Article Cite This: Environ. Sci. Technol. XXXX, XXX, XXX−XXX
pubs.acs.org/est
Quantitative Response-Response Relationships Linking Aromatase Inhibition to Decreased Fecundity are Conserved Across Three Fishes with Asynchronous Oocyte Development Jon A. Doering,*,†,‡ Daniel L. Villeneuve,† Shane T. Poole,† Brett R. Blackwell,† Kathleen M. Jensen,† Michael D. Kahl,† Ashley R. Kittelson,§ David J. Feifarek,† Charlene B. Tilton,§ Carlie A. LaLone,† and Gerald T. Ankley† Downloaded via UMEA UNIV on August 20, 2019 at 12:05:50 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
†
Mid-Continent Ecology Division, U.S. Environmental Protection Agency, Duluth, Minnesota 55804 United States National Research Council, U.S. Environmental Protection Agency, Duluth, Minnesota 55804 United States § Oak Ridge Institute of Science Education, U.S. Environmental Protection Agency, Duluth, Minnesota 55804 United States ‡
S Supporting Information *
ABSTRACT: Quantitative adverse outcome pathways (qAOPs) describe quantitative response-response relationships that can predict the probability or severity of an adverse outcome for a given magnitude of chemical interaction with a molecular initiating event. However, the taxonomic domain of applicability for these predictions is largely untested. The present study began defining this applicability for a previously described qAOP for aromatase inhibition leading to decreased fecundity developed using data from fathead minnow (Pimephales promelas). This qAOP includes quantitative response-response relationships describing plasma 17βestradiol (E2) as a function of plasma fadrozole, plasma vitellogenin (VTG) as a function of plasma E2, and fecundity as a function of plasma VTG. These quantitative response-response relationships simulated plasma E2, plasma VTG, and fecundity measured in female zebrafish (Danio rerio) exposed to fadrozole for 21 days but not these responses measured in female Japanese medaka (Oryzias latipes). However, Japanese medaka had different basal levels of plasma E2, plasma VTG, and fecundity. Normalizing basal levels of each measurement to equal those of female fathead minnow enabled the relationships to accurately simulate plasma E2, plasma VTG, and fecundity measured in female Japanese medaka. This suggests that these quantitative response-response relationships are conserved across these three fishes when considering relative change rather than absolute measurements. The present study represents an early step toward defining the appropriate taxonomic domain of applicability and extending the regulatory applications of this qAOP.
■
INTRODUCTION
molecular initiating event, through a consecutive series of connected key events describing measurable biological changes, to a final adverse effect at a biological-level of regulatory relevance, known as the adverse outcome.13 An AOP is connected by key event relationships which represent response-response or input-output relationships between the molecular initiating event, key events, and the adverse outcome.13,14 The most advanced developments in AOPs, known as quantitative AOPs (qAOPs), include descriptions of quantitative response-response relationships.14−16 A quantitative response-response relationship can predict the probability or magnitude of change in a downstream key event (including the adverse outcome) for a given magnitude of change in an
Ecological risk assessments of chemicals require the extrapolation of toxicities measured in a small number of laboratory model species to the vast diversity of native species of regulatory concern because not all species can be pragmatically evaluated in laboratory toxicity testing. But cross-species extrapolation of toxicities represents a complex challenge for risk assessors and is a probable source of uncertainty because potential differences in sensitivity to chemicals among species could range from a few fold to more than a thousand-fold.1−11 However, the process for assessing risks associated with exposure to chemicals is currently undergoing a fundamental shift from an emphasis on whole-animal testing of apical-level toxicities to a greater focus on conserved mechanistic end points.12 This shift stimulated development of a conceptual framework known as the adverse outcome pathway (AOP).13 An AOP describes the progression from an initial molecularlevel perturbation of a biological system, known as the © XXXX American Chemical Society
Received: Revised: Accepted: Published: A
April 30, 2019 August 2, 2019 August 6, 2019 August 6, 2019 DOI: 10.1021/acs.est.9b02606 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
AOP-driven ecological risk assessments for native species of regulatory concern.
upstream key event (including the molecular initiating event).14−16 This predictive capability gives qAOPs potential utility for quantitative ecological risk assessments.15,17 Quantitative response-response relationships in a qAOP are intended to represent generalized patterns of responses in order to produce models with the most broad taxonomic domain of applicability.15 However, currently, the taxonomic applicability for the predictive relationships captured in qAOP descriptions is almost completely untested. Therefore, the present study investigated the taxonomic domain of applicability for a qAOP describing inhibition of cytochrome P450 aromatase (CYP19) leading to decreased fecundity and population trajectory.15 Aromatase is a steroidogenic enzyme involved in the conversion of androgens to estrogens, and inhibition of aromatase activity by chemicals can result in reduced concentrations of 17β-estradiol (E2).18,19 Less E2 decreases synthesis of vitellogenin (VTG), an egg yolk precursor protein, which is required for females to produce viable eggs.20 Reduced fecundity can cause declines in populations.21,22 To quantitatively predict responses of E2, VTG, fecundity, and population trajectory resulting from inhibition of aromatase, this qAOP incorperates three linked models based on data from dose-response, time-course studies of a single laboratory model species of fish, the fathead minnow (Pimephales promelas).15 Specifically, the qAOP incorporates (1) a mechanistically based hypothalamicpituitary-gonadal (HPG) axis model describing concentrations of E2 and E2-dependent production of VTG, (2) a statistically based oocyte growth dynamics model describing egg development and spawning, and (3) a density-dependent population model describing population trajectory.15 However, the key event relationships linking inhibition of aromatase to decreased fecundity are known to be qualitatively conserved among species of fish, namely decreases in plasma E2 and VTG.23,24 Since key event relationships for inhibition of aromatase leading to decreased fecundity are qualitatively conserved among fishes, it was hypothesized that quantitative responseresponse relationships in this qAOP might also be conserved. Therefore, the present study investigated quantitative response-response relationships in two other laboratory model species of fish which have gonads with asynchronous oocyte development in common with the fathead minnow and therefore share generally similar reproductive characteristics.25 Specifically, spawning groups of Japanese medaka (Oryzias latipes) and zebrafish (Danio rerio) were exposed to the model nonsteroidal aromatase inhibiting chemical, fadrozole, for 21 days. From these reproduction assays, input and output variables in the quantitative response-response relationships for the qAOP could be measured in each species, including concentrations of plasma fadrozole, plasma E2, plasma VTG, as well as fecundity. Previously, a 21 day reproduction assay with fathead minnow exposed to fadrozole that measured the same responses had been performed but was not used in development of the qAOP.20 These data from fathead minnow were used to determine whether the response-response predictions in the qAOP simulate responses measured in Japanese medaka and zebrafish with accuracy comparable to responses measured in fathead minnow. Results of the present study represent an early step toward defining the appropriate taxonomic domain of applicability for this qAOP. Understanding this domain is essential for extending the regulatory applications of this qAOP and potentially enabling predictive,
■
MATERIALS AND METHODS Twenty-One Day Reproduction Assays. A 21 day reproduction assay was performed for Japanese medaka and for zebrafish based on test guidelines described under the Organisation for Economic Cooperation and Development (OECD) Test No. 229: Fish Short Term Reproduction Assay,26 with minor modifications. Sexually mature Japanese medaka (>2 months post hatch) and zebrafish (>5 months post hatch) were acquired from an on-site culture facility at the USEPA laboratory in Duluth, MN. Eight male and female Japanese medaka or five male and female zebrafish were placed together in 20 L aquaria. Zebrafish were contained in plastic cages to prevent cannibalism of eggs. A 16 day chemical-free acclimation period was used to assess egg production of each group of fish. Fadrozole (Chemical Abstracts Service [CAS] 102676−47−1; Novartis, Inc., Summit, NJ) exposures were initiated with four randomly selected replicate aquaria per treatment for Japanese medaka (n = 4) and three randomly selected replicate aquaria per treatment for zebrafish (n = 3). Both species had four treatment groups with nominal concentrations of 0, 2, 10, or 30 μg fadrozole/L. Water samples were collected from each aquaria a minimum of twice weekly for measurement of fadrozole. During both the acclimation and exposure periods, Japanese medaka and zebrafish were maintained in continuously flowing, sand filtered, UV treated Lake Superior water at a flow rate of 45 mL/min at 25 °C with a photoperiod of 16:8 h light/dark. Temperature and dissolved oxygen were monitored daily and remained within acceptable limits during the studies.26 Fish were fed twice daily ad libitum with newly hatched live brine shrimp (Biomarine Aquafauna. Hawthorne, CA). Eggs were collected and counted daily. After 21 days of exposure, fish were euthanized in buffered tricaine methanesulfonate (MS222, 100 mg/L buffered with 200 mg NAHCO3/L; Argent, Richmond, WA). Blood was collected from the caudal vasculature with a heparinized microhematocrit tube and centrifuged to obtain plasma. Plasma samples were stored at −80 °C until analyzed. Animal research protocols were approved by the on-site Animal Care and Use Committee in accordance with Animal Welfare Act regulations and Interagency Research Animal Committee guidelines. Analytical and Biochemical Measurements. Volumes of plasma collected from individual Japanese medaka and zebrafish were too small to perform the desired analytical and biochemical measurements. Therefore, plasma from all eight female Japanese medaka or all five female zebrafish in each aquaria were pooled to create a composite plasma sample for each replicate of each treatment (Japanese medaka, n = 4; zebrafish, n = 3). Plasma and water concentrations of fadrozole were measured with an Agilent 1260 high performance liquid chromatograph (LC) coupled to an Agilent 6410 mass spectrometer (MS) (Agilent, Santa Clara, CA) according to methods described in the Supporting Information (SI). Plasma concentrations of E2 were measured by radioimmunoassay (RIA) according to methods described previously.27 Radioactivity was measured by use of a Tri-Carb 2910 TR liquid scintillation analyzer (PerkinElmer, Waltham, MA). Plasma concentrations of VTG were measured by use of commercially available enzyme-linked immunosorbent assays (ELISAs) with antibodies specific for Japanese medaka (CAT#10009223, B
DOI: 10.1021/acs.est.9b02606 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
Figure 1. Response-response relationship predictions from the qAOP (black dots with fitted line) compared to responses measured in female fathead minnow, Japanese medaka, or zebrafish exposed to fadrozole for 21 days (open circles with standard deviations). Quantitative responseresponse relationships for plasma E2 as a function of plasma fadrozole (A), plasma VTG as a function of plasma E2 (B), and fecundity as a function of plasma VTG (C) are shown. Accuracy of the quantitative response-response relationships for predicting responses measured in fathead minnow, Japanese medaka, and zebrafish are presented (Table 1).
derived in another study of female fathead minnow exposed to fadrozole.28 Plasma vitellogenin concentrations of Japanese medaka and zebrafish were converted from mg/mL to μM using a molecular weight of 146.3 kDa, which was used to develop the quantitative response-response relationships for fathead minnow.29 Responses measured in female fathead minnow, Japanese medaka, and zebrafish were plotted against the corresponding response-response predictions from the qAOP. The accuracy of the qAOP for simulating responses to fadrozole measured in female fathead minnow, Japanese medaka, and zebrafish was evaluated based on mean absolute error (MAE). The MAE is the average of the absolute values of the difference between the predicted and measured responses represented in the same units as the evaluated response.30 The closer the MAE is to zero, the more accurate the model is at predicting empirical responses.30 Statistical analysis was not conducted on MAE values because no suitable test currently exists. Species-specific differences in the basal levels of each measurement and differences in intrinsic sensitivity to inhibition of the aromatase enzyme could potentially affect quantitative response-response relationships. Therefore, these species-specific differences were investigated by normalizing the measurements of each species and plotting the normalized values against the corresponding response-response predictions. The potential effect of differences in basal levels of each
Biosense Laboratories, Bergen, Norway) or zebrafish (CAT#10004995, Biosense Laboratories), and according to the protocol provided by the manufacturer. Absorbance was measured in a Bio-Rad Model 3550 microplate reader (Hercules, CA). Response-Response Prediction Comparisons. Previously developed equations for quantitative response-response relationships describing plasma E2 as a function of fadrozole concentration, plasma VTG as a function of plasma E2, and fecundity as a function of plasma VTG15 were used to generate curves representing response-response predictions for each relationship. Quantitative response-response relationships involving population trajectory were not considered here because life-table parameters for different species can easily be inserted to allow broad taxonomic applicability for this model component.15 Previously, water concentration was used to develop the corresponding quantitative response-response relationship because fathead minnow have a bioconcentration factor (BCF) between water and plasma of approximately 1.0 for fadrozole.15 However, plasma fadrozole concentration was used here to account for possible species-specific differences in adsorption, distribution, metabolism, and excretion (ADME) of fadrozole. Plasma fadrozole was not measured in the 21 day reproduction assay with fathead minnow20, and therefore, plasma concentrations were estimated from measured fadrozole water concentrations using an average BCF of 0.8 C
DOI: 10.1021/acs.est.9b02606 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
Table 1. Mean Absolute Error (MAE) Representing the Average Absolute Value of the Difference between Response-Response Predictions from the qAOP and Measured, Basal Normalized, or Sensitivity Normalized Responses from 21 Day Reproduction Assays for Each Species E2 from Fadrozole (μM) Fathead Minnow Japanese Medaka Zebrafish
Fecundity from VTG (eggs/ female/day)
VTG from E2 (μM)
Measureda
Basalb
Sensitivityc
Measureda
Basalb
Measureda
Basalb
0.0035 0.035 0.0049
0.0034 0.0032 0.0043
0.0034 0.0026 0.0046
64 1.5 × 1011 57
13 35 8.4
1.1 20 4.2
0.77 7.2 4.2
a
MAEs for responses presented in Figure 1. bMAEs for responses presented in Figure 2. cMAEs for responses presented in Figure 3.
Japanese medaka than for fathead minnow and zebrafish (Table 1). Response-Response Predictions Compared to Basal Normalized Responses. Basal levels of plasma E2, plasma VTG, and fecundity measured in female fathead minnow and zebrafish from the 0 μg fadrozole/L treatment group were comparable to the basal levels used in the qAOP (Table 2).
measurement were investigated by normalizing the measurements for each species so that the levels measured in females from the 0 μg fadrozole/L treatment group equal the basal level used in the qAOP. The potential effect of differences in intrinsic sensitivity to inhibition of the aromatase enzyme were investigated by normalizing the plasma fadrozole concentration to account for species-specific differences in the potency of fadrozole. Sensitivity normalization values were derived from previously reported relative species sensitivities to fadrozole based on 50% inhibition concentrations (IC50s) from a standard in vitro aromatase inhibition assay using subcellular fractions of whole tissue homogenates from each species.31 The accuracy of the qAOP for simulating normalized responses to fadrozole measured in female fathead minnow, Japanese medaka, and zebrafish was evaluated based on MAE.
Table 2. Values Used to Normalize the Responses Measured in Each Species to Account for Differences in Basal Levels in Each Measure or Intrinsic Sensitivity of the Aromatase Enzyme.a Basal
■
RESULTS Japanese Medaka and Zebrafish Reproduction Assays. Water concentrations of fadrozole were close to nominal and did not vary greatly during the exposure for either species (Table S1). No fadrozole was detected (detection limit = 0.5 μg fadrozole/L) in any control aquaria over the course of the exposures (Table S1). The average BCF measured for fadrozole in plasma of female Japanese medaka and zebrafish exposed for 21 days was 0.9 and 0.2, respectively (Table S1). These BCFs are comparable to the average BCF of 0.8 measured for fadrozole in plasma of female fathead minnow.28 Exposure to fadrozole caused a concentration-dependent decrease in plasma E2, plasma VTG, and fecundity in Japanese medaka and zebrafish (Figure S1). This concentrationdependent decrease is consistent with responses previously reported in fathead minnow exposed to fadrozole and other nonsteroidal aromatase inhibiting chemicals.20,32,33 Response-Response Predictions Compared to Measured Responses. Response-response predictions from the qAOP describing plasma E2 as a function of plasma fadrozole (Figure 1A), plasma VTG as a function of plasma E2 (Figure 1B), and fecundity as a function of plasma VTG (Figure 1C) simulated responses measured in female fathead minnow exposed to fadrozole within the range of variability typical of reproduction assays, as was shown previously.15 Responseresponse predictions from the qAOP for these three relationships simulated responses measured in female zebrafish exposed to fadrozole with calculated MAEs between predicted and measured responses being comparable between zebrafish and fathead minnow (Figure 1, Table 1). In contrast, the response-response predictions from the qAOP did not simulate responses measured in female Japanese medaka exposed to fadrozole for any of these three relationships (Figure 1). Calculated MAEs between predicted and measured responses for the three relationships were more than 10-fold greater for
Sensitivity
Test Species
E2
VTG
Fecundity
Fadrozole
Fathead Minnow Japanese Medaka Zebrafish
0.86 0.23 1.1
1.0 2.6 0.64
1.1 0.78 1.0
1.0 5.5 0.82
a
Absolute values for each measurement in Japanese medaka and zebrafish are provided (Figure S1).
Consequently, basal normalization had minimal effect on the accuracy with which the qAOP simulated responses to fadrozole in these two species (Figure 2, Table 1). In contrast, basal levels of plasma E2 and fecundity measured in female Japanese medaka from the 0 μg fadrozole/L treatment group were 4.4- and 1.3-fold greater, respectively, than basal levels used in the qAOP, while the basal level of plasma VTG was 0.3-fold that used in the qAOP (Table 2). Response-response predictions from the qAOP for the three relationships simulated basal normalized responses measured in female Japanese medaka exposed to fadrozole with calculated MAEs between predicted and measured responses comparable to the MAEs of fathead minnow and zebrafish (Figure 2, Table 1). Response-Response Predictions Compared to Sensitivity Normalized Responses. Aromatase enzyme of Japanese medaka is 5.5-fold more sensitive to in vitro inhibition by fadrozole relative to aromatase enzyme of fathead minnow, while sensitivity of aromatase enzyme of zebrafish is 0.82-fold that of fathead minnow (Table 2). Normalization of plasma fadrozole to account for these differences in sensitivity of the aromatase enzyme had minimal effect on the accuracy with which the qAOP simulated plasma E2 in Japanese medaka or zebrafish (Figure 3, Table 1).
■
DISCUSSION A qAOP is intended to represent generalized patterns of responses in order to produce models with the most broad taxonomic domain of applicability.15 However, the taxonomic applicability for the quantitative response-response relationships captured in qAOP descriptions is largely untested. D
DOI: 10.1021/acs.est.9b02606 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
Figure 2. Response-response relationship predictions from the qAOP (black dots with fitted line) compared to basal normalized responses measured in female fathead minnow, Japanese medaka, or zebrafish exposed to fadrozole for 21 days (open circles with standard deviations). Quantitative response-response relationships for plasma E2 as a function of plasma fadrozole (A), plasma VTG as a function of plasma E2 (B), and fecundity as a function of plasma VTG (C) are shown. Values used to basal normalize the measured responses are presented (Table 2). Accuracy of the quantitative response-response relationships for predicting responses measured in fathead minnow, Japanese medaka, and zebrafish are presented (Table 1).
Figure 3. Response-response relationship predictions from the qAOP for plasma E2 as a function of plasma fadrozole (black dots with fitted line) compared to sensitivity normalized responses measured in female fathead minnow, Japanese medaka, or zebrafish exposed to fadrozole for 21 days (open circles with standard deviations). Since basal normalization had minimal effect on calculated MAEs for fathead minnow and zebrafish, but significantly decreased MAEs for Japanese medaka (Table 1), the basal normalized plasma E2 response was coupled with the sensitivity normalized plasma fadrozole concentration. Values used to sensitivity normalize the measured responses are presented (Table 2). Accuracy of the quantitative response-response relationships for predicting responses measured in fathead minnow, Japanese medaka, and zebrafish are presented (Table 1).
21 days (Figure 1; Table 1), as was shown previously.15 But, responses linking inhibition of aromatase to decreased fecundity are known to be qualitatively conserved among fishes23,24 and therefore it was hypothesized that quantitative response-response relationships might also be conserved. However, while response-response predictions from the qAOP simulated plasma E2, plasma VTG, and fecundity measured in female zebrafish exposed to fadrozole for 21 days
Therefore, the present study began defining the appropriate taxonomic domain of applicability for a qAOP describing inhibition of aromatase leading to decreased fecundity. Response-response predictions from the qAOP describing plasma E2 as a function of plasma fadrozole, plasma VTG as a function of plasma E2, and fecundity as a function of plasma VTG simulated plasma E2, plasma VTG, and fecundity measured in female fathead minnow exposed to fadrozole for E
DOI: 10.1021/acs.est.9b02606 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
marine medaka (Oryzias melastigma), and four fishes from other families, the sheepshead minnow (Cyprinodon variegatus; Cyprinodontidae), mummichog (Fundulus heteroclitus; Fundulidae), Murray rainbowfish (Melanotaenia fluviatilis; Melanotaeniidae), and cunner (Tautogolabrus adspersus; Labridae) (Table S2). The basal levels measured in Japanese medaka and zebrafish in the present study are similar to basal measurements reported in other studies for either species with other studies showing Japanese medaka having generally greater basal levels of plasma E2 and lesser basal levels of plasma VTG relative to levels in fathead minnow and zebrafish (Table S2). Further, these studies show that measurements of basal levels of plasma E2, plasma VTG, and fecundity are similar among species of the same family, but one or more measurements can differ among species from different families. Prior reviews of reproduction in fishes have noted similar observations among species and families,37,39,40 with differences in basal levels of plasma E2 and VTG being hypothesized as reflecting differences in the number, size, and frequency of the eggs produced.36 However, available measurements of basal levels of plasma E2, plasma VTG, and fecundity from the other families considered here (i.e., the Cyprinodontidae, Fundulidae, Melanotaeniidae, and Labridae) are generally similar to the Cyprinidae, with the exception of fecundity in the Labridae which is around 10-fold greater than that in the Cyprinidae (Table S2). This could suggest that the basal levels of fathead minnow are broadly representative across numerous families of fishes, while differences in basal levels might be relatively rare. But, asynchronous oocyte development is not present in all species of fish and other reproductive characteristics are common which can have dramatically different basal levels of plasma E2, plasma VTG, and fecundity41 and require further investigation of conservation in quantitative response-response relationships. Regardless, in the absence of known basal levels of plasma E2, plasma VTG, or fecundity for a given species, results of the present study suggest that the qAOP could still be applied by only considering relative change from a control group, at least for species with asynchronous oocyte development. The investigated quantitative response-response relationships were found to be conserved across fathead minnow, Japanese medaka, and zebrafish when differences in basal levels of plasma E2, plasma VTG, and fecundity are normalized. However, these three species have comparable sensitivities to aromatase inhibiting chemicals,24,29 but growing evidence suggests that differences in sensitivity to these chemicals among fishes could be as great as 2 orders of magnitude.31,42 These potential differences add uncertainty, and therefore, it could be important to consider differences in sensitivity when making qAOP-based predictions across species. Investigations with other biological pathways have demonstrated that differences in sensitivity to interaction of chemicals with the molecular initiating event can be a critical intrinsic driver of differences in sensitivity to adverse effects of chemicals among species.2,17,43−47 Specifically, this was demonstrated in a previously developed qAOP describing activation of the aryl hydrocarbon receptor (AHR) leading to early life stage mortality in birds and fishes.17 In this qAOP, a greater than 2 orders of magnitude difference in sensitivity could be accounted for through normalizing the response-response relationship based on species- and chemical-specific responses measured in a standard in vitro AHR transactivation assay.17 Analogously, a standard in vitro enzyme inhibition assay using
with comparable accuracy as for fathead minnow (Figure 1; Table 1), responses measured in female Japanese medaka were different (Figure 1; Table 1). Fathead minnow, Japanese medaka, and zebrafish share generally similar reproductive characteristics.25 These three species have gonads with asynchronous oocyte development meaning the gonads contain oocytes at all stages of maturation and which makes them capable of ovulating on a regular basis and over a prolonged period.27,34−36 Also, these three species have comparable sensitivities to inhibition of aromatase based on in vitro aromatase inhibition assays on subcellular fractions of whole tissue homogenates29 and in 21 day reproduction assays performed with fadrozole20 (Figure S1) or with the nonsteroidal aromatase inhibiting chemical, prochloraz.24 However, there are differences among these three species that could affect the quantitative response-response relationships. Fathead minnow and zebrafish are phylogenetically more similar sharing the family Cyprinidae (order Cypriniformes) relative to the Japanese medaka from the family Adrianichthyidae (order Beloniformes). Results of the present study demonstrate conservation in quantitative response-response relationships between the two cyprinids but not for the more distantly related Japanese medaka (Figure 1; Table 1). Basal levels of plasma E2, plasma VTG, and fecundity measured in females from the 0 μg fadrozole/L treatment groups were comparable between fathead minnow and zebrafish (Table S2). In contrast, basal levels of plasma E2 and fecundity are 4.4- and 1.3-fold greater in Japanese medaka than in fathead minnow and zebrafish, and the basal level of plasma VTG of Japanese medaka is 0.3-fold that of fathead minnow and zebrafish (Table 2). Importantly, normalizing basal levels measured in female Japanese medaka to equal those used in the qAOP enables the response-response predictions to accurately simulate plasma E2, plasma VTG, and fecundity measured in female Japanese medaka exposed to fadrozole for 21 days with comparable accuracy as for fathead minnow and zebrafish (Figure 2; Table 1). This demonstrates that quantitative response-response relationships describing plasma E2 as a function of plasma fadrozole, plasma VTG as a function of plasma E2, and fecundity as a function of plasma VTG are conserved across these three fishes when species-specific differences in basal levels of each measurement are accounted for or when considering the relative change compared to control females rather than the absolute measurements. Basal levels of plasma E2, plasma VTG, and fecundity have been extensively characterized in systematic reviews of data from fathead minnow21,34,37 but have not yet been systematically reviewed for Japanese medaka or zebrafish. For fathead minnow, the basal levels measured in the study used here20 and used in the qAOP15 are comparable to the averages of the measurements reported in these systematic reviews21,34,37 and fall within the range of reported measurements (Table S2). In order to gain insight into basal levels of plasma E2, plasma VTG, and fecundity for Japanese medaka and zebrafish, the ECOTOX Knowledgebase (https://cfpub.epa.gov/ecotox/)38 was used to identify studies containing these measurements in these species as well as in other fishes with asynchronous oocyte development. A total of 12 studies were found reporting these measurements in Japanese medaka, while 11 studies were found for zebrafish (Table S2). Further studies were identified that contained some or all of these measurements for another cyprinid, the Chinese rare minnow (Gobiocypris rarus), two other medaka spp., the Java ricefish (Oryzias javanicus) and F
DOI: 10.1021/acs.est.9b02606 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
measurements and using only measures of one of these responses, both of which can be analyzed in plasma and could even be collected through sublethal means for certain species. Therefore, this qAOP developed from data from a single laboratory model test species, the fathead minnow, could have immediate regulatory applications that include several laboratory model test species as well as regionally important native indicator species and economically valuable species.
subcellular fractions of whole tissue homogenates from fathead minnow, Japanese medaka, zebrafish, and 15 other fishes was used to assess sensitivity to inhibition of aromatase by fadrozole and other nonsteroidal chemicals.31 It was proposed that this information could be used to normalize the exposure concentration input for the qAOP in a manner that accounts for species-specific differences in the intrinsic sensitivity of the aromatase enzyme.31 However, the range in sensitivity of the aromatase enzyme among fathead minnow, Japanese medaka, and zebrafish is just 6-fold.31 Consequently, in this case, normalizing for differences in sensitivity resulted in a minimal difference in the accuracy with which the response-response predictions of the qAOP simulated plasma E2 (Figure 3, Table 1). However, normalizing for intrinsic differences in sensitivity to inhibition of the aromatase enzyme might be more important when making qAOP-based predictions for other species with greater differences in sensitivity and warrants further investigation. The qAOP for inhibition of aromatase leading to decreased fecundity which was developed for fathead minnow had some immediate regulatory applications, such as the estimation of a benchmark dose for untested aromatase inhibiting chemicals based on potencies measured in high-throughput chemical screening programs (e.g., USEPA’s ToxCast Program).15 However, the present study extends the taxonomic domain of applicability of this qAOP to include fishes with asynchronous oocyte development. Asynchronous oocyte development is taxonomically well represented and is known to occur in at least 36 families of freshwater and marine fishes, including species of significant ecological and economic importance (Table S3). This newly defined taxonomic domain of applicability significantly extends the potential regulatory applications of the qAOP. For example, this qAOP could now be used to estimate a species-specific benchmark dose for untested aromatase inhibiting chemicals by normalizing differences in intrinsic sensitivity to inhibition of the aromatase enzyme from results of standard in vitro enzyme inhibition assays of subcellular fractions of whole tissue homogenates. This qAOP could especially have important applications centered around cyprinids that share similar basal levels of measurements and for which the qAOP could simulate plasma E2, plasma VTG, and fecundity in multiple species within the range of variability typical of standardized 21 day reproduction assays.26 Strong applicability to cyprinids is potentially important because they are an ideally suited taxon for investigation of the adverse effects of chemicals, with several species already being among the most widely used laboratory model test species and numerous others being routinely used in environmental effects monitoring.23,40,41 Further, the cyprinids are a large family of more than 4000 species which are abundant and widely distributed in the freshwaters of North America, Europe, Asia, and Africa.48 Previously, portions of this qAOP were used in conjunction with a historic database of plasma steroid and fecundity measurements to predict the population status of white sucker ( Catostomus commersonii; Catostomidae) at sites impacted by a pulp and paper mill plant and changes that might occur because of different possible mitigation scenarios.49 The present study builds on this application by demonstrating that the qAOP can simulate responses in numerous fishes, particularly in cyprinids. Therefore, impacts to fecundity through decreased E2 or VTG could potentially be assessed in native cyprinids without the need for a database of historic
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.9b02606. Details on fadrozole chemical analysis, response measures in Japanese medaka and zebrafish, summary of basal levels of each measure from asynchronous fishes, and list of families of asynchronous fishes (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Jon A. Doering: 0000-0001-6866-6456 Daniel L. Villeneuve: 0000-0003-2801-0203 Brett R. Blackwell: 0000-0003-1296-4539 Carlie A. LaLone: 0000-0003-3174-1314 Gerald T. Ankley: 0000-0002-9937-615X Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS Research was supported by the U.S. EPA. Thanks to WanYun Cheng for supplying equations for the quantitative response/ response relationships used in the qAOP, Joe Swintek for statistical support, and Rory Conolly for providing thoughtful review comments on an earlier version of the paper. This paper has been reviewed in accordance with the requirements of the U.S. EPA Office of Research and Development. However, the recommendations made herein do not represent U.S. EPA policy. Mention of products or trade names does not indicate endorsement by the U.S. EPA.
■
REFERENCES
(1) Doering, J. A.; Giesy, J. P.; Wiseman, S.; Hecker, M. Predicting the sensitivity of fishes to dioxin-like compounds: possible role of the aryl hydrocarbon receptor (AhR) ligand binding domain. Environ. Sci. Pollut. Res. 2013, 20, 1219−1224. (2) Doering, J. A.; Farmahin, R.; Wiseman, S.; Beitel, S. C.; Kennedy, S. W.; Giesy, J. P.; Hecker, M. Differences in activation of aryl hydrocarbon receptors of white sturgeon relative to lake sturgeon are predicted by identities of key amino acids in the ligand binding domain. Environ. Sci. Technol. 2015, 49, 4681−4689. (3) Eisner, B. K.; Doering, J. A.; Beitel, S. C.; Wiseman, S.; Raine, J. C.; Hecker, M. Cross-species comparison of relative potencies and relative sensitivities of fishes to dibenzo-p-dioxins, dibenzofurans, and polychlorinated biphenyls in vitro. Environ. Toxicol. Chem. 2016, 35 (1), 173−181. (4) Russom, C. L.; LaLone, C. A.; Villeneuve, D. L.; Ankley, G. T. Development of an adverse outcome pathway for acetylcholinesterase inhibition leading to acute mortality. Environ. Toxicol. Chem. 2014, 33 (10), 2157−2169. (5) Shekh, K.; Tang, S.; Hecker, M.; Niyogi, S. Investigating the role of ionoregulatory processes in the species- and life-stage-specific
G
DOI: 10.1021/acs.est.9b02606 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology differences in sensitivity of rainbow trout and white sturgeon to cadmium. Environ. Sci. Technol. 2018, 52 (21), 12868−12876. (6) Song, M. Y.; Stark, J. D.; Brown, J. J. Comparative toxicity of four insecticides, including imidacloprid and tebufenozide, to four aquatic arthropods. Environ. Toxicol. Chem. 1997, 16, 2494−2500. (7) Song, T.; Doering, J. A.; Sun, J.; Beitel, S. C.; Shekh, K.; Patterson, S.; Crawford, S.; Giesy, J. P.; Wiseman, S. B.; Hecker, M. Linking oxidative stress and magnitude of compensatory responses with life-stage specific differences in sensitivity of white sturgeon(Acipenser transmontanus) to copper or cadmium. Environ. Sci. Technol. 2016, 50 (17), 9717−9726. (8) Wang, Y.; Wang, Q.; Wu, B.; Li, Y.; Lu, G. Correlation between TCDD acute toxicity and aryl hydrocarbon receptor structure for different mammals. Ecotoxicol. Environ. Saf. 2013, 89, 84−88. (9) Wirgin, I.; Waldman, J. R. Resistance to contaminants in North American fish populations. Mutat. Res., Fundam. Mol. Mech. Mutagen. 2004, 552 (1−2), 73−100. (10) Caldwell, D. J.; Mastrocco, F.; Anderson, P. D.; Lange, R.; Sumpter, J. P. Predicted-no-effect concentration for the steroid estrogens estrone, 17β-estradiol, estriol, and 17α-ethinylestradiol. Environ. Toxicol. Chem. 2012, 31 (6), 1396−1406. (11) Caldwell, D. J.; Mastrocco, F.; Hutchinson, T. H.; Lange, R.; Heijerick, D.; Janssen, C.; Anderson, P. D.; Sumpter, J. P. Derivation of an aquatic predicted no-effect concentration for the synthetic hormone, 17 alpha-ethinyl estradiol. Environ. Sci. Technol. 2008, 42 (19), 7046−7054. (12) NRC. Toxicity Testing in the 21st Century: A Vision and a Strategy; The National Academies Press: Washington, DC, USA. 2007. (13) Ankley, G. T.; Bennett, R. S.; Erickson, R. J.; Hoff, D. J.; Hornung, M. W.; Johnson, R. D.; Mount, D. R.; Nichols, J. W.; Russom, C. L.; Schmieder, P. K.; Serrano, J. A.; Tietge, J. E.; Villeneuve, D. L. Adverse outcome pathways: A conceptual framework to support ecotoxicology research and risk assessment. Environ. Toxicol. Chem. 2010, 29 (3), 730−741. (14) Villeneuve, D. L.; Crump, D.; Garcia-Reyero, N.; Hecker, M.; Hutchinson, T. H.; LaLone, C. A.; Landesmann, B.; Lettieri, T.; Munn, S.; Nepelska, M.; Ottinger, M. A.; Vergauwen, L.; Whelan, M. Adverse outcome pathway (AOP) development I: Strategies and principles. Toxicol. Sci. 2014, 142 (2), 312−320. (15) Conolly, R. B.; Ankley, G. T.; Cheng, W.; Mayo, M. L.; Miller, D. H.; Perkins, E. J.; Villeneuve, D. L.; Watanabe, K. H. Quantitative adverse outcome pathways and their application to predictive toxicology. Environ. Sci. Technol. 2017, 51, 4661−4671. (16) Wittwehr, C.; Aladjov, H.; Ankley, G.; Bryne, H.; de Knecht, J.; Heinzle, E.; Klambauer, G.; Landesmann, B.; Luijten, M.; MacKay, C.; Maxwell, G.; Meek, B.; Paini, A.; Perkins, E.; Sobanski, T.; Villeneuve, D.; Waters, K.; Whelan, M. How adverse outcome pathways can aid the development of computational prediction models for regulatory toxicology. Toxicol. Sci. 2017, 155, 326−336. (17) Doering, J. A.; Wiseman, S.; Giesy, J. P.; Hecker, M. A crossspecies quantitative adverse outcome pathway for activation of the aryl hydrocarbon receptor leading to early life stage mortality in birds and fishes. Environ. Sci. Technol. 2018, 52 (13), 7524−7533. (18) Callard, G. V.; Petro, Z.; Ryan, K. J. Phylogenetic distribution of aromatase and other androgen-converting enzymes in the central nervous system. Endocrinology 1978, 103, 2283−2290. (19) Simpson, E. R. Aromatase cytochrome-P450-structure, function and regulation. Endocr. Rev. 1994, 8, A1241. (20) Ankley, G. T.; Kahl, M. D.; Jensen, K. M.; Hornung, M. W.; Korte, J. J.; Makynen, E. A.; Leino, R. L. Evaluation of the aromatase inhibitor fadrozole in a short-term reproduction assay with the fathead minnow (Pimephales promelas). Toxicol. Sci. 2002, 67, 121−130. (21) Ankley, G. T.; Miller, D. H.; Jensen, K. M.; Villeneuve, D. L.; Martinovic, D. Relationship of plasma sex steroid concentrations in female fathead minnows to reproductive success and population status. Aquat. Toxicol. 2008, 88, 69−74. (22) Miller, D. H.; Jensen, K. M.; Villeneuve, D. L.; Jahl, M. D.; Makynen, E. A.; Durhan, E. J.; Ankley, G. T. Linkage of biochemical
responses to population-level effects: a case study with vitellogenin in the fathead minnow (Pimephales promelas). Environ. Toxicol. Chem. 2007, 26 (3), 521−527. (23) Ankley, G. T.; Gray, L. E. Cross-species conservation of endocrine pathways: A critical analysis of tier 1 fish and rat screening assays with 12 model chemicals. Environ. Toxicol. Chem. 2013, 32 (5), 1084−1087. (24) Celander, M. C.; Goldstone, J. V.; Denslow, N. D.; Iguchi, T.; Kille, P.; Meyerhoff, R. D.; Smith, B. A.; Hutchinson, T. H.; Wheeler, J. R. Species extrapolation for the 21st century. Environ. Toxicol. Chem. 2011, 30, 52−63. (25) Ankley, G. T.; Johnson, R. D. Small fish models for identifying and assessing the effects of endocrine disrupting chemicals. ILAR J. 2004, 45 (4), 469−483. (26) OECD. Test No. 229; Fish Short Term Reproduction Assay, OECD Guidelines for the Testing of Chemicals, Section 2, OECD Publishing: Paris, 2012. (27) Jensen, K. M.; Korte, J. J.; Kahl, M. D.; Pasha, M. S.; Ankley, G. T. Aspects of basic reproductive biology and endocrinology in the fathead minnow (Pimephales promelas). Comp. Biochem. Physiol., Part C: Toxicol. Pharmacol. 2001, 266, 475−480. (28) Villeneuve, D. L.; Breen, M.; Bencic, D. C.; Cavallin, J. E.; Jensen, K. M.; Makynen, E. A.; Thomas, L. M.; Wehmas, L. C.; Conolly, R. B.; Ankley, G. T. Developing predictive approaches to characterize adaptive responses of the reproductive endocrine axis to aromatase inhibition: I. data generation in a small fish model. Toxicol. Sci. 2013, 133 (2), 25−233. (29) Cheng, W. Y.; Zhang, Q.; Schroeder, A.; Villeneuve, D. L.; Ankley, G. T.; Conolly, R. Computational modeling of plasma vitellogenin alterations in response to aromatase inhibition in fathead minnows. Toxicol. Sci. 2016, 154 (1), 78−89. (30) Willmott, C. J.; Matsuura, K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Clim. Res. 2005, 30, 79−82. (31) Doering, J. A.; Villeneuve, D. L.; Fay, K. A.; Randolph, E. C.; Jensen, K. M.; Kahl, M. D.; LaLone, C. A.; Ankley, G. T. Differential sensitivity to in vitro inhibition of cytochrome P450 aromatase (CYP19) activity among 18 freshwater fishes. Toxicol. Sci. 2019, 170, 394. (32) Ankley, G. T.; Jensen, K. M.; Durhan, E. J.; Makynen, E. A.; Butterworth, B. C.; Kahl, M. D.; Villeneuve, D. L.; Linnum, A.; Gray, L. E.; Cardon, M.; Wilson, V. S. Effects of two fungicides with multiple modes of action on reproductive endocrine function in the fathead minnow (Pimephales promelas). Toxicol. Sci. 2005, 86, 300− 308. (33) Skolness, S. Y.; Blanksma, C. A.; Cavallin, J. E.; Churchill, J. J.; Durhan, E. J.; Jensen, K. M.; Johnson, R. D.; Kahl, M. D.; Makynen, E. A.; Villeneuve, D. L.; Ankley, G. T. Propiconazole inhibits steroidogenesis and reproduction in fathead minnow (Pimephales promelas). Toxicol. Sci. 2013, 132, 284−297. (34) Iwamatsu, T.; Ohta, T.; Oshima, E.; Sakai, N. Oogenesis in the medaka Oryzias latipes - stages of oocyte development. Zool. Sci. 1988, 5, 353−373. (35) Lawrence, C. The husbandry of zebrafish (Danio rerio): A review. Aquaculture 2007, 269 (1−4), 1−20. (36) Tyler, C. R.; Sumpter, J. P. Oocyte growth and development in teleosts. Rev. Fish Biol. Fisheries. 1996, 6, 287−318. (37) Watanabe, K. H.; Jensen, K. M.; Orlando, E. F.; Ankley, G. T. What is normal? A characterization of the values and variability in reproductive endpoints of the fathead minnow, Pimephales promelas. Comp. Biochem. Physiol., Part C: Toxicol. Pharmacol. 2007, 146, 348− 356. (38) USEPA. ECOTOX User Guide: ECOTOXicology Knowledgebase System. Version 5.0. 2019. http:/www.epa.gov/ecotox/ (retrieved 12/19/2018). (39) Hontela, A.; Stacey, N. E. Chapter 4: Cyprinidae. In: Munro, A. D., Scott, A. P., Lam, T. J. (Eds). Reproductive seasonality in teleosts: Environmental influences. CRC Press, Inc.: Boca Raton, 1990; pp 53− 78. H
DOI: 10.1021/acs.est.9b02606 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology (40) Tyler, C. R.; van der Eerden, B.; Jobling, S.; Panter, G.; Sumpter, J. P. Measurement of vitellogenin, a biomarker for exposure to oestrogenic chemicals, in a wide variety of cyprinid fish. J. Comp. Physiol., B 1996, 166, 418−426. (41) Barrett, T. J.; Munkittrick, K. R. Seasonal reproductive patterns and recommended sampling times for sentinel fish species used in environmental effects monitoring programs in Canada. Environ. Rev. 2010, 18, 115−135. (42) Beitel, S. C.; Doering, J. A.; Patterson, S. E.; Hecker, M. Assessment of the sensitivity of three North American fish species to disruptors of steroidogenesis using in vitro tissue explants. Aquat. Toxicol. 2014, 152, 273−283. (43) Ffrench-Constant, R. H.; Rocheleau, T. A.; Steichen, J. C.; Chalmers, A. E. A point mutation in a Drosophilia GABA receptor confers insecticide resistance. Nature 1993, 363, 449−451. (44) Karchner, S. I.; Franks, D. G.; Kennedy, S. W.; Hahn, M. E. The molecular basis for differential dioxin sensitivity in birds: role of the aryl hydrocarbon receptor. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 6252−6257. (45) Liu, Z.; Williamson, M. S.; Lansdell, S. J.; Denholm, I.; Han, Z.; Millar, N. S. A nicotinic acetylcholine receptor mutation conferring target-site resistance to imidacloprid in Nilaparvata lugens (brown planthopper). Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 8420−8425. (46) Martinez-Torres, D.; Foster, S. P.; Field, L. M.; Devonshire, A. L.; Williamson, M. S. A sodium channel point mutation is associated with resistance to DDT and pyrethroid insecticides in the peachpotato aphid, Myzus persicae (Sulzer) (Hemiptera: Amphididae). Insect Mol. Biol. 1999, 8, 339−346. (47) Mutero, A.; Pralavorio, M.; Bride, J. M.; Fournier, D. Resistance-associated point mutations in insecticide insensitive acetylcholinesterase. Proc. Natl. Acad. Sci. U. S. A. 1994, 91, 5922− 5826. (48) Zardoya, R.; Doadrio, I. Molecular evidence on the evolutionary and biogeographical patterns of European cyprinids. J. Mol. Evol. 1999, 49, 227−237. (49) Miller, D. H.; Tietge, J. E.; McMaster, M. E.; Munkittrick, K. R.; Xia, X.; Ankley, G. T. Assessment of status of white sucker (Catostomus commersoni) populations exposed to bleached kraft pulp mill effluent. Environ. Toxicol. Chem. 2013, 32 (7), 1592−1603.
I
DOI: 10.1021/acs.est.9b02606 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
