Endogenous retroviruses and cancer.

The genomes of vertebrates contain sequences that are similar to present-day exogenous retroviruses. Such sequences, called endogenous retroviruses (ERVs), have resulted from ancestral germ line infections by exogenous retroviruses which have thereafter been transmitted in a Mendelian fashion. By analogy to exogenous tumorigenic retroviruses, ERVs have been implicated in the pathogenesis of cancer. Cumulative evidence from animal models indicates that ERVs may participate in the process of malignant transformation or promote tumor growth, e.g. through insertional mutagenesis or via counteracting tumor immunosurveillance. Here, we review the role of ERVs in tumorigenesis with focus on human ERVs (HERVs) in human cancer. Although available data suggest a potential role of HERVs in human cancers, in particular germ cell tumors, the contributions of HERVs to human tumorigenesis warrant further elucidation. (Part of a multi-author review).

Functional cooperation of Epstein-Barr virus nuclear antigen 2 and the survival motor neuron protein in transactivation of the viral LMP1 promoter.

Epstein-Barr virus nuclear antigen 2 (EBNA2) is essential for viral transformation of B cells and transactivates cellular and viral target genes by binding RBPJkappa tethered to cognate promoter elements. EBNA2 interacts with the DEAD-box protein DP103 (DDX20/Gemin3), which in turn is complexed to the survival motor neuron (SMN) protein. SMN is implicated in RNA processing, but a role in transcriptional regulation has also been suggested. Here, we show that DP103 and SMN are complexed in B cells and that SMN coactivates the viral LMP promoter in the presence of EBNA2 in reporter gene assays and in vivo. Subcellular localization studies revealed that nuclear gems and/or coiled bodies containing DP103 and SMN are targeted by EBNA2. Protein-protein interaction experiments demonstrated that DP103 binds to SMN exon 6 and that both EBNA2 and SMN interact with the C terminus of DP103. Furthermore, a DP103 binding-deficient SMN mutant was released from nuclear gems and/or coiled bodies and further enhanced coactivation. In addition, impaired transactivation of a DP103 binding-deficient EBNA2 mutant was rescued by overexpression of SMN. Testing different promoter constructs in luciferase assays showed that RBPJkappa is required but not sufficient for coactivation by EBNA2 and SMN. Overall, our data suggest that EBNA2 might target spliceosomal complexes by binding to DP103, thereby releasing SMN which subsequently exerts a coactivational function within the RNA-polymerase II transcription complex on the LMP1 promoter.

Incidence and prevalence of infection with human granulocytic ehrlichiosis agent in Germany. A prospective study in young healthy subjects.

BACKGROUND: Only limited data are available on incidence and prevalence of infection with the human granulocytic ehrlichiosis (HGE) agent in a healthy population. MATERIALS AND METHODS: In a prospective study, we tested 361 male soldiers (age 18-29 years) from southwestern Germany for the HGE agent immunoglobulin G (IgG) using an indirect immunofluorescence antibody assay and for Borrelia burgdorferi IgG with an ELISA at the beginning and the end of their 10-month military service. Using a standardized questionnaire, the subjects were asked about clinical symptoms at the beginning and the end of the observation period. RESULTS: Of these 361 subjects, 14.9% were HGE agent IgG positive at study entry. 19 participants (5.3%) seroconverted from IgG negative to positive during the observation period resulting in an incidence rate of 6.4% per year. 20 subjects converted from initially HGE agent IgG positive to negative resulting in a reconversion rate of 6.6% per year. Concurrence of Borrelia IgG and HGE agent IgG was observed in 21.1%, whereas 13.7% were HGE agent IgG positive but Borrelia IgG negative (not significant). Clinical symptoms associated with HGE were not present in seroconverting subjects. CONCLUSION: Infection with the HGE agent occurs frequently in southwestern Germany but was asymptomatic in these young subjects.

Levels of virus-specific CD4 T cells correlate with cytomegalovirus control and predict virus-induced disease after renal transplantation.

BACKGROUND: Immunosuppressive treatment in transplant patients frequently causes infectious complications with cytomegalovirus (CMV). The extent of CMV replication can be followed by a number of diagnostic methods. There is, however, no simple diagnostic tool to assess the quality of the cellular antiviral immune response of an individual patient. This would be of particular importance for therapy decisions, as patients with detectable virus load do not necessarily develop CMV-related disease. Using a rapid whole blood assay, the frequencies of CMV-reactive CD4 and CD8 T cells were followed after renal transplantation to characterize their relative contribution in the containment of CMV infection. METHODS: T cells from transplant patients ands healthy control persons were stimulated with CMV antigen in vitro. Based on specific cellular activation and induction of intracellular cytokines, the frequency of CMV-reactive CD4 and CD8 T cells was determined using flow cytometry. Viral load quantified using the "hybrid-capture" assay. RESULTS: The absence of CMV complications in long-term transplant recipients is reflected by stable virus-specific T-cell frequencies, which do not differ from healthy CMV-positive controls. In contrast, during the first months after transplantation, clinical symptoms are preceded by a decrease in CMV-reactive CD4 T-cell frequencies and an increase in CMV load. CONCLUSIONS: The individual immune response and CMV replication are critically balanced and can be characterized by assesing both viral load and antiviral T cells. Our experimental design allows the identification of patients with sufficient, insufficient, or absent T-cell activity and can serve as diagnostic tool to facilitate decisions on antiviral therapy.

Evaluation of a recombinant line blot for diagnosis of Epstein-Barr Virus compared with ELISA, using immunofluorescence as reference method.

A commercial line blot using recombinant antigens was compared with a commercial ELISA and 'in-house' IFA (reference test). Two panels were evaluated: Panel A was selected to distinguish between primary infections (89), past infections (20) and seronegatives (8) in immunocompetent individuals. In panel B, patients with a high number of reactivations were included: immunosuppressed patients (37), lymphoma (19), nasopharyngeal carcinoma (10), chronic fatigue syndrome (14). Blood donors (43) and cross-reactive sera (29) were added as controls. Line blot and IFA were concordant in 94% of primary infections, 100% of seronegatives and 100% of past infections, similar to ELISA. Results differed significantly with regard to reactivations. When compared with IFA, the incidence of reactivations was overestimated by the blot, 24 and 58% in blood donors and cross-reactive sera, respectively. ELISA showed a similar problems with 21 and 34% indeterminate results, respectively. The line blot is easy to carry out, has a good concordance with the reference IFA for primary infections, and is, therefore, a sufficient choice for distinguishing primary infection from seronegative and past infection. EBV reactivation assessment will require other methods such as EBV viral load.

The Rev/Rex homolog HERV-K cORF multimerizes via a C-terminal domain.

Expression of human endogenous retrovirus K (HERV-K) is associated with germ-cell neoplasia. HERV-K encodes a protein of the Rev/Rex family, cORF, that supports cellular transformation and binds the promyelocytic leukemia zinc finger (PLZF) protein implicated in spermatogenesis. Rev/Rex function invariably depends on multimerization. Here we show that cORF likewise self-associates to form higher-order oligomers. Amino acids (aa) 47-87 in cORF are sufficient, aa 75-87 essential for self-association. Consistently, this domain is predicted to form a hydrophobic alpha-helix that may represent an oligomerization interface. The existence of a dimerization-competent cORF mutant lacking PLZF-binding activity (cORF47-87) suggests a way of dominant negative inhibition of the proposed tumor susceptibility factor cORF.

Rapid whole blood analysis of virus-specific CD4 and CD8 T cell responses in persistent HIV infection.

OBJECTIVES: Upon HIV infection, strong antiviral cytotoxic and helper T cell responses are generated. They are considered to be an important component in the control of HIV viral load. A simple and rapid whole blood assay was established to quantify and simultaneously characterize HIV-reactive CD4 and CD8 cells. The assay was applied to evaluate the effect of antiretroviral therapy on HIV-specific T cell responses. METHODS: Whole blood of 33 HIV-infected individuals was specifically stimulated by HIV-1 Pr55gag, and activation-induced intracellular cytokine expression in CD4 and CD8 T cells was analysed by flow cytometry. RESULTS: HIV-1-specific CD8 and CD4 T cells can be quantified simultaneously. As specific antigen, HIV-1 Pr55gag virus-like particles were superior to soluble protein, especially for the activation of CD8 T cells. In untreated individuals, a high frequency of HIV-specific T cells was observed. The frequency of CD8 T cells was consistently higher than the respective CD4 T cell response, thus demonstrating a dominance in CD8 T cell expansion in persistent HIV infection. Patients on antiretroviral therapy showed a significant reduction in HIV-specific CD4 and, even more strikingly, CD8 T cells. CONCLUSION: The whole blood assay provides a rapid estimate of the total antiviral T cell resources, and is highly suited for a clinical setting. It may thus have widespread applications for the evaluation of vaccination strategies and immunotherapy. Because antiretroviral therapy significantly reduces both HIV-specific cytotoxic and helper T cell responses, future therapeutic strategies should aim at improving cellular antiviral immunity.

Human endogenous retrovirus protein cORF supports cell transformation and associates with the promyelocytic leukemia zinc finger protein.

Human endogenous retrovirus sequences (HERVs) reside in the genomes of primates and humans for several million years. The majority of HERVs is non-coding but a limited set is intact and can express proteins. We have recently identified an almost intact HERV-K(HML-2) provirus on chromosome 7 and have documented that most patients with germ cell tumors (GCTs) display antibodies directed against proteins of HERV-K(HML-2). To address whether these proteins merely represent tumor markers or contribute to neoplastic transformation, we examined the transforming potential of various HERV sequences and studied physical interactions between HERV and cellular proteins by yeast two-hybrid and biochemical assays. cORF, a protein encoded by the C-terminal open reading frame within the env gene, supports tumor growth in nude mice and associates with the promyelocytic leukemia zinc finger protein (PLZF). The interaction domains map between amino acid residues 21 and 87 of cORF, and between residues 245 and 543 of PLZF. PLZF is critical for spermatogenesis in mice. Abnormal spermatogenesis or maturation of gonocytes is thought to predispose humans to the development of germ cell tumors. Thus, cORF of human endogenous retroviruses may contribute to tumor development by interfering with processes during spermatogenesis that involve PLZF.

A rev-like NES mediates cytoplasmic localization of HERV-K cORF.

The human endogenous retrovirus K (HERV-K)-encoded protein cORF has recently been shown to be a functional homolog of the HIV Rev protein. Rev-mediated RNA export requires interaction between a leucine-rich nuclear export signal (NES) in Rev and the nuclear export receptor Crm1/exportin1. Like Rev, cORF binds to Crm1 and cORF-mediated RNA export depends on Crm1 activity. Here we document that mutation of the putative NES in cORF results in trapping of the protein in the nucleus, suggesting that the cORF NES functions in analogy to the Rev NES.

Characterization of DP103, a novel DEAD box protein that binds to the Epstein-Barr virus nuclear proteins EBNA2 and EBNA3C.

The Epstein-Barr virus-encoded nuclear antigens EBNA2 and EBNA3C both interact with the cellular transcription factor RBP-Jkappa and modulate the expression of several shared target genes, suggesting a tight cooperation in latently infected cells. In a survey for additional cellular factors that bind to EBNA2 as well as EBNA3C, we have isolated and characterized DP103, a novel human member of the DEAD box family of putative ATP-dependent RNA helicases. The interaction with DP103 is mediated by amino acids (aa) 121-213 of EBNA2 and aa 534-778 of EBNA3C, regions that are not involved in binding of the viral proteins to RBP-Jkappa. The DP103-cDNA encodes a protein of 824 aa that harbors all of the common DEAD box motifs. Monoclonal antibodies raised against DP103 detect a protein of 103 kDa in mammalian cells that resides in high molecular weight complexes in vivo. We have detected an ATPase activity intrinsic to or closely associated with DP103. By subcellular fractionation, we find DP103 in both a soluble nuclear fraction as well as in the insoluble skeletal fraction. Whereas the protein and its mRNA are uniformly expressed in all tested cell lines, we observed differential expression of the mRNA in normal human tissues.

Functional analysis of different LMP1 proteins isolated from Epstein-Barr virus-positive carriers.

The Epstein-Barr virus (EBV) is the causative agent of infectious mononucleosis and is implicated in the development of several human malignancies. Latent membrane protein 1 (LMP1), an EBV protein with known oncogenic properties, may be important in the pathogenesis of EBV-associated tumors, particularly nasopharyngeal carcinoma (NPC) and Hodgkin's disease (HD). Several reports suggested that sequence variations in the LMP1 gene may define a more aggressive, geographically restricted EBV-genotype. Most mutations in the LMP1 gene described are located within the C-terminus of the protein. However, the effect of these mutations on the biological function of the protein remains widely unknown. Therefore, this study aimed in investigating whether mutations detected in LMP1 genes isolated from different EBV-positive carriers have an effect on the biological function of the protein. For this purpose the LMP1 genes were amplified by nested PCR from DNA out of bone marrow and peripheral blood lymphocytes and sequenced. Three functional assays were performed in order to evaluate the biological activity of the different isolates: activation of the transcription factors NF-kappaB and AP-1 as well as the anchorage independent growth of LMP1 transfected ratl cells in soft agar. The results suggested that whereas differences in the activation of NF-kappaB through the various LMP1 isolates correlated tightly with their different expression levels, the outgrowth of transfected cells in soft agar did not and the transcription factor NF-kappaB therefore appeared not to be the major effector for the transformation of the rodent cell line ratl by LMP1. The various LMP1-isolates also differed in their capacity in activating the transcription factor AP-1. We found no correlation between the transforming ability of the LMPI isolates and activation of AP-1 suggesting that other so far uncharacterized domains also influence the transforming ability of the protein.

A potential NES of the Epstein-Barr virus nuclear antigen 1 (EBNA1) does not confer shuttling.

The Epstein-Barr virus nuclear antigen 1 (EBNA1) is a multifunctional protein involved in the replication and maintenance of the viral episome. We identified a potential Rev-like nuclear export signal (NES) which, however, does not confer the export of EBNA1. In the yeast two-hybrid system EBNA1 does not bind to the nuclear exporter Crm1p. In spite of the RNA-binding ability of EBNA1 and its structural homologies to RNA binding proteins like hnRNP U and/or A1, EBNA1 does not shuttle to the cytoplasm in heterokaryon analysis. We propose the function of the RNA binding of EBNA1 in retaining RNAs to the nucleus.

High prevalence of antibodies against HERV-K10 in patients with testicular cancer but not with AIDS.

Human endogenous retrovirus K10 (HERV-K10) env and gag expression has been detected in placenta, embryonic tissue, and cell lines. By transfection, these sequences have been expressed in insect cells and developed into serological assays, revealing HERV-K10 antibodies in patients with testicular cancer. Patients with AIDS are at an increased risk for testicular cancer and frequently reactivate latent infections. We postulated that HERV-K10 seroprevalence might be increased with HIV infection or AIDS. Stored, frozen serum samples from 52 patients with testicular cancer (8 patients with HIV and 30 patients with samples near the time of diagnosis) and 84 controls (40 patients with HIV) were diluted 1:40 and tested by immunofluorescence against SF158 cells transfected with HERV-K10 env [ENV1.9(+)] or gag (pACGAG). Seroprevalence rates were compared cross-sectionally in cases and controls, excluding those with indeterminate results (3 of 30 cases and 7 of 84 controls), and also were examined longitudinally in the cases before or after diagnosis of testicular cancer. Seroprevalence to HERV-K10 Env or Gag was 17 of 27 testicular cancer patients (63%) around the time of diagnosis, compared to 4 of 77 controls (5%; P < 0.0001). Seroprevalence was similar (50% to 60%) with seminoma, teratocarcinoma, or embryonal carcinoma, and it was not increased with HIV infection in either cases (33%) or controls (3%). HERV-K10 antibodies were detected in 12 of 19 cases (63%) more than 6 months before seminoma diagnosis, as well as in four cases with residual or recurrent malignancy more than 1 month after initial diagnosis. Thus, HERV-K10 antibodies are detected frequently with testicular cancer and seem to resolve rapidly with effective therapy of the malignancy. Antibody reactivity also occurs in approximately 5% of controls, perhaps because of nonspecific or cross-reactive epitopes. HIV and AIDS were not associated with HERV-K10 antibodies, thus, leaving their higher risk of testicular cancer unexplained.

High prevalence of hepatitis G virus (HGV) infections in dialysis staff.

BACKGROUND: Patients on renal replacement therapy, haemodialysis (HD), or after kidney transplantation (TX), are known to be at risk of acquiring blood-borne infections (HBV, HCV). GBV-C/Hepatitis G virus (HGV) has been described recently and is considered to cause blood-borne infections. The aim of this study was to analyse the risk for the medical staff of HD and TX patients to acquire HGV infection. METHODS: Eighty-five HD patients and 86 TX recipients were compared with 49 health-care workers and 64 blood donors as controls. The HGV prevalence was determined by RT-PCR and antibodies to E2 protein. RESULTS: A high prevalence of HGV was found in the medical staff (24%) which nearly corresponded to the prevalence of the patients (TX 36%, HD 25%) but not to the controls (9%). In contrast, the prevalence of HCV was low in the medical staff (2%) and controls (0%) but high in HD (13%) and TX (13%). Age and duration of employment in the department did not significantly influence the HGV prevalence in staff. The number of viraemic subjects in staff was high, possibly indicating a more recent infection. CONCLUSION: An occupational risk for HGV exists in medical staff of dialysis and transplant patients. Further routes of transmission than only parenteral may play a role in this setting.

Human endogenous retrovirus (HERV)-K transcripts in gonadoblastomas and gonadoblastoma-derived germ cell tumours.

Gonadoblastomas are rare tumours of abnormal or dysgenetic gonads, often transforming to invasive seminomatous and nonseminomatous germ cell tumours (GCT). Because of the intimate association of noninvasive and invasive lesions, gonadoblastoma may provide clues as to the molecular pathogenesis of GCT. We studied the expression of the human endogenous retrovirus (HERV)-K gag gene in eight gonadoblastomas arising in phenotypically female patients, including two newborn girls. We also studied testicular biopsies with immature Sertoli cell nodules harbouring neoplastic germ cells, a lesion with morphological resemblance to gonadoblastoma. In five gonadoblastomas, invasive seminoma/dysgerminoma was noted, in two cases with formation of additional GCT components. HERV-K gag transcripts were found with moderate levels in gonocytes of all gonadoblastomas and in neoplastic germ cells in testicular Sertoli cell nodules. All invasive GCT except for teratomas displayed HERV-K transcripts. Thus, expression of HERV-K is induced during fetal or embryonal development and precedes invasive GCT formation. Although the specific role of HERV-K expression remains unknown, the findings place HERV-K expression in an appropriate time frame for it to have a role in the molecular pathogenesis of GCT and suggest a precursor-invasive tumour relationship for ovarian GCT equivalent to the more common carcinoma in situ of the testis and testicular GCT.

Effects of multiple sclerosis and other inflammatory CNS disorders on tick-borne encephalitis serology.

The goal of the present study was to investigate whether a direct association exists between false-positive recognition of IgG antibodies and inflammatory changes in the central nervous system (CNS) and whether inflammatory diseases of the CNS affect the specificity of the enzyme-linked immunosorbent assay (ELISA) of tick-borne encephalitis (TBE) virus. A group of patients (1,815), treated in the Department of Neurology, University Hospital of the Saarland, Homburg/Saar, Germany, were tested forTBE IgG antibodies by ELISA. Several subgroups of patients with and without inflammatory changes in the CSF as well as patients with and without confirmed multiple sclerosis (MS) were investigated. Overall, 4.5% of all the 1,815 patients and 4.8% of the patients with inflammatory changes in the CSF but without MS had TBE IgG antibodies. In the subgroup with inflammatory changes in the CSF and MS, 4.4% of the patients were TBE IgG-positive. In the subgroup without inflammatory changes in the CSF, 3.8% of the patients without MS were TBE IgG-positive and 4.9% of the patients with MS were TBE IgG-positive. The rate of TBE IgG positivity was not significantly different in the subgroups with and without inflammatory changes in the CSF (P = 0.45). The comparison of the subgroups with and without MS showed no significant difference in the TBE IgG titer (P = 0.83) as well. This indicates that the specificity of the ELISA was affected neither by inflammatory changes in the CSF nor by MS.