it difficult to conclude what differences in lipid
composition exist between HIV and host cell plasma membranes. The
precise structural or functional requirements for cholesterol during HIV
exit from the target cell, and the ability of other sterols to substitute thus
are as yet unclear.
The role of target membrane cholesterol in HIV fusion has been tested
by assaying the in vitro fusion of liposomes by peptides with sequences
derived from the N-terminal fusion peptide of the HIV gp41 fusion protein.
These studies demonstrated that the absence of cholesterol in the liposomes
inhibits peptide-induced fusion by ~33%, while peptide-membrane binding
is unaffected (Pereira et al., 1997a; Pereira et al., 1997b). Peptide-dependent
fusion is inhibited by several physiologically relevant modulators of HIV-1
infectivity, such as inclusion of point mutations in the peptide that block
HIV fusion, competition by hexapeptides derived from the N-terminal gp41
fusion peptide, and addition of several antiviral compounds. These results
suggest that the peptide fusion system may accurately reflect an in vivo role
for cholesterol in the target membrane during HIV fusion. However, it is
important to note that such peptide-liposome fusion studies are prone to
possible artifacts and may not reflect the mechanism of biological mem-
brane fusion (Stegmann et al., 1989a). For example, studies with influenza
virus have demonstrated that the membrane insertion observed with a puri-
fied virus peptide from the spike coiled-coil region (Yu et al., 1994) does
not occur during the actual fusion of influenza virus with a target mem-
brane (Durrer et al., 1996). In addition to the potential problems with
peptide-liposome studies, a group using an in vitro virus-liposome fusion
system found that the presence or absence of cholesterol does not affect
the kinetics or efficiency of fusion of HIV-1 or the closely related SIV-1
(Larsen et al., 1990; Larsen et al., 1993). The exact physiological relevance
of the latter study is also not clear, since virus fusion in this system occurred
with target membranes not containing the CD-4 receptor and co-receptor,
the key molecules known to trigger HIV fusion during infection (Berger,
1997; Binley and Moore, 1997). Thus, a better understanding of the role of
cholesterol in HIV entry and exit awaits further study.
The effects of cholesterol on HIV fusion are believed to be exerted via
the virus fusion protein subunit gp41, which contains the putative fusion
peptide at its N-terminus. The HIV gp120 spike protein subunit is respon-
sible for the binding of virus to cells, usually by recognition of the CD4
receptor. However, several groups have identified the glycosphingolipid
galactosylceramide (GalCer) as an alternative receptor to account for infec-
tion of CDCnegative cells such as colonic epithelial, cells, Schwann cells,
and oligodendrocytes (Harouse et al., 1991; Yahi et al., 1992). Infection of
cultured human colonic epithelial or neural cells is blocked by antibodies