PREDICTING HUMAN DEVELOPMENTAL TOXICITY OF PHARMACEUTICALS USING HUMAN STEM-LIKE CELLS AND METABOLOMIC RATIOS
This present invention provides rapid, reproducible, biomarker-based screening methods for the developmental toxicity testing of compounds. The methods are designed to identify the exposure level at which a test compound perturbs metabolism in a manner predictive of developmental toxicity. In particular, the perturbation of two metabolites, ornithine and cystine, is measured, wherein a ratio of the fold change in ornithine to the fold change in cystine of less than or equal to about 0.88 is indicative of the teratogenicity of a test compound.
This application is a divisional application of U.S. application Ser. No. 14/439,723, filed Apr. 30, 2015, and which is a § 371 U.S. National Stage of International Application No. PCT/US2013/067980, filed Nov. 1, 2013, which claims the benefit of U.S. Provisional Applications Ser. No. 61/721,746, filed Nov. 2, 2012, and Ser. No. 61/827,407, filed May 24, 2013, each of which are incorporated by reference herein.
GOVERNMENT FUNDINGThis invention was made with government support under Grant No. IIP-1058355, awarded by the National Science Foundation. The Government has certain rights in the invention.
BACKGROUNDBirth defects are reported in approximately 3% of all human births and are the largest cause of infant mortality in the United States (Hoyert et al., 2006, Pediatrics; 117:168-183). Exposure to toxic chemicals and physical agents is believed to be responsible for approximately 3% of all birth defects (National Research Council, 2000, “Scientific frontiers in developmental toxicology and risk assessment,” Washington, D.C.: The National Academies Press).
It is understood that developmental toxicity can cause birth defects, and can generate embryonic lethality, intrauterine growth restriction (IUGR), dysmorphogenesis (such as skeletal malformations), and functional toxicity, which can lead to cognitive disorders such as autism. There is an increasing concern about the role that chemical exposure can play in the onset of these disorders. Indeed, it is estimated that 5% to 10% of all birth defects are caused by in utero exposure to known teratogenic agents that induce developmental abnormalities in the fetus (Beckman and Brent, 1984, Annu Rev Pharmacol; 24: 483-500). Concern exists that chemical exposure may be playing a significant and preventable role in producing birth defects (Claudio et al., 2001, Environm Health Perspect; 109: A254-A261).
However, this concern has been difficult to evaluate, due to the lack of robust and efficient models for testing developmental toxicity for the more than 80,000 chemicals in the market, plus the new 2,000 compounds introduced annually (General Accounting Office (GAO), 1994, Toxic Substances Control Act: Preliminary Observations on Legislative Changes to Make TSCA More Effective, Testimony, Jul. 13, 1994, GAO/T-RCED-94-263). Fewer than 5% of these compounds have been tested for reproductive outcomes and even fewer for developmental toxicity (Environmental Protective Agency (EPA), 1998, Chemical Hazard Data Availability Study, Office of Pollution Prevention and Toxins). Although some attempts have been made to use animal model systems to assess toxicity (Piersma, 2004, Toxicology Letters; 149:147-53), inherent differences in the sensitivity of humans in utero have limited the predictive usefulness of such models.
Toxicity, particularly developmental toxicity, is also a major obstacle in the progression of compounds through the drug development process. Currently, toxicity testing is conducted on animal models as a means to predict adverse effects of compound exposure, particularly on development and organogenesis in human embryos and fetuses. The most prevalent models that contribute to FDA approval of investigational new drugs are whole animal studies in rabbits and rats (Piersma, 2004, Toxicology Letters; 149: 147-53). In vivo studies rely on administration of compounds to pregnant animals at different stages of pregnancy and embryonic/fetal development (first week of gestation, organogenesis stage and full gestation length). However, these in vivo animal models are limited by a lack of biological correlation between animal and human responses to chemical compounds during development due to differences in biochemical pathways. Species differences are often manifested in trends such as dose sensitivity and pharmacokinetic processing of compounds. According to the reported literature, animal models are approximately 60% efficient in predicting human developmental response to compounds (Greaves et al., 2004, Nat Rev Drug Discov; 3:226-36). Thus, there is a need for human-directed predictive in vitro models.
The thalidomide tragedy in the 1960s emphasized the importance of preclinical developmental toxicity testing, the significant differences among species in their response to potentially teratogenic compounds, and how the developing fetus can be affected by such compounds. Developmental toxicity testing of thalidomide in rodent models did not indicate the compound's teratogenic potential in humans. Over 10,000 children were born with severe birth defects following in utero exposure. Current preclinical models for detecting developmental toxicity have varying degrees of concordance with observed developmental toxicity in humans, with rats and rabbits (the most commonly used species for developmental toxicity testing) having approximately 70-80% concordance to known human teratogens (Daston G P and Knudsen T B, 2010, “Fundamental concepts, current regulatory design and interpretation,” In: Knudsen T B, Daston G P, editors. Comprehensive Toxicology. Vol 12, 2nd ed. New York: Elsevier. p 3-9). These decades-old in vivo animal models require large numbers of animals, kilogram quantities of test compound, and are both time consuming and expensive. Due to the cost and complexity of these models, safety assessments often occur too late in the compound's life cycle for the developer to react to a positive developmental toxicity signal, and can result in the termination of the development of the compound or series. Though these animal models are, and have long been, considered the regulatory gold standard, differences in species response to a compound may lead to missed signals of developmental toxicity and biological misinterpretation. As such, the development of a new generation of tools using human cells for assessment of potential developmental toxicity risk related to chemical exposure is needed. The appropriate tests would also reduce product development time, control costs, and respond proactively to the call to decrease animal use.
Thus, there is a need for a relevant, predictive, accurate, low cost, and rapid human in vitro tests for reliably determining developmental toxicity of pharmaceutical agents and other chemical compounds.
SUMMARY OF THE INVENTIONThe present invention includes a method of classifying a test compound as a teratogen or a non-teratogen, the method including culturing undifferentiated human stem cell-like cells (hSLCs) in the presence of the test compound and in the absence of the test compound; determining the fold change in ornithine, or fragment, adduct, deduct or loss thereof, in the culture media of undifferentiated hSLCs cultured in the presence of the test compound in comparison with hSLCs cultured in the absence of the test compound; determining the fold change in cystine, or fragment, adduct, deduct or loss thereof, in the culture media of undifferentiated hSLCs cultured in the presence of the test compound in comparison with hSLCs cultured in the absence of the test compound; and determining the ratio of the fold change in ornithine, or fragment, adduct, deduct or loss thereof, to the fold change in cystine, or fragment, adduct, deduct or loss thereof, wherein a ratio of less than or equal to about 0.88 is indicative of the teratogenicity of the test compound and a ratio of greater than about 0.88 is indicative of the non-teratogenicity of the test compound.
The present invention includes a method of predicting teratogenicity of a test compound, the method including culturing undifferentiated human stem cell-like cells (hSLCs) in the presence of the test compound and in the absence of the test compound; determining the fold change in ornithine, or fragment, adduct, deduct or loss thereof, in the culture media of undifferentiated hSLCs cultured in the presence of the test compound in comparison with hSLCs cultured in the absence of the test compound; determining the fold change in cystine, or fragment, adduct, deduct or loss thereof, in the culture media of undifferentiated hSLCs cultured in the presence of the test compound in comparison with hSLCs cultured in the absence of the test compound; and determining the ratio of the fold change in ornithine, or fragment, adduct, deduct or loss thereof, to the fold change in cystine, or fragment, adduct, deduct or loss thereof, wherein a ratio of less than or equal to about 0.88 is indicative of the teratogenicity of the test compound and a ratio of greater than about 0.88 is indicative of the non-teratogenicity of the test compound.
The present invention includes a method for validating a test compound as a teratogen, the method including culturing undifferentiated human stem cell-like cells (hSLCs) in the presence of the test compound and in the absence of the test compound; determining the fold change in ornithine, or fragment, adduct, deduct or loss thereof, in the culture media of undifferentiated hSLCs cultured in the presence of the test compound in comparison with hSLCs cultured in the absence of the test compound; determining the fold change in cystine, or fragment, adduct, deduct or loss thereof, in the culture media of undifferentiated hSLCs cultured in the presence of the test compound in comparison with hSLCs cultured in the absence of the test compound; and determining the ratio of the fold change in ornithine, or fragment, adduct, deduct or loss thereof, to the fold change in cystine, or fragment, adduct, deduct or loss thereof, wherein a ratio of less than or equal to about 0.88 is indicative of the teratogenicity of the test compound and a ratio of greater than about 0.88 is indicative of the non-teratogenicity of the test compound.
The present invention includes a method for determining the exposure concentration at which a test compound is teratogenic, the method including culturing undifferentiated human stem cell-like cells (hSLCs) in a range of concentrations of the test compound and in the absence of the test compound; determining the fold change in ornithine, or fragment, adduct, deduct or loss thereof, in the culture media of undifferentiated hSLCs cultured in each concentration of the test compound in comparison with hSLCs cultured in the absence of the test compound; determining the fold change in cystine, or fragment, adduct, deduct or loss thereof, in the culture media of undifferentiated hSLCs cultured in each concentration of the test compound in comparison with hSLCs cultured in the absence of the test compound; and determining the ratio of the fold change in ornithine, or fragment, adduct, deduct or loss thereof, to the fold change in cystine, or fragment, adduct, deduct or loss thereof, for each concentration of test compound, wherein a ratio of less than or equal to about 0.88 at a given concentration of the test compound is indicative of the teratogenicity of the test compound at that given concentration and a ratio of greater than about 0.88 at a given concentration of the test compound is indicative of the non-teratogenicity of the test compound at that given concentration.
In some aspects of the methods of the present invention, cystine, or fragment, adduct, deduct or loss thereof, and/or ornithine, or fragment, adduct, deduct or loss thereof, are identified using a physical separation method. In some aspects, a physical separation method includes mass spectrometry. In some aspects, mass spectrometry includes liquid chromatography/electrospray ionization mass spectrometry.
In some aspects of the methods of the present invention, cystine, or fragment, adduct, deduct or loss thereof, and/or ornithine, or fragment, adduct, deduct or loss thereof, are measured using a colorimetric or immunological assay.
In some aspects of the methods of the present invention, hSLCs includes human embryonic stem cells (hESCs), human induced pluripotent (iPS) cells, or human embryoid bodies.
In some aspects of the methods of the present invention, the hSLCs are cultured at a concentration of the test compound including the test compound's human therapeutic Cmax.
In some aspects of the methods of the present invention, the hSLCs are cultured in a range of concentrations of the test compound. In some aspects, the range of concentrations includes a serial dilution. In some aspects, the range of concentrations includes nine three-fold dilutions. In some aspects, the range of concentrations includes from about 0.04 μM to about 300 μM, about 4 μM to about 30,000 μM, and about 0.0001 μM to about 10μ. In some aspects, the range of concentrations of the test compound includes the test compound's human therapeutic Cmax.
In some aspects of the methods of the present invention, the method further includes detecting one or more additional metabolites associated with hSLCs cultured in the presence of the test compound in comparison with hSLCs cultured in the absence of the test compound. In some aspects, one or more additional metabolite includes arginine, ADMA, cystathionine, and/or a fragment, adduct, deduct or loss thereof. In some aspects, one or more additional metabolites are identified using a physical separation method. In some aspects, a physical separation method includes mass spectrometry. In some aspects, mass spectrometry includes liquid chromatography/electrospray ionization mass spectrometry. In some aspects, one or more additional metabolites are measured using a colorimetric or immunological assay.
In some aspects of the methods of the present invention, the method further includes determining the ratio of the fold change in arginine, or fragment, adduct, deduct or loss thereof, to the fold change in ADMA, or fragment, adduct, deduct or loss thereof, wherein a ratio of less than at least about 0.9 or greater than at least about 1.1 is indicative of the teratogenicity of the test compound and a ratio of greater than at least about 0.9 and less than at least about 1.1 is indicative of the non-teratogenicity of the test compound.
The term “and/or” means one or all of the listed elements or a combination of any two or more of the listed elements.
The words “preferred” and “preferably” refer to embodiments of the invention that may afford certain benefits, under certain circumstances. However, other embodiments may also be preferred, under the same or other circumstances. Furthermore, the recitation of one or more preferred embodiments does not imply that other embodiments are not useful, and is not intended to exclude other embodiments from the scope of the invention. The embodiment(s) described, and references in the specification to “one embodiment,” “an embodiment of the invention,” “an embodiment,” “an example embodiment,” etc., indicate that the embodiment(s) described may include a particular feature, structure, or characteristic, but every embodiment may not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is understood that it is within the knowledge of one skilled in the art to effect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.
The terms “comprises” and variations thereof do not have a limiting meaning where these terms appear in the description and claims. It is understood that wherever embodiments are described herein with the language “comprising,” otherwise analogous embodiments described in terms of “consisting of” and/or “consisting essentially of” are also provided.
Unless otherwise specified, “a,” “an,” “the,” and “at least one” are used interchangeably and mean one or more than one.
In the following description, for purposes of explanation, specific numbers, parameters and reagents are set forth in order to provide a thorough understanding of the invention. It is understood, however, that the invention can be practiced without these specific details. In some instances, well-known features can be omitted or simplified so as not to obscure the present invention.
Also herein, the recitations of numerical ranges by endpoints include all numbers subsumed within that range (e.g., 1 to 5 includes 1, 1.5, 2, 2.75, 3, 3.80, 4, 5, etc.).
Unless otherwise indicated, all numbers expressing quantities of components, molecular weights, and so forth used in the specification and claims are to be understood as being modified in all instances by the term “about.” Accordingly, unless otherwise indicated to the contrary, the numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties sought to be obtained by the present invention.
For any method disclosed herein that includes discrete steps, the steps may be conducted in any feasible order. And, as appropriate, any combination of two or more steps may be conducted simultaneously.
The above summary of the present invention is not intended to describe each disclosed embodiment or every implementation of the present invention. The description that follows more particularly exemplifies illustrative embodiments. In several places throughout the application, guidance is provided through lists of examples, which examples can be used in various combinations. In each instance, the recited list serves only as a representative group and should not be interpreted as an exclusive list.
The present invention provides human-specific in vitro methods for determining toxicity, particularly developmental toxicity, and teratogenicity of pharmaceuticals and other non-pharmaceutical chemical compounds using human stem-like cells (hSLCs). The present invention utilizes hSLCs and metabolomics to provide a predictive, quantitative, all-human in vitro screening method for predicting human developmental toxicity of compounds. The present methods overcome limitations associated with interspecies animal models and provide innovative and robust alternative in vitro model systems to predict developmental toxicity of chemicals. The application of more predictive developmental toxicity screens would reduce the prevalence of birth defects and increase pharmaceutical and chemical safety.
The present invention provides an exposure-based in vitro assay by measuring a metabolic perturbation in the culture media that could be used as an early signal for the potential of developmental toxicity.
With the methods of the present invention, any of a variety of human stem-like cells (hSLCs) may be used to predict developmental toxicity of chemical entities. Human stem-like cells include, but are not limited to, pluripotent, undifferentiated human embryonic stem cells (hESCs), human induced pluripotent (iPS) cells, human embryoid bodies, and hSLC-derived lineage-specific cells.
hESCs are pluripotent, self-renewing cells isolated directly from preimplantation human embryos that recapitulate organogenesis in vitro. Lineage-specific precursor cells are derived from hESCs and have entered a specific cellular lineage, but yet remain multipotent with regard to cell type within that specific lineage. For example, neural precursors have committed to neural differentiation but yet remain unrestricted as to its neural cell type. As used herein, the term “human embryonic stem cells (hESCs)” is intended to include undifferentiated stem cells originally derived from the inner cell mass of developing blastocysts, and specifically pluripotent, undifferentiated human stem cells and partially-differentiated cell types thereof (e.g., downstream progenitors of differentiating hESC). As provided herein, in vitro cultures of hESCs are pluripotent and not immortalized, and can be induced to produce lineage-specific cells and differentiated cell types using methods well-established in the art. hESCs useful in the practice of the methods of the present invention include, but are not limited to, those are derived from preimplantation blastocysts, for example, as described by Thomson et al., in U.S. Pat. No. 6,200,806. Multiple hESC lines are currently available in US and UK stem cell banks. hESCs used may include any of the three hES cell lines, WA01, WA07, and WA09. Previous work has established that an untargeted metabolomics-based evaluation of hES cell spent media following exposure to compounds with known human teratogenicity outcomes produces predictive signatures that can be utilized as a developmental toxicity screen (Kleinstreuer et al., 2011, Toxicol Appl Pharmacol; 257:111-121; and West et al., 2010, Toxicol Appl Pharmacol; 247:18-27, each of which is incorporated herein in its entirety).
Human induced pluripotent stem cells (iPS) cells are a type of pluripotent stem cell artificially derived from a non-pluripotent cell, typically an adult somatic cell, by inducing a forced expression of certain genes. iPS cells are believed to be identical to natural pluripotent stem cells, such as embryonic stem cells in many respects, such as the expression of certain stem cell genes and proteins, chromatin methylation patterns, doubling time, embryoid body formation, teratoma formation, viable chimera formation, and potency and differentiability. iPS cells may be obtained, for example, from adult tissues (such as for example, from cells obtained from the bone the marrow) and by parthenogenesis (see, for example, Vrana et al., 2003, Colloquium; 100, Supp. 1:11911-11916).
Human embryoid bodies are aggregates of cells derived from human embryonic stem cells. Cell aggregation is imposed by hanging drop, plating upon non-tissue culture treated plates or spinner flasks; either method prevents cells from adhering to a surface to form the typical colony growth. Upon aggregation, differentiation is initiated and the cells begin to a limited extent to recapitulate embryonic development. Embryoid bodies are composed of cells from all three germ layers: endoderm, ectoderm and mesoderm.
The cells of the present invention can include hSLC-derived lineage specific cells. The terms “hSLC-derived lineage specific cells,”, “stem cell progenitor,” “lineage-specific cell,” “hSLC derived cell,” and “differentiated cell” as used herein are intended to encompass lineage-specific cells that are differentiated from hSLCs such that the cells have committed to a specific lineage of diminished pluripotency. For example, hSLC-derived lineage specific cells are derived from hSLCs and have entered a specific cellular lineage, but yet remain multipotent with regard to cell type within that specific lineage. The hSLC-derived lineage specific cells can include, for example, neural stem cells, neural precursor cells, neural cells, cardiac stem cells, cardiac precursor cells, cardiomyocytes, and the like. In some embodiments, these hSLC-derived lineage-specific cells remain undifferentiated with regard to final cell type. For example, neuronal stem cells are derived from hSLCs and have differentiated enough to commit to neuronal lineage. However, the neuronal precursor retains “stemness” in that it retains the potential to develop into any type of neuronal cell. Additional cell types include terminally-differentiated cells derived from hESCs or lineage-specific precursor cells, for example neural cells.
With the methods of the present invention, hSLCs may be cultured using methods of cell culture well-known in the art, including, for example, methods disclosed in Ludwig et al. (2006, Nat Methods; 3:637-46), U.S. patent application Ser. No. 11/733,677 (“Reagents and Methods for Using Human Embryonic Stem Cells to Evaluate Toxicity of Pharmaceutical Compounds and other Compounds”), PCT/US2011/029471 and U.S. patent application Ser. No. 13/069,326 (“Predicting Human Developmental Toxicity of Pharmaceuticals Using Human Stem-Like Cells and Metabolomics”), and any of those described herein.
In some aspects of the present invention, hSLCs are maintained in an undifferentiated state prior to and/or during exposure to a test compound. In some aspects of the present invention, hSLCs may be cultured in the absence of a feeder cell layer during exposure to a test compound and/or cultured on feeder cell layer prior to such exposure.
The methods of the present invention profile changes in cellular metabolism that are measured in the spent cell culture medium from hSLCs following compound exposure. This metabolic footprint of the culture medium is a functional measurement of cellular metabolism referred to as the “secretome.” The secretome refers to the metabolites present in the spent media (which may also be referred be herein as “cell culture supernatant,” “culture supernatant,” “supernatant,” “cell supernatant,” “cell culture media,” “culture media,” “cell culture medium,” “culture medium,” “media,” or “medium”) following cell culture. The secretome includes media components, metabolites passively and actively transported across the plasma membrane, intracellular metabolites release upon lysis, and those produced through extracellular metabolism of enzymes. The change in the secretome elicited by test compound exposure relative to untreated cultures produces a metabolic signature of toxicity. The secretome is measured because of several unique qualities for profiling cell culture media; it is very easy to reproducibly sample, minimal handling is required to quench metabolism, it does not destroy the cells that can then be used for other assays, it is amenable to high-throughput evaluation, and strong signals can be measured due to the accumulation of metabolites over time. The ability to measure metabolic changes following compound exposure has identified new biomarkers associated with disruption of human development and provided the opportunity to develop highly predictive models of developmental toxicity based on these changes.
Metabolites include, but are not limited to, sugars, organic acids, amino acids, fatty acids, hormones, vitamins, oligopeptides (less than about 100 amino acids in length), as well as ionic fragments thereof. In some aspects, metabolites are less than about 3000 Daltons in molecular weight, and more particularly from about 50 to about 3000 Daltons.
With the present invention, a fold change in a metabolite in hSLCs cultured in the presence of a test compound in comparison with hSLCs cultured in the absence of the teratogenic compound may be determined. The metabolic effect of a teratogenic compound refers to the difference in one or more metabolites in hSLCs cultured in presence of the teratogenic compound in comparison with hSLCs cultured in absence of the teratogenic compound (or, in some aspects, hSLCs cultured in presence of a known non-teratogenic compound). A metabolite may be differentially expressed, for example, the expression of a metabolite may be increased or decreased when exposed to a teratogenic compound.
In some aspects, a ratio of the fold changes of two metabolites in hSLCs cultured in presence of a test compound in comparison with hSLCs cultured in absence of the teratogenic compound may be determined. For example, with the present invention, it has been determined that altered ratios in the fold changes of ornithine to cystine, asymmetric dimethylarginine (ADMA) to cystine, and/or cystathionine to cystine may be predictive of the developmental toxicity/teratogenicity of a test compound. Any one, two or all three of these ration may be utilized in the determination of the developmental toxicity of a compound.
With the present invention, a change in the secretome elicited by test compound exposure relative to untreated cultures produces a metabolic signature that may be used for measuring cell viability. Changes in cellular metabolism as measured in the spent medium following cell culture are a functional measure of cell health. The change in the secretome elicited by exposure to a test agent relative to untreated cultures produces a metabolic signature that can be used to infer the number of metabolically viable cells present within a cell culture. One or more of the secreted metabolites described herein can be utilized to infer the number viable cells relative to the number of cells in a reference culture “control group.” These metabolites could be utilized to determine the number of viable cells within a cell culture without a requirement to destroy or impact the cells. These metabolites can be used as novel measure of viability that does not require disrupting the growing cells.
With the present invention, a change in the secretome elicited by exposure to a range of concentrations of a test compound relative to untreated cultures may be used to determine the concentration at which a test compound is teratogenic. The teratogenic potential of a compound is associated with the level of exposure to the fetus. Therefore a compound could be considered both teratogenic and non-teratogenic depending on the exposure level. For example, retinol (vitamin A), when taken at or below the Food and Drug Administration (FDA) maximum recommended daily allowance (RDA; 8,000 IU), does not have an adverse effect on the developing fetus. However, high doses of retinol (>25,000 IU/day) have been shown to cause malformations similar to those seen following 13-cis retinoic acid exposure in both experimental animals and humans (Teratology Society, 1987, “Teratology Society position paper: recommendations for vitamin A use during pregnancy,” Teratology; 35:269-275).
In some aspects, the teratogenicity of a compound may be tested at concentrations corresponding to their IC50 or EC50 dose levels, at concentrations corresponding to their circulating dose, at concentrations corresponding to in maternal circulation and/or at concentrations corresponding to the test compound's human therapeutic Cmax. Such dosing recapitulates the exposure level to a developing human embryo in vivo and the toxic or teratogenic effect of the dosing compound on human development.
In some aspects, the teratogenicity of a compound may be tested over a range of concentrations of the test compound. Such a range may include, for example, about 0.04 μM to about 300 μM, about 4 μM to about 30,000 μM, and about 0.0001 μM to about 10 μM. Such a range may include, for example, a serial dilution of, for example, five, six, seven, eight, nine, ten, or more dilutions. Such dilutions may be, for example, two-fold, three-fold, four-fold, five-fold, ten-fold, or more.
With the present invention, individual metabolites and/or ratios of fold changes may be utilized in concordance with cell viability data for the prediction of developmental toxicity. The quickPredict method described herein combines cell culture based evaluation of a nine-point dose curve with a metabolic index to predict the dose at which a test agent may exhibit developmental toxicity and cytotoxicity within a seven day time frame. This assay workflow represent a significant five-fold increase in throughput over traditional ‘omics’ based computational approaches. In the previously described devTox assay (see, for example, PCT/US2011/029471 and U.S. patent application Ser. No. 13/069,326 (“Predicting Human Developmental Toxicity of Pharmaceuticals Using Human Stem-Like Cells and Metabolomics,” West et al., 2010, Toxicol Appl Pharmacol; 247(1):18-27, and Kleinstreuer et al., 2011, Toxicol Appl Pharmacol; 257(1):111-121), stem cells are dosed with a test compound in two steps, (1) at multiple concentrations for cell viability measurements which are performed to determine the optimal dose levels for metabolomics studies that provide a maximum metabolic response with a minimum of cell death, and (2) then after the best concentration was determined, a new batch of cells is then dosed with 3 concentrations derived from the optimal concentration and IC50, the media is collected for LC-MS analysis using both ESI positive and ESI negative ionization polarities. In the QuickPredict methods of the present invention, media is collected from the first step 96-well plates containing the cells dosed at multiple concentrations and is analyzed directly on the mass spectrometer using a much shorter LC gradient (6.5 minutes versus 23 minutes for the previous method), using only positive polarity ESI. In some aspects, the QuickPredict method may utilize a Waters Acquity UPLC BEH Amide 2.1×50 1.7 uM column, rather than a longer Phenomenex Luna HILIC 100×3 mm 1.7 uM column. LC-MS data can be acquired for two 96 well plates (corresponding to 2 test compounds) in 18 hours.
In some aspects, a fold change ratio of other than about 1 is indicative of the teratogenicity of the test compound, for example, a fold change ratio of greater than about 1 (for example, including, but not limited to, about 1.01, about 1.02, about 1.03, about 1.04, about 1.05, about 1.06, about 1.07, about 1.08, about 1.09, about 1.1, about 1.11, about 1.12, about 1.13, about 1.14, about 1.15, about 1.16, about 1.17, about 1.18, about 1.19, about 1.2, about 1.21, about 1.22, about 1.23, about 1.24, about 1.25, about 1.26, about 1.27, about 1.28, about 1.29, about 1.3, about 1.31, about 1.32, about 1.33, about 1.34, about 1.35, about 1.36, about 1.37, about 1.38, about 1.39, about 1.4, about 1.41, about 1.42, about 1.43, about 1.44, about 1.45, about 1.46, about 1.47, about 1.48, about 1.49, or about 1.5) and/or a fold change ratio of less than about 1 (for example, including, but not limited to, about 0.99, about 0.98, about 0.97, about 0.96, about 0.95, about 0.94, about 0.93, about 0.92, about 0.91, about 0.9, about 0.89, about 0.88, about 0.87, about 0.86, about 0.85, about 0.84, about 0.83, about 0.82, about 0.81, about 0.8, about 0.79, about 0.78, about 0.77, about 0.76, about 0.75, about 0.74, about 0.73, about 0.72, about 0.71, about 0.7, about 0.69, about 0.68, about 0.67, about 0.66, about 0.65, about 0.64, about 0.63, about 0.62, about 0.61, about 0.6, about 0.59, about 0.58, about 0.57, about 0.56, about 0.55, about 0.54, about 0.53, about 0.52, about 0.51, or about 0.5).
For example, in some aspects, a fold change ratio of less than about 0.9 and/or greater than about 1.1 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.9 and/or less than about 1.1 is indicative of the non-teratogenicity of the test compound. In some aspects, a fold change ratio of less than or equal to about 0.9 and/or greater than or equal to about 1.1 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.9 and/or less than about 1.1 is indicative of the non-teratogenicity of the test compound.
For example, in some aspects, a fold change ratio of less than about 0.89 and/or greater than about 1.11 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.89 and/or less than about 1.11 is indicative of the non-teratogenicity of the test compound. In some aspects, a fold change ratio of less than or equal to about 0.89 and/or greater than or equal to about 1.11 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.89 and/or less than about 1.1 is indicative of the non-teratogenicity of the test compound.
For example, in some aspects, a fold change ratio of less than about 0.88 and/or greater than about 1.12 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.88 and/or less than about 1.12 is indicative of the non-teratogenicity of the test compound. In some aspects, a fold change ratio of less than or equal to about 0.88 and/or greater than or equal to about 1.12 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.88 and/or less than about 1.12 is indicative of the non-teratogenicity of the test compound.
For example, in some aspects, a fold change ratio of less than about 0.87 and/or greater than about 1.13 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.87 and/or less than about 1.13 is indicative of the non-teratogenicity of the test compound. In some aspects, a fold change ratio of less than or equal to about 0.87 and/or greater than or equal to about 1.13 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.87 and/or less than about 1.13 is indicative of the non-teratogenicity of the test compound.
For example, in some aspects, a fold change ratio of less than about 0.86 and/or greater than about 1.14 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.86 and/or less than about 1.14 is indicative of the non-teratogenicity of the test compound. In some aspects, a fold change ratio of less than or equal to about 0.86 and/or greater than or equal to about 1.14 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.86 and/or less than about 1.14 is indicative of the non-teratogenicity of the test compound.
For example, in some aspects, a fold change ratio of less than about 0.85 and/or greater than about 1.15 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.85 and/or less than about 1.15 is indicative of the non-teratogenicity of the test compound. In some aspects, a fold change ratio of less than or equal to about 0.85 and/or greater than or equal to about 1.15 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.85 and/or less than about 1.15 is indicative of the non-teratogenicity of the test compound.
For example, in some aspects, a fold change ratio of less than about 0.84 and/or greater than about 1.16 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.84 and/or less than about 1.16 is indicative of the non-teratogenicity of the test compound. In some aspects, a fold change ratio of less than or equal to about 0.84 and/or greater than or equal to about 1.16 is indicative of the teratogenicity of the test compound and a fold change ratio of greater than about 0.84 and/or less than about 1.16 is indicative of the non-teratogenicity of the test compound.
A determination of a metabolite, fragment, adduct, deduct or loss thereof, may be identified using a physical separation method. In some embodiments, a metabolite, fragment, adduct, deduct or loss thereof, may be identified using a methodology other than a physical separation method. Such measurement methods may include, for example, colorimetric assays, enzymatic assays, or immunological assays. Immunological assays may include, for example, IF, RIA, ELISA and other immunoassays. Alternatively, certain biomarkers can be identified by, for example, gene expression analysis, including real-time PCR, RT-PCR, Northern analysis, and in situ hybridization.
The term “physical separation method” as used herein refers to method known to those with skill in the art sufficient to produce a profile of changes and differences in small molecules produced in hSLCs, contacted with a toxic, teratogenic or test chemical compound. In some embodiments, physical separation methods permit detection of cellular metabolites including but not limited to sugars, organic acids, amino acids, fatty acids, hormones, vitamins, and oligopeptides, as well as ionic fragments thereof and low molecular weight compounds (preferably with a molecular weight less than 3000 Daltons, and more particularly between 50 and 3000 Daltons). For example, mass spectrometry can be used. In particular embodiments, this analysis may be performed by liquid chromatography/electrospray ionization time of flight mass spectrometry (LC/ESI-TOF-MS). However it will be understood that metabolites as set forth herein can be detected using alternative spectrometry methods or other methods known in the art, including, but not limited to, any of those described herein.
For example, biomarkers are identified by methods including LC/ESI-TOF-MS and/or QTOF-MS. Metabolomic biomarkers are identified by their unique molecular mass and consistency with which the marker is detected in response to a particular toxic, teratogenic or test chemical compound; thus the actual identity of the underlying compound that corresponds to the biomarker is not required for the practice of this invention.
Biomarkers may be identified using, for example, Mass Spectrometry such as MALDI/TOF (time-of-flight), SELDI/TOF, liquid chromatography-mass spectrometry (LC-MS), gas chromatography-mass spectrometry (GC-MS), high performance liquid chromatography-mass spectrometry (HPLC-MS), capillary electrophoresis-mass spectrometry, nuclear magnetic resonance spectrometry, tandem mass spectrometry (e.g., MS/MS, MS/MS/MS, ESI-MS/MS etc.), secondary ion mass spectrometry (SIMS), and/or ion mobility spectrometry (e.g. GC-IMS, IMS-MS, LC-IMS, LC-IMS-MS etc.).
In some aspects, a gas phase ion spectrophotometer may be used. In other aspects, laser-desorption/ionization mass spectrometry may be used to identify biomarkers. For example, modern laser desorption/ionization mass spectrometry (LDI-MS) may be practiced in two main variations; matrix assisted laser desorption/ionization (MALDI) mass spectrometry and surface-enhanced laser desorption/ionization (SELDI). In MALDI, the analyte is mixed with a solution containing a matrix, and a drop of the liquid is placed on the surface of a substrate. The matrix solution then co-crystallizes with the biomarkers. The substrate is inserted into the mass spectrometer. Laser energy is directed to the substrate surface where it desorbs and ionizes the proteins without significantly fragmenting them. However, MALDI has limitations as an analytical tool. It does not provide means for fractionating the biological fluid, and the matrix material can interfere with detection, especially for low molecular weight analytes. In SELDI, the substrate surface is modified so that it is an active participant in the desorption process. In one variant, the surface is derivatized with adsorbent and/or capture reagents that selectively bind the biomarker of interest. In another variant, the surface is derivatized with energy absorbing molecules that are not desorbed when struck with the laser. In another variant, the surface is derivatized with molecules that bind the biomarker of interest and that contain a photolytic bond that is broken upon application of the laser. In each of these methods, the derivatizing agent generally is localized to a specific location on the substrate surface where the sample is applied. The two methods can be combined by, for example, using a SELDI affinity surface to capture an analyte (e.g. biomarker) and adding matrix-containing liquid to the captured analyte to provide the energy absorbing material.
Data from mass spectrometry may be represented as a mass chromatogram. A “mass chromatogram” is a representation of mass spectrometry data as a chromatogram, where the x-axis represents time and the y-axis represents signal intensity. In one aspect the mass chromatogram may be a total ion current (TIC) chromatogram. In another aspect, the mass chromatogram may be a base peak chromatogram. In other aspects, the mass chromatogram may be a selected ion monitoring (SIM) chromatogram. In yet another aspect, the mass chromatogram may be a selected reaction monitoring (SRM) chromatogram. In yet another aspect, the mass chromatogram may be an extracted ion chromatogram (EIC). In an EIC, a single feature is monitored throughout the entire run. The total intensity or base peak intensity within a mass tolerance window around a particular analyte's mass-to-charge ratio is plotted at every point in the analysis. The size of the mass tolerance window typically depends on the mass accuracy and mass resolution of the instrument collecting the data. As used herein, the term “feature” refers to a single small metabolite, or a fragment of a metabolite. In some embodiments, the term feature may also include noise upon further investigation.
A person skilled in the art understands that any of the components of a mass spectrometer, e.g., desorption source, mass analyzer, detect, etc., and varied sample preparations can be combined with other suitable components or preparations described herein, or to those known in the art. For example, a control sample may contain heavy atoms, e.g. 13C, thereby permitting the test sample to be mixed with the known control sample in the same mass spectrometry run. Good stable isotopic labeling is included.
A laser desorption time-of-flight (TOF) mass spectrometer may be used. In laser desorption mass spectrometry, a substrate with a bound marker is introduced into an inlet system. The marker is desorbed and ionized into the gas phase by laser from the ionization source. The ions generated are collected by an ion optic assembly, and then in a time-of-flight mass analyzer, ions are accelerated through a short high voltage field and let drift into a high vacuum chamber. At the far end of the high vacuum chamber, the accelerated ions strike a sensitive detector surface at a different time. Since the time-of-flight is a function of the mass of the ions, the elapsed time between ion formation and ion detector impact can be used to identify the presence or absence of molecules of specific mass to charge ratio. In one aspect, levels of biomarkers may be detected by MALDI-TOF mass spectrometry.
Methods of detecting biomarkers also include the use of surface plasmon resonance (SPR). The SPR biosensing technology may be combined with MALDI-TOF mass spectrometry for the desorption and identification of biomarkers.
A computer may be used for statistical analysis. Data for statistical analysis can be extracted from chromatograms (spectra of mass signals) using softwares for statistical methods known in the art. Statistics is the science of making effective use of numerical data relating to groups of individuals or experiments. Methods for statistical analysis are well-known in the art.
For example, the Agilent MassProfiler or MassProfilerProfessional software may be used for statistical analysis. Or, the Agilent MassHunter software Qual software may be used for statistical analysis. Alternative statistical analysis methods can be used. Such other statistical methods include the Analysis of Variance (ANOVA) test, Chi-square test, Correlation test, Factor analysis test, Mann-Whitney U test, Mean square weighted derivation (MSWD), Pearson product-moment correlation coefficient, Regression analysis, Spearman's rank correlation coefficient, Student's T test, Welch's T-test, Tukey's test, and Time series analysis.
In some aspects, signals from mass spectrometry can be transformed in different ways to improve the performance of the method. Either individual signals or summaries of the distributions of signals (such as mean, median or variance) can be so transformed. Possible transformations include taking the logarithm, taking some positive or negative power, for example the square root or inverse, or taking the arcsin (Myers, Classical and Modern Regression with Applications, 2nd edition, Duxbury Press, 1990).
In some aspects, statistical classification algorithms can be used to create a classification model in order to predict teratogenicity and non-teratogenicity of test compounds. Machine learning-based classifiers have been applied in various fields such as machine perception, medical diagnosis, bioinformatics, brain-machine interfaces, classifying DNA sequences, and object recognition in computer vision. Learning-based classifiers have proven to be highly efficient in solving some biological problems.
As used herein, a “training set” is a set of data used in various areas of information science to discover potentially predictive relationships. Training sets are used in artificial intelligence, machine learning, genetic programming, intelligent systems, and statistics. In all these fields, a training set has much the same role and is often used in conjunction with a test set.
As used herein, a “test set” is a set of data used in various areas of information science to assess the strength and utility of a predictive relationship. Test sets are used in artificial intelligence, machine learning, genetic programming, intelligent systems, and statistics. In all these fields, a test set has much the same role.
“Sensitivity” and “specificity” are statistical measures of the performance of a binary classification test. Sensitivity measures the proportion of actual positives which are correctly identified as such (e.g. the percentage of sick people who are correctly identified as having the condition). Specificity measures the proportion of negatives which are correctly identified (e.g. the percentage of healthy people who are correctly identified as not having the condition). These two measures are closely related to the concepts of type I and type II errors. A theoretical, optimal prediction can achieve 100% sensitivity (i.e. predict all people from the sick group as sick) and 100% specificity (i.e. not predict anyone from the healthy group as sick). A specificity of 100% means that the test recognizes all actual negatives—for example, in a test for a certain disease, all disease free people will be recognized as disease free. A sensitivity of 100% means that the test recognizes all actual positives—for example, all sick people are recognized as being ill. Thus, in contrast to a high specificity test, negative results in a high sensitivity test are used to rule out the disease. A positive result in a high specificity test can confirm the presence of disease. However, from a theoretical point of view, a 100%-specific test standard can also be ascribed to a ‘bogus’ test kit whereby the test simply always indicates negative. Therefore the specificity alone does not tell us how well the test recognizes positive cases. A knowledge of sensitivity is also required. For any test, there is usually a trade-off between the measures. For example, in a diagnostic assay in which one is testing for people who have a certain condition, the assay may be set to overlook a certain percentage of sick people who are correctly identified as having the condition (low specificity), in order to reduce the risk of missing the percentage of healthy people who are correctly identified as not having the condition (high sensitivity). This trade-off can be represented graphically using a receiver operating characteristic (ROC) curve.
The “accuracy” of a measurement system is the degree of closeness of measurements of a quantity to its actual (true) value. The “precision” of a measurement system, also called reproducibility or repeatability, is the degree to which repeated measurements under unchanged conditions show the same results. Although the two words can be synonymous in colloquial use, they are deliberately contrasted in the context of the scientific method. A measurement system can be accurate but not precise, precise but not accurate, neither, or both. For example, if an experiment contains a systematic error, then increasing the sample size generally increases precision but does not improve accuracy. Eliminating the systematic error improves accuracy but does not change precision.
The term “predictability” (also called banality) is the degree to which a correct prediction or forecast of a system's state can be made either qualitatively or quantitatively. Perfect predictability implies strict determinism, but lack of predictability does not necessarily imply lack of determinism. Limitations on predictability could be caused by factors such as a lack of information or excessive complexity.
In some aspects, a method of the present invention may predict the teratogenicity of a test compound with at least about 80% accuracy, at least about 85% accuracy, at least about 90% accuracy, or at least about 95% accuracy.
In some aspects, a method of the present invention may predict the teratogenicity of a test compound with at least about 80% sensitivity, at least about 85% sensitivity, at least about 90% sensitivity, or at least about 95% sensitivity.
In some aspects, a method of the present invention may predict the teratogenicity of a test compound with at least about 80% specificity, at least about 85% specificity, at least about 90% specificity, or at least about 95% specificity.
In some aspects, the methods described herein may utilize cystine determinations alone, or cystine in combinations with any of a variety of other metabolites, including, but not limited to one or more of the metabolites described herein. For example, a determination of a fold change in cystine alone can be used to classify teratogens, using a threshold of at least a 10% increase relative to the reference treatment.
In some aspects, the methods described herein may utilize ornithine determinations alone, ornithine in combinations with any of a variety of other metabolites, including, but not limited to one or more of the metabolites described herein. For example, a determination of a fold change in ornithine alone can be used to classify teratogens, using a threshold of about a 20% increase and/or an 18.5% decrease relative to the reference treatment.
In addition to determining altered ratios in the fold changes of ornithine to cystine, asymmetric dimethylarginine (ADMA) to cystine, and/or cystathionine to cystine, the accuracy of the methods described herein may be improved by further determining the fold change in one or more additional metabolites associated with hSLCs cultured in the presence of the test compound in comparison with hSLCs cultured in the absence of the test compound.
In some embodiments, a method may further include a determination of the ratio of the fold change in arginine, or fragment, adduct, deduct or loss thereof, to the fold change in ADMA, or fragment, adduct, deduct or loss thereof. In some aspects, a ratio of less than at least about 0.9 or greater than at least about 1.1 is indicative of the teratogenicity of the test compound and a ratio of greater than at least about 0.9 and less than at least about 1.1 is indicative of the non-teratogenicity of the test compound. See, for example, PCT/US2011/029471 and U.S. patent application Ser. No. 13/069,326 (“Predicting Human Developmental Toxicity of Pharmaceuticals Using Human Stem-Like Cells and Metabolomics”), West et al., 2010, Toxicol Appl Pharmacol; 247(1):18-27, and Kleinstreuer et al., 2011, Toxicol Appl Pharmacol; 257(1):111-121.
Additional metabolites may include, for example, one or more additional metabolites, two or more additional metabolites, three or more additional metabolites, four or more additional metabolites, five or more additional metabolites, six or more additional metabolites, seven or more additional metabolites, eight or more additional metabolites, nine or more additional metabolites, ten or more additional metabolites, eleven or more additional metabolites, twelve or more additional metabolites, thirteen or more additional metabolites, fourteen or more additional metabolites, or fifteen or more additional metabolites.
One or more additional metabolite may include a metabolite of a metabolic pathway selected from, for example, an alanine, aspartate and glutamate metabolic network; an arginine and proline metabolic network; an ascorbate and aldarate metabolic network; a citrate cycle; a cysteine and methionine metabolic network; a galactose metabolic network; a glutathione metabolic network; a glyoxylate and dicarboxylate metabolic network; a nicotinate and nicotinamide metabolic network; a pantothenate and coenzyme A biosynthesis pathway; a pentose and glucoronate interconversions pathway; a pentose phosphate pathway; a propanoate metabolic network; a pyruvate metabolic network; and/or a vitamin B6 metabolic network.
For example, one or additional metabolite may include a metabolite of the pantothenate and coenzyme A biosynthesis pathway, such as, for example, pyruvate, L-valine, dimethylmalate, pantoate, patothenate, phosphorpatothenoyl-L-cyteine, 5,6-dihydrouracil, N-carbamoyl-β-alanine, and/or coenzyme A.
For example, one or additional metabolite may include a metabolite of the glutathione metabolic network, such as, for example, 5-oxoproline, L-glutamate, glycine, L-γ-glutamylcysteine, glycine, dehydroascorbate, glutathionyl spermine, and/or L-ornithine. For example, one or additional metabolite may include a metabolite of the arginine and proline metabolic network, such as, for example, pyruvate, dimethlarginine, L-arginine, L-citrulline, glutamine, aspartate, L-argosuccinate, guanidino-acetate-phosphate, fumarate, sarcosine, 2-oxoarginine, pyruvate, 5-amino-pentanoate, linatine, pyrrole-2-carbosylate, putrescine, 6-oxo-1,4,5,6-tetrahydronicotinate, 2,6-dihydroxynictinate, fumarate, and/or GABA.
For example, one or additional metabolite may include a metabolite of the nicotinate and nicotinamide metabolic network, such as, for example, 6-oxo-1,4,5,6-tetrahydronicotinate, 2,6-dihydroxynictinate, and/or fumarate.
For example, additional metabolites may include one or more, two or more, three or more, four or more, or five or more additional metabolites selected from cystine, N1-acetyl spermidine, asymmetric dimethylarginine, cystathionine, 2′-deoxyuridine, GABA, malic acid, succinic acid, and aspartic acid.
For example, additional metabolites may include any one or more, any two or more, any three or more, any four or more, any five or more, any six or more, any seven or more, any eight or more, any nine or more, any ten or more, any eleven or more, any twelve or more, any thirteen or more, or any fourteen or more of the additional metabolites selected from methyl sulfonylacetonitrile; aspartic acid, N-acetyl spermidine; dimethyl-L-arginine; L-cystathionine; GABA; fumaric acid; valine; succinic acid; aspartic acid; pantoic acid; the metabolite having m/z of 215.1387, RT of 466, and ESI(+) polarity; the metabolite having m/z of 234.8904, RT of 246, and ESI(+) polarity; the metabolite having m/z of 251.0666, RT of 105, and ESI(+) polarity; and the metabolite having m/z of 403.0839, RT of 653, and ESI(+) polarity. In some aspects, all fold changes in fifteen metabolites is determined. See, Table 11 of PCT/US2011/029471 and U.S. patent application Ser. No. 13/069,326 (“Predicting Human Developmental Toxicity of Pharmaceuticals Using Human Stem-Like Cells and Metabolomics”), each of which is hereby incorporated by reference in its entirety.
The hSLC and metabolomics based methods of the present invention offer a significant advantage over other studies that use mouse or zebra fish-based models to determine toxicity and teratogenicity of chemical compounds.
The methods of the present invention may be used for classifying a test compound as a teratogen or a non-teratogen, for predicting the teratogenicity of a test compound, and/or for validating a test compound as a teratogen. The methods of the present invention may also serve as a high throughput screening tool in preclinical phases of drug discovery. In addition, this approach can be used to detect detrimental effects of environmental (heavy metals, industrial waste products) and nutritional chemicals (such as alcohol) on human development. Further, the methods of this invention can assist pharmaceutical, biotechnology and environmental agencies on decision-making towards development of compounds and critical doses for human exposure. The integration of chemical biology to embryonic stem cell technology also offers unique opportunities to strengthen understanding of human development and disease. Metabolomics of cells differentiated from hSLCs should serve similar roles and be useful for elucidating mechanisms of toxicity and disease with greater sensitivity for particular cell or tissue types, and in a human-specific manner.
Biomarker portfolios produced using the hSLC-dependent methods of this invention may also be used in high throughput screening methods for preclinical assessment of drug candidates and lead compounds in drug discovery. This aspect of the inventive methods produces minimal impact on industry resources in comparison to current developmental toxicology models, since implementation of this technology does not require experimental animals. The resulting positive impact on productivity enables research teams in the pharmaceutical industry to select and advance compounds into exploratory development with greater confidence and decreased risk of encountering adverse developmental effects.
The present invention includes a kit for identifying and/or measuring one or more metabolites. In some aspects, the kit may be for the determination of a metabolite by a physical separation method. In some aspects, the kit may be for the determination of a metabolite by a methodology other than a physical separation method, such as for example, a colorimetric, enzymatic, immunological methodology. In some aspects an assay kit may also include one or more appropriate negative controls and/or positive controls. Kits of the present invention may include other reagents such as buffers and solutions needed to practice the invention are also included. Optionally associated with such container(s) can be a notice or printed instructions. As used herein, the phrase “packaging material” refers to one or more physical structures used to house the contents of the kit. The packaging material is constructed by well known methods, preferably to provide a sterile, contaminant-free environment. As used herein, the term “package” refers to a solid matrix or material such as glass, plastic, paper, foil, and the like, capable of holding within fixed limits a polypeptide. Kits of the present invention may also include instructions for use. Instructions for use typically include a tangible expression describing the reagent concentration or at least one assay method parameter, such as the relative amounts of reagent and sample to be admixed, maintenance time periods for reagent/sample admixtures, temperature, buffer conditions, and the like.
In some aspects, a kit may be a packaged combination comprising the basic elements of a first container comprising, in solid form, a specific set of one or more purified metabolites, as described herein, and a second container comprising a physiologically suitable buffer for resuspending the specific subset of purified metabolites.
The present invention is illustrated by the following examples. It is to be understood that the particular examples, materials, amounts, and procedures are to be interpreted broadly in accordance with the scope and spirit of the invention as set forth herein.
EXAMPLES Example 1 Establishment and Assessment of a New Human Embryonic Stem Cell-Based Biomarker Assay for Developmental Toxicity ScreeningWith this example a metabolic biomarker-based in vitro assay utilizing human embryonic stem (hES) cells was developed to identify the concentration of test compounds that perturbs cellular metabolism in a manner indicative of teratogenicity. This assay is designed to aid the early discovery-phase detection of potential human developmental toxicants. In this study, metabolomic data from hES cell culture media was used to assess potential biomarkers for development of a rapid in vitro teratogenicity assay. hES cells were treated with pharmaceuticals of known human teratogenicity at a concentration equivalent to their published human peak therapeutic plasma concentration. Two metabolite biomarkers (ornithine and cystine) were identified as indicators of developmental toxicity. A targeted exposure-based biomarker assay using these metabolites, along with a cytotoxicity endpoint, was then developed using a 9-point dose response curve. The predictivity of the new assay was evaluated using a separate set of test compounds. To illustrate how the assay could be applied to compounds of unknown potential for developmental toxicity, an additional 10 compounds were evaluated that do not have data on human exposure during pregnancy, but have shown positive results in animal developmental toxicity studies. The new assay identified the potential developmental toxicants in the test set with 77% accuracy (57% sensitivity, 100% specificity). The assay had a high concordance (≥75%) with existing in vivo models, demonstrating that the new assay can predict the developmental toxicity potential of new compounds as part of discovery phase testing and provide a signal as to the likely outcome of required in vivo tests.
This example describes the development of a rapid, reproducible, biomarker-based screen for developmental toxicity testing designed to identify the exposure level at which a test compound perturbs metabolism in a manner predictive of developmental toxicity. Perturbation of two metabolites, ornithine and cystine, in response to the test compound was assessed across nine independent experimental replications to ensure repeatability across experiments and liquid chromatography high resolution mass spectrometry (LC-HRMS) systems. Using the ornithine/cystine ratio (o/c ratio), we developed a rapid, targeted assay that measured changes in metabolism and cellular viability across a 9-point dose response curve to determine the exposure level at which a test compound perturbs metabolism in a manner associated with developmental toxicity potential. To assess the predictivity of the assay for known human teratogens in the training and test sets of compounds, the exposure level where a compound was predicted to have developmental toxicity potential was scored against the compound's human peak plasma in vivo concentration (Cmax) following therapeutic doses. The Cmax value in this case is used as a benchmark exposure level to aid in interpreting the performance of the assay as it is the highest concentration a human would normally be exposed to under therapeutic circumstances and we would expect to detect developmental toxicity at this exposure level.
However, application of the assay in the discovery stage of a compound's development would not require this Cmax information, and a test compound's teratogenic potential is based on the exposure level at which a test compound perturbs metabolism in a manner indicative of teratogenicity. The design and sensitivity of the assay allows for identification of teratogenic potential at non-cytotoxic levels of the test compound, by negating the confounding effects of changes in metabolite abundance due strictly to cytotoxicity. The ability to identify developmental toxicity in the absence of cytotoxicity at a variety of exposure levels is a key strength of the assay and distinguishes it from existing in vitro assays.
Useful Terms and DefinitionsTeratogenicity Threshold. A threshold of metabolic perturbation that is associated with the potential for teratogenesis. The threshold was empirically determined to be 0.88 for the targeted biomarker assay using the training set results. This threshold was applied to all test set and unknown compounds evaluated using the assay.
Ornithine/Cystine Ratio (0/C Ratio). The fold change of ornithine (Orn) for treatment x divided by the fold change of cystine (Cyss) for treatment x.
Teratogenicity Potential. Interpolated exposure level (concentration) of a test compound where the dose response curve for the o/c ratio or cell viability crosses the teratogenicity threshold. Exposure levels greater than this concentration are associated with teratogenicity.
Accuracy. Number of correct predictions divided by the number test compounds evaluated.
Sensitivity. Detection of teratogens, True Positives/(False Negatives+True Positives).
Specificity. Detection of non-teratogens, True Negatives/(True Negatives+False Positives).
Training Set. Set of compounds that have well established human developmental toxicity information used to identify biomarkers of developmental toxicity. This set of compounds was tested in both phases of the study and used to set the teratogenicity threshold.
Test Set. Set of compounds with well-established human developmental toxicity information that were not used to identify the biomarkers, but used to evaluate the predictivity of the biomarkers of developmental toxicity. This set of compounds was used to evaluate the performance of the targeted biomarker assay and the teratogenicity threshold set using the training set.
Application Set. Set of compounds with poorly defined human developmental toxicity information used to demonstrate application of the assay. These compounds are not classified as a teratogen or non-teratogen based on their Cmax since human teratogenicity is unknown at this concentration.
Materials and MethodsDevelopment and evaluation of the targeted biomarker-based assay was conducted in two phases. In the first phase (Phase 1), the predictive potential of two previously identified predictive biomarkers (ornithine and cystine, Kleinstreuer et al., 2011, Toxicol Appl Pharmacol; 257:111-121) was characterized across nine independent experimental replications (experimental blocks) of the training set using untargeted metabolomic methods. In the second phase (Phase 2), the predictive biomarkers were used to develop a rapid turnaround, targeted, exposure-based assay for compound prioritization based on teratogenicity potential. The predictivity of the new assay was evaluated using the original training set as well as an independent test set of compounds.
Test Chemical Selection and Classification. A total of 46 compounds were used to evaluate the ability of ornithine, cystine and the o/c ratio to predict developmental toxicity in two experimental phases. These 46 compounds were divided into three groups, named the training, test, and application sets. The training set was a set of compounds that have well established human developmental toxicity information used to identify biomarkers of developmental toxicity. The test set was a set of compounds with well-established human developmental toxicity information that were not used to identify the biomarkers, but used to evaluate the predictivity of the biomarkers of developmental toxicity. The application set was a set of compounds with poorly defined human developmental toxicity information used to demonstrate application of the assay. These compounds are not classified as a teratogen or non-teratogen based on their Cmax since human teratogenicity is unknown at this concentration.
The training set consisted of 23 well characterized pharmaceutical compounds (11 known human non-teratogens and 12 known human teratogens, Table 2) and was previously used to build a computational model and identify biomarkers predictive of teratogenicity (Kleinstreuer et al., 2011, Toxicol Appl Pharmacol; 257:111-121). This training set was utilized in both experimental phases. To assess the predictive capacity of the targeted biomarker assay developed in these studies, an additional test set of 13 well characterized pharmaceutical compounds (6 known human non-teratogens and 7 known human teratogens, Table 3) was used in the second experimental phase to evaluate the predictivity of the new assay. The final set of compounds (the application set, Table 4) consists of 10 compounds that do not have conclusive developmental toxicity data available on exposure during human pregnancy, but do have animal data available on developmental toxicity potential. A two-class system of compound classification (teratogen and non-teratogen) was applied for assay development, focusing the teratogenicity classification strictly on observed human risk associated with each chemical. Compounds were purchased from Sigma-Aldrich (St. Louis, Mo.), except for amprenavir, bosentan, entacapone (Toronto Research Chemicals, Toronto, Ontario, Canada), lapatinib (Chemie Tek, Indianapolis, Ind.), cidovofir and ramelteon (Selleck Chemicals, Houston, Tex.).
Undifferentiated hES Cell Line Maintenance (Phases 1 and 2). WA09 hES cells were obtained from the WiCell Research Institute (Madison, Wis.) and were maintained in feeder free conditions using mTeSR1 media (StemCell Technologies, Inc., Vancouver, BC, Canada) on hESC-qualified Matrigel (BD Biosciences, San Jose, Calif.) coated 6-well plates. To maintain the undifferentiated stem cell population, differentiated colonies were removed daily through aspiration and media was replaced. Additionally, the hES cells were only used in experiments up to passage 40 and were karyotyped approximately every 10 passages to minimize and monitor the potential for genetic instability. hES cells were passaged at 90-95% confluency (approximately every 7 days) using Versene (Life Technologies, Grand Island, N.Y.). Cell cultures were maintained at 37° C. under 5% CO2.
96-well hES Cell Plating (Phases 1 and 2). All experimental treatments were carried out in 96-well plates. To minimize plating variability and increase reproducibility, hES cells were plated as a single cell suspension and maintained in an undifferentiated state during compound exposure. Prior to plating in the 96-well plates, hES cells were removed from a 6-well plate using TrypLE (Life Technologies). The cells were washed with DMEM/F12 (Life Technologies) and resuspended in mTeSR1 containing 10 μM Y27632 Rho-associated kinase (ROCK) inhibitor (Merck KGaA/Calbiochem, Darmstadt, Germany). The ROCK inhibitor is added to the plating media to increase plating efficiency by decreasing dissociation-induced apoptosis. The inner 60 wells of hESC-qualified Matrigel coated 96-well plates were seeded at a density of 100,000 cells per well. The outer wells of the plate contained an equal volume media to minimize differences in humidity across the plate. Compound exposure began 24 hours after plating.
Phase I hES Cell Compound Exposure. hES cells were treated with a test compound at a single concentration equivalent to the compound's published therapeutic Cmax. The therapeutic Cmax was used because it is considered to be a physiologically relevant exposure level and has been correlated with the developmental effect of the compound (National Research Council, 2000, “Scientific frontiers in developmental toxicology and risk assessment,” Washington, D.C.: The National Academies Press). For six compounds (5-fluorouracil, aminopterin, busulfan, cytosine arabinoside, hydroxyurea and methotrexate) an experimentally determined IC30 was used in place of the Cmax value due to greater than 30% cytotoxicity at the Cmax exposure level. This was done to ensure that enough cells were present at the time of sample collection to provide a signal for LC-HRMS analysis. For test compound exposure, all compound stock solutions, with the exception of valproic acid, were made with DMSO (Sigma-Aldrich). Valproic acid was insoluble in DMSO at the concentrations used in this study, so it was diluted in mTeSR1 containing 0.1% DMSO. Each 96-well plate included media controls with and without test compound, 0.1% DMSO solvent control cells and cells exposed to a single concentration of eight different test compounds (
Phase 2 hES Cell Compound Exposure. The predictivity of the targeted biomarker assay was evaluated in the original training set as well as an independent test set (Tables 2 and 3). The assay was additionally applied to the application set of compounds (Table 4) to demonstrate utility when human teratogenicity is unknown. The standard compound exposure levels used for most compounds were nine, 3-fold dilutions ranging from 0.04 μM-300 μM (
Sample Preparation (Phases 1 and 2). High molecular weight constituents (>10 KDa) of the spent media samples were removed using a Millipore Multiscreen Ultracel-10 filter plate (EMD Millipore, Billerica, Mass.). Prior to sample filtration, the filter plate was washed with 0.1% NaOH to remove a known contaminant polymer. The plate was then rinsed twice with HPLC-grade water to remove residual polymers and NaOH. Spent media samples were added to the washed filter plate. In Phase 1, samples were spiked with 13C6 labeled arginine. Samples were centrifuged at 2,000×g at 4° C. for 200 minutes. The filtrate was collected and concentrated overnight in a Savant High Capacity Speedvac Plus Concentrator. The concentrated sample was resolubilized in a 1:1 0.1% formic acid in water: 0.1% formic acid in acetonitrile mixture containing 13C5 labeled glutamic acid (Cambridge Isotope Laboratories). The 13C labeled compounds were used as internal standards to track preparatory efficiency and track LC-HRMS performance.
Phase 1 Mass Spectrometry. LC-HRMS data was acquired for nine biological replications on three separate LC-HRMS systems with three replications evaluated on each system. Each system consisted of an Agilent 1290 Infinity LC system interfaced either with an Agilent G6520A QTOF high resolution mass spectrometer (QTOF HRMS), an Agilent G6530A QTOF HRMS, or an Agilent G6224A TOF HRMS system (Agilent Technologies, Santa Clara, Calif.). To facilitate separation of biological small molecules with a wide range of structures and to allow increased retention of hydrophilic species, Hydrophilic Interaction Liquid Chromatography (HILIC) was utilized. A Luna HILIC column (Phenomenex, Torrance, Calif.) with dimensions 3×100 mm and 3 μm particle size was used and maintained at 30° C. Sample (2 μL) was injected and the data acquisition time was 23 minutes (min) at a flow rate of 0.5 ml/min, using a 17 min solvent gradient with 0.1% formic acid in water (Solvent A) and 0.1% formic acid in acetonitrile (Solvent B). Electrospray ionization was employed using a dual ESI source. The scan range of the instrument was 70-1600 Da. Data acquisition was performed with MassHunter Acquisition software (version B 04.00, Agilent Technologies) using high-resolution exact mass conditions and each set of samples was run first under ESI positive polarity then under ESI negative polarity conditions.
Phase 2 Mass Spectrometry. Data was acquired to assess the performance of the targeted biomarker assay using two instrument platforms. Ultra high performance liquid chromatography (UPLC)-HRMS data acquisition for each compound was performed using one of two systems. System 1 consisted of an Agilent 1290 Infinity LC system interfaced with an Agilent G6520A QTOF HRMS. System 2 used the same model LC system interfaced with an Agilent G6224A TOF HRMS. A Waters Acquity UPLC BEH Amide 2.1×50 mm 1.7 μm particle size column (Waters, Milford, Mass.) maintained at 40° C. was applied for separation of metabolites. A solvent gradient with 0.1% formic acid in water (Solvent A) and 0.1% formic acid in acetonitrile (Solvent B) at a flow rate of 1.0 ml/min was used and 2 μL of sample was injected. Electrospray ionization was employed using a dual ESI source operated in positive ionization mode only. The mass range of the instrument was set to 60-1600 Da and data was acquired over 6.5 min using MassHunter Acquisition software (version B 04.00). Identification of cystine and ornithine metabolites in samples was previously confirmed by comparison of their collision-induced dissociation mass spectra to reference standards (Sigma-Aldrich).
Peak Detection (Phases 1 and 2). Agilent raw data files were converted to the open source mzData file format using MassHunter Qualitative Analysis software version 5.0 (Agilent Technologies). During the conversion process, deisotoping (+1 charge state only) was performed on the centroid data and peaks with an absolute height less than 200 were excluded from analysis. Peak picking and feature creation were performed using the R package “xcms” (Smith et al., 2006, Anal Chem; 78:779-787). Mass features (peaks) were detected using the centwave algorithm. Deviations in retention times were corrected using the obiwarp algorithm that is based on a non-linear clustering approach to align the data from the LC-MS samples. Mass feature bins or groups were generated using a density based grouping algorithm. After the data had been grouped into mass features, missing features were integrated based on retention time and mass range of a feature bin using the iterative peak filling. Feature intensity is based on the Mexican hat integration values of the feature extracted ion chromatograms.
Ornithine/Cystine Ratio Calculation. In both phases of the study, every 96-well plate of samples contained a reference treatment (0.1% DMSO) to allow compensation for the differences in LC-MS instrument response over time. Relative fold changes were calculated for each metabolite by dividing the integrated area of each sample within a treatment level by the median integrated area of the reference treatment (DMSO) samples to produce a normalized value for both metabolites in each sample within a plate of cell culture samples. The o/c ratio was calculated for each sample in a treatment by dividing the reference normalized value of ornithine by the reference normalized value of cystine. In Phase 2, a four-parameter log-logistic model of dose response was fit using the mean o/c ratio value of each concentration using the R package “drc” (Ritz and Streibig, 2005, J Statistical Software; 12:1-22).
Teratogenicity Threshold Selection (Phases 1 and 2). Classification of teratogenicity was based on the premise that a threshold of metabolic perturbation could be identified for individual metabolites that is associated with developmental toxicity. This threshold of metabolic change is called the teratogenicity threshold and is a measure of the magnitude of metabolic perturbation required to differentiate teratogens from non-teratogens. The teratogenicity threshold was empirically generated for ornithine, cystine, and the o/c ratio by iteration through a range from 10% to 25% change, to identify a one-sided or two-sided asymmetrical threshold that was able to classify the training set with the greatest accuracy and highest sensitivity. In the case of a tie in classification accuracy and sensitivity between one-sided and two-sided thresholds, one-sided thresholds were given priority to favor simplicity. A teratogenicity threshold was determined for each phase of the study, since the assays performed in Phase 1 used only a single concentration of each compound and the targeted biomarker assay developed in Phase 2 utilized an exposure based approach. The teratogenicity threshold was determined in Phase 2 using only the results from the training set. This threshold was then applied to the results from the test and application sets.
Phase 1 Prediction of Developmental Toxicity Potential. A test compound was classified as a developmental toxicant if the mean of the change in the abundance in the treated sample compared to the reference treatment (DMSO) across the nine experimental replications for either metabolite or the o/c ratio exceeded its respective teratogenicity threshold at the concentration tested. The predictive accuracy (correct prediction), sensitivity (true positive rate), and specificity (true negative rate) were based on scoring the predicted result (teratogen or non-teratogen) against the known human teratogenicity of the compound.
Phase 2 Prediction of Developmental Toxicity Potential. For test compounds with unknown developmental toxicity potential, the targeted biomarker assay is utilized to identify the exposure level where a test compound perturbs metabolism in a manner indicative of teratogenicity and does not require any pharmacokinetic information (e.g., Cmax).
In order to assess the predictivity of the assay in the training and test sets, the teratogenicity potential concentrations determined from the o/c ratio and cell viability were used to classify the teratogenicity of the test compound relative to the human therapeutic Cmax concentrations. This approach was not applied to the application set since the developmental toxicity potential of these compounds in humans is unknown. The logic of scoring a test compound as a teratogen or non-teratogen using the human therapeutic Cmax is based on the paradigm that exposure is a critical factor in teratogenesis, and that a known human teratogen would likely perturb cellular metabolism at or below the highest exposure that is likely to occur at the therapeutic circulating levels. If perturbation of the o/c ratio was exhibited at concentrations greater than the compound's Cmax concentration (
Comparison of the Targeted Biomarker Assay to Other Developmental Toxicity Tests. A literature review compared the developmental toxicity prediction of the in vivo rodent and rabbit models and three in vitro screens (the European Centre for the Validation of Alternative Methods (ECVAM)-evaluated mouse embryonic stem cell test (mEST), the zebrafish embryotoxicity test (ZET), and the post-implantation rat whole embryo culture (WEC) test) for the compounds tested in the targeted biomarker assay. The predictions made in these assays using each original author's classification methods were used for comparison and the data was not reinterpreted. The other in vitro systems employ a three class classification system (non-, weak/moderate, and strong teratogens; Brown, 2002, Altern Lab Anim; 30:177-198), compared to the two class system used in this study. Thus, in order to compare the results from the targeted biomarker assay to other models, the predicted results from these assays needed to be modified to a two class system. Compounds that were predicted to be either weak/moderate or strong teratogens were both labeled as a predicted teratogen. The accuracy, sensitivity and specificity were calculated for each assay by scoring the predicted result against the known human teratogenicity. These values were additionally calculated for the targeted biomarker assay for the specific set of compounds that had been tested in the other model system. Concordance between the targeted biomarker assay and the other above-mentioned models was evaluated by comparing the classification of teratogen or non-teratogen within the common treatments of each comparison.
ResultsPhase 1 Model Confirmation and Characterization of Metabolites Predictive of Developmental Toxicity. The first phase of this study was conducted to confirm the predictivity of individual metabolites. Characterization of the predictive metabolites led to the development of the new targeted biomarker assay described in the second phase of this study. Previously, a training set of 23 pharmaceutical compounds (Table 2) was utilized to identify a metabolic signature capable of predicting teratogenicity in vitro (Kleinstreuer et al., 2011, Toxicol Appl Pharmacol; 257:111-121). The metabolites that exhibited a statistically significant change upon treatment with teratogens, and lacked a response in non-teratogens, were characterized for their ability to classify developmental toxicants using a simple fold change threshold. Of these metabolites, ornithine and cystine were identified as metabolites that are representative of the previously applied metabolic signature that was highly predictive of developmental toxicity. The capacity of each of these two metabolites to classify developmental toxicants was characterized by determining a teratogenicity threshold based on the fold change of cells treated with a test compound versus the reference treatment (0.1% DMSO) of each metabolite. The threshold was used to evaluate the classification accuracy of each metabolite within the training set.
Ornithine and cystine each exhibited characteristics amenable to rapid evaluation of the potential for a test compound to perturb metabolism in manner consistent with teratogenicity. Both metabolites are highly abundant in spent cell culture media from hES cells and show changes in their abundance in response to treatment that were reproducibly measured on multiple LC-HRMS instruments. To confirm these initial observations, and the reproducibility of the approach, the metabolites were further evaluated in a study that encompassed 9 independent experimental replications (blocks) of the training set. The secreted metabolite ornithine was able to distinguish teratogens from non-teratogens with 83% accuracy (Table 5) using a two-sided threshold consisting of either an 18.5% decrease or 20% increase in accumulation of ornithine (
(such as 5-fluorouracil, cytosine arabinoside, methotrexate, and valproic acid) but increased when cells were exposed to the related non-cytotoxic teratogens all-trans retinoic acid and 13-cis retinoic acid.
Next, the possibility that the fold changes in the ratio of ornithine and cystine would be more predictive than their individual fold changes was evaluated. When the ornithine fold change was divided by the cystine fold change (i.e., the o/c ratio), the resulting ratio was able to correctly classify 91% (Table 5) of the training set (
Phase 2 Development and Evaluation of a Targeted Biomarker Assay to Predict Developmental Toxicity Associated with Exposure.
Targeted LC-HRMS Method Development. In the second phase of this study, a targeted biomarker-based assay was developed using the metabolites confirmed in Phase 1. Since toxicity is a function of both the chemical agent and exposure level, the high level of predictivity associated with a threshold of toxicity of the o/c ratio provided an opportunity for development of a targeted, rapid, teratogenicity assay. To that end, a short and reproducible analysis method was developed and optimized for fast-turnaround analysis of relative changes in ornithine and cystine abundance in hES cell spent media samples. In contrast, the untargeted metabolomic methods that had been previously used were designed to analyze a wider breadth of small molecules, and thus required a lengthy chromatographic separation. The prior platform also depended upon two data acquisitions for each sample, in positive and negative ionization modes. Focusing on the chromatographic separation, ionization and detection of ornithine and cystine only, a new, targeted method was designed specifically to more rapidly measure the relative changes of these metabolites observed in the hES cell model system. The new UPLC-HRMS method was developed and assessed using spent media samples (prepared as previously described) for added speed, sensitivity, and retention time reproducibility for measurements of ornithine and cystine. This resulted in a significant reduction in assay turn-around time. The data acquisition time for each sample was reduced from 23 to 6.5 minutes, providing a four-fold increase in LC-HRMS throughput. The positive ionization mode was preferentially amenable for detection of these metabolites, thereby eliminating the need for the negative mode, which further reduced the total analysis time by half for each sample batch, thus increasing total instrument throughput eight-fold. Method reproducibility was evaluated across 17 batches performed over 120 days using reference treatment samples (DMSO treated cells). The average CV for the integrated area of the internal standards and endogenous metabolites was <5% and <8%, respectively, demonstrating that the method performs in a reproducible manner.
Identification of the Teratogenicity Threshold. Based on the high classification accuracy achieved in Phase 1 using a defined teratogenicity threshold, a 9-point concentration curve was used to classify developmental toxicity potential based on a range of exposures. The teratogenicity threshold was optimized using the Phase 2 training set data by selecting a threshold that produced the highest accuracy of prediction with the greatest sensitivity. The predicted teratogenicity potential concentration was compared to the therapeutic Cmax to score the performance and classification accuracy of this new assay design (described in
Evaluation of the Targeted Biomarker Assay Performance based on the Test Set Predictions. The teratogenicity threshold identified using the training set was applied to the test set of compounds to assess the predictivity of the targeted biomarker assay developed in this study. The test set consisted of 13 compounds not included in the training set with known human teratogenicity, having FDA pregnancy classifications of B, D and X. The teratogenicity potential concentration of each compound for the o/c ratio was scored against the compound's therapeutic Cmax. The test set was classified with 77% accuracy (100% specificity, 57% sensitivity, Table 7). The o/c ratio incorrectly classified the teratogens bosentan, lapatinib and lovastatin (Table 8,
Comparison of the Ornithine/Cystine Ratio and Cell Viability. Because the metabolites that make up the o/c ratio are measured in spent cell culture media, the treated cells were available to perform cell viability analysis. The cell viability results were compared to the o/c ratio to determine if the change in the ratio was due to cell death or if it was due to metabolic changes unrelated to changes in cell viability. The viability results were evaluated to determine classification performance using an approach similar to the o/c ratio (
When applied to the training and test sets, the o/c ratio was 26% and 15% more accurate, respectively, than viability alone for the prediction of development toxicity (Table 7). Both the o/c ratio and cell viability assay correctly classify non-teratogens with respect to the Cmax having 100% specificity, however they differ in their ability to discriminate teratogens (Table 7). The o/c ratio is 50% more sensitive in the detection of teratogens than viability alone in the training set and 28% more sensitive in the test set (Table 7). Additionally, the o/c ratio is able to classify both cytotoxic and non-cytotoxic teratogens correctly. The decrease in false negatives provided by the o/c ratio is related to the assay's measurement of metabolic perturbation that can occur independent of changes in cell viability.
Highlighted in
Application of the 0/C Ratio to Compounds with Unknown Human Teratogenicity. The targeted biomarker assay was applied to an application set of 10 compounds that have unknown human developmental toxicity outcomes. Since the human developmental toxicity of these compounds is unknown, the Cmax approach (illustrated in
Assay Performance (Comparison to Other Assays). The developmental toxicity predictions based on the o/c ratio for the training and test sets were compared to published results from other model systems (Table 10). The developmental toxicity predictions from the model systems presented in Table 10 for the application set are summarized in Supplementary Table 1. For the combined 36 training and test set compounds, comparisons were made on a model system-by-system basis using only the treatments evaluated in both the targeted biomarker assay and each model system it was being compared to. The results of the comparisons (Table 11) indicate that the o/c ratio described here is a more accurate predictor of human developmental toxicants than the other model systems considered. The increase in accuracy is due to a lower false positive rate (increased specificity) of the o/c ratio in each comparison with significant increase in specificity over other in vitro systems such as mEST and WEC, as well as a moderate gain in sensitivity. Interestingly, the o/c ratio is able to correctly classify the non-teratogens caffeine and retinol and teratogens warfarin and D-penicillamine, where the majority of other model systems fail. There is a high degree of concordance (≥75%) between the teratogenicity prediction of the o/c ratio and the in vivo rodent and rabbit models as well as the ZET (Table 11). Concordance is lower between the o/c ratio and the mEST and WEC (67% and 69%, respectively, Table 11). The reason for lower concordance between the o/c ratio and these in vitro models is due to the high accuracy of the targeted biomarker assay.
The present assay has been developed to address the need for more accurate, rapid, and less expensive alternatives to animal testing. Our goal was to provide toxicologists with a new and biologically germane tool to aid in compound prioritization prior to the currently required in vivo testing and as part of emerging multi-tiered testing strategies. Undifferentiated hES cells represent a simple and elegant test system for modeling a test compound's developmentally toxic effects on human cells at the very earliest stages of development, which in some cases can lead to implications of the compound's effects in later stage fetal development as well. A developmental toxicity test based on hES cells reduces the risk of false-negatives due specifically to inter-species differences in developmental pathways and pharmacokinetics (Scott et al., 2013, Toxicol Lett; 219:49-58). The present example modifies an untargeted metabolomics-based developmental toxicity assay to decrease complexity and increase throughput by focusing on two biologically relevant metabolites that can accurately model human toxic response over a wide range of exposure levels.
This example demonstrates that a certain degree of metabolic perturbation can be used to predict a test compound's potential to cause developmental toxicity. The assay of this example uses a multi-exposure approach that allows for a look at cellular response over a large range of exposure levels. Application of the teratogenicity threshold to this approach allowed the use of changes in metabolism at increasing exposure levels to identify the concentration at which metabolism was altered in a manner indicative of potential teratogenicity. The model created here allows the comparison of changes in a metabolic ratio of ornithine and cystine to cell viability to identify the exposure level where changes in metabolism are likely to lead to teratogenicity and relate it to cell death. The combined evaluation of cell viability and changes in metabolism allow this assay to also identify when exposure could lead to developmental toxicity due to cell death or possible embryo toxicity. The o/c ratio can discriminate between teratogens and non-teratogens with a combined 89% accuracy in the training and test sets using the teratogenicity threshold set in Phase 2 (Table 11).
Analysis of metabolites is a critical process in understanding mechanisms of toxicity since metabolites play critical roles in the maintenance of homeostasis and signaling. Perturbation of individual metabolites has the ability to disrupt normal developmental processes. Alterations in metabolite abundance can occur via mechanisms independent of protein and transcript abundance such as allosteric interaction of a compound or compound's metabolite with an enzyme, defects in post-translational modification, disrupted protein-protein interactions and/or altered transport. Changes in metabolism, as measured in the spent medium of cell culture systems, yield a distinguishable “metabolic footprint,” which is a functional measure of cellular metabolism that can be used to evaluate response to treatment. The perturbation of biochemical pathways that contain ornithine and cystine as reactants or products have been experimentally associated with mechanisms of teratogenesis. Extra-cellularly, or within the secretome measured by our assays, cystine predominates over cysteine due to the oxidative state of the medium. Cystine is rapidly converted to cysteine once it is imported into the intracellular environment and is part of the cystine/cysteine thiol redox couple, a critical component of a cell's regulatory capacity to handle reactive oxygen species (ROS). Its role has been investigated with regard to its capacity to modulate differentiation, proliferation, apoptosis, and other cellular events that may lead to teratogenesis (Hansen, 2006, Birth Defects Res C Embryo Today; 78:293-307). A broad spectrum of teratogens including pharmaceuticals, pesticides, and environmental contaminants are suspected of creating ROS or disrupting cellular mechanisms that maintain the appropriate balance of a cell's redox state, which can lead to adverse effects on developmental regulatory networks as a mechanism of action of developmental toxicity (Hansen, 2006, Birth Defects Res C Embryo Today; 78:293-307; Kovacic and Somanathan, 2006, Birth Defects Res C Embryo Today; 78:308-325). It has been hypothesized that a major mechanism of thalidomide teratogenesis and its species specific manifestation of developmental toxicity is related to ROS related up-regulation of apoptotic pathways during limb formation (Hansen, 2006, Birth Defects Res C Embryo Today; 78:293-307). The measurement of cystine in this assay provides insight into a cell's redox status. When cystine's uptake is perturbed, it can act as a biomarker, indicating a disruption in the cell's ability to signal using ROS related pathways.
The second metabolite in this assay is ornithine, which is secreted by the hES cells during culture. Ornithine is formed as a product of the catabolism of arginine into urea, is critical to the excretion of nitrogen, and is a precursor to polyamines. Catabolism of ornithine is impacted by the teratogen all-trans retinoic acid, which is a suppressor of the transcription of ornithine decarboxylase (ODC), leading to increased ornithine secretion which in turn inhibits polyamine synthesis (Mao et al., 1993, Biochem 295:641-644). It is also clear that ODC plays an important role in development, since a mouse model with ODC knocked out leads to disruption of very early embryonic stages and is lethal to the developing embryo (Pegg, 2009, IUBMB Life; 61:880-894). Alterations in ornithine levels could lead to the disruption in polyamine metabolism, which is critical for cellular growth and differentiation during human development (Kalhan and Bier, 2008, Annu Rev Nutr; 28:389-410).
Only one of the 23 compounds in the training set (diphenylhydantoin) and three of the 13 compounds in the test set (bosentan, lapatinib, and lovastatin) were misclassified in the targeted biomarker assay (Tables 6 and 8). All four of these compounds exhibited a change in the o/c ratio indicative of teratogenicity; however the teratogenicity potential concentration is higher than the therapeutic Cmax, which was set as a marker of biological relevance for exposure level. For discovery compounds that will not have an established Cmax value, these changes in the o/c ratio can be used as a signal regarding the teratogenic potential of the compound. While epidemiological studies have shown an association between diphenylhydantoin and birth defects, there have been no such studies describing the incidence of birth defects following bosentan, lapatinib and lovastatin exposure during pregnancy. No case reports have been published regarding birth defects in infants exposed to bosentan or lapatinib during pregnancy and only a handful of reports describing malformations following lovastatin exposure during early pregnancy (TERIS).
In vivo rat developmental toxicity studies have identified a lowest observed adverse effect level (LOAEL) for lovastatin of 100 mg/kg body weight per day during organogenesis (Lankas et al., 2004, Birth Defects Res B Dev Reprod Toxicol; 71:111-123). Interestingly, this level of exposure results in a Cmax around 1.5 μM (Lankas et al., 2004, Birth Defects Res B Dev Reprod Toxicol; 71:111-123), which is close to the teratogenicity potential identified by the o/c ratio in this study (1.3 μM, Table 7,
For the compounds evaluated in this study, the targeted biomarker assay agrees with in vivo rodent and rabbit studies about 75% of the time (Table 11). There is still significant opportunity to improve the understanding of how to translate compound concentrations from in vitro systems to human exposure levels (Bhattacharya et al., 2011, PLoS One; 6:e20887). The application set was used to demonstrate how the measurement of toxicity potential across an exposure range can put model response into perspective in terms of the overall compound risk when combined with additional assays conducted during a compound's discovery and development. The 10 compounds in this set have unknown human developmental toxicity outcomes, as would any novel compound. The o/c ratio was compared with the available Cmax for the application set of compounds to begin to assess the relevance of the signal of teratogenicity potential for each compound (Supplementary Table 1). The therapeutic Cmax was used to understand the potential exposure level encountered in humans. However, since the human teratogenicity of these compounds is unknown, the Cmax was not used to assess the predictivity of the assay. The application set was meant to demonstrate utility of the targeted biomarker assay for unknown compounds in contrast to assessment of assay performance for compounds with known human teratogenicity (
A significant advantage of the targeted biomarker assay is the use of human cells, derived from an embryo, which are able to recapitulate every cell type in the body and have an unlimited capacity to proliferate in culture. The possibility of species-specific differences in developmental toxicity that may be observed in other in vitro developmental toxicity assays is eliminated. In contrast to the ECVAM-evaluated mEST, the assay presented here does not require differentiation of the hES cells into specific lineages such as embryo bodies or cardiomyocytes. Differentiation into specific lineages may limit an assay's potential for predicting teratogens that affect a different developmental lineage. The assay described herein can correctly classify compounds that are known to affect multiple lineages, including cardiovascular, neural and skeletal (Tables 2 and 3). The targeted biomarker assay provides endpoints which are determined analytically and do not need any subjective interpretation of morphology, as is required by the mEST, post-implantation rat WEC test and ZET. Recent modifications to the mEST have begun to address these limitations by adding additional developmental endpoints (i.e., neural and osteoblast differentiation) and implementing molecular endpoints in place of subjective evaluation (reviewed in Theunissen and Piersma, 2012, Front Biosci; 17:1965-1975). Table 10 presents a comparison of the results of the targeted biomarker assay described here and five other developmental toxicity assays; the targeted biomarker assay has a higher accuracy than the other assays (Table 11). The higher accuracy of the predictions made with the o/c ratio is due to an increase in specificity, or the detection of non-teratogens, over the other assays. It is important to note that differences exist between each of the model systems in the way that compounds are predicted. None of the other assays included in Table 10 classify compounds based on human exposure levels, whereas our classification system directly compares a compound's teratogenicity potential to the known therapeutic Cmax for compounds that have known human developmental toxicity outcomes. When making predictions, the actual exposure levels of a compound likely to be encountered by a fetus are critical. Nine of the 17 human non-teratogens tested in the targeted biomarker assay caused a change in the o/c ratio at exposure levels above the therapeutic Cmax. It is believed that any compound, given at the right dose, at the right time during development, in the right species will be teratogenic (Daston G P and Knudsen T B, 2010, “Fundamental concepts, current regulatory design and interpretation,” In: Knudsen T B, Daston G P, editors. Comprehensive Toxicology. Vol 12, 2nd ed. New York: Elsevier. p 3-9). The ability of the targeted biomarker assay to separate exposure levels that are not indicative of teratogenicity from levels that are indicative of teratogenicity is a key strength of the assay.
Although the targeted biomarker assay described herein shows significant promise in predicting developmental toxicity, hES cells, as with other in vitro models, cannot fully reproduce all events contributing to the disruption of normal human development by exogenous chemicals. In vitro models of toxicity do not include the effects of absorption, distribution, metabolism and excretion (ADME), which may make it difficult to predict how a substance of unknown toxicity will act in vivo. The absence of metabolic activity could partially be overcome by the addition of an exogenous bioactivation system when metabolic activation is required or to test both the parent compound and any known metabolites for developmental toxicity potential. Testing both parent compounds and metabolites can help discern which agent is the proximate teratogen, which is essential to accurately predicting a test compound's developmental toxicity potential. Additionally, maternal-fetal interactions and organogenesis cannot be modeled using an in vitro model. However, one of the advantages of using an in vitro assay is the ability to separate adverse outcomes due to compound versus outcomes due to maternal toxicity from compound exposure. Developmental toxicity testing in cells derived from human embryos is likely to generate more reliable in vitro prediction endpoints than endpoints currently available through the use of animal models, or other in vitro non-human assays given the physiological relevance of hES cells to human development.
This assay can help reduce or eliminate species-specific misinterpretations, reduce need for a second species, and could be included as part of a panel of in vitro assays aimed at defining where potential adverse responses in human populations may exist. Much like other in vitro culture systems that are used to understand potential for target organ toxicity, this assay can assess potential for developmental toxicity. Part of its strength is that this is accomplished across a range of exposure levels. While there is no defined way to project safety margins or fully predict human response based on in vitro data, assays such as this one can help define exposure ranges where response may be expected as well as those where a response would not be expected to occur. Results could then be incorporated into a panel of tests that in aggregate develop an approximation of clinical safety margins. This information could help to drive decisions as to whether a compound should progress along its development path.
Example 1 has also published as Smith et al., 2013, “Establishment and assessment of a new human embryonic stem cell-based biomarker assay for developmental toxicity screening,” Birth Defects Res B Dev Reprod Toxicol; 98(4):343-63, which is hereby incorporated by reference in its entirety.
Example 2 ADMA/Cystine RatioWith the present invention, it has been determined that the analysis of data obtained from a small number of metabolites can serve as very accurate predictors of teratogenicity. As described in Example 1, an algorithm was developed that evaluated the individual predictive capacity of these secreted features and media components with the training set to identify and confirm several key features that could be used to develop a much simplified predictive model. The selection process weighted the predictive capacity of a feature, overall intensity, and peak shape to identify very well behaved features/metabolites that could be measured by targeted LC-MS or even by other detection systems. Several pairs of features and some individual features were identified that could accurately identify at least 90% of the teratogens and non-teratogens in the training and test sets that were used for the development of the devTOX computational models.
In this example, cystine and asymmetric dimethylarginine (ADMA) were selected for the simplified predictive model due to their abundance, ideal peak shapes, and their exhibition of similar performance metrics as the computational model (Table 14) with both showing an accuracy of 93%. This simplified model is based on a ratio of the reference treatment (DMSO) normalized values of ADMA and cystine. This simple ratio is able to differentiate teratogens that generally exhibit a decrease in the ratio relative to non-teratogens. When evaluated across 9 independent replications of the training set it is clearly able to differentiate teratogens from non-teratogens (
Following procedures as described in the previous examples, it was also determined the determination of cystathionine/cystine fold change ratios also provide excellent predictivity and general performance in the rapid teratogenicity screen described herein. This is shown in
Changes in cellular metabolism as measured in the spent medium following cell culture (the secretome) is a functional measure of cell health. The cell culture “secretome” refers to the metabolites present in the spent media or cell culture supernatant following cell culture. The secretome is comprised of media components, metabolites passively and actively transported across the plasma membrane, intracellular metabolites release upon lysis, and those produced through extracellular metabolism of enzymes. The change in secretome elicited by an experimental agent relative to untreated cultures produces a metabolic signature that can be used to infer the number of metabolically viable cells present within a cell culture. We have identified a number of secreted metabolites that can be utilized to infer the number viable cells relative to the number of cells in a reference culture “control group”. We compared a number of secreted metabolites to the results of viability analysis performed using a commercial kit and discovered that a decrease in the relative abundance of the secreted metabolites are directly correlated with measurements of cell viability with a Pearson correlation coefficient greater than 0.86 (P value <<0.001) when cytotoxicity is observed in at least the two highest concentrations of a 9 point concentration curve. These metabolites could be utilized by LC-MS or kit based detection to determine the number of viable cells within a cell culture without a requirement to destroy or impact the cells. These metabolites can be used as novel measure of viability that does not require disrupting the growing cells.
The complete disclosure of all patents, patent applications, and publications, and electronically available material (including, for instance, nucleotide sequence submissions in, e.g., GenBank and RefSeq, and amino acid sequence submissions in, e.g., SwissProt, PIR, PRF, PDB, and translations from annotated coding regions in GenBank and RefSeq) cited herein are incorporated by reference. All headings are for the convenience of the reader and should not be used to limit the meaning of the text that follows the heading, unless so specified. In the event that any inconsistency exists between the disclosure of the present application and the disclosure(s) of any document incorporated herein by reference, the disclosure of the present application shall govern. The foregoing detailed description and examples have been given for clarity of understanding only. No unnecessary limitations are to be understood therefrom. The invention is not limited to the exact details shown and described, for variations obvious to one skilled in the art will be included within the invention defined by the claims.
Claims
1-16. (canceled)
17. A method of determining whether a test compound has developmental toxicity potential, the method comprising:
- quantifying an effect of a test compound on levels of a first metabolite in human cells to obtain a first value;
- quantifying an effect of the test compound on levels of a second metabolite in the human cells to obtain a second value; and
- determining that the test compound has developmental toxicity potential if a ratio of the first value to the second value does not exceed a threshold value.
18. The method of claim 17, wherein the human cells are selected from the group consisting of human embryonic stem cells (hESCs), human induced pluripotent (iPS) cells, human embryoid bodies, and hSLC-derived lineage-specific cells.
19. The method of claim 17,
- wherein the first metabolite is ornithine, or a fragment, adduct, deduct or loss thereof, and
- wherein the second metabolite is cystine, or a fragment, adduct, deduct or loss thereof.
20. The method of claim 19,
- wherein the first value comprises a fold change in levels of the first metabolite in culture media of the human cells cultured in the presence of the test compound in comparison with the human cells cultured in the absence of the test compound, and
- wherein the second value comprises a fold change in levels of the second metabolite in culture media of the human cells cultured in the presence of the test compound in comparison with the human cells cultured in the absence of the test compound.
21. The method of claim 20, wherein the threshold value is 0.88.
22. The method of claim 20,
- wherein the human cells are human embryonic stem cells (hESCs), and
- wherein the threshold value is 0.85.
23. The method of claim 20,
- wherein the threshold value is 0.75, and
- wherein the test compound has developmental toxicity potential at a level of exposure that is not cytotoxic.
24. The method of claim 17, wherein at least one of the quantifying steps comprises a physical separation method.
25. The method of claim 24, wherein the physical separation method comprises mass spectrometry.
26. The method of claim 25, wherein the mass spectrometry comprises liquid chromatography/electrospray ionization mass spectrometry.
27. A method of determining whether a concentration of a test compound is potentially toxic to development, the method comprising:
- quantifying an effect of a concentration of a test compound on levels of a first metabolite in human cells to obtain a first value;
- quantifying an effect of the concentration of the test compound on levels of a second metabolite in the human cells to obtain a second value; and
- determining that the concentration of the test compound is potentially toxic to development if a ratio of the first value to the second value does not exceed a threshold value.
28. The method of claim 27, wherein the human cells are selected from the group consisting of human embryonic stem cells (hESCs), human induced pluripotent (iPS) cells, human embryoid bodies, and hSLC-derived lineage-specific cells.
29. The method of claim 27,
- wherein the first metabolite is ornithine, or a fragment, adduct, deduct or loss thereof, and
- wherein the second metabolite is cystine, or a fragment, adduct, deduct or loss thereof.
30. The method of claim 29,
- wherein the first value comprises a fold change in levels of the first metabolite in culture media of the human cells cultured in the presence of the test compound in comparison with the human cells cultured in the absence of the test compound, and
- wherein the second value comprises a fold change in levels of the second metabolite in culture media of the human cells cultured in the presence of the test compound in comparison with the human cells cultured in the absence of the test compound.
31. The method of claim 30, wherein the threshold value is 0.88.
32. The method of claim 30,
- wherein the human cells are human embryonic stem cells (hESCs), and
- wherein the threshold value is 0.85.
33. The method of claim 30,
- wherein the threshold value is 0.75, and
- wherein the concentration of the test compound is potentially toxic to development but is not cytotoxic.
34. The method of claim 27, wherein at least one of the quantifying steps comprises a physical separation method.
35. The method of claim 34, wherein the physical separation method comprises mass spectrometry.
36. The method of claim 35, wherein the mass spectrometry comprises liquid chromatography/electrospray ionization mass spectrometry.
37. The method of claim 27,
- wherein the quantifying steps are performed for a plurality of concentrations, and
- wherein the determining step is performed for each of the plurality of concentrations.
38. The method of claim 37, wherein one of the plurality of concentrations comprises the test compound's human therapeutic Cmax.
Type: Application
Filed: Dec 18, 2018
Publication Date: May 2, 2019
Inventors: Alan M. Smith (Madison, WI), Paul R. West (Arena, WI), Jessica Palmer (Prairie du Sac, WI)
Application Number: 16/224,150
