November Discussion

George L. Wolff GWOLFF at FDANT.NCTR.FDA.GOV
Thu Nov 9 12:45:23 EST 1995


1) A major problem with using Sprague-Dawley rats in risk assessment 
assays is that they are an outbred stock. They are genetically 
heterogeneous so that selection for "healthy", i.e. larger, animals 
as breeding stock may account for the great increase in body weight 
observed over the years. Inadvertent inbreeding in each breeder's 
colony would differentiate that colony genetically from those of 
other breeders. Therefore, the responses to xenobiotics could be 
expected to differ in assays using SD rats from different colonies. 
In my opinion, this is a serious consideration, more so than the body 
weight increase, with regard to the use of genetically heterogeneous 
rodents such as SD rats for risk assessment assays.

2) Most men and women are not likely to reduce their caloric intake 
voluntarily by 25-40% over a lifetime. Therefore, the use of 
calorically-restricted rodents for determining the carcinogenic 
potential of suspect substances is not necessarily more valid than 
the use of ad lib fed animals. Use of calorically-restricted 
animals may underestimate the risk to humans, while ad lib fed 
rodents may provide a more appropriate estimate of this risk because 
of the prevalence of obesity in many human populations.

3) For carcinogenicity/toxicity risk assessment assays I have long 
proposed using inbred or F1 hybrid genomes in which different 
levels of susceptibility to neoplasia are provided by differential 
expression of a single gene. The background strain genome can be 
chosen for desired tissue-specific sensitivities, while the gene 
enhances that sensitivity to different degrees in genetically 
identical obese and lean phenotypes. The gene is dominant and has 
a coat color marker so that it can be easily maintained by forced 
heterozygosis in combination with a recessive allele. The latter has 
no effect on the animal's responsiveness to carcinogens. Thus, a 
genetically homogeneous assay system can include three easily 
distinguishable phenotypes, each with a different degree of 
response to the test agent, e.g. maximum, some, none. If anyone 
is interested in further details. I would be pleased to provide such 
as well as bibliographic references.

George L. Wolff














Tel: (501) 543-7522
FAX: (501) 543-7635/7662
NCTR       HFT-140    
3900 NCTR Road
Jefferson, AR 72079-9502





More information about the Toxicol mailing list