Faculty Bibliography

Historically, striped bass Morone saxatilis were indigenous to many major drainages of the Gulf of Mexico (Gulf). It is believed that almost all natural populations were depleted by the 1950s and 1960s with the exception of fish in the Apalachicola-Chattahoochee-Flint (A-C-F) river system in northwestern Florida, Georgia, and Alabama. Striped bass of Atlantic ancestry were introduced into the A-C-F system during the 1960s and 1970s to enhance population size. We compared DNA fingerprints of striped bass collected from four Atlantic river systems with those of fish from the A-C-F system to determine if genetic differences still exist. Moderate levels of polymorphism were observed with two probes, the bacteriophage M-13 genome and mouse sequences related to the Drosophila Per gene. Striped bass DNA digested with single restriction enzymes and hybridized to these two probes generated single DNA fragments shared by 71 out of 75 A-C-F fish but not seen in any of 51 Atlantic fish. Heritability of DNA fingerprints was demonstrated from hatchery-raised fish of known parents. We believe that descendants of a genetically distinct ancestral Gulf population of striped bass still exist in the A-C-F and that efforts to maintain the genetic integrity of this population are warranted. Screening of potential A-C-F hatchery brood stock may be used to both maintain the genetic integrity of the Gulf strain and maximize its genetic diversity

Messenger RNA and genomic DNA from Atlantic tomcod (Microgadus tomcod) from the cancer prone Hudson River population and two rivers in Maine were screened for size polymorphisms in cytochrome P-450IA. In northern blot analysis, we found that approximately 10% (16 of 176) of Hudson tomcod exhibited a polymorphism in P-450IA hybridizable mRNA not seen in any tomcod from Maine. All tomcod displayed a 3.0 kb P-450IA mRNA while variant individuals exhibited an additional 2.2 kb P- 450IA mRNA band. Southern blot analysis on these same fish confirmed the genetic basis of this polymorphism. All individuals exhibited a single P-450IA hybridizable DNA fragment while tomcod with the variant genotype also displayed a second smaller (< 1 kb) DNA fragment. The results of multiple restriction enzyme digests on these DNA's suggest that this polymorphism represents a second smaller P-450IA allele in these fish. This is the first demonstration of multiple forms of the cytochrome P-450IA gene in any species of fish

It has been reported that Atlantic tomcod (Microgadus tomcod) from the Hudson River exhibit an extremely high incidence of liver tumors. More than 90% of spawning 2-year-old fish display hepatocellular carcinomas. In contrast, representatives of this species from a relatively pristine environment show a much lower incidence of tumors. Genomic DNA and mitochondrial DNA (mtDNA) were isolated from tomcod from the Hudson River, New York, and the Saco River and Royal River, Maine. We found a statistically significant difference in the frequency of PstI-generated restriction fragment length polymorphisms in the abl cellular oncogene between Hudson and Maine tomcod. Allelic variation was observed at two of the three abl domains scored. A single composite genotype seen in approximately 40% of Hudson River fish was seen in only one Maine fish. This polymorphism enabled us to differentiate a Hudson River population from that encountered in the Maine rivers. This is the first demonstration of a population-specific polymorphism at a cellular oncogene locus in any species. In contrast, no restriction site polymorphisms were seen in mtDNA between the populations. The lack of mtDNA diversity in these fish is consistent with the geological history of the area. In combination, these results suggest that the genetic diversity observed at the c-abl oncogene locus must have been a fairly recent event and that oncogene loci may be particularly sensitive to mutational change.

The frequency of Ha-ras mutations was determined as a function of neoplastic progression in cell lines derived from rat tracheal implants exposed in vivo to 7,12-dimethylbenz[a]anthracene. Restriction fragment-length polymorphism (RFLP) analysis revealed an A----T transversion in the second base of codon 61 in 2 of 11 cell lines. One of the positive cell lines was tumorigenic, but the other was neither tumorigenic nor anchorage independent, thus indicating a lack of correlation between neoplastic stage and ras mutation. Densitometry analysis of the RFLP bands indicated that approximately 50% of the cells within these two heterogeneous populations contained the mutation. Direct sequence analysis of polymerase chain reaction-amplified DNA confirmed these results and did not reveal any other mutations in this region of the Ha-ras gene

Adult Atlantic tomcod collected from the Hudson River slightly north of New York City have an extremely high incidence (55-90%) of histologically defined hepatocellular carcinomas, whereas tomcod from control sites in Maine or Rhode Island exhibit little evidence of this condition. Genomic DNA was isolated from Hudson tomcod tumors and from normal Hudson and Saco River, Maine tomcod livers and tested for transforming activity in the NIH3T3 transfection assay. Six out of nine tumors (66%) tested proved positive. Southern blot analysis of all primary (6/6) transfectant and nude mouse tumor DNAs revealed evidence of an exogenous tomcod K-ras gene, while no activation of the H-ras gene was observed. These studies demonstrate that an outbred population of fishes and inbred mammals suffer genetic alternations at the same oncogene loci and suggest that similar pathways to neoplasia may be operative in both systems. Oncogene activation in naturally exposed feral populations may prove a particularly sensitive marker of environmental degradation in aquatic systems

Rat nasal squamous cell carcinomas induced by inhalation of three direct-acting alkylating agents yielded DNA containing activated oncogenes with no homology to any member of the ras family. The novel NIH 3T3 transforming oncogenes from tumors induced by beta-propiolactone and methylmethane sulfonate are distinct from each other based on restriction analysis. The gene isolated from beta-propiolactone-induced tumors is between 6 and 9 kb in size. None of the tumors induced by dimethylcarbamyl chloride contained positive DNA in the NIH 3T3 focus assay or in the nude mouse cotransfection assay. The rat nasal tumor model is apparently ideally suited for analysis of the roles of carcinogen and tissue specificity in oncogene activation, especially related to novel (non-ras) transforming oncogenes