Use of a Miniaturized Test System for Determining Acute Toxicity of Toxicity Identification Evaluation Fractions

Use of a Miniaturized Test System for Determining Acute Toxicity of Toxicity Identification Evaluation Fractions

ECOTOXICOLOGY AND ENVIRONMENTAL SAFETY ARTICLE NO. 35, 1–6 (1996) 0075 Use of a Miniaturized Test System for Determining Acute Toxicity of Toxicity...

80KB Sizes 2 Downloads 61 Views

ECOTOXICOLOGY AND ENVIRONMENTAL SAFETY ARTICLE NO.

35, 1–6 (1996)

0075

Use of a Miniaturized Test System for Determining Acute Toxicity of Toxicity Identification Evaluation Fractions REBECCA L. POWELL,* E. MICHELLE MOSER,* RICHARD A. KIMERLE,* DAVID E. MCKENZIE,*

AND

MICHAEL MCKEE†

*Environmental Sciences Center and †Ceregen, Monsanto Company, 800 N. Lindbergh, St. Louis, Missouri 63167 Received January 4, 1996

II focuses on identifying the specific chemical(s) responsible for acute or chronic effluent toxicity. Toxicity tests, in conjunction with analytical analysis, are used to identify the toxicants (U.S. EPA, 1991c). Phase III confirms the cause of toxicity. During the identification phase (Phase II) of a TIE, the effluent is first separated into toxic and nontoxic subsamples. This reduces the cost associated with analyzing for a wide range of chemicals and helps with the interpretation of data on effluents containing multiple constituents. One separation technique that is commonly used is high-performance liquid chromatography (HPLC). HPLC can reduce the number of nonpolar organic chemicals associated with the toxicant(s). One limitation of using HPLC, however, is that following fractionation the postcolumn effluent is in the HPLC eluent, which may contain high percentages of organic solvents. The fractions must be diluted so that measured toxicity does not arise from the organic HPLC solvent. If the initial effluent had low toxicity, an increased volume of effluent must be collected and processed in order to have sufficient sample at a high enough concentration to demonstrate toxicity once the fraction is diluted. To conduct definitive, static, acute toxicity tests, current U.S. EPA methods (U.S. EPA, 1991a) recommend a minimum of 1000 ml of test solution for Daphnia magna and 2000 ml for Pimephales promelas. Due to the cost and the time involved in increasing the volume of test solution, researchers are often forced to compromise by reducing the number of replicates and/or test concentrations or the number of species tested. Compromises such as these can increase the uncertainty in the results. An alternate method would be to reduce the volume of test solution required to conduct the toxicity tests. The objectives of this study were, first, to determine if a ‘‘miniaturized’’ test system could be used to determine acute toxicity and, second, demonstrate the usefulness of the miniaturized system when conducting TIEs. The major benefit of the miniaturized test system is that less than 50 ml of final test solution is necessary per test species to conduct a definitive test compared to the 1000 to 2000 ml currently recommended by U.S. EPA. This becomes especially im-

A miniaturized test system was developed and used to determine the acute toxicity of effluent fractions separated by HPLC to Daphnia magna and Pimephales promelas. The miniaturized test system consists of exposing test organisms in 1 ml of test solution using 48-well microtiter plates for the test vessels. Several factors were investigated to determine the acceptability of this test system. These factors included organism biomass to test solution ratio, toxicity of the microtiter plates to the organisms, dissolved oxygen in the test solution, partitioning of the test chemicals to the walls of the test vessels, and dilution of the test solution when the organisms are transferred. Toxicity of four reference chemicals to D. magna and P. promelas was also determined using the miniaturized test systems. It was concluded that the test system could be miniaturized and still provide results comparable to those obtained when standard U.S. EPA test procedures were used. The major benefit of using the miniaturized test system is that less solution is required for conducting a toxicity test. This becomes important when only a small amount of test solution is available, as might occur during a toxicity identification evaluation, after an effluent has been fractionated by HPLC. Other benefits include less space required to conduct a test, less time necessary to prepare test solutions, and a reduced volume of waste for disposal. q 1996 Academic Press, Inc.

INTRODUCTION

Under the guidance of the Clean Water Act, the U.S. EPA and state regulatory agencies control the discharge of toxic substances to waters of the United States. To meet this goal, permits are issued through the National Pollutant Discharge Elimination System (NPDES). If an effluent fails to meet the permit limits, the discharger may be required to conduct a toxicity reduction evaluation (TRE) in which the causative agents of toxicity are characterized and identified and effective control options are explored (U.S. EPA, 1991c). Procedures for characterizing and identifying the toxic components of an effluent are outlined in EPA’s Methods for Aquatic Toxicity Identification Evaluations (U.S. EPA, 1991b, 1993). The toxicity identification evaluation (TIE) procedures are outlined in three phases. In Phase I, the physical/chemical nature of the toxicants is characterized. Phase 1

AID

EES 1442

/

a107$$$$$1

09-26-96 04:09:53

0147-6513/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

eesas

AP: EES

2

POWELL ET AL.

portant in the second phase of the TIE process when trying to identify toxic fractions and the specific components in those fractions. The miniaturized test system would accelerate the pace in which answers regarding toxicity issues are obtained by reducing the time and effort necessary to prepare the fractions. Less space is also required to conduct a test and there is less waste for disposal. This translates to significant cost savings. MATERIALS AND METHODS

Test Organisms The miniaturized test system was developed using two common test organisms, D. magna and P. promelas (fathead minnows). The D. magna were cultured in-house using U.S. EPA-approved methods. Neonates less than 24 hr old were used in the tests. The P. promelas were cultured by a commercial supplier (Florida Bioassay, Florida). Eggs were collected and shipped overnight to the testing facility. Newly hatched fry were used. Test Vessels Microtiter plates, having 48 wells (Costar Corp., Cambridge, MA), were selected for use as the test vessels. Each well holds approximately 2 ml. The plates are constructed of polystyrene and are advertised as ‘‘tissue culture treated.’’ Establishment of Miniaturization Techniques To determine if the miniaturized test system was conducive to conducting toxicity tests, several criteria were examined. These criteria included organism biomass to test solution ratio, toxicity of the microtiter plates, and dissolved oxygen concentration. To determine the organism biomass to test solution ratio, three groups of D. magna and three groups of P. promelas, all less than 48 hr old, were collected, blotted dry to remove excess water, and weighed. Individual weight was determined by dividing the fresh weight of the mass by the number of individuals in each group. Toxicity of the microtiter plates was examined by placing D. magna and P. promelas in wells that contained only dilution water. The number of surviving organisms was determined after 48 hr. The dissolved oxygen concentration of the dilution water in the wells containing organisms was measured at periodic intervals using a Model OM-4 oxygen meter equipped with a micro-oxygen electrode Model MI-730 (Microelectrodes, Inc., Londonderry, NH). Evaluation of Exposure Concentrations To determine if the exposure concentration was significantly influenced in the miniaturized test system, a couple of factors were examined. These included partitioning of test

AID

EES 1442

/

a107$$$$$2

09-26-96 04:09:53

chemicals and the extent of dilution, by culture water, when introducing the organisms. Partitioning of the chemicals to the walls of the test vessels was assessed by using 14Cradiolabeled kepone, at a concentration of 0.85 mg/ml (initial dpm Å 1019). The kepone test solution was pipetted into the wells and at periodic intervals over 48 hr, and 100-ml samples were removed and analyzed by liquid scintillation counting. To determine if the test solutions were being diluted when the test organisms were transferred, 1-ml aliquots of water were pipetted into each of 60 wells. The amount of solution was verified in 20 wells by removing the liquid with a microsyringe. Next, 20 D. magna and 20 P. promelas were transferred into the remaining wells (1 organism per well) and the amount of liquid was measured using a microsyringe. Individual Reference Compounds and TIE Extractions The miniaturized system was first tested using individual ‘‘reference’’ compounds of known toxicity. These compounds included kepone, linear alkylbenzene sulfonate (LAS), pentachlorophenol (PCP), and sodium lauryl sulfate (SLS). The dilution water consisted of 20% Perrier water:80% Milli-Q water that had been aerated overnight. To determine the utility of the miniaturized test system when conducting a TIE, a synthetic effluent was prepared and fractionated using HPLC. The synthetic effluent was composed of 12 chemicals, including acenaphthene, borneol, Captan, 4-chlorocatechol, 3,5-dinitrio-o-toluic acid, kepone, 2-mercaptobenzothiazole, pentachlorophenol, reserpine, 2,4,5,6-tetrachloro-m-xylene, Trifluralin, and triphenyl phosphonium cyclopentadienylide. The synthetic effluent was prepared by first making stock solutions of each chemical in methanol. The stock solutions were then combined to form the synthetic effluent. Chromatographic separation was performed using a Perkin Elmer Series 4 liquid chromatograph (Norwalk, CT) equipped with a 100-ml sample loop and a 250 1 4.6-mm Budick and Jackson (Muskegon, MI) ODS column with 5 mm particle size. Fractions were collected at 1.2-min intervals for each separation using an ISCO Foxy Fraction collector (Lincoln, NB). Chromatographic conditions were 5% methanol/95% water for 3 min with methanol concentration linearly increased, so that at an elapsed time of 10 min the methanol concentration was 100%, which was held at 100% methanol for 30 min. Flow was 0.8 ml/min. Following injections, an average UV absorption chromatogram (at 235l) was produced by summing all individual chromatograms and dividing by the number of injections, in order to take into account retention time variability. Twelve distinct time windows were selected to represent test fractions. The 1.2-min fractions were combined to form the 12 test fractions. Each test fraction was

eesas

AP: EES

3

TOXICITY IDENTIFICATION EVALUATION TEST

FIG. 1. Chromatogram of synthetic effluent showing the 12 test fractions, the concentration factors associated with each fraction, and the test chemicals in each fraction. TPCD, triphenylphosphonium cyclopentadienylide.

reduced in volume and the solvent was exchanged into dimethyl sulfoxide. The final concentration for each test chemical was estimated to be 1 mg/liter. Figure 1 graphically demonstrates the 12 test fractions and the compounds associated with each fraction based on chromatographs established for each stock solution. For a complete list of compounds used in this study along with selected physical properties and listed toxicity values, see Table 1.

Test Procedures Procedures for using the miniaturized test system were based on those developed by U.S. EPA for conducting acute toxicity tests (U.S. EPA, 1991a). For the individual reference test compounds, definitive toxicity tests were conducted using the miniaturized test system to establish dose–response curves. Tests following U.S. EPA’s testing procedures were conducted in parallel. For the synthetic effluent, the HPLC

TABLE 1 Reported Physical and Biological Values for Test Chemicalsa Chemical

CAS No.

Log Kow

WSol (mg/liter)

Acenaphthene Borneol Captan 4-Chlorocatechol 3,5-Dinitrio-o-toluic acid Kepone Linear alkyl benzene sulfonate 2-Mercaptobenzothiazole Pentachlorophenol Reserpine Sodium lauryl sulfate 2,4,5,6-Tetrachloro-m-xylene Trifluralin Triphenyl phosphonium cyclopentadienylide

83-32-9 507-70-0 133-06-2 2138-22-9 28169-46-2 143-50-0 — 149-30-4 87-86-5 50-55-5 151-21-3 877-09-8 1582-09-8

3.92 2.4 (est) 2.35 2.0 (est) 2.1 (est) 3.45 — 1.61 5.01 3.64 — 5.05 (est) 5.05

Insol 28.7 (est)b 3.3 3150 (est) 90 (est) 7.6 (est) Infin Insol 14 73.04 — — 24

— 7–10 3.8 (est) 2.18 (est) 0.69 0.10–29.0 20.1 (est) 0.68 — 1.80–27.0 0.62 (est) 0.56

0.608–1.73 — 0.20 1.58 2.26 (est) 0.34 2.5–5.0 23.1 (est) 0.205 — 6.0–6.90 0.61 (est) 0.105

29473-30-1









a b

Source: AQUIRE and HSDB databases. Estimated value predicted by QSAR program.

AID

EES 1442

/

a107$$$$$2

09-26-96 04:09:53

eesas

AP: EES

D. magna (48 hr; mg/liter) 41

P. promelas (96 hr; mg/liter)

4

POWELL ET AL.

fractions were first screened at 100% for toxicity. Definitive toxicity tests were then conducted on the ‘‘toxic’’ fractions. Toxicity of the initial synthetic effluent to D. magna was also determined. In the miniaturized test system the definitive toxicity tests were conducted by first preparing 50% serial dilutions. Onemilliliter aliquots from each dilution were then pipetted into each of 20 wells. D. magna and P. promelas were then transferred into the wells, one organism per well. Care was taken to minimize the amount of liquid transferred with each organism. The plates were covered with the provided lids and held for 48 hr at 25 { 27C with 16 hr of light. The number of immobilized or dead organisms were counted at 24 and 48 hr and LC50/EC50’s were calculated (Stephan, 1977). RESULTS AND DISCUSSION

Establishment of Miniaturization Techniques The U.S. EPA recommends that the organism biomass should not exceed a level of 0.4 mg fresh wt/ml test solution when the test is conducted at 257C. This is to prevent the organisms from becoming stressed due to overcrowding and/ or decreased concentration of dissolved oxygen. The biomass to test solution ratio is also important in maintaining the concentration of the test chemical. The mean fresh weights for the D. magna and P. promelas were determined to be 0.17 and 0.73 mg/ml, respectively. The P. promelas exceeded the weight limit; however, no adverse effects such as increased death or decreased dissolved oxygen (DO) were observed. It was concluded that the increased body weight of the P. promelas did not significantly influence the results. Test vessels should not contain substances that can be leached or dissolved by aqueous solutions in amounts that are toxic to the test organisms. Before initial use, the vessels should be tested for toxicity by exposing the test organisms in the test system where the material is used. The microtiter plates used throughout this study were proven to be nontoxic to both D. magna and P. promelas. U.S. EPA recommends that the DO concentration within the dilution water should remain above 4.0 mg/liter for warmwater, freshwater species to provide adequate conditions for survival. For the D. magna, DO concentrations remained constant at an average of 9.0 mg/liter during a 72-hr time period. Oxygen in the vessels containing P. promelas dropped slightly from approximately 8.7 to approximately 6.7 during a 48-hr time period, but was still above the recommended level. Evaluation of Exposure Concentrations Partitioning of the test compounds to the wall of the test vessel can significantly reduce the exposure concentration.

AID

EES 1442

/

a107$$$$$2

09-26-96 04:09:53

To provide a ‘‘worst case’’ scenario, partitioning of radiolabeled kepone to the walls of the test vessel was examined. Kepone represented a nonpolar organic compound with a relatively high Kow and low water solubility, characteristics which are favorable for partitioning to container walls. No significant decrease in the number of disintegrations per minute (dpm’s) was noted during a 24-hr time period. This implied that there was no significant partitioning of kepone to the test vessel walls and therefore should not be a problem for most chemicals. Transferring test organisms into the test vessels results in a small amount of culture water also being transferred. With standardized methods this amount is negligible; however, in the smaller vessels it could inadvertently dilute the solution. Using a 1-ml micropipet to fill the vessels resulted in an average of 1.006 { 0.003 ml of test solution in the vessels. After the D. magna were transferred, the volume of test solution was increased to an average of 1.010 { 0.005 ml (0.4% increase). With the P. promelas, the test solution increased to an average of 1.040 { 0.010 ml (3.3% increase). These small differences should not effect the overall performance of the test; however, care must be taken to minimize the amount of culture water transferred with the test organisms. Individual Reference Compounds To ascertain if the miniaturized test system could be used to determine acute toxicity, several toxicity tests were conducted for each of the four individual reference compounds. The mean EC50/LC50 values and their related coefficient of variation (CV) were then calculated to determine the variability between tests (Table 2). For D. magna the CVs using the miniaturized test system ranged from 10% for kepone to 65% for sodium lauryl sulfate. For P. promelas the CVs ranged from 0% for pentachlorophenol to 23% for kepone. These CVs were, with the exception of sodium lauryl sulfate, similar to those obtained when the standard U.S. EPA methods were used (Table 2). To determine if the miniaturized test system yielded comparable toxicity values to those obtained when standard U.S. EPA methods were used, EC50/LC50 values from these two test methods were compared. For most tests, 48-hr EC50/ LC50 values using the miniaturized system were similar (95% confidence intervals overlapped) to those using the standard system (Table 2). When compared to the toxicity values reported in the literature (Table 1), EC50/LC50 values generated using the miniaturized system were either in the same range (LAS for D. magna and P. promelas, SLS for D. magna) or higher (PCP and kepone for D. magna and P. promelas, SLS for P. promelas). TIE Extractions Definitive test results on the synthetic effluent for D. magna indicated an EC50 of 24.5% (Table 3). Following

eesas

AP: EES

5

TOXICITY IDENTIFICATION EVALUATION TEST

TABLE 2 Toxicity Values (EC50/LC50 with Confidence Intervals) Generated Using Miniaturized Test System and Standard EPA Methods Daphnia magna EC50 (mg/liter) Compound

Standard

LAS

5.2 1.5 10 8.8 12.8 7.66

Mean (CV) PCP

2.2 2.5 3.5 2.73

Mean (CV) Kepone

Miniaturized

(3.1–6.3) (0.1–1.0) (5–10) (5–10) (10–20) (57%)

SLS

(1.5–3) (1.5–3) (1.5–6) (25%)

14.1 (6.3–25) 11.2 (6.3–12.5) 12.7 (16%)

Mean (CV)

Standard

Miniaturized

(3.1–6.3) (1.0–10) (5–20) (10–20) (5–20) (48%)

3.9 (2.5–5) 5.6 (2.5–10) 3.4 (0–5)

4.5 (2.5–5) 6.0 (5–10) 4.1 (2.5–5)

4.30 (27%)

4.87 (17%)

1.6 2.1 3.0 2.23

(0.7–3) (1.5–3) (0.8–6) (26%)

1.4 1.6 1.4 1.47

(0.8–1.5) (0.8–3) (0.8–1.5) (8%)

1.1 1.1 1.1 1.10

(0.8–1.5) (0.8–1.5) (0.8–1.5) (0%)

1.6 1.4 1.8 1.60

(1–2) (1–2) (1–2) (10%)

0.3 0.4 0.4 0.37

(0.3–0.5) (0.3–5) (0.3–5) (16%)

0.7 0.5 0.4 0.53

(0.5–1) (0.3–1) (0.3–5) (23%)

5.3 2.1 10.5 12.4 11.7 8.40

1.7 (1–2) 1.5 (1–2) ú1.0 1.60 (9%)

Mean (CV)

Pimephales promelas EC50 (mg/liter)

11.8 (6–12.5) 31.8 (12.5–50) 21.8 (65%)

fractionation of the whole effluent, screening of the individual fractions revealed that toxicity was confined to the seventh fraction (data not provided). In this fraction, there was 100% lethality for both the D. magna and P. promelas. Definitive toxicity tests on fraction 7 yielded an EC50 of 28.3% (estimated 0.28 mg/liter) for D. magna and a LC50 of 21.1% (estimated 0.21 mg/liter) for P. promelas. Fraction 7 was composed of four compounds, three of which (kepone, pentachlorophenol, and Trifluralin) have confirmed toxicity values below 1 mg/liter for both test organisms. Each of these compounds could be responsible for causing the toxicity. The lack of separation of these compounds demonstrates

TABLE 3 Definitive Toxicity Test Results (48-hr) on Synthetic Effluent and Fraction 7

% Concentration (est mg/liter) 100 50 25 12 6 0

AID

(1.0) (0.5) (0.25) (0.12) (0.06) (0.00)

EES 1442

Synthetic effluent D. magna, % dead

D. magna

P. promelas

100 100 45 5 0 0 EC50 Å 24.5%

100 95 30 5 0 0 EC50 Å 28.3%

100 100 70 5 0 0 LC50 Å 21.1%

/

Fraction 7: % Dead

a107$$$$$3

09-26-96 04:09:53

33.8 (25–50)

56.6 (25–100)

the limited sensitivity of the HPLC for separating compounds. Fraction 6, which contained Captan with a confirmed LC50 of 0.20 mg/liter, should also have been toxic to P. promelas. However, this fraction was nontoxic at the screening concentration (estimated 1 mg/liter). One possible explanation for the lack of toxicity is that the P. promelas tests using the miniaturized methods were conducted for 48-hr values versus the reported 96-hr values. CONCLUSIONS

When trying to identify the toxic constituents in an effluent during Phase II of a TIE, one of the limiting factors is generating enough test solution at an appropriate concentration to conduct definitive toxicity tests. As a result, the researcher often needs to modify the standard EPA testing procedures by eliminating a number of test concentrations, reducing the number of replicates, or testing with only one species. Unfortunately, the greater the compromise, the greater the uncertainty in the results. An alternative method for conducting definitive toxicity tests, without increasing the uncertainty, would be to reduce the amount of test solution required for the tests. A miniaturized test system, based on the current U.S. EPA test protocols for static, acute toxicity testing with D. magna and P. promelas, has been described in this paper. The miniaturized system used 48-well microtiter plates for the test vessels

eesas

AP: EES

6

POWELL ET AL.

with each vessel holding 1 ml of test solution. The smaller vessels were determined to be acceptable for testing based on factors such as the organism biomass to test solution ratio, DO concentration, partitioning of the test chemicals to the vessel walls, toxicity of the vessels to the test organisms, and dilution of the test solutions when the organisms are transferred. Toxicity values, as well as the variation among tests, using the miniaturized test system were very similar to those values using the standard U.S. EPA methods. Therefore, it appears that the miniaturized test system can be used to conduct toxicity tests and provide accurate results. The major benefits from miniaturizing the test system is that complete definitive toxicity tests can be conducted when there is only of small amount of test solution available. Within the second phase of the effluent TIE process, this means answers can be obtained sooner with less space required and less waste generated, all of which results in substantial cost savings.

AID

EES 1442

/

a107$$$$$3

09-26-96 04:09:53

REFERENCES Stephan, C. E. (1976). Methods for calculating an LC50. In Aquatic Toxicity and Hazard Evaluation (F. L. Mayer and J. L. Hamerlink, Eds), ASTM STP 634, pp. 65–84. American Society for Testing and Materials, Philadelphia, PA. U.S. EPA (1991a). Methods for Measuring the Acute Toxicity of Effluents and Receiving Waters to Freshwater and Marine Organisms. EPA-600/ 4-90-027. U.S. Environmental Protection Agency, Cincinnati, OH. U.S. EPA (1991b). Methods for Aquatic Toxicity Identification Evaluations, Phase I Toxicity Characterization Procedures, 2nd ed. EPA/600/6-91003. U.S. Environmental Protection Agency, Washington, DC. U.S. EPA (1991c). Technical Support Document for Water Quality-Based Toxins Control. EPA/505/2-90-001. U.S. Environmental Protection Agency, Washington, DC. U.S. EPA (1993). Methods for Aquatic Toxicity Identification Evaluations, Phase II Toxicity Identification Producers for Samples Exhibiting Acute and Chronic Toxicity. EPA/600/R-92/080. U.S. Environmental Protection Agency, Washington, DC.

eesas

AP: EES