Biology Reference
In-Depth Information
where between these two extremes; either could account for the success of this
virus as an opportunistic pathogen.
In contrast, during transplantation and likely accompanying immune suppres-
sion, the cytokine storm induced by transplantation could drive virus reactivation
which is uncontrollable in the absence of a good cytotoxic T cell (CTL) response.
However, studies to address these types of questions are extremely difficult, requir-
ing analyses with naturally, latently infected cells and homologous or nonhomolo-
gous CTLs. Given the frequency of HCMV DNA-positive cells, this would be a
challenging exercise. It is becoming increasingly clear that conducting these and
other analyses pertinent to HCMV latency will require a system for enriching for
HCMV-positive cells in cell populations from naturally infected individuals.
Other Sites of HCMV Latency
Whilst most analyses of HCMV latency have focused on the myeloid lineage, it
must be acknowledged that there may be other cellular sites of HCMV latency in
vivo. Upon virus reactivation, a diverse number of tissues become infected very
rapidly. Whilst this could be attributed to mononuclear cell-mediated dissemination
of the virus, it could also equally be attributed to other sites of reactivation
(Fig. 1c).
CD34
+
stem cells are also believed to be the precursors of some endothelial cells
(Quirici et al. 2001), and latent carriage and/or reactivation in these cells has been
suggested to be involved in atherosclerotic disease (Epstein et al. 1996; see the
chapter by D.N. Streblow et al., this volume). Although the data linking atherosclerosis
and HCMV is circumstantial and contentious (Danesh et al. 1997), HCMV reacti-
vation in endothelial cells (ECs) could result from changes in cellular gene expres-
sion (Maussang et al. 2006; Reinhardt et al. 2006). Countering this interpretation is
the observation that latent HCMV genomes have not been detected in ECs isolated
from saphenous vein tissue of seropositive donors (Reeves et al. 2004). Although
all ECs are likely to arise from the same progenitor, such an analysis does not pre-
clude the possibility that aortic ECs are sites of latent carriage of HCMV (Jarvis
and Nelson 2007). Alternatively, the possibility that a selective segregation of viral
genomes also occurs between venous and aortic ECs, as appears to be the case in
the myeloid linage, cannot be ruled out (Fig. 1c).
Another possibility is that ECs are sites of low-level persistent HCMV infection,
in vivo, rather than sites of true viral latency. Some studies have shown that infected
ECs are sites of virus production which show little sign of cell lysis (Fish et al.
1995, 1998). However, other studies have suggested that the infection of ECs is
fully cytopathic, resulting in complete cell lysis (Kahl et al. 2000). Whether these
differences result from different types of ECs used in the studies or virus strain
variation is unclear. What is clear, though, is that the difficulty in obtaining aortic
tissue from healthy individuals means it is almost impossible to analyse directly
whether these cells are latently or persistently infected.
