Expression of HERV-K, a Human Endogenous Retrovirus, in Hematopoietic Cells.

Background: In contrast to other human endogenous retrovirus families which are highly defective, HERV-K proviruses possess all viral gene types of exogenous retroviruses. In malignancy (tumor cell lines, breast cancer, melanoma as well as leukemias and lymphomas) but also in many normal tissues, including placenta HERV-K mRNA expression has been detected. In hematopoietic stem cells, an interaction of HERV-K with elements from vectors which are used for gene therapy may lead to unforeseen consequences. The purpose of this study was to localize HERV-K mRNA expression in hematopoietic stem cells of different sources and to evaluate a potential association of HERV-expression with hematologic diseases and to define a possible risk of HERV-expression for transplantation.

Methods: Quantitative real time PCR (RTQPCR) was used for detecting HERV-K expression in MNC from blood and/or bone marrow samples of patients with leukemias ( 10 MDS, 16 MPD, 14 AML and 1 CML blast crisis) and MNC from blood (n=9) and lymph nodes (n=3) from lymphoma patients. In addition, 8 blood samples and 2 bone marrow samples from healthy donors as well as 9 cord blood samples and 20 apheresates from lymphoma patients in remission containing PBSC were analyzed. Analysis of the same HERV-K gene (pol) in placenta and maternal blood as well as in genomic DNA from all above mentioned sources served as control. In HERV-K overexpressing samples expression of HERV in specific cells was tested by RNA in situ hybridisation (RISH), and combined staining with a CD34 specific monoclonal antibody.

Results: Relative mRNA levels of HERV-K in apheresates as well as in blood and bone marrow samples of patients with MDS, MPD and AML from subtypes with less than 1% CD34 positive cells were in the range of less than 2% in relation to housekeeping genes. However, in bone marrow samples of two AML and a CML blast crisis (all of them with more than 50% CD34+ cells) as well as in CD34+ cells from cord blood (but not from apheresates or normal bone marrow) as well as in three lymph nodes of patients with B-cell lymphomas, an overexpression of HERV-K was detected and confirmed by RISH. Enhanced expression of HERV-K was detected in cord blood derived apoptotic CD34+ cells, whereas the relative amount of HERV in genomic DNA did not change in samples from different sources.

Conclusions. Our data indicate that HERV-K may be activated in lymph nodes of lymphomas and in CD34+ cells from bone marrow of leukemia patients and from cord blood but not in stem cell apheresates from peripheral blood. In cord blood and malignancy expression appears to be associated with apoptosis. Similar amounts of HERV-K in different samples of genomic DNA of the same cells and the absence of HERV-K activation in normal bone marrow and PBSC indicate that activation of HERV-K is the result of regulation of gene expression rather than amplification. A possible specific role of HERV in the pathogenesis of hematologic diseases and in hematopoiesis (cord blood) remains to be established. Differences between peripheral blood stem cells and cord blood may support a critical stance regarding the use of cord blood derived stem cells for transplantation and gene therapy.