Jump to main content.

In Vitro to In Vivo exposure and effects


This task will provide extrapolation tools, methods, and models needed to utilize data from medium- and high-throughput in vitro screening assays in both human health and ecological risk assessments. Although dose-response relationships are generally considered in the context of whole-organism responses, the principles that underlie these relationships also apply to in vitro testing efforts. Thus, challenges associated with extrapolating in vitro data to the intact organism generally relate to uncertainty regarding toxicologically relevant chemical concentrations in vitro and vivo (dosing uncertainty) or the extent to which effects observed in vitro reflect in vivo effects occurring at the same effective dose (effects uncertainty). Research will address both of these fundamental sources of extrapolation uncertainty. Additional research will be conducted to relate measured contaminant levels in water (e.g., municipal effluents) to in vitro and in vivo bioassay results, thereby linking in vitro, in vivo, and environmental exposures.

Rationale and Research Approach:

A significant challenge when performing in vitro-to-in vivo effects extrapolations is the need to relate the effective in vitro dose to that achieved at the site of action in the intact organism. Currently, in vitro effects information is generally related to a measured or nominal concentration. Typically, the unbound chemical concentration provides a stronger basis for extrapolation, since this is the concentration that is “free” to exert adverse effects. The unbound fraction that exists within an in vitro system is a function of the components of that system and may differ substantially from that in vivo. This discrepancy creates a need to measure the unbound concentration and relate this to the measured or predicted unbound concentration in vivo. We will develop novel methods to measure and control  unbound chemical concentrations in vitro, and then develop a framework to relate  this information to the intact organism. As part of this effort,  we will  use passive sampling methods (Solid Phase Microextraction/SPME)to measure unbound chemical concentrations in commonly employed in vitro metabolizing systems (S9, hepatocytes). Existing algorithms that predict in vivo equilibrium blood and tissue partitioning from inherent chemical properties (e.g., log KOW) will then be used to create the in vitro-in vivo dosimetry linkage. We will develop controlled in vitro dosing methods  by using a passive partitioning phase (typically PDMS-silicon) as the source of chemical for in vitro bioassays. Once perfected, this approach could be used to conduct in vitro experiments at chemical concentrations measured in vivo or predicted from pharmacokinetic modeling efforts.

Uncertainty regarding extrapolation of in vitro effects to the intact organism derives from the incomplete nature of such assays with respect to the presence and activities of relevant biochemical processes, and their ability to represent complex responses of the intact organism. Research is needed to determine the strengths and limitations of in vitro assays so that the resulting data can be confidently related to an expected in vivo outcome. This issue is particularly relevant to the development of  in vitro assays for chemicals that have adverse impacts on the nervous system. Neurophysiological assessments will be performed across three levels of biological complexity--output of neurons in culture, output of brain slices with a maintained neural circuitry, and output in brain recordings from intact animals--to assess impact of chemicals on nervous system function. The ability of chemically-induced alterations in neurophysiology of cell-based assays to predict outcomes measures in more complex systems will be determined. We will select pesticides and other test chemicals  based on mode-of-action and toxicity pathways in consultation with National Center for Computational Toxicology (NCCT) and proposed Systems Biology Projects. Initial experiments will optimize procedures, parameters, and endpoint selection across preparations to allow direct linkages between in vitro and in vivo effects. Once established, we will address application of these models to assess developmental neurotoxicity and tissue dosimetry.

We will perform additional research to address the issue of potential exposure of non-target species to endocrine active pharmaceuticals (EAPs) using an approach that relies on both effects and exposure extrapolation. A need exists to develop methods for measurement of the EAPs to which humans and wildlife are exposed. Novel methods, including the use of passive sampling devices, will be developed to measure EAPs in effluents and receiving waters. Additional research is designed to translate measured exposures into predictions of chemical uptake and accumulation. Of special concern is the role of sex hormone binding globulins (SHBG) in promoting uptake and retention of specific endocrine-active compounds by non-target species (e.g., fish). Although SHBGs are expressed by all vertebrates, their binding properties vary among species. Different species may also differ with respect to their ability to metabolize these compounds. In vitro and in vivo work will be conducted to characterize these species differences. Exposure data will be linked to predicted molecular targets for groups of EAPs (e.g., cellular and biochemical endpoints) and adverse outcomes (e.g., fish reproduction) using in vitro and in vivo approaches. This information will then be used to predict likely adverse outcomes on non-target species based on their environmental exposure and the presence/absence of relevant molecular targets. Finally, we will use in vitro assays, suitable for the evaluation of an EAP alone, or in mixtures,  to evaluate effects of environmentally relevant mixtures of EAPs. These products will provide tools for extrapolation of both effects and exposure information for a high priority class of emerging chemical contaminants.

MED Scientists:

Patrick Fitzsimmons
John Nichols
Alex Hoffman
Kellie Fay


Escher, B.I., C.E. Cowan-Ellsberry, S. Dyer, M.R. Embry , S. Erhardt, M. Halder, J.-H. Kwon, K. Johanning, M.T. Oosterwijk, S. Rutishauser, H. Segner, and J. Nichols. 2011. Protein and lipid binding parameters in rainbow trout (Oncorhynchus mykiss) blood and liver fractions to extrapolate from an in vitro metabolic degradation assay to in vivo bioaccumulation potential of hydrophobic organic chemicals. Chemical Research in Toxicology 24:1134-1143.

Expected Products:




Sep 30, 2013

(2) Novel methods to measure unbound chemical concentrations in vitro and a framework to reduce uncertainty when extrapolating from in vitro to relevant in vivo exposures.

John Nichols

Sep 30, 2014 (5) Guidance document on the use of passive dosing and sampling methods to perform in vitro metabolism assays and extrapolate in vitro to in vivo effects information based on measured unbound chemical concentrations. John Nichols
Sep 30, 2014 (4) Toolbox of analytical chemistry methods that identify and quantify EAPs plus major metabolites in aqueous samples (waste water, surface waters) and plasma. Sampling strategies (e.g, grab samples versus polar organic chemical integrated samplers (POCIS)) for EAPs will be included. John Nichols

Top of page

Research and Development Home || NHEERL Home || Grosse Ile Home

Local Navigation

Jump to main content.