New Horizons in Toxicity Prediction.
Lhasa Limited Symposium Event in Collaboration with
the University of Cambridge - February 2009
A Report by Wendy A. Warr, wendy@warr.com, http://www.warr.com Application of in silico modeling
in guiding alternatives research in skin allergy
Cameron MacKay, Unilever
Assuring the safety of consumer products without the need to conduct
animal tests is a considerable challenge. The mouse local lymph node
assay (LLNA) is now used widely to generate data for assessing the risk
of chemical-induced skin sensitization, but changes in EU legislation
(in the seventh amendment to the EU Cosmetics Directive) have made
developing non-animal approaches to provide the data for skin
sensitization risk assessment a key business need.
Skin sensitization is a complicated multistage process and a single in
vitro assay system is not feasible at present. It is
difficult to know
where to target assays and how to interpret a battery of disjoint
assays. Unilever decided to try applying mathematical models, or
“systems biology”. The purpose was to explain and
elucidate
mechanisms, not to predict sensitization a priori; this is not a QSAR
model. In collaboration with Entelos, Unilever has developed a
large-scale in silico
model of skin sensitization induction (comprising
nonlinear ordinary differential equations) to characterize and quantify
the contribution of implicated pathways to the overall biological
response. Such knowledge is crucial in guiding the development of in
vitro assay development for use in consumer safety risk
assessment.
The model describes the developing immune response in mice over a 7-day
period following exposure to dinitrochlorobenzene (a well known contact
allergen) and includes both epidermal and lymph node cellular processes
implicated in skin sensitization such as cytokine responses, cell
surface marker regulation, cellular migration and proliferation. In
order to populate the model, in
vivo and in
vitro data from the
published literature were used. Some of the data, such as epidermal
cytokine release in response to chemical insult, were used to build
focused submodels of the biology. Cellular data from mouse LLNA were
used in order to ensure that, acting together, these submodels could
model the full system response effectively.
The modeling uncovered a previously underappreciated pathway in skin
sensitization and showed it to be key to the sensitization response.
Additionally, the modeling revealed a number of gaps in both the
current mechanistic knowledge and the available data. Unilever is using
the model to focus and guide its future research in the area of skin
sensitization. The session chairman pointed out the importance of using
a model to develop an in
vitro assay.
Challenges
in predicting metabolism and toxicity with known and abstract targets
Fred Guengerich, Vanderbilt University School of Medicine
The costs of drug development and environmental risk assessment
continue to increase, and the availability of better in silico and in
vitro methods has the potential to yield better and more
economical
predictions. Metabolism issues are key to some toxicities, but an
accurate assessment of the fraction of all drug and other toxicities
due to metabolism is unavailable and, even when metabolism is agreed to
be central, the ensuing biological events are not well understood.
A need exists for better biomarkers and assays to predict not only
bioactivation but also off-target toxicity, immune-related toxicity,
and idiosyncratic reactions.32 It is hard to
predict toxicity because
of lack of understanding of mechanisms. Few protein target structures
are available and there is limited information about linear pathways
and networks, so SAR has to be done with gross endpoints. The issue of
relevance of the parameters used in comparisons is critical in judging
the usefulness of the analyzer. Although much has been learned about
the enzymes involved in the metabolism of many drugs and other
xenobiotics, predictions are not trivial. We do now have crystal
structures for the main five cytochrome P450s. Actual protein
structures are much preferable to homology models, but even when these
are available they often do not predict products accurately.
Boyer and co-workers have studied biotransformation in early drug
discovery. Their SPORCalc system9 is described
above. Ligand-based
methods for in-house screening can be used when complete P450-specific
data are unavailable; SPORCalc compares favorably with docking methods
at picking the top three sites of oxidation. Unfortunately, it is very
hard to predict rates a priori, i.e., to study how fast the compound
will be metabolized.
MacDonald and co-workers have published a strategy for identifying
off-target effects and hidden phenotypes of drugs by directly probing
biochemical pathways that underlie therapeutic or toxic mechanisms.33
Iconix Biosciences (now part of Entelos)34 has
an in vivo
predictive
toxicogenomics paradigm. In an idealized system, principle component
analysis (PCA) could be applied to data from transcriptonomics,
metabolomics and other disciplines.
Guengerich closed by summarizing three more big problems. Is the animal
model relevant to humans for the toxicity issue? Is the in vitro system
relevant to the in vivo
system? Many cell lines lack critical features,
e.g., bioactivation. Finally, what about dose? This is a problem in in
vitro work. What is the human exposure?
