Autism is a highly heritable behavioral disorder. Yet, two decades of genetic investigation have unveiled extremely few cases that can be solely explained on the basis of de novo mutations or cytogenetic abnormalities. Vertical viral transmission represents a nongenetic mechanism of disease compatible with high parent-to-offspring transmission and with low rates of disease-specific genetic abnormalities. Vertically transmitted viruses should be found more frequently in the affected tissues of autistic individuals compared to controls. Our initial step was thus to assess by nested polymerase chain reaction (PCR) and DNA sequence analysis the presence of cytomegalovirus (CMV), Epstein-Barr virus (EBV), herpes simplex virus type 1 (HSV1), herpes simplex virus type 2 (HSV2), human herpes virus 6 (HHV6), BK virus (BKV), JC virus (JCV), and simian virus 40 (SV40) in genomic DNA extracted from postmortem temporocortical tissue (Brodmann areas 41/42) belonging to 15 autistic patients and 13 controls. BKV, JCV, and SV40 combined are significantly more frequent among autistic patients compared to controls (67% versus 23%, respectively; P<.05). The majority of positives yielded archetypal sequences, whereas six patients and two controls unveiled single-base pair changes in two or more sequenced clones. No association is present with the remaining viruses, which are found in relatively few individuals (N

SV40 is a simian polyomavirus that was a contaminant of some viral vaccines administered to people between 1955 and 1962. SV40 DNA has recently been found associated with several types of human tumors, suggesting that the virus is present in humans. We examined sera from patients infected with human immunodeficiency virus type 1 (HIV-1) as well as from HIV-1-negative controls to determine the prevalence of SV40 neutralizing antibodies using a specific plaque reduction assay. We found that 16.1% of HIV-infected patients (n = 236) were seropositive for SV40, as compared to 12.0% of HIV-negative control volunteers (n = 108) and 11.1% of HIV-negative patients (n = 72). These differences were not statistically significant. As individuals born between 1941 and 1962 had the highest chance of having received SV40-contaminated poliovaccines, we analyzed SV40 seropositivity rates based on year of birth. SV40 antibody rates for HIV-infected patients born before 1941, between 1941 and 1962, and after 1962 were 17.1%, 16.3%, and 11.8%, respectively. For the HIV-negative subjects, the rates were 12.5%, 12.0%, and 9.7%, respectively. There was no correlation between SV40 seropositivity and either the stage of disease in HIV-infected patients or the race/ethnicity. Also, there was no correlation between the presence of SV40 neutralizing antibody and the titer of neutralizing antibody to human polyomavirus BKV. The SV40 seropositivity rates in the patients born between 1941 and 1962 may be explained by the likelihood of those individuals having received SV40-contaminated vaccines, but the detection of SV40 neutralizing antibody in individuals born after 1962 (with no risk of having received contaminated vaccines) is significant. Although cross-reactive antibodies might theoretically contribute to the observed reactivities, these results suggest that SV40 neutralizing antibodies are present in certain individuals and raise the possibility that SV40 continues to infect humans long after vaccines were freed from contamination.

Some polio vaccines prepared from 1954 to 1961 were contaminated with infectious SV40. It has been assumed that all polio vaccines were SV40 free in the United States after 1961 and in other countries after 1962. Following a WHO requirement that was prompted by the detection of SV40 in some human tumors, we conducted a multilaboratory study to test for SV40 polio vaccines prepared after 1961. Vaccine samples from 13 countries and the WHO seed were initially tested by PCR. The possible presence of intact and/or infectious SV40 DNA in PCR-positive samples was tested by transfection and infection of permissive CV-1 cells. All results were verified by immunohistochemistry, cloning, and sequencing. All the vaccines were SV40 free, except for vaccines from a major eastern European manufacturer that contained infectious SV40. We determined that the procedure used by this manufacturer to inactivate SV40 in oral poliovirus vaccine seed stocks based on heat inactivation in the presence of MgCl2 did not completely inactivate SV40. These SV40-contaminated vaccines were produced from early 1960s to about 1978 and were used throughout the world. Our findings underscore the potential risks of using primary monkey cells for preparing poliovirus vaccines, because of the possible contamination with SV40 or other monkey viruses, and emphasize the importance of using well-characterized cell substrates that are free from adventitious agents. Moreover, our results indicate possible geographic differences in SV40 exposure and offer a possible explanation for the different percentage of SV40-positive tumors detected in some laboratories.