Chemicals, not viruses, more likely cause of catfish tumors

Regarding Mark Saltis’ letter, Tumors in bullheads: Environmental toxins or common virus?” (May 2006), we have the following response.

First, our report has never focused attention on PCBs as a possible cause of the tumors. There are no laboratory studies with PCBs that suggest that they initiate tumors in fish. Rather than being carcinogenic in fish, PCBs have inhibited experimental induction of liver cancer in rainbow trout by aflatoxin (Journal of Toxicology and Environmental Health 13: 649-657, 1984).

In fact, bullheads are often not very good accumulators of PCBs, especially in comparison to channel catfish. The Maryland Department of the Environment recently collected bullheads for contaminant analyses from the South River, but these data are not yet available.

Second, several studies have focused on a viral cause for bullhead tumors. In our view, this is still an open question. Using electron microscopy, Poulet and Spitsbergen (1996) did not detect viral particles in the skin tissue of 8 brown bullheads with tumors (papillomas). Poulet et al. (1993) was not successful in transmitting skin tumors between bullheads after inoculating tumor-free fish with cell-free filtered homogenates, live cells or a cell-line.

They did, however, identify RNA-dependent DNA polymerase activity but were not able to determine if it was cellular or viral in nature. To date, no bullhead retrovirus has been verified. Thus, it is incorrect to state, as Saltis does, “Although the virus has yet to be identified, these bullhead retroviruses result in papillomas that later develop into squamous carcinomas.”

The 1941 study was in brown bullhead not channel catfish as stated (Journal of Experimental Medicine 74:397-408 and 22 plates). The brown bullheads were from contaminated rivers within Philadelphia. Tumor cells were only transplanted to the immunoprotected anterior chamber of the eye and cornea. Transplanted tumor cells were observed by direct microscopic examination to grow and persist. These features are consistent with a chemical etiology. They neither confirm nor exclude a virus.

Viruses are often passengers in tumors without contributing to the tumorigenesis.

Viruses in transforming families do not necessarily produce tumors. For example, channel catfish herpes virus (CCHV) does not cause tumors in channel catfish and CCHV does not infect brown bullhead.

Retrovirus-associated epidermal skin proliferations or tumors in walleye, carp and other species are not invasive and have a plaque-like morphology in contrast to the masslike invasive orocutaneous squamous cell carcinoma of brown bullhead.

Third, in our report, we place equal or greater emphasis on the 20 percent prevalence of liver tumors than the 53 percent prevalence of skin tumors in the South River. There is no evidence for a viral cause of liver tumors in brown bullheads.

In the Great Lakes, Baumann (2002) suggested that liver tumor prevalence greater than about 5 percent and skin tumor prevalence of about 12 percent be used as criteria to distinguish between highly contaminated Areas of Concern and less contaminated Areas of Recovery.

Thus, both the liver and skin tumor prevalence in the South River were about four times Baumann’s suggested criteria.

Fourth, the comparison of tumor prevalence in brown bullheads with that in channel catfish has nothing to do with PCBs. It is well-known that brown bullheads are more likely to develop liver and skin tumors than channel catfish. Part of the difference may be due to differences in how the two species metabolize benzo(a)pyrene, a polynuclear aromatic hydrocarbon (PAH), into the carcinogenic intermediate (BaP-didhydrodiol) and in the degree to which DNA-adducts form. Willett et al. (2000) found that brown bullheads generate more of the carcinogenic metabolite than do channel catfish. Ploch et al. (1999) found that brown bullhead had higher concentrations of altered DNA (DNA-adducts) than did channel catfish after similar exposure to PAHs. DNA adducts are an early stage in the cancer process.

Fifth, bullheads are not a small, short-lived species but in fact may live as long as 18 years. Trautman (1981) in “Fishes of Ohio,” states that the largest specimen was 47.8 cm and weighed 1.8 kg.

Based on the established links between PAH exposure and the development of liver and skin tumors in bullheads, we are proposing to focus initially on chemical causes. Depending on the results of this research, we may, in future studies, include analyses for viral or other etiological agents.

Fred Pinkney
U.S. Fish and Wildlife Service
Chesapeake Bay Field Office
Annapolis, MD

John Harshbarger
George Washington University Medical Center
Washington, D.C.