                                                              				SRC TR-13-060

                                       
                                       
                                       
                                       
                                       
                                       
                 HPV3 TEST RULE DATA ADEQUACY REVIEW (FINAL):
             TOXICITY TO ALGAE, CHRONIC TOXICITY TO DAPHNIA, AND 
                            CHROMOSOMAL ABERRATIONS
                                       
Propanoic acid, 2-methyl-, 3-(benzoyloxy)-2,2,4-trimethylpentyl ester, CASRN 22527-63-5
                                       
                                       
                                 Prepared by:
                                       
      Amy Nguyen, Jennifer Rayner, Teresa Manyin, and William L. Richards
                                       
                                       
                 Defense and Environmental Solutions Division
                                   SRC, Inc.
                             7502 Round Pond Road
                           North Syracuse, NY 13212
                                       
                                       
                           Contract No. EP-W-09-027
                             Task FG007.3.009.006
                                       
                                       
                                 Submitted to:
                                       
                            OCSPP/OPPT/RAD (7403M)
                     U.S. Environmental Protection Agency
                           1200 Pennsylvania Avenue
                            Washington, D.C. 20460
                                       
                         Bryan Lobar, Project Officer
                     David Brooks, Work Assignment Manager
                 Amy Benson, Alternate Work Assignment Manager
                                       
                              September 12, 2013

                               TABLE OF CONTENTS

A.  SOURCE MATERIAL USED FOR ASSESSMENT	1
B.  EVALUATING DATA ADEQUACY	1
C.  SUMMARY OF DATA ADEQUACY FOR TESTING REQUIRED BY TEST RULE	2
D.  DATA FOR SIDS ENDPOINTS	5
D-1.  Toxicity to Aquatic Plants	5
D-2.  Chronic Toxicity to Aquatic Invertebrates	12
D-3.  Genetic Toxicity  -  Chromosomal Aberrations	21
E.  DATA ADEQUACY SUMMARY TABLE	26


A.  SOURCE MATERIAL USED FOR ASSESSMENT

This document evaluates toxicity tests for propanoic acid, 2-methyl-, 3-(benzoyloxy)-2,2,4-trimethylpentyl ester (CASRN 22527-63-5) submitted by Eastman Chemical Company (on June 20, 2013) to EPA in response to the final test rule, "Testing of Certain High Production Volume Chemicals; Third Group of Chemicals" (76 FR 65385, October 21, 2011; referred to herein as HPV3; available online at http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPPT-2009-0112-0086 as of September 12, 2013).  

The HPV3 test rule requires the following toxicological testing for CASRN 22527-63-5: C1 (consisting of acute toxicity to fish, acute toxicity to Daphnia, and toxicity to algae testing if log Kow < 4.2 or chronic toxicity to Daphnia and toxicity to algae testing if log Kow >= 4.2), D (acute mammalian toxicity test), E1 (bacterial reverse mutation test), E2 (chromosomal aberration or micronucleus test), and F1 (combined repeated dose toxicity study with the reproduction/developmental toxicity screening test).  Since the log Kow of CASRN 22527-63-5 has been experimentally determined to be >= 5.4 (based on measured log Kow values of 5.4, 5.5, and 5.8 for the three main components of the test item), which is > 4.2, the C1 aquatic toxicity testing requirement is considered to include chronic toxicity to Daphnia and toxicity to algae testing.  

The June 20, 2013 submission from Eastman Chemical Company ("Eastman"), available online at http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPPT-2009-0112-0145 as of September 12, 2013, contained the following toxicological test reports:
   * "Benzoflex 1046: Algal Growth Inhibition Test", submitted for the C1 testing requirement 
   * "Benzoflex 1046: Daphnia magna Reproduction Test", submitted for the C1 testing requirement 
   * "Benzoflex 1046: Chromosome Aberration Test in Human Lymphocytes In Vitro", submitted for the E2 testing requirement 

These newly submitted data are evaluated for adequacy herein.  No other toxicity data are evaluated in this report.  

B.  EVALUATING DATA ADEQUACY

Data available for SIDS toxicity endpoints are evaluated for adequacy in Section D.  Each study was summarized according to a standard format and assigned a Klimisch reliability code based on its scientific merit and conduct, using professional judgment.  To determine whether data were adequate to satisfy SIDS endpoints, study details were compared to current standard test guidelines to establish whether the study met current testing requirements.  If the data available for the endpoint were determined to be inadequate, a basis for that conclusion is supplied.  A table summarizing data availability/adequacy for each SIDS endpoint is included on the last page of this document (Section E).

C.  SUMMARY OF DATA ADEQUACY FOR TESTING REQUIRED BY TEST RULE

Justification for Use of Test Substance:
The test rule classifies propanoic acid, 2-methyl-, 3-(benzoyloxy)-2,2,4-trimethylpentyl ester (CASRN 22527-63-5) as a Class 1 Chemical Substance for which testing must be conducted with material having a purity of 99% or greater.  However, in a letter to EPA dated February 20, 2012 (available online at http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPPT-2009-0112-0107 as of September 12, 2013), Eastman proposed the use of a bulk reaction product of the following purity: 60-65% propanoic acid, 2-methyl-, 3-(benzoyloxy)-2,2,4-trimethylpentyl ester (CASRN 22527-63-5), 20-25% 1,3-pentanediol, 2,2,4-trimethyl-1,3-dibenzoate (CASRN 68052-23-3), and 10-15% 2,2,4-trimethyl-1,3-pentanediol diisobutyrate (CASRN 6846-50-0).  

Eastman's rationale for the use of this mixture (the "bulk reaction product") was based on the fact that "The manufacture of benzoate ester for commercial use has been as a bulk reaction and there is no readily available process to synthesize the chemical substance by direct reaction to achieve the purity required for Class 1 substances.  Eastman believes that use of the bulk reaction product which is formulated into finished products for commercial sale will more accurately represent exposure and use of the benzoate ester and avoid a costly and technically uncertain purification procedure."  Given that the major constituent (60-65%) of the bulk reaction product is the test rule substance, propanoic acid, 2-methyl-, 3-(benzoyloxy)-2,2,4-trimethylpentyl ester (CASRN 22527-63-5), and the remaining two major components (CASRNs 68052-23-3 and 6846-50-0) can be considered close analogs of CASRN 22527-63-5, EPA approved Eastman's plan to conduct aquatic and mammalian toxicity tests on this mixture via email from P. Campanella (EPA) to G. Shrum (Eastman) on April 19, 2012.  This mixture, referred to as "Benzoflex 1046" in the submitted study reports, can be considered a supporting substance for CASRN 22527-63-5; as such, the toxicity of Benzoflex 1046 is expected to be representative of the toxicity of CASRN 22527-63-5. 

Evaluation of Toxicological Data Submitted on June 20, 2013:
Data submitted for the chromosomal aberration test for Benzoflex 1046, supplied on June 20, 2013 by Eastman and reviewed herein, are adequate to satisfy the E2 testing requirement promulgated in the final HPV3 test rule.  

Data submitted for chronic toxicity to Daphnia and toxicity to algae for Benzoflex 1046, supplied on June 20, 2013 by Eastman, are not considered adequate to satisfy the C1 testing requirement for CASRN 22527-63-5 promulgated in the final HPV3 test rule.  Both studies are considered unreliable because the concentrations tested were less than the water solubility limit (1.01 mg/L at 20°C) of the test item and the tests failed to identify EL50 values for algae and daphnids.  In the algae limit test, the initial (time = 0) measured test substance concentration of the WAF (nominal loading rate of 100 mg/L) was 0.38 mg/L, which is 62% less than the water solubility limit.  In the daphnid static-renewal test, measured test concentrations of "fresh" test solutions (sampled at the time of solution renewal) only approached the limit of water solubility on Day 0, when measured test substance concentrations of 0.814 mg/L and 0.927 mg/L were reported for nominal loading rates of 10 and 100 mg/L, respectively; measured test substance concentrations were 0.521, 0.529, and 0.546 mg/L for the remaining three loading rates of 1, 3.2, and 32 mg/L, respectively, on Day 0.  The measured concentrations of the "fresh" test solution samples collected on days 5, 9, 14, and 19 of the daphnid test ranged from 50 to 100% less than the water solubility limit.  No explanation was provided in either report for the difference between the measured concentrations of "fresh" test solutions and the reported water solubility limit.

Additional deficiencies of the toxicity to algae and chronic toxicity to Daphnia tests include the following:
   (1) The solubility, stability, and volatility of the test substance in the test media were not determined prior to the initiation of the definitive test in either study as recommended by ASTM and OPPTS guidelines.
   (2) Test substance concentrations measured following the preparation of test solutions were inconsistent within and between studies, although the method of mixing the WAFs was reported to be the same in both studies.  In the algae study, the initial measured concentration of the 100 mg/L loading rate WAF was 1.23 mg/L in the range-finding test, but was 0.38 mg/L in the definitive test.  In the daphnid study, large variations in measured test substance concentrations were observed in WAFs at test solution renewal; for example, the measured concentrations of "fresh" solutions of WAFs at a nominal loading rate of 100 mg/L were 0.927, 0.0522, 0.489, 0.467, and 0.0573 mg/L on Days 0, 5, 9, 14, and 19, respectively.  No explanations for the variability among measured concentrations of "fresh" test solutions were provided in the individual study reports.
   (3) The test reports did not provide an explanation for the decrease in measured test substance concentrations of 38-99% during the tests.  In the algae definitive test, measured concentrations were 0.38 and 0.00638 mg/L at 0 and 96 hours, respectively; the decrease in concentration was reported to be due to the unstable nature of the test item.  In the daphnid study, measured concentrations of "old" test solutions (sampled prior to solution renewal) were 38% to 99% less than the respective measured "fresh" concentrations, but no explanation was provided.  No data were provided to demonstrate instability of the test substance in the test media.
   (4) Algal growth rate in control cultures was not constant during the 96-hour exposure period.  After 72 hours, a marked decline in growth rate was observed in controls.  No explanation was provided by the study author for this decrease.  
   (5) In the chronic Daphnia study, the test item was detected in control samples collected on Days 0, 2, 7, 12, 16, and 21, suggesting unreliable analytical techniques for measuring concentrations or inadequate quality control.  

Therefore, the algae and daphnid studies should be repeated to assure that CASRN 22527-63-5 is tested up to its solubility limit in a test medium appropriate for algae and Daphnia, respectively.  It is recommended that the study design for further toxicity testing of algae and Daphnia includes the conduct of preliminary testing to determine what procedures are necessary to assure that the test substance reaches maximum possible solubility in the intended test medium (i.e., dilution water, culture media) at each concentration that is to be tested in the main toxicity study.  Procedures for preparation of WAF solutions should also be refined to minimize variability among measured test substance concentrations of fresh test solutions at renewal.  In addition, stability trials should be conducted to determine if the test substance significantly degrades during the intended testing period (algae) or renewal cycle (Daphnia).  It is recommended that test concentration measurements be conducted more frequently (every 24 hours) to better characterize exposures during the test period.  The study reports should contain detailed procedures for storage and handling conditions of prepared WAFs prior to use, as well as detailed procedures for test solution sampling, storage, and analysis.

Evaluation of Previously Submitted Toxicological Data:
In a previous submission (dated February 26, 2013; available online at http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPPT-2009-0112-0138 as of September 12, 2013), Eastman supplied data from acute mammalian toxicity and gene mutation tests.  These data were reviewed in a separate report (SRC Technical Report #13-050) and were determined to be adequate to satisfy the D and E1 testing requirements promulgated in the HPV3 test rule.

Remaining HPV3 Test Rule Toxicity Testing Requirements:
Data have not yet been submitted for the following toxicity testing requirement:  combined repeated dose toxicity study with the reproduction/developmental toxicity screening test (F1 testing).  In a letter dated June 20, 2013 (available online at http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPPT-2009-0112-0145 as of September 12, 2013), Eastman provided an update on testing status.  At that time, Eastman noted that the test was underway and the finalized report would be submitted when it was available.

Summary of Data Adequacy for Toxicological Endpoints:
Data submitted by Eastman for the chronic toxicity to Daphnia and toxicity to algae endpoints are not considered adequate; these tests need to be repeated.  Adequate data have been submitted by Eastman for acute mammalian toxicity, gene mutation, and chromosomal aberrations endpoints.  Data have not yet been submitted for repeated-dose, reproductive, and developmental toxicity endpoints.


D.  DATA FOR SIDS ENDPOINTS

D-1.  Toxicity to Aquatic Plants

Summary of endpoint:  Not Adequate

Study 1:  Algal growth inhibition toxicity test in Pseudokirchneriella subcapitata (Vryenhoef, 2013)

Title:  Benzoflex 1046: Algal growth inhibition test 
	
This study is:  Adequate		 Not Adequate

This study is not considered adequate because the test was not conducted up to the water solubility limit (1.01 mg/L) and the test failed to identify an EL50 for algal growth.  The measured concentration of test item at the start of the definitive test was 0.38 mg/L, which is 62% lower than the water solubility limit.  The solubility, stability, and volatility of the test substance in the growth medium were not determined prior to the initiation of the algae test, as recommended by ASTM and OPPTS guidelines.

TEST SUBSTANCE   
		
Identity:  Benzoflex 1046; supporting substance for propanoic acid, 2-methyl-, 3-(benzoyloxy)-2,2,4-trimethylpentyl ester (CASRN 22527-63-5)

Remarks:  Clear colorless liquid; Batch No. V046212201; Composition: 68.8% propanoic acid, 2-methyl-, 3-(benzoyloxy)-2,2,4-trimethylpentyl ester (CASRN 22527-63-5), 19.5% 1,3-pentanediol, 2,2,4-trimethyl-1,3-dibenzoate (CASRN 68052-23-3), 7% 2,2,4-trimethyl-1,3-pentanediol diisobutyrate (CASRN 6846-50-0), 4.2% miscellaneous esters, 0.5% volatiles, and 0.02% moisture.

METHOD

Method/guideline followed:  OPPTS 850.5400 ("Algal Toxicity, Tiers I and II") and ASTM E1218-04 ("Standard Guide for Conducting Static Toxicity Tests with Microalgae") 

Test type:  Toxicity to plants (algae)  -  static  

GLP compliant?:  Yes

Year study was performed:  2013

Test species:  Green algae (Pseudokirchneriella subcapitata) strain CCAP 278/4; obtained from the Culture Collection of Algae and Protozoa, SAMS Research Services Ltd, Scottish Marine Institute, located in Oban, Argyll, Scotland 

Analytical monitoring:  Yes 

Exposure duration:  96 hours

Concentrations tested:  The test item, Benzoflex 1046, a multi-component mixture, was prepared as a water accommodated fraction (WAF) due to its low aqueous solubility.
Range-finding test: 
Loading rate (WAFs):  0, 10, and 100 mg/L.  Measured concentrations for the 100 mg/L WAF loading rate at 0 and 96 hours:  1.23 and 0.0454 mg/L, respectively.   The mean measured concentration (calculated by the EPA reviewer) was 0.64 mg/L. 
Definitive test: 
Loading rate (WAFs):  0 and 100 mg /L.  Measured concentrations for the 100 mg/L WAF loading rate at 0 and 96 hours:  0.380 and 0.00638 mg/L, respectively.  The mean measured concentration (calculated by the EPA reviewer) was 0.19 mg/L. 

Test method/conditions remarks:  
"The culture medium was prepared using reverse osmosis purified deionized water and the pH [was] adjusted to 7.5 +- 0.1 with 0.1N NaOH or HCl.  The water hardness (calculated) of the culture medium was 0.15 mmol/L (= 15 mg/L as CaCO3)." [Vryenhoef 2013, p. 31]

In pre-study tests, the test item was stirred in deionized reverse osmosis water for 23, 47, or 95 hours to examine the effect of increased stirring time on the concentration of dissolved test item.  In these tests, stirring the test item at a loading rate of 100 mg/L for longer than 23 hours (47 or 95 hours) did not substantially increase the amount of dissolved test substance, as indicated by total organic carbon (TOC) measurements.  TOC values for the aqueous phase were 1.20, 1.22, and 1.28 mg carbon/L (corrected for control) for stirring periods of 23, 47, and 95 hours, respectively. 

Range-Finding Test Procedure: 
In the range-finding test, Pseudokirchneriella subcapitata cells were exposed to a series of nominal loading rates of 10 and 100 mg/L.  Negative control cells were not exposed to the test item.  

"The test was conducted in 250 mL glass conical flasks each containing 100 mL of test preparation and plugged with polyurethane foam bungs to reduce evaporation.  Two replicate flasks were prepared for each control and test concentration." [Vryenhoef 2013, p. 12]

"Amounts of test item (20 and 200 mg) were each separately added to the surface of 2 liters of culture medium to give the 10 and 100 mg/L loading rates respectively.  After the addition of the test item, the culture medium was stirred by magnetic stirrer using a stirring rate such that a vortex was formed to give a dimple at the water surface.  The stirring was stopped after 23 hours and the mixtures [were] allowed to stand for 1 hour...  The aqueous phase or WAF was removed by mid-depth siphoning (the first 75-100 mL discarded) to give the 10 and 100 mg/L loading rate WAFs.  Microscopic inspection of the WAFs showed no micro-dispersions or undissolved test item to be present." [Vryenhoef 2013, p. 12]

An aliquot of each of the WAFs was inoculated with algal suspension, resulting in an initial cell density of approximately 1 x 10[4] cells/mL.

"The flasks were... incubated ...at 24 +- 2 °C under continuous illumination (intensity approximately 7000 lux) provided by warm white lighting (380 -730 nm) and constantly shaken at approximately 150 rpm for 96 hours." [Vryenhoef 2013, p. 12]

Cell densities were determined at 0 and 96 hours using a Coulter[(R)] Multisize Particle Counter.

"Samples of each loading rate WAF were taken for chemical analysis at 0 and 96 hours in order to determine the stability of the test item over the test duration.  All samples were stored at approximately -20 °C prior to analysis." [Vryenhoef 2013, p. 12]

Definitive Test Procedure:  
Based on the results of the range-finding test, a loading rate of 100 mg/L was selected for the definitive (limit) test.  Negative control cells were not exposed to the test item.

"An amount of test item (200 mg) was added to the surface of 2 liters of culture medium to give the 100 mg/L loading rate.  After the addition of the test item, the culture medium was stirred by magnetic stirrer using a stirring rate such that a vortex was formed to give a dimple at the water surface.  The stirring was stopped after 23 hours and the mixture [was] allowed to stand for 1 hour...  The aqueous phase or WAF was removed by mid-depth siphoning (the first 75-100 mL discarded) to give the 100 mg/L loading rate WAF." [Vryenhoef 2013, pp. 12-13]

"At the start of mixing the 100 mg/L loading rate WAF was observed to have formed a clear colorless media column with oily globules of test item floating on the surface and settled on the base of the mixing vessel.  After stirring, and following a 1-hour standing period, the WAF was observed to have formed a clear colorless media column with an oily slick of test item floating on the surface and a few oily globules dispersed throughout.  Microscopic examination of the WAF showed there to be no micro-dispersions or globules of test item present." [Vryenhoef 2013, p. 19]

Exposure flasks were 250 mL glass conical flasks, each containing 100 mL of test solution.  Six replicates were tested for both the control and 100 mg/L WAF loading rate.  

"Pre-culture conditions gave an algal suspension in log phase growth characterized by a cell density of 3.30 x 10[5] cells per mL.  Inoculation of 1 liter of test medium with 30.3 mL of this algal suspension gave an initial nominal cell density of 1 x 10[4] cells per mL..." [Vryenhoef 2013, p. 13]

"The flasks were plugged with polyurethane foam bungs and incubated ... at 24 +- 1 °C under continuous illumination (intensity approximately 7000 lux) provided by warm white lighting (380 - 730 nm) and constantly shaken at approximately 150 rpm for 96 hours."  [Vryenhoef 2013, p. 13]

"Samples were taken at 0, 24, 48, 72 and 96 hours and the cell densities determined using a
Coulter[(R)] Multisizer Particle Counter." [Vryenhoef 2013, p. 13]

The pH was measured at the start and end of the study and temperatures within the incubator were recorded hourly.

"Samples were taken from the control and the 100 mg/L loading rate WAF test group (replicates R1  -  R6 pooled) at 0 and 96 hours for quantitative analysis.  All samples were stored at approximately -20 °C prior to analysis.  Duplicate samples were taken at each occasion and stored at approximately -20 °C for further analysis if necessary." [Vryenhoef 2013, p. 14]

Test concentrations were analyzed using high performance liquid chromatography (HPLC) with UV detection (wavelength of 230 nm).  The limit of quantitation [LOQ] was 0.0036 mg/L.

"A Student's t-test incorporating Bartlett's test for homogeneity of variance (Sokal and Rohlf 1981) was carried out on the growth rate, yield and biomass integral data after 96 hours for the control and the 100 mg/L loading rate to determine any statistically significant differences between the test and control groups.  All statistical analyses were performed using the SAS computer software package (SAS 1999-2001)." [Vryenhoef 2013, p. 16]

"The results of the test were considered valid if the following criteria were met:
   * The cell concentration of the control cultures must increase by a factor of at least 16 after 72 hours (ASTM Guidelines) and by a factor of at least 100 after 96 hours (OPPTS
      Guidelines).
   * The coefficient of variation in replicate control cultures after 72 hours must not exceed 7% for average specific growth rate and 20% for yield (ASTM Guidelines).
   * The coefficient of variation of the cell density values in replicate control cultures must not exceed 20% after 72 and 96 hours (OPPTS Guidelines)."  [Vryenhoef 2013, p. 16]

RESULTS

Endpoint value(s): The following results were based on WAF loading rates: 

Growth Rate, Yield, and Biomass:  
96-hour EL50 > 100 mg/L loading rate 
96-hour NOEL = 100 mg/L loading rate
96-hour LOEL > 100 mg/L loading rate

The following values are based on mean measured concentrations (definitive test) that correspond with nominal loading rates:

Growth Rate, Yield, and Biomass:  
96-hour EC50 > 0.19 mg/L
96-hour NOEC = 0.19 mg/L 
96-hour LOEC > 0.19 mg/L 

Results remarks:  
Range-Finding Test Results:
"Chemical analysis of the 100 mg/L loading rate WAF at 0 hours showed a measured test concentration of 1.2 mg/L was obtained.  A significant decline in measured test concentration [96%] was observed at 96 hours to 0.045 mg/L.  This decline was considered [by the study author] to be due to the unstable nature of the test item." [Vryenhoef 2013, p. 17]

The percent inhibition of "yield/biomass integral" was 22% and 28% at loading rates of 10 and 100 mg/L, respectively.  No inhibition of growth rate was observed at the nominal loading rates of 10 and 100 mg/L.  

Definitive Test Results:
"At the start of the test all control and 100 mg/L loading rate WAF test cultures were observed to be clear colorless solutions.  After the 96-hour test period all control and test cultures were observed to be dark green dispersions." [Vryenhoef 2013, p. 19]

"Analysis of the 100 mg/L loading rate WAF test preparation at 0 hours ... showed [that] a measured test concentration of 0.38 mg/L was obtained.  A significant decline in measured test concentration [98%] was observed at 96 hours to 0.0064 mg/L.  This decline in measured concentration was considered [by the study author] to be due to the unstable nature of the test item." [Vryenhoef 2013, p. 20]

"[T]he growth rate, yield and biomass integral of Pseudokirchneriella subcapitata (CCAP 278/4) were not affected by the presence of the test item over the 96-Hour exposure period." [Vryenhoef 2013, p. 18]

The percent inhibition values for growth rate, yield, and biomass integral were 1%, 0%, and 2%, respectively, after 96 hours at a nominal loading rate of 100 mg/L.  No statistically significant differences (p >= 0.05) between the control and 100 mg/L loading rate WAF test group were observed and, therefore, the No Observed Effect Loading Rate (NOEL) for growth rate, yield, and biomass was 100 mg/L based on the WAF loading rate.

"All test and control cultures were inspected microscopically at 96 hours.  There were no abnormalities detected in any of the control or test cultures." [Vryenhoef 2013, p. 19]

"Temperature was maintained at 24 +- 1 °C throughout the test." [Vryenhoef 2013, p. 19]  The pH ranged from 7.1 to 7.8.

The study author considered the test to be valid based on the following:
   * The cell density in control cultures increased by a factor of 122 after 72 hours and a factor of 206 after 96 hours. 
   * The coefficient of variation for average specific growth rate and yield in controls after 72 hours was 1% and 4%, respectively. 
   * The coefficient of variation for cell density values in controls was 4% after 72 hours and 6% after 96 hours.

CONCLUSIONS

The 96-hour NOEL was 100 mg/L based on WAF loading rates.  The 96-hour EL50 values for growth rate, yield, and biomass were > 100 mg/L based on WAF loading rates.

STUDY RELIABILITY 

[3] Not reliable.  This study is considered unreliable because the test was not conducted up to the water solubility limit (1.01 mg/L) and the test failed to identify an EL50 for algal growth.  The measured concentration of test item at the start of the definitive test was 0.38 mg/L, which is 62% less than the water solubility limit.  The solubility, stability, and volatility of the test substance in the growth medium were not determined prior to the initiation of the algae test, as recommended by ASTM and OPPTS guidelines.  
  
Additional deficiencies include the following:
   (1) Analytical measurements demonstrated disparities between the prepared test concentrations in the range-finding and definitive tests.  The initial (time = 0) test substance concentration of the 100 mg Benzoflex 1046/L loading rate WAF sample in the definitive test was >3-fold lower than that in the range-finding test.  At test initiation, measured test substance concentrations of the 100 mg/L loading rate WAF were 1.23 mg/L and 0.380 mg/L for the range-finding and definitive tests, respectively.  The author did not provide an explanation for the difference between the measured concentrations of the WAFs used in the range-finding and definitive tests.  Although similar procedures were described for the WAF preparations for the range-finding and definitive tests, the discrepancy in measured concentrations at test initiation raises concerns about the consistency and reproducibility of the methods used to prepare the test solutions for the two tests and/or the analytical methods used to determine test substance concentrations.  The study report did not provide a detailed description of the procedures for storage and handling of WAF preparations.  Missing information included, for example, whether the WAF was stored prior to use and, if so, for how long and under what conditions.  
   (2) During the 96-hour exposure period, measured test substance concentrations declined by more than 98% during the definitive study and by more than 96% in the range-finding study.  In the definitive test, measured concentrations of the 100 mg Benzoflex 1046/L loading rate WAF samples were 0.380 and 0.00638 mg/L at 0 and 96 hours, respectively.  The reason for the decline in test item concentrations during the course of the study was not addressed in detail by the author and is unclear.  The author stated that the test item has low solubility in water and was unstable, but did not provide data to demonstrate instability of the test substance in the test media. 
   (3) Growth rate in control cultures was not constant during the exposure.  After 72 hours, a marked decline in growth rate was observed in controls.  The average specific growth rate in control cultures from 72 to 96 hours (0.022 cells/mL/hour) was less than one third of the growth rate observed during the first 72 hours (0.068 cells/mL/hour).  No explanation was provided by the study author for this decrease.  

REFERENCE

Vryenhoef, H.  2013.  Benzoflex 1046: Algal growth inhibition test.  Harlan Laboratories Ltd., Shardlow, Derbyshire, UK, Project No. 41202705.  Sponsor: Eastman Chemical Company, Kingsport, TN.  Study report available online at http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPPT-2009-0112-0145 as of September 12, 2013.

Text in quotes is taken directly from Vryenhoef (2013).

D-2.  Chronic Toxicity to Aquatic Invertebrates
Summary of endpoint:  Not adequate

Study 1:  Chronic toxicity study in Daphnia magna (Harris, 2013)

Title:  Benzoflex 1046: Daphnia magna reproduction test

This study is:	 Adequate		 Not Adequate

This study is not considered adequate because the test was not conducted up to the water solubility limit of the test item (1.01 mg/L) and the test failed to identify an EL50 for chronic toxicity to daphnids.  Measured test concentrations of "fresh" test solutions (sampled at the time of solution renewal) only approached the limit of water solubility on Day 0, when measured test substance concentrations of 0.814 mg/L and 0.927 mg/L were reported for nominal loading rates of 10 and 100 mg/L, respectively; measured test substance concentrations were 0.521, 0.529, and 0.546 mg/L for the remaining three loading rates of 1, 3.2, and 32 mg/L, respectively, on Day 0.  The measured concentrations of the "fresh" test solution samples collected on days 5, 9, 14, and 19 of the daphnid test ranged from 50 to 100% less than the water solubility limit.  No explanation was provided for the difference between the measured concentrations of "fresh" test solutions and the reported water solubility limit.  The solubility and stability of the test substance in the dilution water were not determined as recommended by ASTM E1193-97. 

TEST SUBSTANCE   
		
Identity:  Benzoflex 1046; supporting substance for propanoic acid, 2-methyl-, 3-(benzoyloxy)-2,2,4-trimethylpentyl ester (CASRN 22527-63-5)

Remarks:  Clear colorless liquid; Batch No. V046212201; Composition: 68.8% propanoic acid, 2-methyl-, 3-(benzoyloxy)-2,2,4-trimethylpentyl ester (CASRN 22527-63-5), 19.5% 1,3-pentanediol, 2,2,4-trimethyl-1,3-dibenzoate (CASRN 68052-23-3), 7% 2,2,4-trimethyl-1,3-pentanediol diisobutyrate (CASRN 6846-50-0), 4.2% miscellaneous esters, 0.5% volatiles, and 0.02% moisture.

METHOD

Method/guideline followed:  Method was designed to be compatible with ASTM Standard E1193-97 ("Standard Guide for Conducting Daphnia magna Life-Cycle Toxicity Tests")

Test type:  Chronic toxicity to Daphnia  -  Semi-Static

GLP compliant?:  Yes

Year study was performed:  2013

Test species:  Daphnia magna, less than 24 hours old; obtained from laboratory cultures maintained in-house at Harlan Laboratories, Ltd., in Shardlow, Derbyshire, UK

Analytical monitoring:  Yes

Exposure duration:  21 days

Concentrations tested:  The test item, Benzoflex 1046, a multi-component mixture, was prepared as a water accommodated fraction (WAF) due to its low aqueous solubility.  
Loading rates (WAFs):  0, 1.0, 3.2, 10, 32, and 100 mg/L  
Mean measured concentrations (calculated by the EPA reviewer):  0.14, 0.13, 0.18, 0.12, and 0.21 mg/L

Test method/conditions remarks:  Deionized reverse osmosis water was used to prepare the test medium (Elendt M7 medium).  The pH of the prepared medium was 8.0 +- 0.2.

In pre-study tests, the test item was stirred in deionized reverse osmosis water for 23, 47, or 95 hours to examine the effect of increased stirring time on the concentration of dissolved test item.  In these tests, stirring the test item at a loading rate of 100 mg/L for longer than 23 hours (47 or 95 hours) did not substantially increase the amount of dissolved test substance, as indicated by total organic carbon (TOC) measurements.  TOC values for the aqueous phase were 1.20, 1.22, and 1.28 mg carbon/L (corrected for control) for stirring periods of 23, 47, and 95 hours, respectively.

"Amounts of test item (20, 64, 200, 640 and 2000 mg) were each separately added to the surface of 20 liters of Elendt M7 to give the 1.0, 3.2, 10, 32 and 100 mg/L loading rates, respectively.  After the addition of the test item, the Elendt M7 was stirred by magnetic stirrer using a stirring rate such that a vortex was formed to give a dimple at the water surface.  The stirring was stopped after 23 hours and the mixtures [were] allowed to stand for 1 hour...  The aqueous phase or WAF was removed by mid-depth siphoning (the first approximate 75-100 mL discarded) to give the 1.0, 3.2, 10, 32 and 100 mg/L loading rate WAFs." [Harris 2013, pp. 12-13]

"At the start of the mixing period the 1.0, 3.2, 10, 32, and 100 mg/L loading rates were observed to be clear colorless water columns with an oily slick of test item on the surface and oily globules of test item dispersed throughout.  After 23 hours stirring and a 1-hour standing period the 1.0, 3.2, 10, 32, and 100 mg/L loading rates were observed to be clear colorless water columns with oily globules of test item on the surface and on the base of the vessel.  Microscopic inspection of the WAFs showed no micro-dispersions or undissolved test item to be present [with the exception of the 100 mg/L loading rate WAF on day 9].  After siphoning and for the duration of the test, the 1.0, 3.2, 10, 32, and 100 mg/L loading rates were observed to be clear, colorless solutions." [Harris 2013, p. 17]

"On day 9, microscopic examination of the 100 mg/L loading rate WAF showed oily globules of test item to be present in the water column and therefore it was considered justifiable to remove the WAF by filtering through a glass wool plug (2-4 cm in length).  Microscopic examination after filtering showed the glass wool plug had removed all test item from the water column." [Harris 2013, p. 17]  "This difference on day 9 may have been due to the media being stirred slightly faster, causing more test item to be drawn into the water column." [Harris 2013, p. 13]  

Daphnia magna (10 replicates per test concentration, each containing a single daphnid) were exposed to WAFs of the test item at nominal loading rates of 0, 1.0, 3.2, 10, 32, and 100 mg/L for a period of 21 days.  The WAF test solutions were renewed 3 times per week (on days 0, 2, 5, 8, 9, 12, 14, 16, and 19).  

Test vessels were 150 mL glass flasks containing 100 mL of test medium and covered with a plastic lid to reduce evaporation.  The daphnids were fed algae and Tetramin[(R)] flake food suspensions daily.

"The flasks were maintained at approximately 20 °C with a photoperiod of 16 hours light (681 to 965 lux) and 8 hours darkness with 20 minute dawn and dusk transition periods...  Each vessel was randomly assigned to a position in the laboratory.  The test vessels were not aerated.  The diluent water only was aerated prior to use." [Harris 2013, p. 13]

The size and general condition of the parental daphnids were evaluated daily.  The length of each surviving parental daphnid was determined at the end of the test.  The number of daphnids with eggs or young in the brood pouch and the number of discarded unhatched eggs were recorded daily.  The number of live and dead daphnids of the adult (parental, P1) and filial (F1) generations were also recorded daily.  

"Young daphnids were considered to be dead if no sign of movement was apparent during microscopic examination.  Adult Daphnia which were unable to swim for approximately 15 seconds after gentle agitation (i.e., immobile) were considered to be dead." [Harris 2013, p. 14]

"Temperature of the test preparations and light intensity were recorded daily throughout the test. Dissolved oxygen concentrations, pH and temperature were recorded before and after each test media renewal...  Measurements were made on one replicate for each test concentration.  The water hardness, alkalinity and conductivity of the control and the highest surviving test concentration in the old media were measured once per week.  The room temperature and the temperature in one control vessel were recorded continuously throughout the test." [Harris 2013, p. 14]

"Water samples were taken from the control and each surviving test group (replicates pooled) for quantitative analysis.  Samples of the fresh test preparations were taken on Days 0, 5, 9, 14 and 19 and of the expired test preparations on Days 2, 7, 12, 16 and 21.  Samples were stored at approximately -20 °C prior to analysis.  Duplicate samples were taken and stored at approximately -20 °C for further analysis if necessary." [Harris 2013, p. 14]

"The test item concentration in the test samples was determined by high performance liquid chromatography with UV detection (HPLC/UV) [at a wavelength of 230 nm] using an external standard.  The test item gave a chromatographic profile consisting of two peaks." [Harris 2013, p. 44]

"[T]he limit of quantitation [LOQ] was determined to be 0.0036 mg/L."  [Harris 2013, p. 48]

"For the estimation of the "Lowest Observed Effect Loading Rate" (LOEL) and the "No Observed Effect Loading Rate" (NOEL), the numbers of live young produced per adult over the duration of the test for the control, and 1.0, 3.2, 10, 32 and 100 mg/L loading rate WAF test groups were compared using one way analysis of variance incorporating the Williams test for differences between treatment means when several dose levels are compared with a zero dose control...  Results from the control and 1.0, 3.2, 10, 32 and 100 mg/L loading rate WAF test groups [for] Daphnia length data, determined for the surviving daphnids on termination of the test, were compared using one way analysis of variance incorporating Bartlett's test for homogeneity of variance... and Dunnett's multiple comparison procedure for comparing several treatments with a control...  All statistical analyses were performed using the SAS computer software package (SAS 1999-2001)." [Harris 2013, p. 15]

The test was considered valid if the following criteria were met:
   * Control mortality was <= 30%.
   * Dissolved oxygen content was > 3 mg O2/L.
   * Control group pH did not deviate by > 1.5 units.
   * The mean number of live young produced per surviving adult daphnid in the control group was >= 60 after 21 days.
   * The coefficient of variation for the control group was <= 25%.
   * The first offspring in the control group were born before a maximum of nine days.

RESULTS

Endpoint value(s): The following values are based on WAF loading rates:
21-day EL50 (mortality/immobility):  > 100 mg/L loading rate 
21-day EL50 (reproduction):  > 100 mg/L loading rate
21-day EL50 (growth):  > 100 mg/L loading rate

21-day NOEL (mortality/immobility):  100 mg/L loading rate
21-day NOEL (reproduction):  10 mg/L loading rate
21-day NOEL (growth):  1.0 mg/L loading rate

21-day LOEL (mortality/immobility):  > 100 mg/L loading rate
21-day LOEL (reproduction):  32 mg/L loading rate
21-day LOEL (growth):  3.2 mg/L loading rate

MATC (growth):  1.79 mg/L loading rate

The following values are based on mean measured concentrations that correspond with nominal loading rates:
21-day EC50 (mortality/immobility):  > 0.21 mg/L 
21-day EC50 (reproduction):  > 0.21 mg/L 
21-day EC50 (growth):  > 0.21 mg/L 

21-day NOEC (mortality/immobility):  0.21 mg/L 
21-day NOEC (reproduction):  0.18 mg/L* 
21-day NOEC (growth):  0.14 mg/L* 

21-day LOEC (mortality/immobility):  > 0.21 mg/L 
21-day LOEC (reproduction):  0.12 mg/L*
21-day LOEC (growth):  0.13 mg/L* 

MATC (growth):  0.13 mg/L loading rate

*Based on mean measured concentrations that correspond with the reported LOEL and NOEL values, the LOEC values for reproduction and growth are less than the NOEC values.  LOEC values, by definition, must be greater than NOEC values, suggesting that the LOEC and NOEC values identified above are not reliable.  The large fluctuations in measured test substance concentrations for each nominal loading rate preclude the accurate estimation of LOEC and NOEC values.  

Results remarks:  
Measured concentrations are summarized in Table 1.  Based on combined "old" and "fresh" samples for the 0 (dilution control), 1, 3.2, 10, 32, and 100 mg/L groups, mean measured concentrations were < LOQ, 0.14 (14% of nominal), 0.13 (4% of nominal), 0.18 (2% of nominal), 0.12 (0.4% of nominal), and 0.21 (0.21% of nominal) mg/L, respectively.  The test substance was detected in several control samples, but an explanation was not provided for these findings.  

            Table 1.  Measured Concentrations of Test Substance[a]
Time point (day)
                         Nominal concentration (mg/L)

                                       0
                                       1
                                      3.2
                                      10
                                      32
                                      100
0 (fresh)
                                 0.00414/     
                                  < LOQ[a]
                                     0.521
                                     0.529
                                     0.814
                                     0.546
                                     0.927
2 (old)
                                   0.00372/
                                  < LOQ[a]
                                     0.301
                                     0.330
                                    0.0163
                                    0.0102
                                    0.0441
5 (fresh)
                                   < LOQ
                                     0.14
                                    0.0475
                                     0.509
                                    0.0168
                                    0.0522
7 (old)
                                   0.00441/
                                  < LOQ[a]
                                   < LOQ
                                   < LOQ
                                    0.00517
                                   < LOQ
                                   < LOQ
9 (fresh)
                                   < LOQ
                                   < LOQ
                                   < LOQ
                                    0.0870
                                     0.061
                                     0.489
12 (old)
                                    0.0115/
                                  < LOQ[a]
                                   < LOQ
                                   < LOQ
                                    0.0209
                                    0.00371
                                    0.00486
14 (fresh)
                                   < LOQ
                                    0.0888
                                     0.121
                                     0.281
                                     0.151
                                     0.467
16 (old) 
                                    0.0192/
                                  < LOQ[a]
                                   < LOQ
                                   < LOQ
                                     0.018
                                    0.0123
                                    0.00943
19 (fresh)
                                   < LOQ
                                     0.331
                                     0.242
                                    0.0513
                                     0.342
                                    0.0573
21 (old)
                                    0.0180/
                                  < LOQ[a]
                                   < LOQ
                                   < LOQ
                                    0.0114
                                    0.00588
                                   < LOQ
Mean measured concentration (mg/L)[b]
                                     0.14
                                     0.13
                                     0.18
                                     0.12
                                     0.21
[a] Duplicate sample stored at -20°C prior to analysis 
[b] Mean of combined fresh and old samples
LOQ = Limit of Quantitation (0.0036 mg/L)

"Analysis of the test preparations showed variable measured concentrations on a day-to-day basis.  This variation is common in the analysis of WAFs of low solubility, complex reaction mixtures.  The results of the analysis did however show a decline in measured concentrations between the fresh and old test media samples on each preparation/exposure period." [Harris 2013, p. 19]

"No significant mortalities occurred at the 1.0, 3.2, 10, 32 and 100 mg/L loading rate WAF test concentrations throughout the test.  There was one [adult] daphnid observed to be immobilized in the 32 mg/L loading rate WAF group, however, since no more than 10% mortality occurred, this is considered not to have affected the outcome of the test." [Harris 2013, p. 18]

"After 21 days the length of each surviving adult [of the parental generation] was determined...  The results showed that there were no statistically significant differences (p >= 0.05) between the control and the 1.0 mg/L loading rate WAF test group in terms of length of the daphnids after 21 days exposure to the test item...  However, there was a significant difference [in daphnid length at the end of exposure] between the control and 3.2, 10, 32 and 100 mg/L loading rate WAF test concentrations." [Harris 2013, p. 18]

Mean daphnid lengths at day 21 in the control, 1.0, 3.2, 10, 32, and 100 mg/L WAF loading rate groups were 4.4 +- 0.13, 4.4 +- 0.16, 4.2 +- 0.13, 4.1 +- 0.20, 4.0 +- 0.12, and 3.8 +- 0.20 mm, respectively.  Daphnid lengths, expressed as percentages of control values, were 100%, 95%, 93%, 91%, and 86% at WAF loading rates of 1, 3.2, 10, 32, and 100 mg/L, respectively.

"After 21 days there were no statistically significant differences between the control and the 1.0, 
3.2 and 10 mg/L loading rate WAF test groups in terms of the number of live young produced per adult.  The 32 and 100 mg/L loading rate WAF test groups showed a statistically significant difference from the control and the remaining test groups after 21 days in terms of producing fewer numbers of live young per adult." [Harris 2013, p. 18]

The mean numbers of live young produced per adult by day 21 in the control, 1.0, 3.2, 10, 32, and 100 mg/L WAF loading rate groups were 120.2 +- 17.3, 133.9 +- 10.8, 134.2 +- 16.6, 124.3 +- 17.7, 106.0 +- 18.8, and 99.9 +- 12.2, respectively.  Live young per adult, expressed as percentages of control values, were 111%, 112%, 103%, 88%, and 83% at WAF loading rates of 1, 3.2, 10, 32, and 100 mg/L, respectively.

"[A]n assessment made at each media renewal showed the "filial" daphnids produced by all the test groups were in the same general condition as the young produced by the controls over the duration of the test...  There were no unhatched eggs or dead young in all control and treatment groups surviving to maturation." [Harris 2013, p. 19]

Water temperatures measured daily in one replicate test chamber per concentration ranged from 19 to 21 °C.  Dissolved oxygen concentrations were maintained at 7.7 to 9.1 mg O2/L throughout the test period (remained >= 85% of saturation).  Measurements of pH in the test chambers ranged from 7.3 to 8.3 throughout the test period.  Light intensity ranged from 681 to 965 lux throughout the test period.  Ranges of water hardness, conductivity, and alkalinity measurements in the control and 100 mg/L nominal loading rate treatments during the study were 216-260 mg CaCO3/L, 680-833 uS/cm, and 39.6-51.6 mg CaCO3/L, respectively.

The results of the test were considered valid based on the following: 
   * Control mortality was 0%.
   * Dissolved oxygen content was > 7.7 mg O2/L.
   * The maximum deviation in pH in the control group was 1.0 units.
   * The mean number of live young produced per surviving adult daphnid in the control group was 120 after 21 days.
   * The coefficient of variation for the control group was 14.4%.
   * The first offspring in the control group were recorded on day 8.

CONCLUSIONS 

Daphnia magna exposed to the test item (Benzoflex 1046 mixture) at nominal loading rates of 1.0 to 100 mg/L for 21 days showed no treatment-related reductions in survival/immobility.  Treatment-related reductions in growth (length) of the surviving daphnids of the parental generation were seen at WAF nominal loading rates >= 3.2 mg/L.  Treatment-related effects on reproduction, as measured by a reduction in the number of live young per daphnid, occurred in the 32 and 100 mg/L WAF loading rate groups.  The NOEL and LOEL for this study, based on the most sensitive endpoint tested (growth), were 1.0 and 3.2 mg/L, respectively, based on WAF loading rates.  The MATC was calculated to be 1.79 mg/L based on WAF loading rates.  The 21-day EL50 values for adult mortality/immobility, reproduction, and growth were > 100 mg/L based on WAF loading rates. 

STUDY RELIABILITY 

[3] Not reliable.  This study is considered unreliable because the test was not conducted up to the water solubility limit of the test item (1.01 mg/L) and the test failed to identify an EL50 for chronic toxicity to daphnids.  Measured test concentrations of "fresh" test solutions (sampled at the time of solution renewal) only approached the limit of water solubility on Day 0, when measured test substance concentrations of 0.814 mg/L and 0.927 mg/L were reported for nominal loading rates of 10 and 100 mg/L, respectively; measured test substance concentrations were 0.521, 0.529, and 0.546 mg/L for the remaining three loading rates of 1, 3.2, and 32 mg/L, respectively, on Day 0.  The measured concentrations of the "fresh" test solution samples collected on days 5, 9, 14, and 19 of the daphnid test ranged from 50 to 100% less than the water solubility limit.  No explanation was provided for the difference between the measured concentrations of "fresh" test solutions and the reported water solubility limit.  The solubility and stability of the test substance in the dilution water were not determined as recommended by ASTM E1193-97. 

Additional deficiencies included the following: 
   (1) The test item was detected in control samples collected on Days 0, 2, 7, 12, 16, and 21, suggesting unreliable analytical techniques for measuring concentrations or inadequate quality control.  
   (2) Measured test substance concentrations in WAFs at test solution renewal were highly variable (see Table 1 above).  For example, the measured concentrations of "fresh" solutions of WAFs at a nominal loading rate of 100 mg/L were 0.927, 0.0522, 0.489, 0.467, and 0.0573 mg/L on Days 0, 5, 9, 14, and 19, respectively.  No explanations for the variability among measured concentrations of "fresh" test solutions were provided in the study report.  This variability raises concerns about the consistency and reproducibility of the methods used to prepare the test solutions during the test and/or the analytical methods used to determine test substance concentrations.  The study report did not provide a detailed description of the procedures for storage and handling of WAF preparations.  Missing information included, for example, whether the WAF was stored prior to use and, if so, for how long and under what conditions.  The inconsistencies among test item concentrations for individual WAF loading rates leads to uncertainty in the reported endpoint values.  
   (3) Substantial declines in measured test substance concentrations were observed between test solution renewals.  Measured concentrations of "old" test solutions (sampled prior to solution renewal) were 38% to 99% less than the respective measured "fresh" concentrations.  No explanation for these declines was provided in the study report and no analysis was conducted to determine the cause.  No data were provided to demonstrate instability of the test substance in the test media.
   (4) The combination of highly variable test substance concentrations at test solution renewal and steep declines in test substance concentration between renewals resulted in large variations in measured test substance concentrations at each WAF loading rate over the 21-day exposure period.  The ranges for the measured concentrations ("fresh" and "old" combined) were < LOQ-0.521, < LOQ-0.529, 0.00517-0.814, < LOQ-0.546, and < LOQ-0.927 mg/L, respectively, for nominal loading rates of 1, 3.2, 10, 32, and 100 mg/L.  The mean measured concentrations (0.14, 0.13, 0.18, 0.12, and 0.21 mg/L, respectively, for nominal loading rates of 1, 3.2, 10, 32, and 100 mg/L) did not consistently increase with increasing loading rate concentrations.  As a result, reliable NOEC and LOEC values for reproduction and growth cannot be identified.
   (5) The 100 mg/L loading rate WAF was filtered through a glass wool plug on day 9 to remove microscopic particles of the test item in the water column.  The use of glass wool for removal of undissolved test material is not recommended.  Instead, the use of a separatory funnel is suggested.
   (6) The study did not include data on observations of first- and second-generation daphnids for abnormal development or aberrant behavior as recommended by section 12.10.5 of ASTM E1193-97. 
 
REFERENCE

Harris, S. 2013.  Benzoflex 1046: Daphnia magna reproduction test.  Harlan Laboratories Ltd., Shardlow, Derbyshire, UK, Project No. 41202706.  Sponsor: Eastman Chemical Company, Kingsport, TN.  Study report available online at http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPPT-2009-0112-0145 as of September 12, 2013.

Text in quotes is taken directly from Harris (2013).

D-3.  Genetic Toxicity  -  Chromosomal Aberrations

Summary of endpoint:  Adequate 

Study 1:  Chromosome aberration assay in human lymphocytes (Bowles, 2013)

Title:  Benzoflex 1046: Chromosome aberration test in human lymphocytes in vitro

This study is:	 Adequate		  Not Adequate

TEST SUBSTANCE   

Identity:  Benzoflex 1046; supporting substance for propanoic acid, 2-methyl-, 3-(benzoyloxy)-2,2,4-trimethylpentyl ester (CASRN 22527-63-5)

Remarks:  Clear colorless liquid; Batch No. V046212201; Composition: 68.8% propanoic acid, 2-methyl-, 3-(benzoyloxy)-2,2,4-trimethylpentyl ester (CASRN 22527-63-5), 19.5% 1,3-pentanediol, 2,2,4-trimethyl-1,3-dibenzoate (CASRN 68052-23-3), 7% 2,2,4-trimethyl-1,3-pentanediol diisobutyrate (CASRN 6846-50-0), 4.2% miscellaneous esters, 0.5% volatiles, and 0.02% moisture.

METHOD

Method/guideline followed:  Compatible with OECD TG 473 ("In Vitro Mammalian Chromosome Aberration Test"); Method B10 of Commission Regulation (EC) No. 440/2008

Test type:  In vitro mammalian chromosome aberration test

GLP compliant?:  Yes; with the following exception: "No analysis was carried out to determine the homogeneity, concentration or stability of the test item formulation.  The test item was formulated within two hours of it being applied to the test system; it is assumed that the formulation was stable for this duration." [Bowles 2013, p. 3]

Year study performed:  2012

Test species:  Human peripheral blood lymphocytes 

Metabolic activation:  With (rat S9 microsomal fraction) and without

Concentrations tested:  
Experiment 1: 4-hour exposure followed by 20-hour incubation without test item:
0 (vehicle control)*, 5*, 10*, 20*, 40, 60, and 80 ug Benzoflex 1046/mL (without S9)
0 (vehicle control)*, 10, 20, 40*, 80*, 120*, and 160 ug Benzoflex 1046/mL (with S9)
Experiment 2: 
Without S9, 24-hour continuous exposure:
0 (vehicle control)*, 5, 10, 20*, 40*, 60*, and 80 ug Benzoflex 1046/mL
With S9, 4-hour exposure followed by 20-hour incubation without test item
0 (vehicle control)*, 5, 10, 20*, 40*, 80*, and 120 ug Benzoflex 1046/mL
*Concentrations selected for metaphase analysis

Statistical methods used:  "The frequency of cells with aberrations excluding gaps and the frequency of polyploid cells was compared, where necessary, with the concurrent vehicle control value using Fisher's Exact test." [Bowles 2013, p. 15]

Vehicle:  Dimethyl sulfoxide

Test method/conditions remarks:  Human lymphocytes were collected from a blood sample isolated from the peripheral circulation of a screened volunteer.  Approximately 48 hours after mitogenic stimulation with phytohemagglutinin, duplicate cultures were treated with the test item or appropriate control in the presence or absence of S9 metabolic activation at 37°C.  

"The test item was... dissolved in dimethyl sulphoxide and serial dilutions [were] prepared.  The molecular weight of the test item was unknown and, therefore, designated to be 500 (because it was a mixture of multiple chemicals).  The maximum dose level [in the preliminary toxicity test] was 5000 ug/mL, which was calculated to be equivalent to the 10 mM maximum recommended dose level." [Bowles 2013, p. 9]

"The test item was formulated within two hours of it being applied to the test system." [Bowles 2013, p. 10]

"There was no significant change in pH when the test item was dosed into dimethyl sulphoxide and the osmolality did not increase by more than 50 mOsm." [Bowles 2013, p. 9]

The S9 homogenate was produced from homogenized liver samples collected from male rats that had received three daily oral doses of a phenobarbitone (80 mg/kg-bw/day) and β-naphthoflavone (100 mg/kg-bw/day) mixture prior to S9-mix preparation on the fourth day.  S9 batches were routinely verified for activation of known mutagens in the Ames assay.

Preliminary Toxicity Test:
"A preliminary toxicity test was performed on cell cultures using a 4-hour exposure time with and without metabolic activation followed by a 20-hour recovery period, and a continuous exposure of 24 hours without metabolic activation.  The dose range of test item used was 19.53, 39.06, 78.13, 156.25, 312.5, 625, 1250, 2500 and 5000 ug/mL.  Parallel flasks, containing culture medium without whole blood, were established for the three exposure conditions so that test item precipitate observations could be made.  Precipitate observations were recorded at the beginning and end of the exposure periods." [Bowles 2013, p. 12]

Chromosome Aberrations Assay:
Experiment 1:  Lymphocytes were exposed to the test item or control for 4 hours without or with S9 (2%), followed by a 20-hour incubation period.  Concentrations without S9 activation selected for metaphase analysis were 0 (vehicle control), 5, 10, and 20 ug Benzoflex 1046/mL.  Concentrations with S9 activation selected for metaphase analysis were 0 (vehicle control), 40, 80, and 120 ug Benzoflex 1046/mL.  

Experiment 2:  Lymphocytes were exposed to the test item or control continuously for 24 hours without S9, or exposed to the test item for 4 hours with S9 (1%) followed by a 20-hour incubation period.  Concentrations without S9 activation selected for metaphase analysis were 0 (vehicle control), 20, 40, and 60 ug Benzoflex 1046/mL.  Concentrations with S9 activation selected for metaphase analysis were 0 (vehicle control), 20, 40, and 80 ug Benzoflex 1046/mL.  

Lymphocytes were arrested in mitosis with demecolcine (Colcemid, 0.1 ug/mL) two hours before scheduled harvest.  The cells were treated with 0.075M hypotonic KCl, fixed with methanol/glacial acetic acid (3:1 v/v), mounted on microscope slides, and stained with 5% Giemsa stain.  Slides were randomly coded using a computerized number generator. 

"The slides were checked microscopically to determine the quality of the metaphases and also the toxicity and extent of precipitation, if any, of the test item.  These observations were used to select the dose levels for mitotic index evaluation." [Bowles 2013, p. 14]

"A total of 2000 lymphocyte cell nuclei were counted and the number of cells in metaphase recorded and expressed as the mitotic index and as a percentage of the vehicle control value." [Bowles 2013, p. 14]

"Where possible the first 100 consecutive well-spread metaphases from each culture were counted, where there were approximately 30 to 50% of cells with aberrations, slide evaluation was terminated at 50 cells.  If the cell had 44-48 chromosomes, any gaps, breaks or rearrangements were noted according to the simplified system of Savage (1976) recommended in the 1983 UKEMS guidelines for mutagenicity testing.  Cells with chromosome aberrations were reviewed as necessary by a senior cytogeneticist prior to decoding the slides." [Bowles 2013, p. 14]

"...[C]ells with 69 chromosomes or more were scored as polyploid cells and the incidence of polyploid cells (%) reported." [Bowles 2013, p. 14]

"A positive response was recorded for a particular treatment if the % cells with aberrations, excluding gaps, markedly exceeded that seen in the concurrent control, either with or without a clear dose-relationship.  For modest increases in aberration frequency a dose response relationship is generally required and appropriate statistical tests may be applied in order to record a positive response." [Bowles 2013, p. 30]

Positive Control Substances: 
Experiment 1: Without S9: Mitomycin C at 0.4 ug/mL; With S9: Cyclophosphamide at 5 ug/mL
Experiment 2: Without S9: Mitomycin C at 0.2 ug/mL; With S9: Cyclophosphamide at 5 ug/mL

RESULTS

Genotoxicity result:  Negative

Cytotoxic concentration:  4- and 24-hour exposures without S9: 78.13 ug/mL; 4-hour exposures with S9: 156.25 ug/mL

Results remarks:  Positive and negative control responses were appropriate.

Preliminary Toxicity Test Results:
Precipitate was observed at >= 156.25 ug/mL in cultures in the 4-hour and 24-hour exposure groups without S9, and at >= 312.5 ug/mL in the 4-hour exposure group with S9.  Hemolysis was observed in the 4-hour exposure group at >= 312.5 ug/mL with S9 (but not in the 4-hour exposure group without S9) and in the 24-hour continuous exposure group without S9 at >= 78.13 ug/mL.

"Microscopic assessment of the slides prepared from the exposed cultures showed that metaphase cells were present up to 78.13 ug/mL in the absence of S9 and 156.25 ug/mL in the presence of S9, respectively.  There were no scorable metaphases present at dose levels above these stated in any exposure group...  [T]he test item demonstrated steep toxicity curves in all three exposure groups which made achieving optimum toxicity very difficult." [Bowles 2013, p. 16]

Chromosomal Aberrations Assay:
The test item did not induce any statistically significant increases in the frequency of cells with chromosome aberrations or in the number of polyploid cells in either Experiment 1 or 2.  

Experiment 1:  Scorable metaphases were present at concentrations up to 40 ug/mL without S9 and up to 120 ug/mL with S9.  However, "[t]he toxicity observed at 40 ug/ml in the absence of S9 [i.e., a 77% inhibition of the mitotic index compared with the control] was considered to be excessive and precluded the dose from chromosome analysis."  [Bowles 2013, p. 17]  No precipitate or hemolysis was observed in the blood cultures at the end of the exposure period either in the presence or absence of S9.    

Mitotic indices were inhibited both with and without S9.  Mitotic indices (% of control) were 99, 93, 62, and 23%, respectively, at 5, 10, 20, and 40 ug/mL for the 4-hour exposure without S9.  Mitotic indices (% of control) were 102, 100, 98, and 54%, respectively, at 20, 40, 80, and 120 ug/mL for the 4-hour exposure with S9.

Experiment 2:  Scorable metaphases were present at the maximum test item concentration of 80 ug/mL without S9 and at 80 ug/mL with S9.  However, "[t]he toxicity observed at 80 ug/ml in the absence of S9 [i.e., a 77% inhibition of the mitotic index compared with the control] was considered to be excessive and precluded the dose from chromosome analysis."  [Bowles 2013, p. 18]  No precipitate or hemolysis of the test item was observed at the end of exposure in the absence of S9; however, hemolysis was observed at 120 ug/mL in the presence of S9.  

Mitotic indices were inhibited both with and without S9.  Mitotic indices (% of control) were 102, 87, 64, and 23%, respectively, at 20, 40, 60, and 80 ug/mL for the 24-hour continuous exposure without S9.  Mitotic indices (% of control) were 136, 100, 87, and 42%, respectively, at 10, 20, 40, and 80 ug/mL for the 4-hour exposure with S9.
 
CONCLUSIONS

The test item, Benzoflex 1046, did not induce chromosomal aberrations in human peripheral blood lymphocytes in vitro.

STUDY RELIABILITY

[2] Reliable with restrictions.  Missing study details included sex of blood donor, the cell density seeded at the start of the experiment, justification for choice of vehicle, and stability of the test item in the vehicle.

REFERENCE

Bowles, A.  2013.  Benzoflex 1046: Chromosome aberration test in human lymphocytes in vitro.  Harlan Laboratories Ltd., Shardlow, Derbyshire, UK, Project No. 41201918.  Sponsor: Eastman Chemical Company, Kingsport, TN.  Study report available online at http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPPT-2009-0112-0145 as of September 12, 2013.

Text in quotes is taken directly from Bowles (2013).

E.  DATA ADEQUACY SUMMARY TABLE 
                                       
                                   Endpoint
                          Study Information Available
                                  Acceptable
                                   Data Gap
Aquatic Effects
                            Acute Toxicity to Fish
                                      No
                                       - 
                                     - [1]
                    Acute Toxicity to Aquatic Invertebrates
                                      No
                                       - 
                                     - [1]
                   Chronic Toxicity to Aquatic Invertebrates
                                      Yes
                                      No
                                      Yes
                          Toxicity to Aquatic Plants 
                                      Yes
                                      No
                                      Yes
Health Effects
                                Acute Toxicity
                                      Yes
                                      Yes
                                      No
                            Repeated-Dose Toxicity 
                                     No[2]
                                       - 
                                      Yes
                                 Gene Mutation
                                      Yes
                                      Yes
                                      No
                            Chromosomal Aberrations
                                      Yes
                                      Yes
                                      No
                             Reproductive Toxicity
                                     No[2]
                                       - 
                                      Yes
                            Developmental Toxicity
                                     No[2]
                                       - 
                                      Yes

[1] Acute toxicity to fish and acute toxicity to aquatic invertebrate testing are not required because the measured log Kow of CASRN 22527-63-5 is < 4.2.
[2] According to Eastman's letter to EPA dated December 21, 2012, Eastman expected initiation of a definitive combined repeated dose toxicity study with the reproduction/developmental toxicity screening test in the second quarter of 2013.




