Biology Reference
In-Depth Information
One major difference between viral latency and reactivation is characteristically
defined by the absence of lytic gene transcription during latent carriage of virus.
Although transcripts arising from the major IE region of HCMV have been
detected during latency (Kondo and Mocarski 1995; Kondo et al. 1996) (see Sect.
4 below), the major IE transcripts IE72 or IE86 are not expressed in naturally
latent CD34
+
cells or monocytes (Taylor-Wiedeman et al. 1994; Mendelson et al.
1996). Indeed, it is only upon terminal differentiation of these cells to mature
macrophage or dendritic cell phenotypes that viral lytic gene expression is
observed, which, under certain conditions, can result in complete reactivation and
release of infectious virus (Taylor-Wiedeman et al. 1994; Soderberg-Naucler
et al. 1997; Soderberg-Naucler et al. 2001; Reeves et al. 2005b). However,
attempts to dissect the mechanisms of HCMV latency and reactivation have
been hampered by the frequency of seropositive cells in vivo and a lack of a
robust tissue culture model which allows a more thorough, large-scale analysis of
natural latency.
Models of HCMV Latency Using Experimental Infection
The low frequency of latently infected cells in vivo has resulted in a number of
studies of HCMV latency which have been used in experimental infection of cord
blood CD34
+
cells, fetal liver CD34
+
cells (GM-Ps), G-CSF mobilised CD34
+
cells and CD34
+
cells isolated from bone marrow aspirates (Kondo et al. 1994;
Minton et al. 1994; Hahn et al. 1998; Maciejewski and St Jeor 1999; Goodrum
et al. 2002; Slobedman et al. 2002; Reeves et al. 2005a). These experimentally
infected latent model systems can result in 20%-90% of cells carrying latent virus
depending on the cell type, virus strain and multiplicity of infection used.
Consequently, such experimental latency model systems result in an increase in
the level of latently infected cells in the experimental population and have made
it possible to perform more comprehensive analyses which can then be tested in
naturally latently infected cells.
Although it is not possible to fully review the wealth of data obtained from these
studies here (for reviews see Streblow and Nelson 2003; Bego and St Jeor 2006;
Sinclair and Sissons 2006), a number of instructive observations have been made.
In general, they suggest that long-term carriage of viral genomes during latency
occurs in the absence of any significant viral gene expression and the carriage of
latent genomes appears to be specific to certain cell populations which include the
precursors of monocytes (CD34/CD33/CD14) and dendritic cells (CD34/CD33/
CD1a) (Hahn et al. 1998). Furthermore, reactivation of lytic gene expression
requires terminal differentiation of such progenitors to macrophages or dendritic
cells (Maciejewski and St Jeor 1999; Reeves et al. 2005a). Thus, there is good
agreement between studies on experimental and natural latency and, consequently,
these experimental models have been used extensively to address one of the more
intriguing aspects of HCMV latency: latent viral gene expression.
