An official website of the United States government.

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Developing Dose Response Information for Microbial and Chemical Risk Assessments


The long-term objectives of this research proposal are to develop dose response information for use in risk assessment of biological and chemical contaminants. These contaminants often occur in our food or water supply, and in many cases, information on the potential health impact to humans is lacking. For regulatory agencies such as USDA, FDA and EPA to make decisions that are protective of human health without needlessly increasing the cost or decreasing food supplies to the consumer, better dose-response models are needed. To understand how well animal and/or in vitro model systems estimate human risk from exposure to a hazardous agent, several pieces of information are needed. First, mechanistic studies are needed to understand how an agent causes pathogenicity or toxicity in the model system and whether the agent causes illness in humans using the same pathway. The better the understanding of the mechanism of pathogenicity or toxicity, the more accurate the estimates of risk. Second, studies are usually done in invertebrates or rodents to determine how an agent affects a whole animal. Third, studies need to be completed in animal models that biochemically and physiologically resemble humans. Fourth, all of these methods need to be compared to understand how accurately they predict risk to humans, and if they do not, why not.

To accomplish these objectives, the following specific aims are proposed: <OL> <LI> Investigate the mechanisms of how chemicals affect embryonic limb development using tissue from chick embryos to establish stem cell lines. Using cartilage-specific molecules to monitor normal cell differentiation, the effects of treatments with various toxicants will be tested for their effects on overall cell growth. Molecular techniques will be used to determine how agents affect programmed cell death (apoptosis) and whether chemical insult results in the same sequence of events as normal physiological apoptosis. <LI> Determine the virulence of six strains of L. monocytogenes in a mouse study. The information will be compared to an on-going primate study for similarities in virulence and whether the mouse study provides predictive dose-response information for primates and humans. <LI> Develop dose response models for L. monocytogenes using information collected from a FDA-sponsored animal feeding study. The data will be used to develop dose response information and subsequently to predict infectious dose to pregnant human. <LI> Compare the model systems for their contribution to the overall dose-response information needed to predict risk to humans.

More information

NON-TECHNICAL SUMMARY: Biological and chemical contaminants often occur in our food or water supply, and in many cases, information on the potential health impact to humans is lacking. The purpose of this project is to use tissue cultures and animal models to provide information for predicting human health impact from exposures to biological and chemical contaminants.


APPROACH: The purpose of this study is to use both in vitro and animal models to develop dose-response information for use in human health risk assessments. In the first series of experiments, in vitro methods will be used to examine the mechanism by which chemicals disrupt normal growth and development in stem cells from chick embryos. In the second set of experiments, a mouse bioassay will be used to determine virulence among several strains of L. monocytogenes which have previously been used in a primate model. Information from the primate model will be compared to the mouse model and both will be compared for their ability to predict illness in humans. In the first series of experiments, a micromass cell culture system will be used to investigate the mechanisms by which exogenous exposure to retinoic acid, a derivative of vitamin A, results in skeletal malformations. Because undifferentiated cells can be manipulated to differentiate into one particular cell type, the micromass tissue culture system provides an excellent method to study molecular and cellular events occurring after toxicant insult. Tissue cultures will be used to complete mechanistic studies using all-trans retinoic acid as a model compound because of previously published information on its relationship to apoptosis. Because human trials for L. monocytogenes are unethical due to the potential fatal outcome for susceptible individuals, the second and third series of experiments will use animal surrogates to develop data for extrapolation to humans. Development of a bioassay system is needed to bridge the gap between studies in laboratory rodents and primates. The mouse bioassay system would also help in comparing results from our current primate study to the mouse bioassay system. The results of this project will aid in extrapolation of data between species and development of a more accurate dose-response model for use in risk assessment for human listeriosis. A panel of six L. monocytogenes strains used for the an on-going animal surrogate study and from a hot dog isolate from humans will be tested for pathogenicity and infectivity in the mouse. Dose response data from the primate study is currently being obtained and the data collected will be used for quantitative risk assessment for humans. Previous studies in our laboratory established the pregnant monkey as very similar to humans in susceptibility and adverse pregnancy outcome after exposure to L. monocytogenes. The information gathered in the first three specific aims will be used to estimate risk to humans. The data will be examined for its contribution to human health risk assessment, and several dose response models will be used tested for their "goodness-of-fit" to the data. In previous studies, dose-response models were examined for their ability to predict infectious dose using several feeding studies for microbial contaminants. These same models will be tested using the data gathered from the three previous studies to fit dose-response information to predict human health outcomes.

Smith, Mary
University of Georgia
Start date
End date
Project number
Accession number