No abstract

Summary CD34+ progenitor cells can harbour latent human cytomegalovirus (HCMV); however, the mechanisms of HCMV latency remain unclear. We have investigated the effects of the haematopoietic lineage restriction on the establishment and spread of the latent HCMV to progeny cells. In vitro‐infected and latently‐infected haematopoietic progenitor cells derived from HCMV seropositive donors were studied. The presence of HCMV DNA in bone marrow progenitor (BMP) cells was determined by single colony polymerase chain reaction and fluorescent in situ hybridization (FISH). The presence of CMV DNA was found to be restricted to myeloid progenitors and the percentage of HCMV‐infected cells was lower in naturally‐infected cells than in in vitro‐infected cells. Erythroid differentiation resulted in an abortive infection with persistence of the viral nucleic acids in red cell precursors. In BMP cells from HCMV seronegative donors, HCMV DNA was localized in the nucleus. Bone marrow progenitors in the presence of granulocyte‐macrophage colony stimulating factor (GMCSF) maintained HCMV DNA for extended periods of time. No viral production could be detected throughout the culture but the comparison of the numbers of latently‐infected cells prior to and after the culture suggests that proliferation of haematopoietic progenitor cells may lead to the expansion of latently‐infected cells.

Apart from congenital human cytomegalovirus (HCMV) infection, manifest HCMV disease occurs primarily in immunocompromised patients. In allogeneic bone marrow transplantation, HCMV is frequently associated with graft failure and cytopenias involving all hematopoietic lineages, but thrombocytopenia is the most commonly reported hematologic complication. The authors hypothesized that megakaryocytes (MK) may be a specific target for HCMV. Although the susceptibility of immature hematopoietic progenitors cells to HCMV has been established, a productive viral life cycle has only been linked to myelomonocytic maturation. The authors investigated whether HCMV can also infect MK and impair their function. They demonstrated that HCMV did not affect the thrombopoietin (TPO)-driven proliferation of CD34+ cells until MK maturation occurred. MK challenged with HCMV showed a 50% more rapid loss of viability than mock-infected cells. MK and their early precursors were clearly shown to be susceptible to HCMV in vitro, as evidenced by the presence of HCMV in magnetic column-purified CD42+ MK and 2-color fluorescent staining with antibodies directed against CD42a and HCMV pp65 antigen. These findings were confirmed by the infection of MK with a laboratory strain of HCMV containing the β-galactosidase (β-gal) gene. Using chromogenic β-gal substrates, HCMV was detected during MK differentiation of infected CD34+ cells and after infection of fully differentiated MK. Production of infectious virus was observed in cultures infected MK, suggesting that HCMV can complete its life cycle. These results demonstrate that MK are susceptible to HCMV infection and that direct infection of these cells in vivo may contribute to the thrombocytopenia observed in patients infected with HCMV.