PERVs present quite a different situation and level of concern since they are inherited as part of the host genome and, therefore, cannot be removed easily from donor animals. All pigs contain multiple (around 50) PERV proviruses in their genome, at least several of which encode infectious virus. PERVs are gammaretroviruses, closely related to MLV, that can be classified into three subtypes, A, B, and C, based on their envelope gene sequences (Takeuchi et al., 1998). Subtypes A and B can infect many types of human cells in culture. Subtype C is much less infectious for humans. Most breeds of pig carry proviruses capable of yielding infectious virus of all three subtypes. Although most pigs carry about the same number of proviruses in their DNA, there is considerable diversity in location, implying that their insertion into the genome must have occurred rather recently (on an evolutionary time scale). Based on extensive experience with related endogenous proviruses of mice, it is highly likely that the majority of proviruses contain some sort of genetic defect, and that only a small number are responsible for release of infectious virus. Taken together with the polymorphism in the presence or absence of specific proviruses, it might well be possible to breed animals lacking infectious proviruses for use as xenotransplant donors.
PERVs have not yet been shown to cause disease (or even viremia) in pigs or any other species in which they have been detected. Nor has their presence been detected (by polymerase chain reaction, PCR, or serology) in more than 150 human recipients of pig cells or tissues (Paradis et al., 1999), although a low level of infection of recipient cells can be observed in immunodeficient mice transplanted with porcine islets of Langerhans (Van der Laan et al., 2000). Nevertheless, given the release of viruses infectious to human cells by many types of pig cells; the close similarity of these viruses to viruses known to cause cancer, immunodeficiency, and other diseases in mice and cats; the well-known adaptability and variability of retroviruses; and the example of the rapid worldwide spread of HIV and AIDS, there is serious concern that the novel association between pig and human tissues might create novel evolutionary opportunities for the virus, leading to the appearance of a new pathogen. Although such a pathogen could have serious long-term adverse consequences for the transplant recipient, this issue is not an area of concern since it is far outweighed by the potential benefits of the transplant. The real issue of concern is that the xenotransplant setting might provide the opportunity for the virus to evolve into a pathogen that also could be transmitted from one individual to another efficiently enough to create a new epidemic disease.
Such an evolutionary pathway would require a series of events, each increasingly improbable, as indicated by the scale shown in Table 3.4 (J. P. Stoye, 2001). As implied by the table, it is virtually certain that many cells in the transplant would express infectious PERV following transplantation, and it