A Small Dose of Toxicology: The Health Effects of Common Chemicals

Free download. Book file PDF easily for everyone and every device. You can download and read online A Small Dose of Toxicology: The Health Effects of Common Chemicals file PDF Book only if you are registered here. And also you can download or read online all Book PDF file that related with A Small Dose of Toxicology: The Health Effects of Common Chemicals book. Happy reading A Small Dose of Toxicology: The Health Effects of Common Chemicals Bookeveryone. Download file Free Book PDF A Small Dose of Toxicology: The Health Effects of Common Chemicals at Complete PDF Library. This Book have some digital formats such us :paperbook, ebook, kindle, epub, fb2 and another formats. Here is The CompletePDF Book Library. It's free to register here to get Book file PDF A Small Dose of Toxicology: The Health Effects of Common Chemicals Pocket Guide.

Using the nominal concentration as a PoD dose metric will lead to an underestimate of the health risk of the chemical, a bias that is undesirable and should be avoided by all means. Moreover, in addition to dose, chemical toxicity is also a function of exposure time. In clinical settings, drug toxicity can correlate with accumulative dose such as the case of doxorubicin-induced congestive heart failure Thus combination of nominal concentration and exposure duration has also been used to more comprehensively characterize PoD dose metrics To address the concern of in vitro dosimetry selection, Groothius et al.

Real-world chemical exposures are complex, leading to dynamical tissue dosimetry. For accidental exposures, the internal chemical concentrations can rise acutely in a short period of time.

  • Materials Handbook!
  • Low Doses of Hormonelike Chemicals May Have Big Effects.
  • Health Effects of Toxic Chemicals.
  • Six-Word Lessons For Project Managers: 100 Lessons to Make You a Better Project Manager (The Six-Word Lessons Series).
  • Symptoms of Pesticide Poisoning.

For chronic environmental exposures, depending on the exposure scenario and clearance characteristics, the internal chemical concentrations can be episodic or relatively constant within a long period of time albeit possibly changing slowly. For clinical applications, drugs are administrated using a variety of regimens, resulting in rapidly fluctuating plasma and tissue concentrations within minutes, hours, and days.

Simple in vitro assays cannot easily duplicate either of the above scenarios. Microfluidic technology such as organ-on-a-chip devices can have a more controlled delivery of the test chemical to the cultured cells or organoids to better mimic clinical applications and environmental exposures, but currently it is challenging to scale up in a high-throughput, economical manner to cover the chemical space and toxicity pathway space to be tested To bridge the data gap between in vitro and in vivo kinetics and correct for the resulting difference in the PoD chemical dose metrics, computational toxicity pathway models and likely in vitro kinetic models are needed Figure 1.

The former is to mathematically reconstruct the relevant biochemical circuits operating in the cells as a dynamical system by using coupled differential equations These differential equations mechanistically describe the KEs in the toxicity pathway of the chemical's AOP, including the interactions and regulations between key biochemical nodes across different omic scales. In a dynamical pathway model, model parameters typically include transcription and translation rates, mRNA and protein half-lives, binding affinities for ligand-protein, protein-protein, and protein-DNA interactions, apparent Hill coefficient for ultrasensitive responses, and Michaelis-Menten constants for enzymatic reactions 23 , 38 — A useful pathway model will mechanistically capture the perturbation dynamics of the underlying biochemical circuits from the MIE through downstream signaling events.

For in vitro testing where organoids or 3-D tissues are used, computational virtual tissue models capturing cell-to-cell interactions that give rise to tissue structures will be needed, especially for understanding developmental processes 41 , These virtual tissue models normally use an agent-based modeling approach where the behavior of each agent, representing a cell, can be described by predetermined rules or through mechanistically-based toxicity pathway modeling as described above Figure 1. Schematic illustration of the workflow of computational approaches supporting dose-response modeling and in vivo extrapolation based on in vitro testing data.

To calibrate these toxicity pathway models, the in vitro kinetic information of the test chemical is needed as model's input. Therefore the chemical concentration, particularly the free concentrations, in the culture media and cells during the course of the in vitro experiment should be determined whenever possible. As high-throughput non-depletive methods of quantifying chemicals in miniaturized assays improve, determining chemical concentration kinetics is becoming increasingly practical In some cases, an in vitro kinetic model describing the chemical concentration changes in the culture medium and cells over time can be constructed Figure 1.

Uncertainty Factors

Research efforts have been under way to construct in vitro kinetic models for validating and predicting chemical concentration variations in widely used cell lines 30 , 45 — The Virtual Cell Based Assay VCBA project in Joint Research Centre in European Union aims to predict time-dependent concentrations of a test chemical both in the culture medium and within the cells by using ordinary differential equations that incorporate the physicochemical properties of the chemical and metabolic and cellular properties of the cell lines.

The determined free concentrations in the culture medium can be different from the nominal concentrations by several orders of magnitude, emphasizing the importance of characterizing chemical distribution in in vitro assays 46 , These in vitro kinetic models can be used to predict cultural and intracellular chemical concentration changes over time for initial chemical dose or repeated dose not experimentally tested.

The in vitro kinetics predicted or actually measured will then be used as input to calibrate the toxicity pathway or virtual tissue models.

Language selection

The model output will be the dynamics of the key biochemical nodes and other cellular metrics identified collectively as the PoD biomarker set, many of which are experimentally measured with the in vitro assays. Model parameters are calibrated through minimizing the difference between the model output and in vitro assay data obtained at different time points for different initial concentrations of the chemical tested. This calibration is essentially a training process for the dynamical toxicity pathway model. An ultimate goal of constructing the in vitro kinetic and toxicity pathway models and calibrating them based on data from the in vitro assay protocols is to help identify the true in vivo PoD chemical concentrations or other relevant dose metrics in target tissues under realistic or anticipated human exposure scenarios Figure 1.

For certain environmental exposure settings where the exposure is chronic and the chemical clearance is slow so its plasma and tissue concentrations are relatively constant, static chemical concentrations can be virtually applied to drive the calibrated computational toxicity pathway models to reach or surpass the predetermined PoD to predict the in vivo chemical concentration in the target tissue that would lead to an adverse outcome. For clinical applications or other environmental exposure settings where the chemical tissue concentrations are episodic and constantly changing, dynamic chemical concentrations mimicking these conditions, likely aided by toxicokinetic models, will be applied to the computational toxicity pathway models to predict the dosing scenario that would result in the predetermined toxicity PoD in the model To determine appropriate in vitro PoDs in the new toxicity testing framework, identifying key, experimentally measureable biochemical and cellular events is a critical step.

The choices of in vitro PoD have been so far arbitrary and many of which are chosen for convenience due to conventional usage. For example, AC 50 , and IC 50 are routinely used to evaluate chemical effects on signaling events such as receptor binding, reporter gene expression, enzyme inhibition, as well as ATP depletion and cell viability. While using these biomarkers are valuable in ranking the potency of chemicals and prioritizing them for further testing, their usage as in vitro PoD to infer adverse health outcome is questionable due to lack of quantitative correlations.

In certain cases, the benchmark-dose BMD approach or its variants are used to define in vitro PoD This approach considers the shape of the entire in vitro concentration-response curve and variability. Moreover, changes in many of the cellular PoD metrics are not only a function of chemical concentrations, but also a function of the duration of chemical treatment. An obvious example is the cell viability assay, where incubation with a chemical for a longer time generally results in more cell deaths This time-dependency generally leads to a left-ward shifting of the concentration-viability curve, and as a result, a different PoD concentration regardless of using IC 50 or BMD For many transcriptionally-mediated cellular stress responses, induction of the stress genes peaks at various times depending on the genes and chemical concentration, thus concentration-response data obtained at a single time point can barely represent the overall response profile 39 , Therefore, it is unrealistic to define a PoD solely based on a single cellular metric obtained at a single time point.

Rather, PoDs should contain dynamic information, i. Defining the scope and degree of cellular perturbations that culminate in adverse outcomes is a challenging task requiring basic knowledge on biological robustness and homeostasis, such as how cells handle stress to continuously function and survive in the face of adversity Adaptation is a salient feature of biological systems and is often framed as a determinant of PoD when the adaptive capacity is reached 7. It is commonly thought that if cells have adapted to the stress posed by a chemical and as a result restored cellular homeostasis, the chemical at the tested concentration can be considered safe.

This view however poses many questions about the relevance of using a PoD so derived in human risk assessments. The immediate question is what aspect of the cellular state should be looked at to determine whether and how well the cells have adapted. Is it the general cellular heath state, stress pathway status, cell growth or the cell's specialized function? If cells have adapted well with the general cellular heath state as measured by biomarkers such as ROS, ATP, viability, and LDH largely indistinguishable from unstressed cells, are their specialized functions equally preserved or nonetheless compromised 56?

Shah et al.

What Is The Deadliest Substance On Earth? Toxicity Comparison

In the study, cells with trajectories that move irreversibly away from the basal state were deemed to exceed a tipping point while those with recovering trajectories are still within or returning to the bound of the normal state. It has been shown that epigenetic changes can occur after recovering from stress, which can be inherited through several cell generations to transcriptionally affect future stress response, as demonstrated in C.

It is unclear whether this post-stress epigenetic memory is a general phenomenon. Transcriptional induction of suites of stress genes is the hallmark of cellular stress responses to chemical insults and has been proposed as a general method for chemical surveillance and in vitro screening A typical stress pathway Figure 2A involves a sensor molecule to detect the cellular state change, a transcription factor that can be specifically activated by the senor molecule or by the state change directly, and a battery of stress proteins which are induced transcriptionally and participate in reactions that restore the perturbed cellular state, such as altered ROS, DNA damage, ATP levels, to homeostasis These canonical stress response pathways are organized primarily as negative feedback loops and have been constructed into mathematical pathway models to understand their dose-response behaviors 60 — With an integral control or proportional control with high loop gains, the cellular state maintained by the feedback loops can exhibit threshold phenomena in response to chemical perturbations, where the threshold point corresponds to the stress intensity that causes the maxing out of the induction of stress genes 60 , This threshold chemical concentration where maximal induction of stress genes occurs often demarcates the beginning of deterioration of general cell heath as determined by assays such as cell viability and LDH release, suggesting that stress gene expression that increases concurrently with increasing chemical stress intensity is responsible for maintaining cell survival 39 , If cell survival is used as a biomarker for PoD, a mathematical model of the relevant stress pathway can be constructed and calibrated to predict PoD by simulating maximal stress gene induction.

Figure 2. Schematic illustration of PoD resulting from perturbation of stress pathways and bistable gene networks. A A simplified view of a cellular stress response network where posttranslational control dashed arrow increases the activities of stress proteins and transcriptional control activated by transcription factor TF increases the abundance of stress proteins. B Chemical concentration-dependent transition of the stress response.

At low chemical concentrations, the activities of basal, preexisting stress proteins are augmented through posttranslational control to maintain homeostasis of the cellular state and specialized cell function and fitness are intact. When the chemical concentration reaches a level that maximizes the activities of the preexisting stress proteins, transcriptional control is initiated to increase the abundance of stress proteins to continue to maintain homeostasis and survival.

C Cellular phenotype can be determined by the state of a gene network which can be perturbed by chemicals. D If the gene network forms a bistable switch, the bifurcation point of the network defines a PoD. For chemical concentrations below the PoD concentration, cells remain in the normal state; for chemical concentrations above the PoD concentration, cells switch to the adverse state. Gene expression levels can exhibit different variabilities depending on the cellular state. When normal-state cells are exposed to chemical concentrations well below the PoD concentration, the variability of gene expression within the bistable gene network is small blue dots on the far left of the curve, representing gene expression levels in single cells.

When normal-state cells are close to the PoD, gene expression variability dramatically increases blue dots close to PoD. Once cells switch to the adverse state, gene expression variability decreases again blue dots on the top curve. Refer to main text for further details. Despite the above considerations, it is questionable whether loss of adaptation as a result of gene induction saturation and subsequent decline in cell survival can be used as proper biomarkers for predicting in vivo adverse outcomes Emerging evidence suggests that they may be too insensitive for this purpose for reasons below.

  • Chemical Mixtures: Considering the Evolution of Toxicology and Chemical Assessment.
  • Expertise. Insights. Illumination..
  • Images of the Wildman in Southeast Asia: An Anthropological Perspective.
  • Philosophy of Mathematics: Structure and Ontology.
  • About the Health Effects Fact Sheets | Hazardous Air Pollutants | US EPA.

Cellular stress response is a highly energy-demanding process, as a number of stress proteins are being synthesized de nova to levels several to tens of folds of their basal levels 66 — Bioenergetically constrained, cells under stress initiate global translational reprogramming and metabolic reprogramming to preserve energy for the stress response for cell survival while sacrificing other non-survival-essential programs. Through this mechanism and others, the induction of stress genes is unaffected However, due to the global translation inhibition, the synthesis of proteins involved in specialized cell functions which are unlikely essential to cell survival at the moment of stress is likely halted.

In addition to translational reprogramming, metabolic reprogramming may also occur to optimize energy distribution in the face of cellular stress 77 , In the case of oxidative stress, the flux through the pentose phosphate pathway is promoted to synthesize more NADPH as reducing agent to handle the stress, which takes the carbon flux away from the glycolysis and downstream Krebs cycle, resulting in less energy available for other cellular processes Therefore, as a result of stress-induced translational and metabolic reprogramming that accompanies the onset of transcriptional induction of stress genes, cells enter a survival mode to avoid being killed, with their specialized functions and other nonessential activities sacrificed or suspended temporarily This mode of cellular stress response suggests that the beginning of transcriptional induction of stress genes may define a fitness-relevant PoD Figure 2B.

In keeping with this notion, a recent ecological study with brook trout, a cold-water fish, showed that the threshold water temperature inducing the onset of heat shock protein expression in the gill is consistent with the average water temperature limiting the fish population in the field, suggesting a connection between thermal stress response and fitness This mode of action raises the question on the mechanisms cells utilize to counter stress in the absence of transcriptional induction of stress genes.

For one, this is because stress response through transcription is too slow to be initiated to handle transient, but sometimes detrimental, stresses Moreover, the transcriptional network can be insensitive to low-level stresses due to the relatively small feedback loop gain near the basal condition where transcription factor-independent constitutive expression of stress genes can be dominant Instead, there appear to be a number of posttranslational control mechanisms cells can engage to promote their anti-stress capacity, by activating pre-existing stress proteins that are normally inactive Figure 2A The activation can be both through allosteric regulation or posttranslational modification such as phosphorylation, oxidation, and acetylation 84 — These processes are fast in responding and demand much less energy than de novo protein synthesis.

Therefore, it is possible that cells adapt to stresses through a two-tiered process depending on stress intensity At low stress levels, cells use posttranslational regulation of basal, pre-existing stress proteins to enhance their activities to maintain homeostasis while their specialized functions are unaffected because there is no global translation inhibition and metabolic reprogramming yet Figure 2B. At higher stress levels, all the pre-existing stress proteins are activated and their abundance needs to be increased—through initiating transcriptional induction—to continue to maintain homeostasis and survival.

Concomitant with the initiation of transcriptional control, cells enter survival mode, and stress-induced translational and metabolic reprogramming diverts energy and molecular resources away from the maintenance and synthesis of non-survival-essential proteins, including those executing specialized cellular functions At very high stress levels where transcriptional control is maxed out, cell viability starts to decrease with increasing cell death.

Therefore, the transition or switching of cells from posttranslational control to transcriptional control may demarcate a functional PoD, where specialized cell functions start to deteriorate or other activities conducive to fitness starts to slow or pause. As a result of this two-tiered operation, it may require us to shift the monitoring focus from measuring transcriptional changes to measuring posttranslational modifications as biomarkers to more sensitively detect PoD that is associated with functional alterations. Concomitantly, computational toxicity pathway models describing both the short, fast posttranslational feedback loop and the slow, transcriptional feedback loop need to be constructed to simulate the transition from posttranslational to transcriptional control.

Such pathway models, calibrated based on in vitro assays measuring transcriptomic and proteomic responses, will allow us to extrapolate functional PoD for tissue chemical concentrations encountered in real world exposures. The idea that transcriptional alteration can be associated with adverse outcomes has also emerged recently from toxicogenomic studies.

Since the RNA microarray technology and subsequently next-generation sequencing became available and affordable, many animal studies have been conducted to examine transcriptomic changes at confirmed or suspected target organs or tissues in response to chemical exposures 87 — Many of the tested chemicals are legacy chemicals, such as pharmaceutical compounds and well-studied environmental toxicants, with known apical endpoint toxicities. The primary purpose of these toxicogenomic studies was to explore whether detecting transcriptomic changes within a short exposure time and with fewer animals can replace the traditional apical endpoint tests.

Initially it was thought that transcriptomic changes may be a very sensitive biomarker for this purpose, i. However, the vast majority of these studies have shown, surprisingly, that there is basically a temporal and dose concordance between transcriptional changes in the most sensitive pathways and the adverse health outcomes in the animals, for both cancer and non-cancer endpoints 90 — These findings are in agreement with the two-tiered cellular stress response profile discussed above, where if a chemical dose is high enough to induce transcriptional alterations, the cell's specialized function or other activities conducive to the fitness of the organism may be compromised, resulting in adverse outcomes.

While toxicity pathway alteration at the transcriptional level can be both the cause and result of adverse outcomes, these toxicogenomic studies suggest that examining transcriptomic changes may not be early and sensitive enough to predict and avert toxicity. Although these studies were done primarily in animals, they provide important clues to guide toxicity pathway-based in vitro testing, with respect to the omic tiers that should be examined.

For chemicals inducing cellular stress responses, proteomic and metabolomic changes may act as sensitive biomarkers for detection. Transitioning of toxicity pathways and AOPs from normal to adverse states, as driven by an MIE such as receptor binding, can be regarded in some cases as switching of the underlying dynamical system from one attractor state to another Converging evidence from many fields, including physics, ecology, climate, finance, biology, and psychology, has revealed that certain generic features exist in complex dynamical systems and can be exploited to predict the imminence of abrupt state transitions such as species vanishing in ecosystems, financial crisis, or climate change 96 , Although often driven by slowly-varying external or internal factors, the transition can occur abruptly through a saddle-node or other types of bifurcation corresponding to a tipping point A common feature of a dynamical system that is approaching such tipping point is that its rate of recovery from small, transient perturbation by random noise slows down dramatically As a result of this critical slowing down, the state of the system i displays an increased temporal auto-correlation and ii fluctuates more dramatically increased variation around the current attractor state.

Both of these features have been utilized to successfully predict critical transition in many fields — Stable cellular fates or phenotypes can be regarded as attractor states of a complex dynamical system underpinned by gene regulatory networks Figure 2C and transitions between cell states are driven by physiological signals or exogenous chemicals It has been postulated that cancers are the result of normal cells falling irreversibly into cancer attractor states that are either preexisting or created by mutations or other carcinogenic drivers As a result, the expression pattern of these network genes will exhibit certain statistical features near the tipping point , , , despite that their mean expression levels can remain largely indistinguishable from the normal, healthy attractor state: it is expected that i gene expression variance will increase dramatically across time and between cells; ii expression correlation between genes in the same network will increase dramatically due to mutual feedback regulation; iii similarity between individual cells, defined by the gene expression vector, will decrease.

So the pre-tipping point can be predicted by a composite index derived from the above statistics of network genes. Practically this can be achieved by analyzing gene expression data obtained from in vitro RNAseq or real-time RT-PCR assays performed at single-cell resolution. In dynamical systems theory, such irreversibility may be mediated through a bistable switch where the system can adopt one of the two possible states under the same conditions The primary network motif of bistable switches is a positive or double-negative feedback loop.

Such feedback loops in gene networks have been shown to play essential roles in mediating many cellular phenomenon, including proliferation, differentiation, and apoptosis It defines an unambiguous threshold beyond which the cellular system will switch irreversibly to a different state, which can be another physiological state, or in the event of toxicity, an adverse outcome state. Environmental and pharmaceutical chemicals have been shown to interfere with these bistable-switching processes 38 , Computational modeling of these feedback networks capable of alternative, contrasting attractor states is important for interpreting single-cell resolution transcriptomic and proteomic data and predicting PoD for chemicals perturbing bistable-switching toxicity pathways.

A Small Dose of Toxicology: The Health Effects of Common Chemicals

For human risk assessment based on the advocated in vitro testing approach, an obvious challenge is how to interpret these in vitro assay results in the in vivo context and inform safety assessment for human populations. The biological scope of an AOP can vary depending on the chemicals and their adverse outcomes. Some AOPs can be very local, i. For these localized disease outcomes, for instance, cancer and certain liver toxicity, the in vitro PoD dose metrics derived from cell or organoid-based assays may be directly applicable to in vivo situations with minor adjustment.

What Makes Chemicals Poisonous

For disease outcomes clearly involving whole-body physiology, such as disorders in blood pressure, body temperature, hormones, metabolism, and immunity, perturbations at one target site can be systemically compensated through feedback or feedforward regulations that involve other body parts to maintain homeostasis. For example, the local effect of an endocrine disrupting chemical that inhibits thyroid hormone secretion can be compensated, within a limit, by the hypothalamic-pituitary-thyroid HPT axis feedback through enhancing TSH secretion While this PoD gap between in vitro and in vivo situations may be addressed to some extent by future organ-on-a-chip technology that can create a mini HPT system, dynamic computational models of the endocrine feedback axis will be able to bridge the gap by taking the in vitro assay results as the input and predict in vivo hormonal changes as the output.

Computational HPT models for both rats and humans have been constructed and applied to predict the in vivo effects on thyroid hormones of several thyroid disruptors such as thyroperoxidase inhibitors and sodium-iodide symporter inhibitors — A computational model of the endocrine system can also simulate cyclic hormone dynamics and incorporate population variability into the relevant physiological processes to conduct risk assessment based on in vitro testing data. Bois et al. Therefore, for the risk assessment of endocrine disruptors, we should aim to model endocrine systems by integrating different biological scales of the relevant AOPs and utilize these models to predict in vivo effects by taking in vitro toxicity pathway perturbation data Developing these quantitative AOP models capable of risk prediction will aid regulatory decisions Similarly, in the pharmaceutical research front, quantitative systems toxicology plays an emerging but important role in predicting drug toxicity to support drug development based on in vitro , animal, and clinical studies.

To this end, many mathematical models describing the underlying pathophysiology of drug adverse effects have been developed for predicting renal toxicity, gastrointestinal toxicity, arrhythmia, and liver injury — Compared to the nascent effort in toxicity pathway and systems toxicology modeling, toxicokinetic or pharmacokinetic modeling of chemical fates within the human or animal bodies has been around for a number of decades.

Physiologically-based toxicokinetic PBTK or pharmacokinetic PBPK modeling has been used to understand and predict the absorption, distribution, metabolism, and excretion ADME for environmental and industrial chemicals since the s Challenges abound, PBTK models play an increasingly important role in chemical health risk assessment More recently, PBPK modeling has seen a resurgence in the pharmaceutical industry, as drug developers aiming to achieve more accurate efficacy and safety are increasingly focusing on compound concentrations in the target tissues and cells and in the meantime hoping to achieve individualized precisions , Neither tissue-specific chemical concentrations nor inter-individual variability can be easily addressed with traditional, compartmental PK models; in contrast PBPK or PBTK models are structurally superior by modeling chemical ADME based on human anatomy and physiology.

Since the publication of the NRC report advocating for in vitro assay-based toxicity testing, applying PBTK models to reverse dosimetry, i. Researchers are eager to apply PoD concentrations derived from low- or high-throughput in vitro assays to existing or newly developed PBPK models to back-extrapolate external exposure levels that would produce equivalent target tissue concentrations Figure 1.

As an essential computational modeling component in the TT21C framework, toxicokinetic IVIVE is finding its applications in both environmental chemical risk assessment and drug development based on results from in vitro assays 29 , Readers can refer to excellent reviews indicated above. In addition to labeling chemicals by their potential hazards, the ultimate goal of toxicity testing is to provide safety assessment for the human population, protect the majority of the public by regulating exposure limit of chemicals of concern, and provide tools for personalized risk assessment.

Thus, addressing the issue of human population variability in response to environmental exposures is a complex integrating process involving convergence of multiple data streams in the above areas If relying on in vitro human cell and organoid studies mainly, how can we inform public health risk better than applying the default 10 or fold uncertainty factors that are routinely used for inter-species extrapolation from animal studies?

The data gap and challenge in using in vitro assays to inform population health risk are multi-faceted. First and at least for now, most of the in vitro assays in development utilize existing cell lines of either animal or human origin. Many of these cell lines are mutant, cancerous cells that have been immortalized or otherwise transformed, so it is questionable how representative they are for an average response of healthy individuals. Despite this caveat, efforts have been under way to utilize human Epstein-Barr virus EBV -transformed lymphoblastoid cell lines derived from individuals in the 1, Genomes Project.

These diversity cell lines are routinely used in the pharmaceutical industry for drug screening , but a subset of them were tapped, through using high-throughput assays measuring cytotoxicity, ATP, and caspase levels, for understanding human population variability in chemical toxicity , More recently, the effort has been further expanded to include over a thousand cell lines representing nine populations from five continents to characterize human inter-individual variability for a number of chemicals These studies, while not only characterizing the response variability, also help to identify through genome-wide association analysis the primary gene polymorphisms responsible for the heterogeneous cellular responses.

Compared to immortalized human cell lines, primary human cells are closer to mimicking in vivo conditions. Numerous studies have been conducted demonstrating individual variability using primary cells. For instance, umbilical cord blood-derived cells were exploited to examine individual variability in the proliferative response to low-dose radiations , Primary human B lymphocytes from different donors were utilized to characterize the inter-individual variability in the immunotoxic effect of dioxin on the antibody secretion response and to ascertain how the variability may affect the linearization of the population-average dose response curves Many of the studies with primary human cells use blood cells for their easy access; lack of easy access to solid living human tissues limits the usage of primary cells for toxicity testing.

In the past decade, however, with the discovery of induced pluripotent stem cells iPSC and maturing experimental protocols to differentiate iPSCs to desired cell lineages, the opportunity of addressing human individual variability and susceptibility using in vitro approaches has improved considerably For instance, recently iPSC-derived cardiomyocytes from healthy donors have been used to study the inter-individual variability in the cardiotoxicity of pharmaceutical compounds While iPSC-derived cells may not be able to fully recapitulate the behavior of the corresponding cells in vivo , as our understanding of the cell differentiation process improves and the differentiation protocols are further refined, this approach presents the most promising future in generating—in large quantity—different types of cells representing different human races and individuals.

Regardless the availability of cells representing human populations, to address population variability experimentally, the testing space as defined by the product of individuals, cell types, toxicity pathways, key events, and chemicals is astronomically huge and can be economically prohibitive. Developing experimental assays and further reducing their costs aside, it is appealing and more economical to use computational modeling of toxicity pathways, virtual tissues, and AOPs to explore individual variability and human population risk.

In theory, once an average, mechanistically-based model for a toxicodynamic process is developed and validated, individual responses can be cheaply explored computationally by varying relevant model parameters based on defined distributions. Each combination of parameter values randomly drawn from these distributions would represent a human individual. The challenge lies in how to come up with the parameter distributions that can represent a human population. Depending on the granularity of the computational model, i. For a toxicity pathway model simulating cellular responses, parameter differences between cells from different individuals can be due to genetic, epigenetic differences which affect the transcription, translation, degradation, and activity of the proteins involved.

For organism-level models, such as the endocrine system, individual difference can be captured in parameters governing hormone synthesis, release, metabolism, and feedback regulation, etc. To determine the most important parameters contributing to individual variability, sensitivity analysis can be performed to scan for parameters that affect the model behavior the most In addition, key determinants for heterogeneous individual responses can be gleaned from genome-wide association studies GWAS and epigenome-wide association studies EWAS of in vitro assays that identify differences among individuals in key genes and the modifications in their regulatory regions , — Such information can be incorporated into relevant computational models to explore population variability in silico.

GWAS and EWAS studies can also help identify important information on joint distributions of certain physiological parameters, which occur through co-evolution or influence by some common yet unknown factors. Compared to toxicity pathway modeling of toxicodynamics, simulating toxicokinetics through PBTK modeling to recapitulate population variabilities is in a far more advanced stage. This is largely because 1 individual variabilities in anatomical parameters, such as body weight, cardiac output and tissue volume, which play a significant role in the ADME of a chemical, can be well documented and defined in a human population.

As far as chemical-specific parameters are concerned, their distributions can be estimated in several ways , For human data-rich compounds, such as drugs, these parameter distributions can be back-calculated in a posterior fashion. For parameter-rich compounds obtained through experimental measurement, distributions can be assumed a priori. Lastly, with a Bayesian PBTK modeling approach, prior parameter distributions can be further constrained through comparing PBTK model output with the measured chemical tissue concentrations to arrive at posterior distributions that are less uncertain , The full power of computational modeling lies in combining TK and TD models to address the exposure-to-outcome continuum along the aggregate exposure pathway AEP and AOP frameworks , Recently Bois et al.

Each of the 19 chapters has: 1 a website address for a slide presentation, 2 a list of web sites for [a] European, Asian, and international agencies, [b] North American agencies, and [c] non-governmental organizations from which one can obtain further information, and 3 a list of references approximately per chapter. The book is organized into four sections. The second section has 12 chapters on toxic agents.

These include alcohol, caffeine, nicotine, pesticides, lead, mercury, arsenic, metals, solvents, radiation, animal and plant toxins, and environmental contaminants. The third section has three chapters including: 1 neurotoxicology, 2 cancer and genetic toxicology, and 3 pregnancy and developmental toxicology. The last section has two chapters concerned with applied toxicology: 1 toxins in the home, and 2 risk assessment and risk management.

In the nearby farmland, trucks and crop dusters sprayed DDT and other pesticides in great, puffy clouds that we kids sometimes rode our bikes through, holding our breath and feeling very brave. Today the air is clear, and the river free of effluents—a visible testament to the success of the U. But my Axys test results read like a chemical diary from 40 years ago.

A Small Dose of Toxicology

My blood contains traces of several chemicals now banned or restricted, including DDT in the form of DDE, one of its breakdown products and other pesticides such as the termite-killers chlordane and heptachlor. The levels are about what you would expect decades after exposure, says Rozman, the toxicologist at the University of Kansas Medical Center.

My childhood playing in the dump, drinking the water, and breathing the polluted air could also explain some of the lead and dioxins in my blood, he says. I went to college at a place and time that put me at the height of exposure for another set of substances found inside me—PCIBs, once used as electrical insulators and heat-exchange fluids in transformers and other products. PCBs can lurk in the soil anywhere there's a dump or an old factory. About miles kilometers downstream is the city of Poughkeepsie, where I attended Vassar College in the late s. PCBs, oily liquids or solids, can persist in the environment for decades.

In animals, they impair liver function, raise blood lipids, and cause cancers. Some of the different PCBs chemically resemble dioxins and cause other mischief in lab animals: reproductive and nervous system damage, as well as developmental problems. But until then, GE legally dumped excess PCBs into the Hudson, which swept them all the way downriver to Poughkeepsie, one of eight cities that draw their drinking water from the Hudson. In , a mile GE has spent million dollars on the cleanup so far, dredging up and disposing of PCBs in the river sediment under the supervision of the EPA. It is also working to stop the seepage of PCBs into the river from the factories.

Birds and other wildlife along the Hudson are thought to have suffered from the pollution, but its impact on humans is less definitive. One study in Hudson River communities found a 20 percent increase in the rate of hospitalization for respiratory diseases, while another, more reassuringly, found no increase in cancer deaths in the contaminated region. But among many of the locals, the fear is palpable. Ed Fitzgerald of the State University of New York at Albany, a former staff scientist at the state department of health, is conducting the most thorough study yet of the health effects of PCBs in the area.

He says he has explained to Prevost and other residents that the risk from the wells was probably small because PCBs tend to settle to the bottom of an aquifer. Eating contaminated fish caught in the Hudson is a more likely exposure route, he says. I didn't eat much Hudson River fish during my college days in the s, but the drinking water in my dorm could have contained traces of the PCBs pouring into the river far upstream.

Or maybe not.

A Small Dose of Toxicology: The Health Effects of Common Chemicals - PDF Free Download

Back home in San Francisco, I encounter a newer generation of industrial chemicals—compounds that are not banned , and, like flame retardants, are increasing year by year in the environment and in my body. Sipping water after a workout, I could be exposing myself to bisphenol A, an ingredient in rigid plastics from water bottles to safety goggles. Bisphenol A causes reproductive system abnormalities in animals. My levels were so low they were undetectable—a rare moment of relief in my toxic odyssey. And that faint lavender scent as I shampoo my hair?

Credit it to phthalates, molecules that dissolve fragrances, thicken lotions, and add flexibility to PVC, vinyl, and some intravenous tubes in hospitals. The dashboards of most cars are loaded with phthalates, and so is some plastic food wrap. Heat and wear can release phthalate molecules, and humans swallow them or absorb them through the skin.

Because they dissipate after a few minutes to a few hours in the body, most people's levels fluctuate during the day. Like bisphenol A, phthalates disrupt reproductive development in mice. An expert panel convened by the National Toxicology Program recently concluded that although the evidence so far doesn't prove that phthalates pose any risk to people, it does raise "concern," especially about potential effects on infants. I scored higher than the mean in five out of seven phthalates tested.

One of them, monomethyl phthalate, came in at Leo Trasande speculates that some of my phthalate levels were high because I gave my urine sample in the morning, just after I had showered and washed my hair. My inventory of household chemicals also includes perfluorinated acids PFAs —tough, chemically resistant compounds that go into making nonstick and stain-resistant coatings.

In animals these chemicals damage the liver, affect thyroid hormones, and cause birth defects and perhaps cancer, but not much is known about their toxicity in humans. Long-range pollution left its mark on my results as well: My blood contained low, probably harmless, levels of dioxins, which escape from paper mills, certain chemical plants, and incinerators. In the environment, dioxins settle on soil and in the water, then pass into the food chain. They build up in animal fat, and most people pick them up from meat and dairy products.

And then there is mercury, a neurotoxin that can permanently impair memory, learning centers, and behavior. Coal-burning power plants are a major source of mercury, sending it out their stacks into the atmosphere, where it disperses in the wind, falls in rain, and eventually washes into lakes, streams, or oceans. There bacteria transform it into a compound called methylmercury, which moves up the food chain after plankton absorb it from the water and are eaten by small fish. Large predatory fish at the top of the marine food chain, like tuna and swordfish, accumulate the highest concentrations of methylmercury—and pass it on to seafood lovers.

For people in northern California, mercury exposure is also a legacy of the gold rush years ago, when miners used quicksilver, or liquid mercury, to separate the gold from other ores in the hodgepodge of mines in the Sierra Nevada. Over the decades, streams and groundwater washed mercury-laden sediment out of the old mine tailings and swept it into San Francisco Bay. I don't eat much fish, and the levels of mercury in my blood were modest.

But I wondered what would happen if I gorged on large fish for a meal or two. Both were caught in the ocean just outside the Golden Gate, where they might have picked up mercury from the old mines. That night I ate the halibut with basil and a dash of soy sauce; I downed the swordfish for breakfast with eggs cooked in my nonstick pan. Twenty-four hours later I had my blood drawn and retested. My level of mercury had more than doubled, from 5 micrograms per liter to a higher-than-recommended Mercury at 70 or 80 micrograms per liter is dangerous for adults, says Leo Trasande, and much lower levels can affect children.

It's a lot harder to dodge the PBDE flame retardants responsible for the most worrisome of my test results. My world—and yours—has become saturated with them since they were introduced about 30 years ago. Scientists have found the compounds planetwide, in polar bears in the Arctic, cormorants in England, and killer whales in the Pacific.