Specific Roles for Lipids in Virus Fusion and Exit Examples from the Alphaviruses - Fusion of Biological Membranes and Related Problems - page 420

(Gaedigk-Nitschko and Schlesinger, 1991; Gaedigk-Nitschko et al., 1990;
Ivanova and Schlesinger, 1993).
It is apparent that lateral spike-spike interactions are also critical for
SFV assembly at the plasma membrane (Kielian, 1995). Cross-linking
analysis shows that the spike protomer is organized as a trimer on the
surface of infected cells (Rice and Strauss, 1982). Although the cellular
site of spike trimerization is not clear, trimer interactions are important
in virus exit. A non-budding spike protein mutant with an E2 tail negative
for capsid binding could be rescued by the formation of mixed trimers
with the wt spike protein (Ekstrom et al., 1994). In addition, several
examples indicate that El-E2 dimer interactions are also critical for
virus assembly and budding. The E2 cytoplasmic tail does not interact with
the nucleocapsid unless the E2 subunit is complexed with El (Barth
and Garoff, 1997). SFV spike proteins containing mutations within the puta-
tive El fusion peptide form highly unstable E1-E2 dimers (Duffus et al.,
1995). These mutant spike proteins are efficiently synthesized and trans-
ported to the plasma membrane where they associate with the nucleocap-
sid, but show a strong and thermoreversible defect in virus exit (Duffus
et al., 1995).
In summary, efficient alphavirus exit requires key viral components:
the virus spike and nucleocapsid, the specific interaction of the E2 tail with
nucleocapsid, the expression of 6K, and correct lateral spike protein inter-
actions. The role of cholesterol in virus exit will be discussed in section 3.2
below.

3. THE ROLE OF CHOLESTEROL IN THE
ALPHAVIRUS LIFECYCLE

3.1. Role of Cholesterol in Fusion

3.1.1. In Vitro Cholesterol Requirements

A role for cholesterol in alphavirus membrane interactions was first
detected in studies of virus-liposome binding by Mooney et al. (Mooney
et al., 1975), who found that SIN bound to liposomes in a low pH and
cholesterol-dependent reaction. As discussed above (section 2.2.3), such
virus-liposome association is indicative of either El-membrane interaction
prior to fusion and/or the actual fusion reaction itself. Once the analysis of
SFV endocytic uptake indicated that virus fusion was triggered by low pH,
the in vitro fusion of SFV with liposomes was characterized using content
mixing assays (White and Helenius, 1980). Fusion with liposomes composed