Greater than 90% of all individuals have been exposed to EBV
, and this exposure usually occurs relatively early in life before the development of most breast cancers. Based on this fact, it might be logical to conclude that EBV could never cause or exacerbate breast cancers since almost all patients with these tumors have been previously exposed to the virus. The error in this logic lies in the fact that EBV infections, and the accompanying viral burden, can vary dramatically between individuals
[24, 41–45]. It has been suggested that more recent and sensitive methods of EBV quantification need to be utilized as a prognostic tool for EBV-associated diseases
[41, 42, 44]. Stated simply, even though most individuals have been exposed to EBV, each person can have very different latent and replicating viral burdens. Because of this difference, it has been postulated that some individuals will demonstrate significant immune dysregulation in their struggle to keep such latent viral infections under control
[46–48]. If true, it is easy to imagine how EBV might be capable of exacerbating diseases in some individuals and not others. Such a relationship has been suggested for the autoimmune diseases, multiple sclerosis
[23, 24], systemic lupus erythematosus
[24, 49], and rheumatoid arthritis
[24, 50], among others.
Investigating the relationship between EBV infection and breast cancer has been the subject of numerous investigations
[2–5]. Some studies have demonstrated the presence of EBV genes or proteins in breast cancer biopsies
[9–14], while other studies have not detected the virus
[15–20]. One criticism of studies that have detected EBV genes is the low level of such expression, which questions the small percentage of tumor cells which might harbor virus and whether virus in non-tumor cells can be excluded. Some studies have correlated antibody reactivity against EBV antigens with the development of breast cancers, suggesting a recent or ongoing immune response against this viral infection. Other studies have suggested that there is a relationship between a subset of very aggressive breast cancers and the presence of EBV
. Whether the presence of this virus has a direct contribution, an indirect contribution, or no contribution to the etiology of breast cancers in a subset of patients remains unclear
Here we investigated the possibility that the presence of a latent gammaherpesvirus might exacerbate disease in a transplantable breast cancer using mouse models. Combining HV-68 latency
[26–28] with mice developing 4 T1 mammary tumors
 allowed a direct analysis of whether this gammaherpesvirus could infect breast carcinoma cells in vivo, and whether disease was exacerbated when compared to uninfected animals. While primary tumor growth did not vary between these groups (Figure
2B), the health of mice infected with HV-68 began to decline dramatically during days 30 to 44 post 4 T1 transplantation (Figure
1). When HV-68 infected mice began to succumb at day 44 (Figure
1), the remaining mice were euthanized for analyses. It then became apparent why HV-68 infected mice were losing weight and were becoming moribund as metastatic lesions in the lungs and secondary tumors were accumulating in these animals (Figures
4). Compared to uninfected mice, we concluded that harboring latent HV-68 resulted in an exacerbation of metastatic disease during developing 4 T1 mammary tumors. Increased expression of pan-cytokeratin (Figure
5) and VEGF-A (Figure
6) was consistent with this conclusion.
The availability of sensitive assays to detect lytic and latent HV-68 allowed us to question whether primary or secondary tumor tissue contained virus. Tissue homogenates from tumors did not harbor latent or infectious viral particles (data not shown) when cultured on permissive cell lines
[29, 30, 38]. Neither the presence of a representative lytic transcript (ORF 65) nor a representative latent transcript (K3) could be detected in any tumor tissue (Figure
7). This RT-PCR assay can detect less than one viral transcript in one microgram of total RNA
[29, 39, 40]. In addition, viral DNA could not be detected (data not shown). Therefore it is highly unlikely that HV-68 or HV-68 infected cells are present in tumor tissues at a level that could explain the increased metastatic disease that was observed. These results do not support the notion that EBV exists within breast cancer cells themselves as a mechanism for exacerbated disease.
The most likely explanation for HV-68 exacerbated metastatic disease is an indirect one. Gammaherpesviruses, like EBV and HV-68, establish latency in B lymphocytes and other antigen presenting cells
[1, 26]. Reactivation from latency permits lytic virus to infect a variety of cells while stimulating an immune response that limits viral replication in immunocompetent individuals
[1, 26]. Whether factors expressed by virally infected cells, or the immune response that controls viral replication, or the immune dysregulation that permits the virus to persist, are responsible for HV-68 induced metastases is not presently clear. Such indirect mechanisms for EBV-exacerbated breast cancer disease in patients have not been investigated. Unfortunately, such definitive studies will be difficult to perform in patients. However the present studies suggest a useful model for investigating mechanisms responsible for gammaherpesvirus-exacerbated cancer.
One possible outcome for demonstrating gammaherpesvirus-exacerbated disease is the notion that anti-viral therapies would then be one co-treatment option for some breast cancer patients. Surprisingly, there is not much compelling data to suggest the presence of replicating EBV in studies using breast cancer biopsies
[2–5]. Most viral RNAs and proteins which have been detected represent those expressed by latent virus. If replicative virus is not present during the course of developing breast cancer, most anti-viral therapies would not be effective
. Furthermore, there have been no studies to address the ability of developing breast cancer to induce EBV reactivation from latency in vivo. Using this mouse model of HV-68 infection, it will be possible to directly address such questions. In particular, it will be important to determine whether developing breast cancer can induce gammaherpesvirus reactivation and at what times during metastatic disease that replicative virus can be detected. If such future studies demonstrate viral reactivation during tumorigenesis, then the possibility that anti-herpesvirus therapies might limit virus-induced exacerbation of transplantable or spontaneous breast cancers could also be explored. Finally, the mechanisms responsible for HV-68 induced exacerbation of metastatic disease must be defined. While such studies would be difficult to perform in patients, the ability to use HV-68 as a model of gammaherpesvirus latency and reactivation will allow definitive studies to be performed.