Here, we assessed an experimental vaccine candidate for its efficacy to eliminate autologous HERV-K positive cells in vivo. We chose a vaccine candidate known to elicit strong T cell responses in vivo. The vaccine candidate is based on the modified vaccinia virus Ankara (MVA), a highly attenuated vaccinia virus strain with a high safety profile suitable for clinical application in immunosuppressed patients
[16, 17]. MVA does not replicate in human cells but shows high protein expression and, consequently, has a very good safety profile without compromising vaccination efficiency. As shown before, MVA vaccination generates antigen specific cellular immunity that is believed to be responsible to clear antigen expressing cells
Vaccine efficacy testing of the recombinant MVA-HKcon was performed in a surrogate mouse model, using a syngeneic mouse tumor cell line which was genetically engineered to express HERV-K Gag. Since HERV-K expression is normally repressed in humans, it is expected that humans are not tolerant to HERV-K gene products and so our simplistic animal model might be appropriate as well. Vaccinations were performed in a therapeutic setting after the establishment of tumor nodules in the mice testing two tumor application routes. Subcutaneous tumor growth and the formation of pulmonary metastasis in the mice after i.v. application were significantly reduced. These data indicate a therapeutic effect of the HERV-K GAG-directed vaccination.
Two therapeutic applications can be envisioned for this vaccine candidate. First, as described before for a HERV-K Env expressing MVA, MVA-HKcon could be used as a tumor vaccine
. However, in addition HERV-K Gag might be used as a surrogate target to develop an HIV vaccine. Despite significant progress and several clinical trials, a safe and effective AIDS vaccine is still elusive. The virus evolved unique ways of evading the immune system, and the human body seems to be incapable of mounting an effective immune response against it. The retrovirus HIV is hypervariable due to its high mutation rate and the ability to recombine; for vaccine development this means aiming at a moving target. Recently, several reports indicate that HERV-K is activated by HIV infection (see review by van der Kuyl
 and cell-mediated immune responses directed against HERV-K are effective in lowering HIV-1 viral loads and correlated with control of HIV-1 viremia
[14, 15, 19]. In addition, antibodies directed against HERV-K were detected in 70 to 80% of patients with HIV viremia. In contrast, only 2% of normal healthy controls were tested anti-HERV-K positive
. Then again the absence of HERV-specific antibodies in AIDS patients has been reported as well
A direct link between HERV-K activation and HIV infection was suggested recently
. Gonzalez-Hernandez and colleagues reported that addition of recombinant HIV-1 Tat protein to Jurkat cells caused a 13-fold increase in HERV-K gag RNA transcripts and a 10-fold increase in treated primary lymphocytes
. However, the correlation between HIV infection and HERV-K expression is still very controversial. Clear data showing HERV-K expression in HIV-infected cells from patients are not yet available and expression of HERV-K in CD8+ cells cannot be excluded
. In addition, HERV-K pol expression has been described after herpes virus, hepatitis B and C virus infections
, indicating a more general mechanism of activation. Quantification of HERV-K RNA in HIV-infected patients under HAART showed a correlation of HERV-K and HIV viral load, however an effect of HAART on HERV-K cannot be excluded
. In general, no significant correlation has been described between HIV viral load, CD4+ T cell counts and HERV-K protein titers
In contrast, Garrison et al. demonstrated the stimulation of HERV-specific T cell responses in HIV-positive participants by ELISPOT detection
. Interestingly, HERV-specific T-cell responses inversely correlated with HIV-1 plasma viral load
. Another study showed that HIV infected individuals (Long-term nonprogressors (LTNPs) or elite controllers), who control HIV-1 viremia without highly active antiretroviral therapy (HAART) had stronger and broader HERV-specific T cell responses than HAART-suppressed patients, virologic noncontrollers, immunologic progressors, and uninfected controls. In addition, the magnitude of the anti-HERV T cell response inversely correlated with HIV-1 viral load and associated with higher CD4 T cell counts in untreated patients. This suggests a beneficial effect of anti-HERV immunity in the control of chronic HIV-1 infection
. Moreover, a HERV-K–specific CD8+ T cell clone has been described to be able to eliminate cells infected with a panel of globally diverse HIV-1, HIV-2, and SIV isolates in vitro, indicating that HERV-K–specific T cell responses might be involved in the control of HIV-1 infections
Although still discussed controversially, a lot of data indicate that HIV-1 infection leads to the expression of otherwise repressed HERV-K and consequently to the stimulation of HERV-specific immune responses which in turns might help to control the HIV infection. However, still basic issues need to be clarified. Are HERV-K epitopes expressed significantly and exclusively in HIV-infected cells; T cells as well as monocytes. Another open issue is the next step towards clinical development. Side effects of an HERV-K-specific vaccination can only be studied in humans.
Recent observations of HERV-K expression in embryonic and induced pluripotent stem cells
 might also be of concern, however embryonic stem cells are located in an immune protected tissue and should not be recognized by HERV-K-specific immune responses. However, potential hazardous HERK-K expression in hematopoietic stem cells still needs to be analyzed. Primates encode very similar endogenous retroviral genomes (SERV) and could be used as a model for safety studies
[28–31]. Vaccination of rhesus macaques which carry SERV-K with SERV-K Gag or Env induced T cell responses without vaccine-related pathogenicity
HERV-K Env expression has also been observed in diverse types of human tumors. We recently showed proof of principle that vaccination directed against the HERV-K envelope protein had anti-tumor activity
. A HERV-K Gag directed vaccine might therefore be used as a tumor vaccine and might have in addition an application as a HIV vaccine.