Volker Specke

Genetic alterations of the long terminal repeat of an ecotropic porcine endogenous retrovirus during passage in human cells

Human-tropic porcine endogenous retroviruses (PERV) such as PERV-A and PERV-B can infect human cells and are therefore a potential risk to recipients of xenotransplants. A similar risk is posed by recombinant viruses containing the receptor-binding site of PERV-A and large parts of the genome of the ecotropic PERV-C including its long terminal repeat (LTR). We describe here the unique organization of the PERV-C LTR and its changes during serial passage of recombinant virus in human cells. An increase in virus titer correlated with an increase in LTR length, caused by multiplication of 37-bp repeats containing nuclear factor Y binding sites. Luciferase dual reporter assays revealed a correlation between the number of repeats and the extent of expression. No alterations have been observed in the receptor-binding site, indicating that the increased titer is due to the changes in the LTR. These data indicate that recombinant PERVs generated during infection of human cells can adapt and subsequently replicate with greater efficiency.

Differences in release and determination of subtype of Porcine endogenous retroviruses produced by stimulated normal pig blood cells

Objective: Porcine endogenous retroviruses (PERVs) are of particular concern with xenotransplantations using pig cells, tissues or organs as they are present in the genome of all pig strains and are able to infect human cells in vitro. However, it remains unclear whether PERV particles will be produced in vivo and whether they may infect xenotransplant recipients. Since normal pig peripheral blood mononuclear cells (PBMCs) may be transmitted together with the transplanted organ, the production of PERVs by stimulated PBMCs was studied in vitro. Methods: To simulate antigen-induced activation of PBMCs, phytohaemagglutinin (PHA), a T cell mitogen, and the phorbol ester O-tetradecanoylphorbol-13-acetate (TPA), a tumour promoter, were used. Virus release was estimated by measuring reverse transcriptase (RT) activity and by RT-PCR of pelleted viruses. Results: Treatment of pig PBMCs with PHA or TPA induced the release of PERVs. For the first time, a correlation between the extent of proliferation of pig PBMCs and PERV production was shown. In addition, PERV release by non-proliferating cells and differences in virus production between stimuli as well as between different pig strains and individuals of one strain were observed. Subtype analysis revealed the release of the three subtypes PERV-A, PERV-B and PERV-C. In contrast to murine endogenous retroviruses, PERVs were induced by PHA alone. Conclusion: The data suggest that the PBMCs transmitted within a xenotransplant may release PERV. These data also suggest that pig strains producing low amounts of virus could be more suitable for xenotransplantation.

Virus safety in xenotransplantation: first exploratory in vivo studies in small laboratory animals and non-human primates.

For xenotransplantation, the transplantation of animal cells, tissues and organs into human recipients, to date, pigs are favored as potential donors. Beside ethical, immunological, physiological and technical problems, the microbiological safety of the xenograft has to be guaranteed. It will be possible to eliminate all of the known porcine microorgansims in the nearby future by vaccinating or specified pathogen-free breeding. Thus, the main risk will come from the porcine endogenous retroviruses (PERVs) which are present in the pig genome as proviruses of different subtypes. PERVs will therefore be transmitted, with the xenograft, to the human recipient. PERVs can infect numerous different types of human primary cells and cell lines in vitro and were shown to adapt to these cells by serial passaging on uninfected cells. Furthermore, PERVs have high homology to other retroviruses, such as feline leukemia virus (FeLV) or murine leukemia virus (MuLV), which are known to induce tumors or immunodeficiencies in the infected host. To evaluate the potential risk of a trans-species transmission of PERV in vivo, naive and immunosuppressed rats, guinea pigs and minks were inoculated with PERV and screened over a period of 3 months for an antibody reaction against PERV proteins or for the integration of proviral DNA into the genomic DNA of the hosts cells. Furthermore, we inoculated three different species of non-human primates, rhesus monkey (Macaca mulatta), pig-tailed monkey (Macaca nemestrina) and baboon (Papio hamadryas) with high titers of a human-adapted PERV. To simulate a situation in xenotransplantation, the animals received a daily triple immunosuppression using cyclosporine A, methylprednisolone and RAD, a rapamycin derivative, presently under development by Novartis. None of the small laboratory animals or the non-human primates showed production of antibodies against PERV or evidence of integration of proviral DNA in blood cells or cells of several organs, 3 months after virus inoculation, despite the observation that cells of the animals used in the experiment were infectible in vitro. This apparent difference in the outcome of the in vitro and the in vivo data might be explained by an efficient elimination of the virus by the innate or adaptive immunity of the animals.

Porcine endogenous retroviruses (PERVs): Generation of specific antibodies, development of an immunoperoxidase assay (IPA) and inhibition by AZT

Abstract: Xenotransplantation may be associated with the risk of transmission of microorganisms. In particular, the porcine endogenous retroviruses (PERV) have raised concerns as in vitro experiments show susceptibility of human cells for PERV infection. However, it remains unclear whether PERVs are able to infect transplant recipients in vivo and whether they are pathogenic. It is therefore essential that the risks are evaluated and for this purpose specific and sensitive screening methods for PERVs have to be developed. We generated specific antibodies against all major structural proteins of PERV and developed several assays which allow antibodies against PERV to be detected as indirect evidence of infection. For direct detection of PERV production, reverse transcriptase (RT) assays were used. PCR methods were used to detect provirus integration and the presence of viral mRNA. Here we present an immunoperoxidase assay (IPA), which would allow the detection of viral proteins in infected cells as well as antibodies against PERV in the serum of an infected host. The specificity of the sera used in the assay was determined by several methods, including immunoelectron microscopy, and the sensitivity of the assay was compared with other methods. This IPA was used to detect PERV infection in in vitro experiments for evaluation of the virus host range, for titrating the virus and for testing anti‐viral properties of AZT. Using this method it was shown that AZT inhibits replication of PERV. This IPA may be very useful for the surveillance of preclinical and clinical xenotransplantations.

Characterisation of a human cell-adapted porcine endogenous retrovirus PERV-A/C

BACKGROUND Porcine endogenous retroviruses (PERVs) pose a potential risk for xenotransplantation using pig cells, tissues or organs. A special threat comes from viruses generated by recombination between human-tropic PERV-A and ecotropic PERV-C. Serial passages of a recombinant PERV-A/C on human 293 cells resulted in increased infectious titers and a multimerization of transcription factor binding sites in the viral long terminal repeat (LTR). In contrast to the LTR, the sequence of the env gene did not change, indicating that the LTR represents the determinant of high infectivity. MATERIAL/METHODS The virus was further propagated on human cells and characterized by different methods (titration, sequencing, infection experiments, electron microscopy). RESULTS Further propagation on human 293 cells resulted in deletions and mutations in the LTR. In contrast to low-titer viruses, the high-titer virus was infectious for cells from non-human primates including chimpanzees. Scanning electron microscopy revealed clustering of budding virions at the cell surface of infected human cells and transmission electron microscopy indicated that the virus infects them via receptor-mediated endocytosis. CONCLUSIONS After propagation of PERV on human cells without selection pressure, viruses with different LTR were generated. High titer PERV was shown to infect cells from non-human primates. The experiments performed here simulate the situation in vivo and give an extended characterization of human cell-adapted PERVs.

Productive infection of a mink cell line with porcine endogenous retroviruses (PERVs) but lack of transmission to minks in vivo

Summary. Porcine endogenous retroviruses (PERVs) are considered a special risk for xenotransplantation because they are an integral part of the porcine genome and are able to infect cells of numerous species including humans in vitro. Among these cells, the mink lung epithelial cell line Mv1Lu could be productively infected with PERV. Provirus integration was detected by PCR, expression of viral proteins was shown by immunostaining and reverse transcriptase was detected in cell supernatants. PERV produced from mink cells could infect both, uninfected mink Mv1Lu cells and uninfected human 293 cells, with considerably higher virus production by human cells. Typical type C retroviruses were observed in PERV-infected mink cells using electron microscopy together with numerous multivesicular body (MVB)-like structures containing small virus-like particles, not present in uninfected mink cells. These MVBs could be stained with PERV-specific serum. In an attempt to establish a small animal model, PERV grown on mink cells was inoculated into adult and newborn American minks. Neither antibody production against PERV nor integration of viral DNA or production of viral proteins in tissues of different organs could be detected 12 weeks post virus inoculation, indicating that PERV infection had not occurred.

No in vivo infection of triple immunosuppressed non-human primates after inoculation with high titers of porcine endogenous retroviruses.

Abstract: Background:  Porcine endogenous retroviruses (PERVs) released from pig tissue can infect selected human cells in vitro and therefore represent a safety risk for xenotransplantation using pig cells, tissues, or organs. Although PERVs infect cells of numerous species in vitro, attempts to establish reliable animal models failed until now. Absence of PERV transmission has been shown in first experimental and clinical xenotransplantations; however, these trials suffered from the absence of long‐term exposure (transplant survival) and profound immunosuppression.

Porcine Endogenous Retroviruses PERV-A and PERV-B Infect neither Mouse Cells in vitro nor SCID Mice in vivo

Objective: Porcine endogenous retroviruses (PERVs) pose a risk for xenotransplantations using pig materials as they are present in the genome of all pigs and are able to infect human cells in vitro. Until recently, transmission of PERVs in vivo was only described in severe combined immunodeficient (SCID) and nude mice inoculated with PERV-producing cells. However, in this series of experiments microchimerism could not be excluded. To overcome this problem, the risk of PERV infection was addressed in a similar way but using cell-free inoculation of mouse cells in vitro and SCID mice in vivo. Methods: Mouse cell lines and primary cells were incubated in vitro with PERV-A, with a recombinant PERV-A/C and with PERV-B. Provirus integration was assessed by PCR. Reverse transcriptase activity was measured in the cell supernatants. SCID mice were inoculated in vivo with cell-free virus at high titers. Results: None of the mouse cell lines and primary cells could be infected by PERV and no provirus integration was observed in different organs of the inoculated SCID mice. Conclusion: The data indicate that PERV-A, PERV-A/C and PERV-B could not infect different mouse cells. These data correlate with the recent finding that mouse cells lack a functional receptor for PERV-A. Although the receptor for PERV-B is still unknown, these data suggest that previously reported PERV transmissions to SCID and nude mice in vivo might be due to microchimerism or pseudotyping with murine viruses and indicate that normal mice are an inappropriate model for the study of PERV infection and pathogenesis.