Abstract
Tox21 is a transformative effort on the part of several U.S. Federal agencies (the National Toxicology Program/National Institute of Environmental Health Sciences [NTP], the National Institutes of Health (NIH) Chemical Genomics Center/National Human Genome Research Institute [NCGC; now called the NIH Center for Translational Therapeutics, National Center for Advancing Translational Sciences] the Environmental Protection Agency's (EPA) National Center for Computational Toxicology, and recently the Food and Drug Administration) that are partnering to fundamentally change the science of safety toxicology. These agencies bring a comprehensive suite of capabilities and are working diligently together to develop, evaluate, and ultimately implement a new safety assessment paradigm. Toxicologic pathology has an important ongoing role in establishing the validity of this transformation, and may ultimately benefit as a discipline through an enhanced understanding of chemically induced disease mechanisms.
Keywords
The Tox21 program (see http://ntp.niehs.nih.gov/go/28213; http://www.epa.gov/ncct/toxcast/; and http://nctt.nih.gov/27543703) moves away from a primary reliance on experimental animal studies to utilization of state-of-the-art high-throughput screening (HTS) methods using cells and biochemical assays and in vivo studies using lower organisms (e.g., zebrafish embryos,
An additional front for advancing this effort is to develop the techniques that will allow for the routine utilization of induced pluripotent stem cells (iPS) or human embryonic stem cells (HuES) in toxicity testing. Optimally, human stem cells will be differentiated in culture under conditions that (1) retain their functional characteristics, including the capacity to metabolize xenobiotics, (2) represent life stages presumably susceptible to a wide range of toxicants, and (3) provide an opportunity to monitor chemically induced alterations in patterns of normal development.
Although rapid progress has been achieved with a wide range of required technologies, a similar paradigm shift will be needed in the way we view and interpret this new information stream with respect to how it can be best used in assessing the safety of chemicals and ultimately the protection of public health. This requires consideration of this new information as it relates to, or perhaps in the context of, our current animal-based assessments. A key difference is temporal. Tox21 cell-based systems measure immediate cellular responses and provide information on longer-term phenotypic outcomes only by inference. Significant efforts are underway to establish predictive links between immediate cellular responses detected using in vitro HTS systems to apical outcomes such as cancer, reproductive and developmental deficits, and other noncancer end points measured in the animal models that currently form the basis for our safety assessment studies.
My prediction is that the current efforts will likely prove unsatisfactory. Why? The simplest explanation is one of probabilities and is not really different from the issues we face in cross-species extrapolations of hazard and risk. Although the toxicologic pathology community has refined their art and science to a high degree, the human relevance of the “apical phenotypic outcome” in rodents is still dependent on how well the model system reflects human physiology and biochemistry and captures genetic diversity with respect to the agent under study. For example, recent unpublished National Toxicology Program/National Institute of Environmental Health Sciences (NTP) studies looking at the kinetics of benzene absorption, distribution, metabolism, and elimination across a range of mouse strains from different genetic backgrounds, coupled with toxicity studies using the new diversity outcross (J: DO) mouse model from the Jackson Laboratory (http://jaxmice.jax.org/strain/009376.html), are revealing a much larger than anticipated range of metabolic capabilities and interanimal susceptibility to toxicity. This perhaps should not be too surprising, as strain-specific sensitivities have been shown with many other agents through the years (Bradford et al. 2010; Harrill et al. 2009). Traditionally, the NTP and industry have relied by necessity on the same few rat and mouse strains with very limited genetic diversity for longer-term toxicity and carcinogenicity studies. These results potentially provide little insight into either the range of diverse responses to an agent across the human population or the immediate molecular responses to an agent in in vitro cell systems, especially considering the fact that Tox21 focuses on using human cells whenever possible.
The NIEHS has recently undertaken a planning exercise that generated a number of strategic goals. One is to identify and understand fundamental shared mechanisms or common biological pathways (e.g., inflammation, epigenetic changes, oxidative stress, mutagenesis, etc.) underlying a broad range of complex diseases, to enable the development of broadly applicable prevention and intervention strategies. Recognition that perturbation of common biological pathways and mechanisms underlie many diverse manifestations of toxicity and disease offers a new way to think about how we go about better identifying substances of human health concern.
But before this can happen, we need to do a better job at understanding the relationship between patterns of immediate biological response and phenotypic outcomes that result from these shared deleterious biological processes stemming from both acute and long-term in vivo exposures. The NTP has been amassing resources and tools to address these temporal questions. The NTP archives currently house materials from over 2,000 studies, including over 7.5 million slides, 4.6 million paraffin blocks, and over 74,000 frozen specimens. We currently have the capabilities to molecularly mine these materials through new techniques (e.g., laser capture microdissection), and by recent advances allowing formalin fixed, paraffin-embedded tissues to be used for genomic studies (Roberts et al. 2007; Bourzac et al. 2011).
The piece that has remained missing until very recently is the immediate in vivo response. The molecular profile of a tumor or other chronic disease phenotype may hold no discernable traces of the events that initiated that process. For these reasons, the NTP recently acquired, and is making publically available, DrugMatrix®, a rat toxicogenomic database (https://ntp.niehs.nih.gov/drugmatrix). This information resource was developed by Entelos and was used primarily by pharmaceutical companies to assist in drug development. Six hundred and thirty-seven unique chemicals, mostly drugs but also over 100 known toxicants, were given to Sprague Dawley rats in acute and repeated dose regimens, and gene expression profiles were generated from target organs, including liver and several other organs. This database is the logical starting point to draw associations between immediate genomic effects in vivo, with immediate effects in HTS studies, and long-term phenotypic and genomic outcomes.
So, where does this new paradigm leave the toxicologic pathologist? Will this paradigm shift make toxicologic pathologists obsolete? It is doubtful that this need cause immediate worry for pathologists as there are several key activities that will likely take many years to play out. One is an enhanced and intensive effort to correlate immediate and late phenotypic outcomes with molecular signatures in in vivo studies. This is the logical next step in advancing the classification systems for cancers and other noncancer pathologies and will deepen our understanding of the molecular events that both initiate and develop the phenotypic consequences of toxic insults or disease processes. Whether this can be best achieved through additional training for pathologists or through enhanced collaborations with “molecular” toxicologists and bioinformaticists remains to be seen. Second, there is a clear need to better understand the interaction of chemicals with chronic disease processes that accompany aging or other lifestyle “conditions” such as obesity-related type 2 diabetes, atherosclerosis, and cancer. Chemicals that modify these “normal” aging and disease processes make up an increasingly important segment of toxicology research, as the acute bad actors are either well understood or phased out of the human environment. Ultimately, the toxicologic pathologist can play an optimal role if his or her diagnoses of cancer and noncancer end points are well aligned with human disease pathology, and supported by a mechanistic understanding of common disease processes, such that they can be relied upon to help predict the molecular events responsible for the lesions observed.
Footnotes
*This is an opinion article submitted to the Regulatory Forum and does not constitute an official position of the Society of Toxicologic Pathology or the journal Toxicologic Pathology. The views expressed in this article are those of the author and do not necessarily represent the policies, positions, or opinions of his respective agencies and organizations. The Regulatory Forum is designed to stimulate broad discussion of topics relevant to regulatory issues in toxicologic pathology. Readers of
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The author received no financial support for the research, authorship, and/or publication of this article.