Document 1463136

Appendix N PBPK Modeling of Ethylbenzene Exposure of Infants Via Breastmilk

FINAL 15 February 2006

Prepared by Lisa M. Sweeney, Ph.D., DABT and Michael L. Gargas, Ph.D. The Sapphire Group Dayton, Ohio For The Ethylbenzene Panel American Chemistry Council Arlington, VA

Final___________________________________________________________________ Appendix N. PBPK Modeling of Ethylbenzene Exposure of Infants Via Breastmilk INTRODUCTION The transfer of ethylbenzene from maternal tissue to infants via ingestion of breastmilk (Section 6) was determined using PBPK modeling. Details of the model that are specific to this scenario are described in this Appendix. METHODS Model Description The PBPK model for nursing mothers was developed by modifying the existing PBPK model of Haddad et al. (2000). The Haddad et al. model was developed based on blood concentrations in volunteers exposed to EB by inhalation. The choice of this model and its predictive performance are described in greater detail in Appendix P. To ensure that the model application was consistent with EPA default exposure guidelines, the default maternal body weight of 71.8 kg was used and total ventilation was adjusted. Alveolar ventilation was assumed to be equal to 0.6 × total ventilation. The model for a nursing mother was constructed by adding a mammary compartment. Milk was assumed to be in equilibrium with mammary tissue and venous blood exiting the mammary tissue. Blood flow to mammary tissue was estimated as 7 percent of cardiac output, based on the model of Fisher et al. (1997). The blood flow to the richly perfused tissues was reduced from the value used by Haddad et al. (2000) to ensure mass balance. The volume of mammary tissue was taken from ICRP (1975). Milk flow rate was assumed to be constant throughout the nursing periods, at a rate sufficient to produce the average mother’s milk ingestion for a 0-12 month infant specified by EPA (2002) (0.688 L/d). An average infant body weight of 8.5 kg was used (EPA, 2002) to normalize dose to body weight. The milk:blood partition coefficient for EB was estimated as 3, based on analogy to the structurally similar compounds benzene, toluene, xylenes, and styrene, which had measured milk:blood partition coefficients of 2.04-2.98 (Fisher et al., 1997). A mammary:blood partition coefficients was also required for the model. This value was derived by calculation of an estimated mammary:air partition coefficient, which was divided by the blood:air partition coefficient to derive the estimated mammary:blood partition coefficient. The mammary: air partition coefficient for EB was estimated based on the lipid content of human adipose tissue and mammary tissue (Duck, 1990). Partitioning to adipose was approximated by the partitioning to the lipid fraction (71.4 percent) alone. Mammary partitioning was likewise assumed to be primarily to the lipid fraction (30.9 percent). Thus the mammary: air partition coefficient for EB was estimated as adipose: air partition coefficient × (mammary lipid %/adipose lipid %) = 1556 × (30.9/71.4) = 673.

N-1

Final___________________________________________________________________ Activity/Location Schedules for EB Exposed Mothers The model incorporates scheduling of nursing, eating, and moving among environments with different EB concentrations. These schedules are summarized in Table 1. Table 1. Schedules for EB Exposed Mothers Production Worker Schedule Time Activity location 07:45 nurse home 08:00 vehicle 08:45 production 11:45 nurse home/indoor air 12:00 home/indoor air 12:45 production 15:15 vehicle 16:00 nurse home 16:15 outdoors 17:45 home 21:45 nurse home 22:00 home

At-Home Mother Schedule Time Activity location 07:45 nurse home 08:00 meal 1 home 11:45 nurse home 12:00 meal 2 home 12:30 vehicle 14:00 outdoors 15:30 nurse home 15:45 home 18:00 meal 3 home 21:45 nurse home 22:00 home

Breast milk ingestion by the infant was assumed to occur during four episodes per day, with durations of 15 minutes per episode, described in the model as a pulse function. The average daily amount of time spent in different environments was specified for two groups of nursing mothers, at-home adults and production workers. The exact schedules for nursing and daily activities that were used in the modeling were developed by the PBPK modeling/risk assessment contractor. These schedules incorporate the total specified time for different environments and practical considerations. For example, motor vehicle occupancy (driving) occurs immediately before and after work. Nursing in the production area or in a motor vehicle is not likely to occur, so nursing was not scheduled in these locations. Dietary ingestion by the mother was neglected for the production worker because of the small contribution it makes to the mother’s exposure. Dietary ingestion for the at-home mother was described as occurring in three meals, consisting of 20, 40, and 40 percent of the total intake. Each meal was assumed to last 30 minutes, and the ingestion was described as a pulse function. Uptake from the stomach to the liver blood supply was modeled as first order, occurring at a rate equal to that determined for rats dosed with EB in corn oil (Faber et al., 2006). The inhalation exposure, ingestion, and nursing schedules were incorporated in the model using pulse functions. Central tendency and upper-bound estimates for dietary intake and EB concentrations in the different environments were taken from Chapters 6 and 7. The non-occupational exposure concentrations were those derived for urban smokers (those

N-2

Final___________________________________________________________________ with the highest exposures outside the workplace). Infant doses from breastfeeding were based on model simulations in which mothers had been exposed to EB under the specified daily schedule for over two weeks, in order to establish equilibrium. RESULTS A sample time course of mammary venous blood concentration for a mother with occupational exposure to EB is provided in Figure N-1. Breastmilk is assumed to be in equilibrium with the mammary venous blood. Note that the time starts at 336 hrs (two weeks), for establishment of equilibrium.

Worker exposure--upper bound

Mammary venous blood concentration (mg/L)

0.018 0.016 0.014 0.012 0.01 0.008 0.006 0.004 0.002 0 336

340

344

348

352

356

360

Time (hr)

Figure N-1. Predicted mammary venous blood concentrations for worker exposure to “upperbound” concentrations of ethylbenzene. Filled squares are used to highlight blood concentrations during nursing periods. Daily EB ingestion from breastmilk was calculated for breastfed infants with mothers who are urban smokers using the PBPK model. Results are summarized in Table N-2.

N-3

Final___________________________________________________________________ Table N-2. Calculation

Infant Exposure to EB via Breastmilk Ingestion—PBPK Model

Central Tendency Upper Bound

Daily Ingested EB Dose (mg/kg/d) Child of At-Home Mother Child of Production Worker -6 9.1 × 10 1.8 × 10-4 2.0 × 10-5 1.7 × 10-3

N-4

Final___________________________________________________________________ REFERENCES Duck, FA. (1990). Physical properties of tissue. Academic Press. San Diego, CA. Faber WD, Roberts LS, Stump DG, Tardif R, Krishnan K, Tort M, Dimond S, Dutton D, Moran E, Lawrence W. (2006). Two generation reproduction study of ethylbenzene by inhalation in Crl-CD rats. Birth Defects Res B Dev Reprod Toxicol. In press. Epub 2005 Dec 8 Fisher, J; Mahle, D; Bankston, L; Greene, R; Gearhart, J. (1997). Lactational transfer of volatile chemicals in breast milk. AIHA J. 58:425-431. Haddad S, Charest-Tardif G, Tardif R, Krishnan K. (2000). Validation of a physiological modeling framework for simulating the toxicokinetics of chemicals in mixtures. Toxicol Appl Pharmacol. 167:199-209. International Commission on Radiological Protection (ICRP) (1975). Report of the Task Group on Reference Man. ICRP No. 23. Pergamon Press, Inc. Elmsford, New York. U.S. Environmental Protection Agency (2002). Child-Specific Exposure Factors Handbook. Interim Report. EPA-600-P-00-002B. National Center for Environmental Assessment, Washington, DC.

N-5

Appendix O Robust Summaries for Selected Studies

1

ETHYLBENZENE TOXICITY HAZARD KEY STUDIES [RS - 1] Acute Oral Toxicity Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

Not specified

Method:

Experimental (Non-regulatory)

Type:

LD50

GLP:

Pre-GLP

Year:

1962

Species/Strain:

Rat / Carworth-Wistar

Sex:

Male

#/dose:

5

Vehicle:

Unknown (water, corn oil, or a solution of 25% sodium 3,9-diethyl-6tridecanol sulfate “Tergitol Penetrant 7”)

Route of Administration:

Oral gavage

Doses/Concentrations:

Logarithmic series differing by a factor of 2

Dose Volume Administered:

Single dose

Post Dose Observation Period:

14 days

Results (LD50):

5.46 g/kg bwt (5.09 – 5.86) (This value is noted in some secondary sources as 4.7 g/kg bwt, presumably due to misinterpretation of the study table finding and inappropriate conversion of the value from mL/kg to g/kg. The table value for ethylbenzene is given as 5.46 (5.09 – 5. 86)* under a column header units of mL/kg; the * footnote at the end of the table, however, notes “as gm/kg in a suitable vehicle”, so conversion is not appropriate).

Remarks:

The animals were 4 – 5 weeks of age and weighed 90 -120 g and were not fasted prior to dosing. The most probable LD50 value and the fiducial range were estimated by the method of Thompson using the tables of Weil. The figures in parentheses show limits of ± 1.96 standard deviations.

2

Conclusion:

The acute oral LD50 for ethylbenzene in male rats is 5.46 g/kg bwt (5.09 – 5.86).

Data Quality:

2 – Reliable study with restrictions. Study is pre-GLP but sufficiently documented and meets generally accepted scientific principles.

Reference:

Smyth Jr., H.F., Carpenter, C.P., Weil, C.S., Pozzani, U.C. and Striegel, J.A. (1962). Range finding toxicity data: List VI. Am. Ind. Hyg. Assoc. J. 23:95-107.

3

[RS - 2] Acute Dermal Toxicity Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

Not specified

Method:

Experimental (Non-regulatory)

Type:

LD50

GLP:

Pre-GLP

Year:

1962

Species/Strain:

Rabbit / New Zealand White

Sex:

Male

#/dose:

4

Vehicle:

Unknown (water, corn oil, or a solution of 25% sodium 3,9-diethyl-6tridecanol sulfate “Tergitol Penetrant 7”)

Route of Administration:

Skin

Doses/Concentrations:

Logarithmic series differing by a factor of 2

Dose Volume Administered:

Single dose

Post Dose Observation Period:

14 days

Results (LD50):

17.8 mL/kg bwt (Equivalent to 15.3 g/kg bwt)

Remarks:

The animals weighed 2.5 – 3.5 kg. Technique similar to the oneday cuff method of Draize and associates. The fur was removed from the entire trunk by clipping and the dose was retained beneath an impervious plastic film. The animals were immobilized during the 24-hour contact period, after which the film was removed. The most probable LD50 value was estimated by the method of Thompson using the tables of Weil.

Conclusion:

The acute dermal LD50 for ethylbenzene in male rabbits is 17.8 mL/kg bwt (15.3 g/kg bwt).

Data Quality:

2 – Reliable study with restrictions. Study is pre-GLP but sufficiently documented and meets generally accepted scientific principles.

4

Reference:

Smyth Jr., H.F., Carpenter, C.P., Weil, C.S., Pozzani, U.C. and Striegel, J.A. (1962). Range finding toxicity data: List VI. Am. Ind. Hyg. Assoc. J. 23:95-107.

5

[RS - 3] Acute Inhalation Toxicity Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

Not specified

Method:

Experimental (Non-regulatory)

Type:

LC50

GLP:

Pre-GLP

Year:

1962

Species/Strain:

Rat / Carworth-Wistar

Sex:

Males or Females

#/dose:

6

Vehicle:

Air

Route of Administration:

Inhalation

Doses/Concentrations:

In an essentially logarithmic series differing by a factor of 2

Exposure Time:

4 hours

Post Dose Observation Period:

14 days

Results (LC50):

4000 ppm

Remarks:

The animals’ age and weight ranges were not provided. Inhalation was by metered vapor concentrations conducted with flowing streams of vapor prepared by proportioning pumps. Concentrations recorded were nominal and not analytically verified. Results given as fractional mortality among 6 rats within 14 days. 4000 ppm ethylbenzene administered for 4 hours yielded 14 day mortality in 3 of 6 rats.

Conclusion:

The acute inhalation LC50 for ethylbenzene in rats is 4000 ppm.

Data Quality:

2 – Reliable study with restrictions. Study is pre-GLP but sufficiently documented and meets generally accepted scientific principles.

Reference:

Smyth Jr., H.F., Carpenter, C.P., Weil, C.S., Pozzani, U.C. and Striegel, J.A. (1962). Range finding toxicity data: List VI. Am. Ind. Hyg. Assoc. J. 23:95-107.

6

[RS - 4] Genetic Toxicity - In Vitro Gene Mutation: Bacterial Reverse Mutation Assay Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

> 99%

Method:

EU Annex V.B.14. OECD Guideline 471 - Genetic Toxicology: Salmonella typhimurium Reverse Mutation Assay

Type:

Bacterial Reverse Mutation Test

System of Testing:

Bacteria

GLP:

Yes

Year:

1992

Species/Strain:

S. typhimurium / TA97, TA98, TA100, TA1535, TA1537

Metabolic Activation:

With and Without

Species/cell type:

Male Sprague-Dawley Rat and Syrian Hamster / Liver; Aroclor 1254-induced (S-9 fraction)

Concentrations: Tested

10 to 10000 µg/plate

Vehicle

Distilled water

Remarks for Test Conditions:

Control plates were set up with solvent alone and with an appropriate known positive control compound. The S9 fractions were prepared from Aroclor-induced rats and hamsters. Ethylbenzene was tested initially in a toxicity assay with a range of test concentrations to establish the appropriate dose range for the mutagenicity assay. Ethylbenzene was tested at half-log dose intervals up to 10000 µg/plate. The mutagenicity assay was then performed based on the results of the toxicity assay taking into account the effect on cell viability and any possible positive increases in mitotic gene conversion. Control plates were set up with solvent alone and with the positive control compounds in the absence of S-9 activation (sodium azide for TA1535 and TA100; 9aminoacridine for TA97 and TA1537; 4-nitro-o-phenylenediamine for TA98) and with S-9 metabolic activation (2-aminoanthracene for all strains). Ethylbenzene was designated non-mutagenic only after it had been tested in strains TA98, TA100, TA1535, TA97 and TA1537 without activation, and with 10% and 30% rat and hamster S-9 activation.

Results:

Negative

7

Remarks: Conclusion:

Ethylbenzene did not induce reverse gene mutations in bacteria.

Data Quality:

1 – Reliable without restrictions. No circumstances occurred that would have affected the quality or integrity of the data

Reference:

Zeiger, E., Anderson, B., Haworth, S., Lawlor, T. and Mortelmans, K. (1992). Salmonella mutagenicity tests: V. Results from the testing of 311 chemicals. Environmental and Molecular Mutagenesis, Volume 19 (Supplement 21):2-141.

8

[RS - 5] Genetic Toxicity - In Vitro Gene Mutation: Bacterial Reverse Mutation Assay Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

> 99%

Method:

Not specified

Type:

Bacterial Reverse Mutation Test

System of Testing:

Bacteria

GLP:

Yes

Year:

1985

Species/Strain:

Escherichia coli WP2, WP2uvra

Metabolic Activation:

With and Without

Species/cell type:

Rat / Liver (S9 Fraction)

Concentrations: Tested

0.2 to 2000 µg

Vehicle

DMSO

Remarks for Test Conditions:

The article reports on the laboratory’s testing of 41 compounds over a 5 year period. The test methods were modified over time with one bacterial assay method used from 1975-1980 and variations to this method employed after 1980. The tests for ethylbenzene were conducted during 1979-1980; hence the first method is presumed to have been used and is described below. The method used was the Plate-Incorporation Assay, using S9 microsomal fraction obtained from a rat liver homogenate from rats pre-treated with Aroclor 1254. A range of amounts of test compound were tested (0.2, 2, 20, 500 and 2000 µg/plate) both in the presence and absence of S9 mix. Overnight broth cultures were washed and resuspended in phosphate buffer pH 7.0. The suspension was then distributed in 2 mL volumes into universal containers and 20 µL test compound solution was added (-S9). For the tests incorporating microsomal activation (+S9), 0.5 mL S9 mix was added to each 2 mL bacterial suspension culture together with 25 µL test compound solution. All cultures were incubated at 37ºC for 1 hour before 0.1 mL volumes were seeded onto minimal agar plates with the appropriate amino acid supplement. Appropriate dilutions were placed onto nutrient agar to determine the numbers of survivors. The plates were then incubated at 37ºC before the colonies were counted.

9

In the mutation assay control plates were set up with the solvent alone and with a known positive control compound (specific compound not stated). All tests were carried out in quadruplicate. Two replicate assays were carried out on different days in order to confirm the reproducibility of results. Data were interpreted on the basis of a consistent doubling of the spontaneous reversion frequency confirmed by a dose-response relationship. A positive response was given in cases where the number of induced revertants was less than twice the spontaneous rate, but a reproducible dose-related increase in revertants was detected. Results:

Negative

Remarks: Conclusion:

Ethylbenzene did not induce reverse gene mutations in bacteria.

Data Quality:

1 – Reliable without restrictions. No circumstances occurred that would have affected the quality or integrity of the data.

Reference:

Dean, B.J., Brooks, T.M., Hodson-Walker, G. and Huston, D.H. (1985). Genetic toxicology testing of 41 industrial chemicals. Mutat. Res. 153:57-77.

10

[RS - 6] Genetic Toxicity - In Vitro Gene Mutation: Yeast Mitotic Gene Conversion Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

> 99 %

Method:

Not specified

Type:

Yeast Mitotic Gene Conversion Assay

System of Testing:

Saccharomyces cerevisiae

GLP:

Yes

Year:

1985

Species/Strain:

Saccharomyces cerevisiae

Metabolic Activation:

With and Without

Species/cell type:

Rat / Liver (S9 Fraction)

Concentrations: Tested

0.01 to 5 mg/mL

Vehicle

DMSO

Remarks for Test Conditions:

Yeast cells were grown in log phase, washed and re-suspended in 2.5 strength YEPD broth at a concentration of 10 x 106 cells/mL. The suspension was then divided into 1.9 mL amounts in 30 mL universal containers and 0.1 mL of the test compound solution was added (-S9). For the experiments with metabolic activation (+S9), 0.1mL of test compound was added to 1.6 mL of yeast cell suspension, together with 0.3 mL S9 mix (prepared from the livers of Aroclor-induced rats). The cultures were incubated with shaking at 30 ºC for 18 hours. Aliquots were then plated onto YM plates supplemented with either histidine or tryptophan to determine the number of prototrophs as each locus, and dilutions were spread onto YEPD plates to determine cell viability. Initially a range of concentrations of test compound (0.01, 0.1, 0.5, 1.0, and 5.0 mg/mL) were tested as solubility allowed. A second experiment was then performed based on the initial test results taking into account the effect of the chemical on cell viability and any possible positive effect. The test compound was considered to increase the rate of mitotic gene conversion if there was a reproducible dose-related increase in the number of prototrophs per 106 survivors together with an increase in the number of prototrophs per plate.

11

Results:

Negative

Remarks: Conclusion:

Ethylbenzene did not induce mitotic gene conversion in yeast.

Data Quality:

1 – Reliable without restrictions. No circumstances occurred that would have affected the quality or integrity of the data.

Reference:

Dean, B.J., Brooks, T.M., Hodson-Walker, G. and Huston, D.H. (1985). Genetic toxicology testing of 41 industrial chemicals. Mutat. Res. 153:57-77.

12

[RS - 7] Genetic Toxicity - In Vitro Gene Mutation: Mammalian Cell Mutation Assay Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

Not specified

Method:

Not specified

Type:

Cell Mutation Assay at the Thymidine Kinase Locus (tk+/-)

System of Testing:

Mouse Lymphoma Cells L 5178Y

GLP:

Not specified

Year:

1988

Species/Strain:

Mouse Lymphoma Cells L 5178Y

Metabolic Activation:

Only Without

Species/cell type:

Not applicable

Concentrations: Tested

10 to 80 µg/mL (highest non-lethal concentration)

Vehicle

DMSO

Remarks for Test Conditions:

An initial toxicity test was conducted to measure cell population expansion. Experimental Design: The experiment consisted of the following groups: vehicle control, four cultures; positive control, two cultures; at least 5 test compound concentrations, two cultures per concentration. Positive controls included ethyl methanesulphonate (EMS; 250 µg/mL) and methyl methansulphonate (MMS, 15 µg/mL). Trial I without S9 mix: 10, 20, 40, 80, 160 µg/mL Trial 2 without S9 mix: 20, 40, 60, 80, 100 µg/mL Mutation Experiment: Exposure: Each exposed culture consisted of 6 x 106 cells in a final volume of 10 mL Fischer’s medium containing 5% heat-inactivated horse serum (F5P) in a 30 mL screw-cap plastic tube. The tube was incubated for 4 hours on a horizontal axis roller drum rotating at 10 rpm. At the end of the incubation period, the cells were sedimented by centrifugation at 500 g.av. for 10 minutes, washed, and finally resuspended in 10 mL Fischer’s medium containing 10% heat-

13

inactivated horse serum (F10P). The cell suspensions (3 x 105 cells/mL) were incubated for a 2 day expression period, the cell population density being adjusted back to 20 mL of 3 x 105 cells/mL after 24 hours. After 48 hours, the cell population densities were estimated and culture volumes containing 3 x 106 cells adjusted to 15 mL with F10P, giving a cell population density of 2 x 105. Cloning Efficiency: A 0.1 mL sample of the cell suspension was withdrawn and diluted 1:100. Three 0.1 mL samples (200 cells) of the diluted cultures were transferred to 30 mL tubes, mixed with 25 mL cloning medium containing 10% heat inactivated horse serum containing 0.35% Noble agar and poured into 90 mm Petri plates. Mutant Selection: Three aliquots (each containing 106 cells) of the remaining culture were distributed to 30 mL tubes, mixed with 10 mL cloning medium to give final concentrations of 0.35% Noble agar and 3 ug trifluorothymidine/mL, then poured into 90 mm Petri plates. Incubation: The agar was gelled at 4ºC for 5-10 minutes, and then the plates were incubated for 11-14 days in 5% CO2:95% air at 37ºC. Colony Counting: Colonies were counted using an automated colony counter. Calculations: The Relative Total Growth and Mutant Fraction were calculated. A test was considered positive when, out of 3 trials, a positive trial was reproducible. Statistics: Statistical analyses consisted of a dose-trend test and a variance analysis of pair-wise comparisons of each dose against the vehicle control. Results:

Positive

Remarks:

Ethylbenzene was mutagenic in two experiments without S9 mix only at the highest non-lethal concentration (80 ug/mL). At this concentration there was significant cytotoxicity, with the relative total growth in two trails being 34 or 13% of the control level. Average Mutant Fraction Trial 1 Trial 2 DMSO 60 31 10 µg/mL 56 20 µg/mL 53 43 40 µg/mL 67 29 60 µg/mL 43 80 µg/mL 589* 150* 100 µg/mL lethal 160 µg/mL lethal EMS 149* 225* MMS 145* 107* * Statistically significantly higher (p < 0.05) than control

Conclusion:

Ethylbenzene induced mutations in mouse lymphoma cells only at the highest non-lethal and cytotoxic concentration.

14

Data Quality:

2 – Reliable with restrictions. Protocol not current on appropriate toxicity levels and evaluation criteria. Positive results accompanied by substantial increases in cytotoxicity that complicate interpretation.

Reference:

McGregor, D.B., Brown, A., Cattanach, P., Edwards, I., McBride, D., Riach, C. and Caspary, W.J. (1988). Responses of the L5178Y tk+/tk- mouse lymphoma cell forward mutation assay. III: 72 coded chemicals. Environ. Mol. Mutag. 12:85-154.

15

[RS - 8] Genetic Toxicity - In Vitro Gene Mutation: Mammalian Cell Mutation Assay Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

99.7%

Method:

OECD Guideline 476 – In Vitro Mammalian Cell Gene Mutation Test

Type:

Cell Mutation Assay at the Thymidine Kinase Locus (tk+/-)

System of Testing:

Mouse Lymphoma Cells L 5178Y (soft agar method)

GLP:

Yes

Year:

2000

Species/Strain:

Mouse Lymphoma Cells L 5178Y

Metabolic Activation:

With and Without

Species/cell type:

Male Wistar Rat / Liver; Phenobarbital and β-Naphthoflavoneinduced (S-9 fraction)

Concentrations: Tested

8.6 to 100 µg/mL (without S9 mix) 68.8 to 900 µg/mL (with S9 mix)

Vehicle Remarks for Test Conditions:

DMSO The study investigated the potential of ethylbenzene to induce mutations at the thymidine kinase (tk) locus in L5178Y tk+/- mouse lymphoma cells using six independent experiments, a treatment period of 4 hours, an expression and growth period of 72 hours and two parallel cultures. According to an initial range finding cytotoxicity test for the determination of the experimental doses and taking into account the cytotoxicity actually found in the first main experiment, the following doses were evaluated: Experiment I without S9 mix: 8.6, 17.2, 34.4, 68.8 µg/mL with S9 mix: 68.8, 137.5, 275, 550, 825 µg/mL Experiment II without S9 mix: 45, 60, 75, 90, 100 µg/mL with S9 mix: 300, 450, 600, 750, 900 µg/mL The third experiment was carried out selecting the same dose

16

ranges as in the second experiment to clarify the results of the first two experiments. However, due to excessively strong toxic effects with S9 mix, this experiment part (with S9 mix) had to be terminated prior to the generation of experimental data. Experiment III without S9 mix: 45, 60, 75, 90, 100 µg/mL Experiments IV and V which were performed solely in the presence of metabolic activation also had to be terminated right after treatment since very few surviving cells were detectable. Therefore experiment VI was carried out with a dose range adjusted to the toxic effects observed in experiments III to V. Experiment VI with S9 mix: 25, 50, 75, 100, 150 µg/mL Concurrent negative and solvent controls (DMSO) were performed. Positive control substances were methylmethane sulfonate (without metabolic activation) and 3-methylcholanthrene (with metabolic activation). Evaluation of Results/Statistical Analyses - The test substance was classified as positive if it induced either a reproducible concentration-related increase in the mutant frequency or a reproducible positive response for at least one of the test points. A test substance producing neither a reproducible concentrationrelated increase in the mutant frequency nor a reproducible positive response at any of the test points is considered non-mutagenic in this system. As per OECD Guideline No. 476, no statistical evaluation of the data was performed. Results:

Ambiguous

Remarks:

No precipitation of the test article was observed up to the maximal concentration evaluated. Experiment I: Mutagenic effects were observed but only at toxic concentrations without S9 mix (34.4 and 68.8 µg/mL) and with metabolic activation (825 µg/mlL. Experiment II: For confirmation of the results of the 1st experiment, a second was carried out. The dose range was adjusted to cover the critical dose range more closely and increasing doses were selected for a better demonstration of a possible dose-response relationship. In the second experiment mutagenicity was not observed, i.e. the findings of the first experiment were not reproduced. Relevant toxic effects were observed at 90 µg/mL and above in the absence and at 900 µg/mL and above in the presence of S9 mix. Experiment III (without S9 mix) and Experiment VI (with S9 mix): There was no increase in the mutation frequency up to the maximum doses tested either without S9 mix or after the addition of a metabolizing system.

17

In the third experiment strong toxic effects were observed at 100 µg/mL without metabolic activation. In the sixth experiment relevant toxicity occurred at 150 µg/mL. At higher concentrations exceedingly strong toxic effects precluded the evaluation of results. According to the author the striking shift of toxicity in the presence of S9 mix may be based upon different batches of S9 used (differences in the content of lipids may be responsible, since whenever a test item binds to lipids shifts in toxicity are likely to occur because the concentration of free substance available to the cells is also different). The negative controls (untreated and vehicle) gave mutant frequencies within the range expected for the L5178Y cell line. Both positive controls showed a distinct increase in induced total mutant colonies and an increase of the relative quantity of small versus large colonies. Conclusion:

Ethylbenzene gave an ambiguous result in the mouse lymphoma assay. The author considered ethylbenzene to be non-mutagenic in the mouse lymphoma assay since the effects observed in the first experiment were not reproduced in two additional experiments carried out independently of each other. The findings of the first experiment were thought to be caused by toxicity-related secondary effects and hence not indicative a true mutagenic potential of the test substance.

Data Quality:

3 – Not reliable. The protocol used in this assay is not a standard protocol. The 3-day expression period is considered suboptimal.

Reference:

Wollny, H.E. (2000). Cell mutation assay at the thymidine kinase locus (TK+/-) in mouse lymphoma cells (soft agar method) with ethylbenzene. RCC-CCR Project No. 635300. RCC-Cytotest Cell Research GmbH, Germany. Sponsored by the Styrenics Steering Committee, CEFIC, Brussels, Belgium.

18

[RS – 9] Genetic Toxicity - In Vitro Gene Mutation: Mammalian Cell Mutation Assay Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

99.85% supplied by Sigma, St. Louis, Missouri (Lot number: 01353MC)

Method:

OECD Guideline 476 – in Vitro Mammalian Cell Gene Mutation Test (1997) USEPA OPPTS 870.5300 (1998) EC, B.17 (2000)

Type:

Cell Mutation Assay at the Thymidine Kinase Locus (tk+/-)

System of Testing:

Mouse Lymphoma Cells L5178Y in culture

GLP:

Yes

Year:

2006

Species/Strain:

Mouse Lymphoma Cells L5178Y

Metabolic Activation:

With and Without

Species/cell type:

S9 liver homogenates prepared from Aroclor® 1254-induced male Sprague-Dawley rats, purchased from a commercial source.

Concentrations: Tested Vehicle Remarks for Test Conditions:

10 to120 µg/mL ( without S9) 10 to 120 µg/mL (with S9) The study investigated the potential of ethylbenzene to induce mutations at the thymidine kinase (tk) locus in L5178Y tk+/- mouse lymphoma cells using three experiments. Assay A1- Preliminary Toxicity Assay In a preliminary toxicity assay, the test material was evaluated at concentrations ranging from 4.2 to 1060 µg/mL in the absence and presence of metabolic activation system (S9). The highest concentration represents the limit dose of 10 mM and exceeded the solubility of the test material in the treatment medium. In the absence of S9, ethylbenzene was excessively toxic at the five highest concentration levels (i.e. 66.25, 132.5, 265, 530 and 1060 µg/mL) as measured by Day 2 relative suspension growth (RSG). The remaining cultures had day 2 RSG value ranging from 52 to 107%. In the presence of S9, excessive toxicity was observed at the 132.5 µg/mL concentration level and above as measured by Day 2 RSG. The remaining concentration levels had RSG values ranging from 15 to 95%. Based upon the results of this assay,

19

concentrations in the range of 10 to 120 µg/mL were selected for the initial mutagenicity assay both in the absence and presence of S9. Assay B1 – Initial Mutagenicity Assay In the initial mutagenicity assay in the absence of S9, cultures treated with ethylbenzene showed excessive toxicity at concentration levels from 60 to 120 µg/mL as assessed by day 2 RSG and were not available for cloning. The remaining concentration levels (from 10 to 50 µg/mL) showed moderate to no toxicity and were selected for cloning. The relative total growth (RTG) in the test material treated cultures varied from 9 to 107%. In the presence of S9, cultures treated with ethylbenzene displayed excessive toxicity at concentrations of 80, 100 and 120 µg/mL. The RTG value in the remaining ethylbenzene treated (from 10 to 70 µg/mL) cultures varied from 21 to 104%. Assay C1- Confirmatory Mutagenicity Assay Based on the initial mutagenicity assay, the following concentrations were selected for the confirmatory mutagenicity assay – 10, 30, 38, 42, 46, 50, 54, 60 and 70 µg/mL in the absence of S9. Day 2 RSG values indicated excessive toxicity at 54, 60 and 70 µg/mL concentration levels. In the remaining cultures, day 2 RSG among these concentration levels, ranged from 15 to 99%. In the presence of S9, cultured treated with 90µg/ml of ethylbenzene displayed excessive toxicity as determined by day 2 RSG. Cultures treated with 82 µg/mL had individual RTG values of 9 and 10%. The mutant plates for these cultures were not enumerated since the average RTG value of the two replicates was 10% the test substance was considered negative in this assay. The test substance was considered equivocal in this assay if there was a significant increase in mutant frequency that met the criteria for a positive response only at RTG values > 10% and < 20%, or there was no evidence of increase in mutant frequency at RTG values >20%. As per OECD Guideline # 476, no statistical evaluation of the data was performed.

20

Results:

Negative.

Remarks:

There was no increase in mutant frequency above 95 x 10-6, the average of the concurrent controls, and there was no a positive dose related linear trend at any of concentrations of ethylbenzene evaluated in this assay with or without metabolic activation. Cultures treated with the positive control chemical induced a positive response as compared to the solvent control. The solvent control values were within the range of the laboratory historical data. All criteria for a valid assay were satisfied.

Conclusion:

Based upon results of the initial and confirmatory mutagenicity assays, ethylbenzene was considered to be non-mutagenic in the absence and presence of metabolic activation in this in vitro mouse lymphoma (L5178Y tk+/-) forward mutation assay.

Data Quality:

1 – Reliable without restrictions. No circumstances occurred that would have affected the quality or integrity of the data

Reference:

Seidel, S.D., Schisler, M.R. and Kleinert, K.M. (2006). Evaluation of Ethylbenzene in the Mouse Lymphoma (L5178YTK+/-) Forward Mutation Assay. Toxicology & Environmental Research and Consulting, The Dow Chemical Company, Study ID: 051157.

21

[RS - 10] Genetic Toxicity - In Vitro Chromosome Aberrations Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

Not specified

Method:

According to Galloway et al., Environ. Mol. Mutagen 10 (Suppl. 10): 1-175, 1987

Type:

Sister Chromatid Exchange Assay

System of Testing:

Chinese Hamster Ovary Cells

GLP:

Yes

Year:

1999

Species/Strain:

Chinese Hamster Ovary Cells

Metabolic Activation:

With and Without

Species/cell type:

Male Sprague-Dawley Rat / Liver; Aroclor 1254-induced (S-9 fraction)

Concentrations: Tested

75.5, 99.5, 125 µg/mL (without S9 mix) 125, 137.5, 150 µg/mL (with S9 mix)

Vehicle Remarks for Test Conditions:

DMSO Each test consisted of concurrent solvent and positive controls and 4 doses of ethylbenzene; the high dose was limited by toxicity. Cultures were handled under gold lights to prevent photolysis of bromodeoxyuridine-substituted DNA. S9 was prepared from Aroclor 1254 induced male Sprague-Dawley liver. In the SCE test without S9, Chinese hamster ovary (CHO) cells were incubated for 26 hours with ethylbenzene in supplemented McCoy’s 5A medium. Bromodeoxyuridine (BrdU) was added 2 hours after culture initiation. After 26 hours, the medium containing ethylbenzene was removed and replaced with fresh medium plus BrdU and Colcemid, and incubation was continued for 1.5 hours. Cells were then harvested by mitotic shake-off, fixed, and stained with Hoechst 33258 and Giemsa. In the sister chromatid exchange (SCE) test with S9, cells were incubated with ethylbenzene, serum free medium, and S9 for 2 hours. The medium was then removed and replaced with medium containing serum and BrdU and no ethylbenzene, and incubation proceeded for an additional 25.8 hours, with Colcemid present for the final 2 hours. Harvesting and staining were the same as for cells treated without S9. All slides

22

were scored blind and those from a single test were read by the same person. Fifty second-division metaphase cells were scored for frequency of SCEs/cell from each dose level. Evaluation of Results/Statistical Analyses - Statistical analyses were conducted on the slopes of the dose-response curves and individual data points. An SCE frequency of 20% above the concurrent solvent control value was chosen as a statistically conservative positive response. An increase of 20% or greater at any single dose was considered weak evidence of activity; increases at two or more doses resulted in a determination that the trial was positive. A statistically significant trend in the absence of any responses reaching 20% above background was considered an equivocal response. Results:

Negative

Remarks:

Dose(µg/mL)

Relative Change of SCEs/Chromosome

-S9 75.5

- 0.82

99.5

- 6.31

125

7.54

Mitomycin C 0.001

39.81

+S9 125

5.95

137.5

0.58

150

- 1.89

Cyclophosphomide 0.35

36.03

Conclusion:

Ethylbenzene did not induce sister chromatid exchanges in Chinese hamster ovary cells in vitro.

Data Quality:

1 – Reliable without restrictions. No circumstances occurred that would have affected the quality or integrity of the data

Reference:

National Toxicology Program. (1999). Toxicology and carcinogenesis studies of ethyl-benzene (CAS No. 100-41-4) in F344/N rats and B6C3F1 mice (Inhalation studies) (Tech. Rep. Ser. No. 466; NIH Publ No. 99-3956), National Toxicology Program, U.S. Dept. of Health and Human Services. Research Triangle Park, NC

23

[RS - 11] Genetic Toxicity - In Vitro Chromosomal Aberrations Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

Not specified

Method:

According to Galloway et al., Environ. Mol. Mutagen 10 (Suppl. 10): 1-175, 1987

Type:

Cytogenetic Assay, Chromosomal Aberrations

System of Testing:

Chinese Hamster Ovary Cells

GLP:

Yes

Year:

1999

Species/Strain:

Chinese Hamster Ovary Cells

Metabolic Activation:

With and Without

Species/cell type:

Male Sprague-Dawley Rat / Liver; Aroclor 1254-induced (S-9 fraction)

Concentrations: Tested

75, 100, 125 µg/mL (without S9 mix) 75, 100, 125 µg/mL (with S9 mix)

Vehicle Remarks for Test Conditions:

DMSO Each test consisted of concurrent solvent and positive controls and 4 doses of ethylbenzene; the high dose was limited by toxicity. Cultures were handled under gold lights to prevent photolysis of bromodeoxyuridine-substituted DNA. S9 was prepared from Aroclor 1254 induced male Sprague-Dawley liver. In the chromosomal aberrations test without S9, cells were incubated in McCoy’s 5A medium with ethylbenzene for 8.5 hours; Colcemid was added and incubation continued for 2 hours. The cells were then harvested by mitotic shake-off, fixed, and stained with Giemsa. For the chromosomal aberrations test with S9, cells were treated with ethylbenzene and S9 for 2 hours, after which the treatment medium was removed and the cells were incubated for 8.5 hours in fresh medium, with Colcemid present for the final 2 hours. Cells were then harvested in the same manner as for the treatment without S9. Cells were selected for scoring on the basis of good morphology and completeness of karyotype (21 ± 2 chromosomes). All slides were scored blind and those from a single test were read by the same person. One hundred first-division metaphase cells were

24

scored at each dose level. Classes of aberrations included simple (breaks and terminal deletions), complex (rearrangements and tranlocations), and other (pulverized cells, despiralized chromosomes, and cells containing 10 or more aberrations). Evaluation of Results/Statistical Analyses - Chromosome aberration data were presented as percentage of cells with aberrations. To arrive at a statistical call for a trial, analyses were conducted on both the dose response curve and individual data points. For a single trial, a statistically significant difference for one dose point and a significant trend were considered weak evidence for a positive response; significant differences for two or more doses indicated the trial was positive. A positive trend test in the absence of a statistically significant increase at any dose level resulted in an equivocal call. Results:

Negative

Remarks:

Compound

Dose (µg/mL)

Cells with Aberrations (%)

-S9 DMSO Ethylbenzene

3 75

1

100

3

125

5

1

22

Mitomycin C +S9 DMSO Ethylbenzene

3 75

4

100

1

125

1

Cyclophosphamide 50

36

Conclusion:

Ethylbenzene did not induce chromosomal aberrations in Chinese hamster ovary cells in vitro.

Data Quality:

1 – Reliable without restrictions. No circumstances occurred that would have affected the quality or integrity of the data.

Reference:

National Toxicology Program. (1999). Toxicology and carcinogenesis studies of ethyl-benzene (CAS No. 100-41-4) in F344/N rats and B6C3F1 mice (Inhalation studies) (Tech. Rep. Ser. No. 466; NIH Publ No. 99-3956), National Toxicology Program, U.S. Dept. of Health and Human Services. Research Triangle Park, NC

25

[RS –12] Genetic Toxicity - In Vitro Micronucleus Assay Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

> 99% supplied by the National Toxicology Program, USA

Method:

Experimental (Non – regulatory) - published in Kerckaert, GA., R.Brauninger, R.A. LeBoeuf and R.J. Isfort. (1996). Use of the Syrian hamster embryo cell transformation assay for carcinogenicity prediction of chemicals currently being tested by the National Toxicology Program in rodent bioassays. Env. Health Perspectives. 104(Suppl.5):1075-1084.

Type:

Induction of micronucleus in the Syrian hamster embryo (SHE) cell in vitro micronucleus assays

System of Testing:

Syrian Hamster Embryo Cells

GLP:

Not specified

Year:

1997

Species/Strain:

Syrian Hamster Embryo Cells

Metabolic Activation:

Without

Species/cell type:

Syrian Hamster Embryo Cell Culture

Concentrations: Tested

25,50,100 and 200 μg/mL

Vehicle

DMSO

Remarks for Test Conditions:

The cells were seeded at 1x106 cells/T-25 flask for control, and chemically-treated cultures. After approximately 24 hours, the cells were exposed to the test chemicals and cytochalasin B (3 μg/mL in DMSO) for 24 hours. The final concentration of solvent is approximately 1.3% (1% solvent +0.3% DMSO from the cytochalasin B). Colchicine (0.25 μg/mL or 0.5 μg/mL), or cyclophosphamide (5 μg/Ml) were used as positive controls. After a 24 hour treatment period, the media was aspirated off and the cells were collected by trypsinization. An aliquot of cells were counted to determine the number of live cells (determined by trypan blue exclusion) as a measure of toxicity. The remaining cells were suspended in 370C 0.075 M KCL for 5-10 minutes. The cells were collected by centrifugation and fixed in at least two changes of cold ( 4O C) 25:1 methanol/acetic acid. The cells were then dropped on either dry or wet slides, air dried and stained for 1-5 minutes in a 10% Giemsa solution in Gurr buffer. At each treatment group, 500 cells were analyzed to determine the percentage of binucleated

26

cells and 1000 binucleated cells were analyzed to determine the number of micronucleated cells. In case of ethylbenzene, fewer binucleated cells were scored due to a decrease in the number of scorable binucleated cells. Only cells with distinct cytoplasm and distinct binucleation were analyzed for the presence of micronuclei. Only micronuclei that were entirely inside the cytoplasm, separate from the main nucleus, less than approximately one-third the size of the main nuclei, and non-refractile were recorded. The number of micronucleated binucleated cells (MNBC):

Results:

DMSO -

28/1000

25μg/mL -

54/1000

50μg/mL -

58/1000

100μg/mL -

71/1000

200μg/mL -

15/250

Positive

Remarks: Conclusion:

Ethylbenzene induced a significant increase in the frequency of micronucleated binucleated cells at all tested concentrations.

Data Quality:

2 – Reliable with restrictions. Study is non-GLP but sufficiently documented and meets generally accepted scientific principles.

Reference:

Gibson, D.P., Brauninger R., Shaffi H.S., Kerckaert G.A., LeBoeuf R.A.,Isfort R.J. and Aardema M.J. (1997). Induction of micronuclei in Syrian hamster embryo cells: Comparison to results in the SHE cell transformation assay for National Toxicology Program test chemicals. Mutation Research 392:61-70.

27

[RS - 13] Genetic Toxicity - In Vivo Mammalian Erythrocyte Micronucleus Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

> 99 %

Method: Type:

Micronucleus Assay (as described in MacGregor et al., Fundam. Appl. Toxicol. 14:513-522, 1990)

System of Testing:

Mouse, Peripheral Blood Erythrocytes

GLP:

Yes

Year:

1999

Species/Strain:

Mouse / B6C3F1

Sex:

Male and Female

#/sex/dose:

8 – 10

Route of Administration:

Inhalation

Doses/Concentrations:

500, 750, 1000 ppm – Vapor

Exposure Time:

6 hours/day, 5 days/week for 13 weeks

Remarks for Test Conditions:

Peripheral blood samples were obtained from male and female B6C3F1 mice at the end of the 13 week NTP toxicity study [see RS17 for study details]. Smears were immediately prepared and fixed in absolute methanol, stained with a chromatin-specific fluorescent dye mixture of Hoechst 33258/pyronin & and coded. Slides were scanned to determine the frequency of micronuclei in 1000 polychromatic erythrocytes (PCEs) and 10000 normochromatic erythrocytes (NCEs) in each animal of each dose group. The criteria of Schmid (Chemical Mutagens. Principles and Methods for their Detection, A. Hollaender, Ed., Vol 4, pp.31-53. Plenum Press, New York 1976) were used to define micronuclei, with the additional requirement that the micronuclei exhibit the characteristic fluorescent emissions of DNA (blue with 360 nm and orange with 510 mn ultraviolet illumination); the minimum size limit was approximately one-twentieth the diameter of the NCE cell. In addition, the percentage of PCEs among the total erythrocyte population was determined. Evaluation of Results/Statistical Analyses - Log transformation of the NCE data, testing for normality by the Shapiro-Wilk test, and testing for heterogeneity of variance by Chochran’s test were performed before statistical analyses. The frequency of

28

micronucleated cells among NCEs was analyzed by analysis of variance using the SAS GLM procedure. The NCE data for each dose group were compared with the concurrent solvent control using Student’s t-test. The frequency of micronucleated cells among PCEs was analyzed by the Cochran-Armitage trend test, and individual dose groups were compared to the concurrent solvent control by Kastenbaum-Bowman’s binomial test. The percentage of PCEs among total erythrocytes was analyzed by an analysis of variance on ranks (classed by sex), and individual dose groups were compared with the concurrent solvent control using a t-test on ranks. Results:

Negative

Remarks:

Dose (ppm) Mean Micronucleated Cells/1000 cells PCEs PCEs

NCEs

(%)

0

2.18

1.54

2.22

500

2.04

1.68

3.13

750

1.90

1.90

1.97

1000

1.21

1.59

2.02

Female 0

1.54

0.92

1.74

500

2.64

1.01

1.83

750

1.87

1.32

1.85

1000

1.01

1.12

1.80

Male

Conclusion:

Ethylbenzene did not induce micronuclei formation in bone marrow erythrocytes of mice following treatment up to the maximum tolerated concentration.

Data Quality:

1 – Reliable without restrictions. No circumstances occurred that would have affected the quality or integrity of the data.

Reference:

National Toxicology Program. (1999). Toxicology and carcinogenesis studies of ethyl-benzene (CAS No. 100-41-4) in F344/N rats and B6C3F1 mice (Inhalation studies) (Tech. Rep. Ser. No. 466; NIH Publ No. 99-3956), National Toxicology Program, U.S. Dept. of Health and Human Services. Research Triangle Park, NC

29

[RS - 14] Genetic Toxicity - In Vivo DNA Repair Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

99.7%

Method:

OECD Guideline 486 - Unscheduled DNA Synthesis With Mammalian Liver Cells In Vivo

Type:

Unscheduled DNA Synthesis

System of Testing:

Mouse, Liver

GLP:

Yes

Year:

2001

Species/Strain:

Mouse / B6C3F1

Sex:

Male and Female

#/dose:

4 - 5 animals per sex per dose

Route of Administration:

Inhalation

Doses/Concentration:

Male mice – 500, 1000 ppm – Vapor Female mice – 375, 750 ppm – Vapor

Exposure Time:

6 hours

Remarks for Test Conditions:

Ethylbenzene was evaluated, using an autoradiographic technique, for its ability to induce unscheduled DNA synthesis (UDS) in the liver of B6C3F1 mice after a single 6 hour inhalation exposure (male mice: 500 and 1000 ppm; female mice 375 and 750 ppm). 1000 ppm and 750 ppm being the maximum tolerated concentration for each sex based on the patterns of clinical signs and lethalities observed over a 4 day period in a preliminary study. The concentration of ethylbenzene in generated atmosphere was determined at 8 sampling time points during the 6 hour exposure. The analytical chamber samples were analyzed using gas chromatography equipped with a flame ionization detector. A positive control, N-nitrosodimethylamine (N-DMA)(10 mg/kg bwt; oral) and a negative vehicle control group were run in parallel. Hepatocytes were isolated immediately after exposure, cultured in the presence of tritiated thymidine and subsequently examined for UDS following autoradiography. Evaluation of Results/Statistical Analyses - Data collected included the mean nuclear grain count, the mean cytoplasmic grain count,

30

the mean net nuclear grain count, and the percentage of cells in repair. A mean net nuclear grain count greater than zero was considered to be indicative of a UDS response in that animal. No statistical analyses were applied. Results:

Negative

Remarks:

The values recorded for the mean net nuclear grain counts and the percentages of cells in repair clearly show that ethylbenzene did not induce DNA repair, as measured by UDS, at any dose level investigated in either sex. The positive control induced marked increases in UDS compared to the vehicle control values. Dose (ppm) Mean Net Nuclear Grain Count Male

Mean % Cells in Repair

0

- 6.1

0

500

- 6.6

1

1000

- 4.9

1

N-DMA (10 mg/kg) 13.1

79

Female 0

- 6.3

1

375

- 6.6

1

750

- 6.5

1

N-DMA (10 mg/kg) 24.3

89

Conclusion:

Ethylbenzene did not induce DNA repair (as measured by unscheduled DNA synthesis) in mouse liver following treatment up to the maximum tolerated concentration.

Data Quality:

1 – Reliable without restrictions. No circumstances occurred that would have affected the quality or integrity of the data.

Reference:

Clay, P. (2001). Ethylbenzene: In vivo mouse liver unscheduled DNA synthesis assay. Central Toxicology Laboratory. CTL/SM0998/REG/REPT. Sponsored by the Styrenics Steering Committee, CEFIC, Brussels, Belgium.

31

[RS - 15] Repeated Dose Toxicity Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

99.7 %

Method:

Not Specified

Type:

Repeated Dose Inhalation Toxicity Study

GLP:

Not specified but presumed Yes

Year:

1989

Species/Strain:

Rat / Fischer 344

Sex:

Male and Female

#/sex/dose:

5

Vehicle:

Air

Route of Administration:

Inhalation

Exposure Period and Frequency of Treatment:

6 hours/day, 5 days/week, for 4 weeks

Doses/Concentrations:

100, 400, 800 ppm – Vapor

Control Group:

Yes

Remarks for Test Conditions:

Animals and Maintenance Species and strain: rat, Fischer 344 (Charles River Breeding Laboratories, Inc, Kingston, NY) Age on receipt: Approximately 6 weeks old Acclimation period: 17 days Housing: individually housed in suspended stainless-steel wire mesh cages Diet: Certified Rodent Chow No. 5002 (Purina), access ad libitum expect during the exposure period Water: access ad libitum expect during the exposure period Environment: Temperature averaged 72 ± 3 ºF and humidity ranged between 30 and 70% Body Weight at first treatment – Range between 158 – 178 g (males) and 124 – 138 g (females) Chambers Rats were exposed in common 10 m3 stainless steel and glass inhalation chambers with mice and rabbits (results reported separately). Ethylbenzene concentrations in each chamber were

32

monitored at hourly intervals, six times per day. Assessments Animals were observed twice daily for obvious clinical signs and underwent detailed examinations once each day after treatment. Body weights were measured weekly, and the weights of the major organs were recorded at termination. Ocular examinations were conducted on all animals immediately prior to exposure and at termination. Blood was collected via venipunture of the orbital sinus. Blood samples were subjected to hematological and clinical chemistry evaluation. Urinalyses were performed on all animals. At termination, animals were exsanginated under ether anesthesia. Over 30 tissues from each of the high-exposure and control animals were subjected to histopathological examination. Evaluation of Results/Statistical Analyses All parameters were evaluated for homogeneity of variance by Bartlett’s test. When homogeneous, analysis of variance was conducted using the F distribution to assess significance. If the overall F statistic was significant, Dunnett’s test was used to compare specific treatment groups to the control. The nonparametric Kruskall-Wallis test was used when variances were heterogeneous, and Dunn’s summed-rank test was used to compare treated groups to controls. Results:

NOAEL – 800 ppm, NOEL – 100 ppm (A NOAEL of 400 ppm was assigned by the author; however, the assignment was based on effects that the authors did not consider adverse, hence 800 ppm is the more appropriate study NOAEL and 100 ppm the study NOEL)

Remarks:

Chamber Concentrations Actual mean exposure concentrations achieved in the chambers throughout the study were 99, 382, and 782 ppm ethylbenzene. All results are presented by target inhalation level. Survival, Clinical Signs, Body Weight There were no treatment effects on survival or body weight gain. At 800 ppm, rats exhibited sporadic lacrimation and salivation. Clinical Pathology There were no treatment effects on clinical chemistry. Small but statistically significant increases in platelet counts occurred in male rats exposed to 800 ppm ethylbenzene. Also at 800 ppm, male rats showed a marginal increase and female rats a statistically significant increase in mean total leukocyte counts. The leukocytosis did not notably affect the differential white cell count. Pathology There were no treatment effects on gross and microscopic pathology. Exposure to 800 ppm resulted in an approximate 20% and 13% (p< 0.01) increase in relative (to body weight) liver weights in females and males, respectively. Female rats that received 400 ppm ethylbenzene exhibited about 7% increases (p< 0.05) in

33

relative liver weights; whereas, the male relative liver weights were not significantly different from controls. The authors interpreted the liver changes as adaptive metabolic response due to the absence of liver histopathology or abnormal clinical chemistry. Conclusion:

An increase in liver weight but no toxic effects were observed in rats that inhaled ≥ 400 ppm ethylbenzene vapor for 4 weeks. Also present at 800 ppm ethylbenzene were sporadic lacrimation and salivation and slight changes in blood cell counts.

Data Quality:

1 – Reliable without restrictions. Study well documented and meets generally accepted scientific principles. No circumstances occurred that would have affected the quality or integrity of the data.

Reference:

Cragg, S.T., Clarke, E.A., Daly, I.W., Miller, R.R., Terrill, J.B. and Ouellette, R.E. (1989). Subchronic inhalation toxicity of ethylbenzene in mice, rats, and rabbits. Fundam. Appl. Toxicol. 13:399-408.

34

[RS - 16] Repeated Dose Toxicity Test Substance:

Ethylbenzene

CAS RN:

100-41-4

Purity:

99.7%

Method:

Not Specified

Type:

Repeated Dose Inhalation Toxicity Study

GLP:

Not specified but presumed Yes

Year:

1989

Species/Strain:

Mouse / B6C3F1

Sex:

Male and Female

#/sex/dose:

5

Vehicle:

Air

Route of Administration:

Inhalation

Exposure Period and Frequency of Treatment:

6 hours/day, 5 days/week, for 4 weeks

Doses/Concentrations:

100, 400, 800 ppm – Vapor

Control Group:

Yes

Remarks for Test Conditions:

Animals and Maintenance Species and strain: mouse, B6C3F1 (Charles River Breeding Laboratories, Inc, Kingston, NY) Age on receipt: 7 weeks old Acclimation period: 17 days Housing: individually housed in suspended stainless-steel wire mesh cages Diet: Certified Rodent Chow No. 5002 (Purina), access ad libitum expect during the exposure period Water: access ad libitum expect during the exposure period Environment: Temperature averaged 72 ± 3 ºF and humidity ranged between 30 and 70% Body Weight at first treatment – Range between 19 – 24 g (males) and 17 – 21 g (females) Chambers Mice were exposed in common 10 m3 stainless steel and glass inhalation chambers inhalation chambers with rats and rabbits (results reported separately).

35

Assessments Animals were observed twice daily for obvious clinical signs and underwent detailed examinations once each day after treatment. Body weights were measured weekly, and the weights of the major organs were recorded at termination. Ocular examinations were conducted on all animals immediately prior to exposure and at termination. Blood was collected via venipunture of the orbital sinus. Blood samples were subjected to hematological evaluation. Clinical chemistry and urinalyses were not performed. At termination, animals were exsanginated under ether anesthesia. Over 30 tissues from each of the high-exposure and control animals were subjected to histopathological examination. Evaluation of Results/Statistical Analyses All parameters were evaluated for homogeneity of variance by Bartlett’s test. When homogeneous, analysis of variance was conducted using the F distribution to assess significance. If the overall F statistic was significant, Dunnett’s test was used to compare specific treatment groups to the control. The nonparametric Kruskall-Wallis test was used when variances were heterogeneous, and Dunn’s summed-rank test was used to compare treated groups to controls. Results:

NOAEL – 800 ppm, NOEL – 400 ppm (A NOAEL of 400 ppm was assigned by the author; however, the assignment was based on effects that the authors did not consider adverse, hence 800 ppm is the more appropriate study NOAEL and 400 ppm the study NOEL)

Remarks:

Chamber Concentrations Actual mean exposure concentrations achieved in the chambers throughout the study were 99, 382, and 782 ppm ethylbenzene. All results are presented by target inhalation level. Survival, Clinical Signs, Body Weights There were no treatment-related effects on survival, clinical signs, or body weight gain. Clinical Pathology There were no treatment-related effects on hematology. Pathology There were no treatment related effects on gross and microscopic pathology. In the mice that received 800 ppm ethylbenzene, liver weights relative to body weight were not statistically significantly different in males or females; but absolute liver weight was increased in females (about 15%; p 99%

Method:

Generally meets the requirements of OECD Guideline 413: Subchronic inhalation Toxicity: 90-day study, with the following exceptions: feed consumption not measured, ophthalmic examinations not conducted, and adrenals not weighed.

Type:

Repeated Dose Inhalation Toxicity Study

GLP:

Yes

Year:

1992

Species/Strain:

Rat / Fischer 344/N

Sex:

Male and Female

#/sex/dose:

10

Vehicle:

Air

Route of Administration:

Inhalation

Exposure Period and Frequency of Treatment:

6 hours/day, 5 days/week for 13 weeks

Doses/Concentrations:

100, 250, 500, 750, 1000 ppm – Vapor

Control Group:

Yes

Remarks for Test Conditions:

Animals and Maintenance Species and strain: rat, Fischer 344/N (Taconic Farms, Inc., Germantown, NY) Age on receipt: Approximately 5 weeks old Acclimation period: 12 days Housing: individually housed in suspended stainless-steel wire mesh cages Diet: NIH 07 Open Diet (Zeigler Bros., Inc.), access ad libitum expect during exposure Water: access ad libitum expect during the exposure period Environment: Temperature averaged 75 ± 3 ºF Age at first treatment: 7 weeks Chambers Concentrations of ethylbenzene in the inhalation chambers were monitored by an automatic sampling system coupled to a gas

41

chromatograph equipped with a flame ionization detector. Each study chamber atmosphere was analyzed at least once per hour during the 6 hour exposure. Animals and Treatment Groups of 10 rats of each sex were exposed to ethylbenzene for 6 hours per day, 5 days per week for 92 (female rats) and 93 (male rats) days. Controls were exposed to filtered air. Ten additional rats/sex were included at each exposure level to provide blood samples for clinical pathology (after blood collection on day 23, these rats were sacrificed and no tissues were retained). Assessments Blood for clinical chemistry and hematology was collected on study days 4 and 23, and at week 13 from the retroorbital sinus of male and female rats anesthetized with CO2. Animals surviving to the end of the study were humanely sacrificed with CO2. The heart, right kidney, liver, lung, right testis, and thymus were weighed. A necropsy was performed on all core study animals. Organs and tissues were examined for gross lesions. Tissues were preserved in 10% neutral buffered formalin and routinely processed for histopathological examination. Sperm morphology and vaginal cytology evaluations were performed according to methods described by Morrissey et al. (Fundam. Appl. Toxicol. 11:343-358, 1988). Evaluation of Results/Statistical Analyses Analysis of organ weight, serum chemistry, hematologic, and male reproductive system data was carried out to assess the significance of pair wise comparisons between dosed and chamber control groups, using nonparametric multiple comparison procedures. Jonckheere’s test was used to evaluate the significance of doseresponse trends to determine whether Dunn’s or Shirley’s test was more appropriate for pair wise comparisons. Results:

NOAEL – 1000 ppm, NOEL – 100 ppm

Remarks:

Chamber Concentrations Actual mean exposure concentrations achieved in the chambers throughout the study were 99.4, 246, 498, 740, and 975 ppm ethylbenzene. All results are presented by target inhalation level. Survival, Clinical Signs, Body Weights There were no treatment-related deaths or clinical signs. Male and female rats at 1000 ppm had mild body weight depression (5-7%) which was not statistically significant. Clinical Pathology There were no treatment-related effects on hematology or clinical chemistry (except lower serum alkaline phosphatase). Serum alkaline phosphatase was decreased in a dose-related manner for both males (statistically significant, p