Biology Reference
In-Depth Information
CD21/CD35 complexes may act as receptors for HIV-1, in addition to CD4. In
our system of C
0
-ADE infection of B cells, the blockade of CD4 by four dif-
ferent mAb and CD4-HSA protein systematically results in partial inhibition of
infection. The level of inhibition is greater in B cells with detectable CD4 (de-
tected by ¯ow cytometry) than in B cells with undetectable CD4. In contrast,
the inhibition induced by CD21 mAb and CD35 mAb is as strong in CD4
ÿ
as
in CD4
B-cell populations ( Legendre et al., 1996). This provides strong evi-
dence that the CD21/CD35 complex, in addition to concentrating antibody-
coated virions at the surface of FDC and B cells such that a threshold is
reached allowing CD4-dependent infection, is also a true HIV receptor on B
cells, mediating CD4-independent HIV entry. Triggering of the CD21/CD35
complexes may also generate intracellular signals that increase HIV-1 repli-
cation after virus entry (Matsumoto et al., 1991). Indeed, CD21 mAb and
aggregated complement fragments increase B-cell activation and proliferation
whereas CD35 mAb increase B-cell di¨erentiation (Tedder et al., 1997; Weiss
et al., 1987).
In our in vitro infection model, we investigated whether signals, other-
wise important for B-cell di¨erentiation within lymphoid organs, facilitated the
entry of HIV-1 and favored its replication in tonsilar B cells. We observed that
p24 production was greater if B cells were activated for 2 days by incubation
with phorbol esters or a combination of CD40 mAb and IL-4 before infection.
In contrast, levels of p24 production remained low after BCR triggering. Post-
infection, p24 production is strongly increased by the combination of IL-4 and
IL-2, with or without CD40 mAb (Gras et al., 1993, 1996). Similarly, Poulin et
al. achieved the in vitro infection of peripheral blood B cells stimulated for 10
days in the presence of CD40L-expressing cells and IL-4, by high amounts of
free HIV-1 virions ( Poulin et al., 1994). In these conditions, they obtained 30±
50% infected B cells, whereas we only obtained 3±10% CD22
-infected B cells
in our model, as shown by polymerase chain reaction ( PCR) analysis of pro-
viral sequences and membrane expression of surface gp120 ( Fig. 5.1). The
higher percentage of infected B cells in Poulin's study correlated with higher
percentages of B cells entering the G1 phase of the cell cycle: 70±80%, versus
less than 30% in our model ( Banchereau et al., 1994). Using the same infection
protocol, Moir et al. recently showed that the HIV-1 infection of B cells re-
quires CD4 and CXCR4 as HIV-1 receptors and T lymphotropic or dual
strains of HIV-1 (Moir et al., 1999). Although CXCR4 is expressed on all
peripheral B cells, only 20±30% of CD40 plus IL-4-activated B cells are
CXCR4
, and only 15% of these activated B cells are also CD4
. Intracellular
p24 staining clearly showed that only one third of the CD4
B cells in Moir's
study (i.e., 5% of total B cells) were infected with HIV-1 (Moir et al., 1999).
This percentage is similar to our 1% of B cells that actively replicated HIV-1
(Gras et al., 1996) ( Fig. 5.1). From these studies, it seems clear that (i) strong
and sustained activation of B cells is required for in vitro HIV-1 B-cell infection;
(ii) phorbol ester and CD40 triggering, which strongly stimulate nuclear factor
(NF)-kB activation, favor viral replication, whereas short-term BCR trigger-
