How Biotech Laboratory In Vitro Study Advances Modeling of Bioaccumulation

February 21, 2020

Written By

Sue Cielinski

Mathematical modeling is used to describe the biotransformation of a chemical substance. At the biotech laboratory at KJ Scientific we are using state of the science in vitro technology to determine metabolism and predict bioaccumulation for your chemical analysis needs while also improving model uncertainty for more trusted outcomes.

“Modeling brings people of so many disciplines together – biology, chemistry, geography – it’s a hub of information explaining the dynamics that are at work in the environment.” -Liisa Toose, Modeler, Research Associate of Arnot Research and Consulting Inc.

“Sometimes you can’t just rely on your yourself to explain a model, you need to talk and collaborate in an interdisciplinary scientific environment, which is always evolving and thus requires an open mindset.” – Ester Papa, Ph.D., Associate Professor, QSAR Research Unit – University of Insubria, Varese, Italy.

For discovery and safety of new chemicals scientists conduct ADME (absorption, distribution, metabolism and elimination) studies to assess the effects of exposure of a chemical to an organism. The fate of a chemical in the body is determined by how much it is absorbed; where it is distributed; how it is metabolized, through what mechanism; and, where it is eliminated. How much, how fast, and how extensive the chemical is processed in the body are important too. These studies are performed in vivo (in whole living animal) and in vitro (“in the glass” or outside the organism). A couple of common examples for in vitro metabolism study include the use of whole liver cryopreserved hepatocytes and liver sub-cellular fractions (e.g. S9 fractions, microsomes).

Models describe and predict the behavior of a toxic substance in an animal body, for example, what parts (compartments) of the body a chemical may tend to enter (e.g. fat, liver, spleen, etc…) They also estimate whether or not the chemical is expected to be metabolized or excreted and at what rate. These models also simulate the processes that lead to toxicity at the level of the organism over time. For this purpose, in vivo data are used and models are created to predict the effects of synthetic or natural chemical substances in various tissues and organs in animal species and in humans. The models strive to be mechanistic and automatic by mathematically translating the anatomical, physiological, physical, and chemical phenomena involved in the complex ADME processes. The models can be predictive but also interpreted by using statistical tools for more complex models.

Theoretically, these models depend on the anatomical and physiological structure of the body, and to a certain extent, on biochemistry. They are usually multi-compartment models, with compartments corresponding to identified organs or tissues, with interconnections corresponding to blood or lymph flows. A system of differential equations (defining a relationship between a physical quantity and rate of change) can be created for concentration or quantity of substance in each compartment. The changing parameters are represented by blood flows, organ volumes, pulmonary ventilation rate, etc… This information is applied to assess substance-specific health risks and carried out by national and international agencies responsible for regulating health and safety.

Models Used to Predict Bioaccumulation of Chemicals

The assessment of bioaccumulation is important in evaluating risks that chemicals may pose to humans and the environment. It also is a large focus concerning government regulation. Chemicals that accumulate may produce high levels in the body that could lead to toxicity. Therefore, substances with high potential to bioaccumulate are of serious concern especially if they are persistent and difficult to remove in the environment and very toxic to humans and animals. However, up until recently, methods to predict bioaccumulation were not available. Today more progress is being made on this front.

Physiological parameters are measured based on in vitro and in silico (by computer modeling or computer simulation) study. These studies determine bioaccumulation and the potential of distributing or impeding uptake of chemicals in tissues and organs, especially in the liver. These physiological processes are described using mathematical models, and these models can be used to predict tissue concentration over time, predicting adverse effects. Key relevant parameters that are often derived from models are absorption rate and extent, and clearance (the ability to remove a substance from an organ). The models include the kinetics of metabolism by the liver that is capable of biotransforming the chemical compound.

Several different factors are used to assess the potential for the substance to bioaccumulate. Certain studies determine these factors and are known as the bioconcentration factor (BCF), bioaccumulation factor (BAF) and the biomagnification factor (BMF). To simplify, the BCF is derived from a test with fish or fish tissue comparing the concentration of the substance that accumulates with the concentration in the water. The BMF is found from using a test with fish and the concentration of the substance obtained from consuming their prey. The BAF considers uptake by both respiration and diet and compares concentration in the organism to that in the water in field studies.

Internationally, the regulation of assessing bioaccumulation of chemicals is mainly based on the BCF in aquatic species. BCFs can be determined by a laboratory test under controlled conditions according to the Organisation for Economic Co-operation and Development Test Guideline – OECD TG 305 and other Federal agency laboratory testing guidelines. Under these test conditions the BCF is determined either under steady conditions comparing the amount of the substance in the fish to the water concentration or when measuring the active rate of uptake and elimination of the substance. BCFs are generally agreed to be sufficient to determine the bioaccumulative properties of a chemical. However, some scientists suggest other factors such as the selected test species used, and how the BCF is experimentally calculated may influence BCF values.

Using in vitro trout liver S9 sub-cellular fraction and the cryopreserved hepatocyte assay KJ Scientific’s environmental biotech laboratory analyzes metabolism of chemicals and predicts bioaccumulation of chemicals. KJ Scientific uses the established standardized OECD TGs (TG 319A and 319B) for measuring in vitro biotransformation rates using S9 and hepatocyte assay systems from rainbow trout liver tissue. Matter of fact, KJ Scientific worked with companies and organizations, in collaboration with SC Johnson, DuPont, Procter and Gamble, Dow Chemical, U.S. Environmental Protection Agency (EPA), Fraunhofer Institute and Givaudan, and with top scientists (in a global Ring Trial) to standardize and validate the process of in vitro metabolism (of a test chemical) using fish liver cells or sub-cellular fractions (S9). This validation process resulted in the creation of two OECD test guidelines using rainbow trout cryopreserved hepatocytes (OECD TG 319A) and liver S9 sub-cellular fractions (OECD 319B). In order to determine chemical uptake in the liver cell (in vitro hepatic or intrinsic clearance) the maximum rate of metabolism is determined using a mathematical equation (slope of the metabolic rate or parent compound depletion) rather than making a direct measurement of the products of metabolism.

Use of In Vitro Intrinsic Clearance to Predict BCF

The in vitro intrinsic clearance rates determined with rainbow trout hepatocytes and/or liver S9 fraction are extrapolated to estimate the rate of metabolism of a whole-body fish (in vivo) using an in vitro to in vivo extrapolation (IVIVE) model. In order for this metabolic rate to be meaningful concerning chemical exposure, other areas of chemical uptake/elimination through gills, feces and diet also need to be calculated. IVIVE, a mass-balance model, incorporates these parameters and is used to evaluate the total fate of the exposure of the chemical to the organism.

For an aquatic animal like a fish, the BCF equals the concentration of the chemical in the fish liver divided by the concentration of the chemical in the surrounding water. The metabolic rate is then used as an input to established mass-balance models for rainbow trout to predict the BCF. The BCF is an important measurement that regulators use to decide if a chemical goes to market or not. Around the world, regulators have different standards that substances must meet. A biotech laboratory should be familiar with the regulatory standards of your industry to help you throughout the testing process.

For a test chemical, at our biotech laboratory at KJ Scientific, we determine the in vitro hepatic metabolism (intrinsic clearance) with the hepatic metabolism assay (OECD TG 319A and 319B) using in vitro liver S9 subcellular preparation and cryopreserved hepatocytes from rainbow trout and other species such as regulatory species (bluegill, carp, fathead minnow and largemouth bass). As part of our process we measure the hepatic metabolic rate over a specified period of time taking into account the properties of the test chemical such as solubility (Kow). Other data, such as the fish’s body weight from which the extracted cells (S9 and/or hepatocytes) are obtained, the protein concentration for the S9 fraction/or cell count for the hepatocytes, reaction rate and incubation temperature are used to predict the BCF.

Some uncertainty does exist in both the IVIVE and BCF models. The models and underlying assumptions that present these uncertainties continue to be evaluated in related on-going research. As of note, predicting BCFs based on in vitro data is especially desirable when compared to in vivo testing since it reduces significantly the number of animals used for testing, and both cost and duration of the experiments. By providing high quality metabolism and BCF predictions our biotech lab at KJ Scientific is also providing data to improve the accuracy of the models that use this data as model input. The quality of the results or output of a model is only at most as good as the overall quality of the input data that is provided. This is critical and cannot be dismissed. As Prof. Ester Papa points out – “The data quality and quantity used in models is very important, you need to be confident about the quality data provided by a reference lab or curate the data quality well if it originates from other sources.”

In Silico Models

In contrast to mathematical models, different types of in silico (non-testing) methods have been developed to also characterize and predict toxic outcomes in humans and the environment. Analogous to the phrases in vivo and in vitro, in silico, is an expression used to signify “performed on computer or via computer simulation.” Therefore, in silico toxicology in its broadest sense means “anything that we can do with a computer in toxicology.” These methods are already used for regulatory purposes and it is anticipated that their role will be much more prominent in the near future.

Computational toxicology is an area of very active development and great potential. It is difficult to define exactly, as today practically all toxicological research and risk assessment have major in silico components. In 2003, the U.S. EPA defines in silico toxicology as the “integration of modern computing and information technology with molecular biology to improve agency prioritization of data requirements and risk assessment of chemicals”.

In silico toxicology differs from traditional toxicology in many ways, but perhaps the most important is that of scale. Scale, in the numbers of chemicals that are studied, breadth of endpoints and pathways covered, levels of biological organization examined, and range of exposure conditions considered all at once. The pharmaceutical industry developed a major portion of in silico technology for use in drug discovery.

Several different kinds of in silico methods have been developed and applied in academia and the pharmaceutical industry to model pharmacokinetic and toxicological hypotheses. These in silico methods include databases, different kinds of quantitative structure-activity relationship (QSAR) methods, molecular modeling approaches, machine learning, data mining, network analysis tools, and data analysis tools using computers.

To further explain, QSAR is a technique that tries to predict the activity, reactivity, and properties of an unknown chemical based on analysis of an equation connecting the structures of molecules with how they behave. Specifically, biological activity can be described quantitatively as the concentration of a substance required to give a certain biological response. QSAR models can also predict the activities of new chemicals.

QSARs can correlate molecular and biological activity of chemical biotransformation relevant to bioaccumulation. Recent research has led to development of a screening-level QSAR model for predicting biotransformation rates based on chemical structure by comparing bioaccumulation of hundreds of chemicals and dietary absorption bioaccumulation data to estimate in vivo biotransformation. This type of information is integrated into a QSAR model that then predicts biotransformation rates based on chemical structure essentially for all types of chemicals. The EPA is already implementing this QSAR in bioaccumulation models.

Despite this progress, biotransformation remains one of the greatest uncertainties in the prediction of bioaccumulation of chemicals in fish. The biotransformation rate data collected by the biotech laboratory at KJ Scientific from in vitro metabolizing systems using fish hepatocytes or liver S9 sub-cellular fractions can be used to refine in silico bioaccumulation prediction models. A model that includes more high-quality data, then improves the quality of the model. This allows the “data to tell for itself” instead of relying on assumptions and weak correlations. By providing data, especially, high-quality data, the biotech laboratory at KJ Scientific is improving the accuracy of these models.

KJ Scientific therefore is a reliable partner for industry, professional modelers, and regulators by providing data and trust worthy BCF predictions for chemical products. Our BCF predictions are arrived from our data obtained through state of the science in vitro S9 fraction and liver hepatocyte metabolism studies. KJ Scientific is sharing BCF values with expert model developers therefore adding a value toward improving bioaccumulation modeling quality and reliability. In order for these models to become better at making predictions more information on biotransformation for more chemicals needs to be integrated into the models. By using in vitro study to collect metabolism data from liver tissue, S9 fraction and hepatocytes, the biotech laboratory at KJ Scientific can also accomplish data collection for metabolism including BCFs, relatively easily and accurately without use of whole animal testing. KJ Scientific encourages other biotech laboratories, universities and industry also to share their data with modelers.

Modeling Partnerships with KJ Scientific

KJ Scientific’s biotech laboratory is partnering with Dr. Jon Arnot, President and Principal Scientist at Arnot Research and Consulting Inc. (ARC). ARC has developed a unique model with various stakeholders to integrate and evaluate BCF and other lines of evidence to assess chemicals for bioaccumulation concerning aquatic and terrestrial organisms. This model is called the Bioaccumulation Assessment Tool (BAT). The BAT especially deals with the quality of the data on metabolism to achieve better decisions when extrapolating in vivo outcomes when using in vitro input data. The BAT is downloadable on the ARC website and provided to interested parties at no cost.

Liisa Toose, a Research Associate of ARC and collaborator with KJ Scientific, modestly describes herself as a programmer with background in geography and fate modeling. She works closely with Dr. Jon Arnot to make sense of the state-of-the-science and notes that Jon says to her that “BAT is your baby.” She says her work in modeling makes predictions about chemicals by “synthesizing data to break down biological and chemical processes in order to quantify what is most important to estimate uptake in the liver, gut and gills.” Liisa states that in vitro data, like what KJ Scientific provides, “is so great because there is a lot of it and it is versatile; you can look at the livers from all types of organisms which provide many pieces of the puzzle.” She also stresses that personal communication and relationships with other scientists are important in her work and she sees the value “to get input, make the connections and find the synergies.”

KJ Scientific produces quality data, which provides good opportunity to collaborate with Prof. Ester Papa, of the QSAR Research Unit in Environmental Chemistry and Ecotoxicology at University of Insubria in Varese, Italy. The QSAR Unit’s research focuses on the development of QSAR models to predict the environmental effects and properties of chemicals of concern, such as organic environmental pollutants and pharmaceuticals. Their main goal is to generate robust models and to validate QSAR predictions for the use of screening and prioritization of hazardous compounds. The QSAR Research Unit is developing free tools (i.e., QSARINS and QSARINS-Chem) dedicated to the development, validation and application of models based on QSARs, and support risk assessment of existing and not yet synthesized chemicals.

Prof. Ester Papa, who works with QSARs since 2000, articulates passionately, “I am fascinated with QSAR modeling, it’s a special niche in the environmental field, it is interesting though not really something popular and sometimes not very well understood, which is a good reason to keep developing and improving this exciting field of research.” She asserts some of the main benefits of the QSAR approach are it, supports risk assessment of chemicals, is an alternative to and reduces animal testing, and saves cost by reducing the amount of testing. Prof. Papa stresses, “It is very helpful when we can provide support for regulators and scientists who don’t have experience with QSAR, since these models are sometimes complicated.” She further emphasizes “Unfortunately, some users don’t understand the background behind QSAR models, it’s not just press-a-button and a result will roll out!” In order for regulators and industry to have confidence in models she stresses the availability of the five OECD principles for the valid use of QSAR for regulatory purposes. In essence, models need to be clear on what is predicted, transparent and reproducible, have a reasonable and logical application, show a good fit of the data, as well as provide an explanation of how they work. She encourages, “be open to approaches that other people have, understand their point of view so you can see how to make better models.”

So foremost, the high-quality data provided by the biotech laboratory at KJ Scientific can facilitate the growth in reliability of your current and future model outcomes. KJ Scientific uses in vitro metabolism using liver S9 fraction and hepatocytes to provide your companies and researchers with accurate data on metabolism and bioaccumulation to meet your professional needs. KJ Scientific offers this collaboration to modelers and other scientists to assist predicting bioaccumulation of metabolized chemicals.

KJ Scientific is the World's Most Innovative Biotech Lab

Discover More About Our Products and Services by Requesting a Webinar

Learn More