The susceptibility of hematopoietic progenitor cells to infection by human cytomegalovirus (HCMV) was investigated using several strains of HCMV, including the recombinant strain RC256. RC256 is derived from the laboratory strain Towne and contains the Escherichia coli LacZ gene coding for beta-galactosidase (beta-gal) regulated by an early HCMV promoter. Expression of LacZ allowed the detection of HCMV in individual hematopoietic cells. Clonogeneic bone marrow (BM) progenitors, including CD34+ cells, could be infected with HCMV and would then form normal hematopoietic colonies. By polymerase chain reaction (PCR) amplification of DNA, HCMV could be detected in both erythroid and myeloid colonies. LacZ activity was observed predominantly in cells of myelomonocytic lineage. When cells derived from HCMV-infected progenitors were cocultivated with permissive human fibroblasts, infectious virus expressing LacZ was recovered. Although no characteristic HCMV cytopathology was observed in BM colonies, high virus to cell ratios resulted in a moderate inhibition of colony formation. Since infected hematopoietic progenitors can harbor HCMV for weeks and through several differentiation steps in culture, we postulate that in vivo these cells may serve as a reservoir of latent virus and contribute to HCMV dissemination.

Clinical experience and laboratory data suggest that human cytomegalovirus (HCMV) is present in peripheral blood of seropositive individuals in a latent or persistent state and can be transmitted via blood products and be reactivated in seropositive imunocompromised patients. The pathophysiology of HCMV latency and the nature of HCMV interaction with hematopoietic cells remains unknown. In this study, we investigated the infection of bone marrow (BM) progenitor cells and their progeny as a model of HCMV latency. A clinical isolate and the recombinant laboratory strain Towne/lox containing the Escherichia coli β galactosidase (β-gal) gene regulated by immediately early (IE) HCMV promoter were used to infect highly purified CD34+ cells. Although the infection of these cells with a clinical isolate was associated with an inhibition of proliferation by 59%, an expansion of progeny derived from these cells was possible. Polymerase chain reaction analysis and staining for β-gal have shown that HCMV persisted in infected BM CD34+ cells and their progeny for up to 4 weeks. However, IE and late gene products (mRNA and protein) were detected only late in the course of infection and their expression correlated with terminal macrophage differentiation of the CD34+-derived progeny. Although early infection of CD34+ progenitor cells was not productive (as shown by the plaque assay), infectious virus could be recovered from the terminally differentiated cultures. BM progenitor cells may serve as a reservoir of the latent virus with limited transcription. Proliferation and monocytic maturation of infected progenitors may lead to the numerical expansion of HCMV-infected cells, which serve as a source of HCMV dissemination and reactivation.

No abstract

Formalin fixation and paraffin embedding present the standard procedures for conserving clinical tissues for histological analysis. However, molecular analysis is impaired by the cross linking properties of formalin. The PAXgene tissue system (PreAnalytix, Switzerland) is a new formalin-free tissue collection device. AIMS: In this study we aimed to evaluate this new tissue preservation technique in comparison with formalin fixation and fresh frozen tissue samples. METHODS: 12 melanoma biopsy samples were divided and fixed simultaneously with formalin, PAXgene or fresh frozen in liquid nitrogen and analysed with regard to morphology, immunohistochemistry, DNA and RNA content and quality. Markers of melanocytic differentiation and tumour cell proliferation were used. RESULTS: Morphology was well preserved in PAXPE samples. However, 5 out of 11 immunohistochemical markers showed significantly lower overall staining and staining intensity with PAXPE tissues in comparison with formalinfixed, paraffin-embedded (FFPE). Increasing membrane permeability through adding a detergent did proportionally increase staining intensity in PAXPE samples. Amplification of different mRNA amplicons showed a direct relationship with the size of the amplicon with greater template integrity observed in PAXPE samples. Sequencing and mutational analysis of DNA samples were comparable for all the different fixation methods, while the level of DNA fragmentation seemed to be lower in PAXPE compared with FFPE tissues. CONCLUSIONS: The switch from formalin to PAXgene fixation would require a re-evaluation of immunohistochemical markers and staining procedures originally developed for FFPE tissues. Our data demonstrate that PAXPE fixation offers some advantages concerning molecular analysis. However, these advantages would not justify substituting formalin fixation in any routine pathology laboratory.