Biology Reference
In-Depth Information
ing, which decreases CXCR4 expression, is less e¨ective. CXCR4-associated
signaling may also be important for the maintenance of HIV replication (Cheng
et al, 2000); (iii) B cells can only be infected by T-lymphotropic or dual strains
of HIV-1, via CXCR4. The switch toward lymphotropic strains is also favored
by high concentrations of IL-4 (Valentin et al., 1998). This suggests that B cells
may be infected by HIV-1 late in the disease, accompanying the transition to
the symptomatic phase and follicular involution.
In our model, the proliferation and TNFa production of infected B cells
were found to increase in the presence of CD40 mAb and IL-4, with or without
IL-2, whereas the highest levels of p24 production were observed in the pres-
ence IL-2 and IL-4, with or without CD40 mAb. In vivo, IL-2 production is
stronger during the early phases of the disease and decreases progressively later
on, whereas IL-4 production progressively increases. Thus, at all stages of the
disease, HIV-1 replication may be sustained by these cytokines. IL-10 decreases
p24 production induced by most of these stimuli by 34±72%, but only inhibits
IL-4-mediated and IL-2-mediated cell proliferation (Gras et al., 1996). IL-10-
mediated inhibition of HIV-1 replication has also been reported in macro-
phages, in which it seems to counteract NF-kB activation ( Wang et al., 1995).
IL-10 is produced during the late phase of the disease and may therefore
antagonize IL-2- or IL-4-mediated HIV replication without decreasing B-cell
proliferation or Ig production. Thus, our results show that cell proliferation,
cytokine production, and HIV-1 replication in B cells are under the control of
di¨erent pathways. In contrast to Poulin et al., who showed a signi®cantly
lower level of proliferation for their HIV-infected B cells ( Poulin et al., 1994),
we found the levels of proliferation of HIV-infected and mock-infected cells to
be similar. Nevertheless, it is possible that the decrease in cell proliferation was
undetectable due to the low percentage of infected B cells we obtained. We also
observed no syncytia formation in our model, whereas Moir et al. detected
syncytia but only after isolation of the CD4
fraction (Moir et al., 1999). In
vivo, syncytium formation and decrease in cell proliferation might result in the
rapid death of HIV-1-infected B cells, contributing to follicle involution ob-
served in HIV
patients.
Although B cells are susceptible to HIV-1 infection in vitro and support
active HIV-1 replication, the in vivo infection of B cells has not yet been dem-
onstrated. This is intriguing given the strong expression of CXCR4 on most B
cells and the presence of CD4 on at least a proportion of B cells. The best
g
ÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐÐ
fected B cells; 2, mock-infected B cells, day 4 postinfection; 3,c
0
-ADE-infected B cells, day 4 post-
infection; T1±6, 10-fold dilutions of 8
E
5-LAV cell DNA in PBMC DNA. T1 contains 17,300 pro-
viral copies in 1 mg of DNA. Representative of three independent experiments. (F ) Flow cytometric
analysis of HIV-1 gp120 envelope protein expression on c
0
-ADE-infected B cells. Phorbol ester-
activated B cells were submitted to HIV infection in c
0
-ADE conditons and cultured for 2 days in
the presence of cytokines. Indirect immuno¯uorescence of gp120 was performed with 110.4 mAbs.
B cells were stained with CD19 mAb and mouse IgG as positive and negative control, respectively.
1, Infected B cells, 1 h postinfection; 2, uninfected B cells, day 2 postinfection; 3, infected B cells,
day 2 postinfection. Representative of two experiments.
