Biology Reference
In-Depth Information
subsets, and were not correlated with clinical parameters, including the stage of
HIV disease. In some cases, similar investigations by other groups even failed to
reveal signi®cant repertoire changes ( Boyer et al., 1993; Posnett et al., 1993).
However, because all of these early cross-sectional studies did not take into
consideration the outbred nature of human populations and MHC-type de-
pendence of the repertoire, the signi®cance of the observed variations in TCR b
chain expression levels was misappreciated. Indeed, to be able to reveal relevant
repertoire di¨erences, such comparisons need to be drawn between subjects
who are at least partly haplo-identical (Gulwani-Akolkar et al., 1991). One
such approach was to study a cohort of HIV-discordant monozygotic twins, in
whom HLA identity was by de®nition complete. Repertoire analysis revealed
signi®cant di¨erences in the levels of expression of multiple TCRBV families,
both in CD4
and CD8
T cells, and the incidence of these perturbations ap-
peared to increase along with the progression of HIV disease ( Rebai et al.,
1994). Similar di¨erences were never observed in pairs of uninfected twins, and
therefore had to be related to opportunistic infections or HIV infection itself.
An alternative approach involved simultaneous testing of several distinct lym-
phoid subcompartments (i.e., peripheral blood, spleen, and lymph nodes)
within the same HIV-infected individual. In two such studies, repertoire typing
by PCR or ¯ow cytometry revealed signi®cant di¨erences in the expression
levels of several TCRBV subsets (Soudeyns et al., 1993, 1994). Because HIV
replicates predominantly within peripheral lymphoid organs during chronic in-
fection (Embretson et al., 1993; Pantaleo et al., 1991, 1993), it was presumed
that the origin of the repertoire disparities observed between these lymphoid
compartments was related to regional di¨erences in HIV-1 viral load, thus
supporting the notion that HIV infection led to qualitative changes in the
composition of T-cell subsets.
An independent line of evidence in support of the existence of an HIV-
associated SAg was provided by Posnett and colleagues. In accordance with
the MMTV model ( Held et al., 1993), these investigators hypothesized that
TCRBV-speci®c SAg-mediated activation of T cells would result in increased
HIV replication within these cell subsets. Indeed, HIV-1 viral load was found
to be signi®cantly greater in CD4
T cells expressing TCRBV12S1 than in
those expressing TCRBV6S7 (Laurence et al., 1992). This subset-speci®c viral
``reservoir'' was only observed in presence of CMV coinfection, suggesting the
existence of a CMV-associated herpesvirus SAg acting to enhance HIV repli-
cation in trans ( Dobrescu et al., 1995a, b). However, because no preferential
expansions of TCRBV12S1 T cells were observed in these subjects, the signi®-
cance of these observations in the context of HIV pathogenesis remains unclear.
Most of these early reports were, to some degree, consistent with the exis-
tence of an HIV-encoded or HIV-associated SAg. Indeed, as the TCR reper-
toire is remarkably stable through time, TCRBV-speci®c perturbations are not
seen in healthy individuals, or at least not to any comparable extent. However,
the apparent randomness in the identity of TCRBV subsets that showed per-
turbations in HIV infection became increasingly di½cult to reconcile with the
