Di-tert-butyl 3,3,5-trimethylcyclohexylidene diperoxide

Bioaccumulation: aquatic / sediment

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

bioaccumulation in aquatic species, other

Remarks:

Test conducted to a pre (OECD319) guideline protocol

Type of information:

experimental study

Remarks:

Conducted to GLP and based on the same principle as the current OECD 319 of BCF calculations being adapted for metabolism using the Gobas Model

Adequacy of study:

key study

Study period:

Not specified, but the date of study initiation was 2007-06-23

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: see 'Remark'

Remarks:

Although there is currently no guideline for this study type, the method is recognised as an animal alternative under REACH. The report provides details on the method and on the validity of the results, the procedures used are clearly described (when taking into account the articles in reference) and the results may be used in a weight of evidence approach.

Reason / purpose for cross-reference:

reference to same study

Reason / purpose for cross-reference:

reference to other study

Qualifier:

according to guideline

Guideline:

other: Trout S9 in vitro metabolism assay

Version / remarks:

Pre OECD 319. Documented to GLP and conducted in Triplicate.

Deviations:

yes

Remarks:

Conducted a 2mg/mL protein an at 20 ºC

Principles of method if other than guideline:

CE cowan Ellsberry, SD Dyer, S Erhardt, MJ Bernhard, AL Roe, ME Dowty, AV Weisbrod. 2008. Approach for extrapolating in vitro metabolism data to refine bioconcentration factor estimates. Chemosphere 70:1804-1817. These authors are referenced by the OECD 319 Guideline due to the basic methodology forming the basis of the guideline. Test material at a concentration of 10 µm was spiked to active and heat inactivated S9 that was incubated at 20ºC for the 2 h test duration.

GLP compliance:

yes

Remarks:

2008-05-15

Specific details on test material used for the study:

Details on properties of test surrogate or analogue material (migrated information):
Appearance (physical state, color): Clear liquid
Molecular Formula: C17H34O4
Molecular Weight: 302.46
CAS: 6731-36-8
Supplier Degussa Initiators
Purity 98%
LOT 354400716

Radiolabelling:

no

Details on sampling:

Sample were taken at 6 time points throught the test in triplicate (3 seperately prepared replicates) for both the active and inactive S9 incubations.

Vehicle:

yes

Details on preparation of test solutions, spiked fish food or sediment:

The test compound (TMCH) was added to a concentration of 10 μM as a 1 mM stock prepared in ethanol using a 50 μL aliquot. Each reaction mixture was prepared as a total of 5 mL. Three active and inactive reaction solutions were seperately prepared and then incubated seperately incubated at approximately 20ºC on a temperature controled shaking platform shaker over a 2 hour period. A non spiked inactive replicate was also prepared to investigate the effects of the matrix on the analysis.

Test organisms (species):

other: trout S9 , Procedure and source fully described in report

Details on test organisms:

The fish were supplied by the Oden State Fish Hatchery located in Alanson, MI and ranged in weight from 35 to 300 gms each. They were held for one week under standard aquaculture conditions before harvesting tissue. The fish were anesthetized with MS222 until immobile prior to sacrificing. The liver was dissected from the carcass, weighed and immediately flash-frozen in liquid nitrogen. The livers were stored at approximately -80 °C until processing further. The trout livers were, minced, and mixed with buffer in a 4:1 ratio of buffer to liver (v:w). The buffer (pH 7.4) was composed of 50 mM potassium phosphate, 0.15 M potassium chloride, and 0.2 M sucrose. The tissue was homogenized on ice using Teflon tipped pestle with glass
mortar. The homogenate was placed into chilled centrifuge tubes and centrifuged at 12,000 g for 30 min at 1-3°C. The supernatant (S9 fraction) was carefully removed and transferred to a chilled bottle or beaker and frozen in 2 mL aliquots on dry ice and transferred to an -80 °C freezer for storage. The total protein was determined using the Pierce BCA™ method (Pierce, Rockford, IL).

Route of exposure:

aqueous

Justification for method:

minimised test method used to support BCF estimates based on Kow

Remarks:

According to pre OECD319 guideline protocol Invitro adaption of BCF

Test type:

other: Test material spiked via solvent decline of test concentrations measured over 2 hours in comparison to active control.

Water / sediment media type:

other: S9 in Tris buffer with added co factors

Total exposure / uptake duration:

120 min

Hardness:

Not applicable in vitro test

Test temperature:

20º

pH:

7.4

Dissolved oxygen:

Not applicable in vitro test

TOC:

Not applicable in vitro test

Salinity:

Not applicable in vitro test

Conductivity:

Not applicable in vitro test

Details on test conditions:

Heat deactivated control reaction solutions were prepared in triplicate with trout liver S9 fraction inactivated by placing a 50 mL centrifuge tube of the diluted S9 preparation in boiling water for approximately 5 minutes. Data generated by the incubation of the heat deactivated S9 with the test substance were used to evalaute for any non-biological dissipation processes such as binding or hydrolysis. A single deactivated reaction solution containing no TMCH was also prepared to evaluate matrix interference during the analysis. Aliquots of 0.5 mL were removed from each reaction solution containing active S9 at 0, 20, 45, 60, 90 and 120 minutes. Aliquots of 0.5 mL were removed from the reaction solutions containing heat-deactived S9 and no test materials at 0, 60, and 120 minutes. The aliquots were added to an equal volume (0.5 mL) of ice-cold hexane and placed on an ice-bath for 30 minutes. Precipitated proteins were removed from the solution by centrifugation.

Nominal and measured concentrations:

Nominal concentration : 10 µM

Reference substance (positive control):

no

Details on estimation of bioconcentration:

The BCF was predicted based on the mass-balance model described by Arnot and Gobas (2003, 2004). To extrapolate from intrinsic clearance rates in the in vitro tests to the kMET, which is used to refine the predicted BCF value, a number of physical and physiological parameters describing fish are required.

Type:

BCF

Value:

>= 442.73 - <= 765.76 L/kg

Basis:

other: in vitro extrapolation

Calculation basis:

kinetic

Metabolites:

Not identified

Details on results:

body rate constant of metabolism (kMET) using the technique described by Cowan et al.,(2007). The kMET was then used as input in the Gobas bioaccumulation model to provide an estimate of metabolism when calculating BCF. The original BCF for TMCH calculated using the Gobas model and assuming no metabolism was 46,097 kg/day. With the addition of metabolism BCF ranged from 422 to 765 kg/day. The BCF with the addition of metabolism is significantly lower than 2000 and would be expected to not be considered bioaccumulative using regulatory criteria.

Reported statistics:

Descriptive statistics were used, i.e., mean ± standard deviation. All calculations in the data were conducted using Microsoft Excel spreadsheets and data in full precision mode (15 digits of accuracy).

kMET Calculation	Trout w/ arterial hepatic blood flow	Trout with arterial and portal hepatic blood flow
CLm (ml/hr/gm of protein)	8.67E+01	8.67E+01
CLi -- Intrinsic Clearance in liver (ml/hr/Kg)	5.96E+01	5.96E+01
CLi -- Intrinsic Clearance in liver (L/d/Kg)	1.43E+00	1.43E+00
CLh -- Hepatic Clearance incorporating blood flow (L/d/Kg))	7.18E-01	1.25E+00
kMET(1/day)	0.128	0.224
BCF Calculation
K1 - uptake rate (L/Kg day)	1.00E+02	1.00E+02
K2- rate of elimination via resp (1/day)	2.75E-05	2.75E-05
Ke - rate of chemical egestion via fecal material (kg/kg/day)	1.35E-03	1.35E-03
Kg - growth dilution (1/day)	7.96E-04	7.96E-04
BCF = k1/(k2+ke+kg+km) incorporating kMET	765.76	442.73
BCF with kMET= 0	46097.54	46097.54

Validity criteria fulfilled:

not applicable

Conclusions:

The BCF value of the substance was determined with and without metabolism being taken into account. The calculation is based on the Arnot Gobas BCF mass balance model run with and without metabolism. When metabolic rate (Kmet) is set to 0, the model calculates a BCF ot 46097. When Kmet is experimentally determined using two methods: arterial hepatic and arterial hepatic and portal, blood flow extrapolation further to a trout hepatocyte in vitro study, BCFs are calculated as 766 and 443 L/Kg respectively.

Executive summary:

Activity in the active control exceeded that of the inactive replicate. This in theory allows correction for the abiotic activity. The current guideline allows inactive depletion to be subtracted from active this approach to be applied to calculate the biological depletion rate. The depletiation curves still fall relitively close together and the competant authorities have indicated that the level of uncertaintly that actual biotransformation is occurring is too high. A flatter inactive replicate is desirable. For this reason improvements in the approach would need to be applied for full acceptance. In conclusion an indication of possible biotransformation is given but is not conclusive.

Reference 2

Endpoint:

bioaccumulation in aquatic species, other

Remarks:

Spiking Test to fish tissue

Type of information:

experimental study

Adequacy of study:

supporting study

Study period:

not specified, but the date of study initiation was 2008-02-29

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Good quality study with GLP statement and substance identity showing the impact of contact time with fish tissue on the recovery of the organic peroxide CAS#6731 -36 -8 and the production of degradation products.

Reason / purpose for cross-reference:

reference to same study

Reason / purpose for cross-reference:

reference to other study

Principles of method if other than guideline:

not applicable

GLP compliance:

yes

Remarks:

2008-11-12

Radiolabelling:

no

Vehicle:

yes

Test organisms (species):

other: rainbow trout, but specie was not specified

Route of exposure:

aqueous

Test type:

not specified

Water / sediment media type:

not specified

Total exposure / uptake duration:

60 min

Nominal and measured concentrations:

Nominal concentrations : 10 and 1 µg/g of CAS#6731-36-8
Measured concentrations : The concentration of CAS#6731-36-8 in samples spiked at 10 µg/g ranged from a maximum of 10.56 µg/g at T=0min to a minimum of 6.41 µg/g at 45 minutes. The concentration of CAS#6731-36-8 gin sample spiked at 1.0 µg/g varied more widely than those dosed at 10 µg/g; the concentration ranged from a maximum of 1.14 µg/g at T=0min to a minimumof 0.49 µg/g at 60 minutes. See table 1 in "Any other information on results incl. tables" field.

Details on estimation of bioconcentration:

No estimation of bioconcentration was performed during the study. However, the objective of this study was to evaluate the impact of contact time on the recovery of the organic peroxide CAS#6731-36-8 from freshly prepared and potentially metabolically active fish tissue.

Table 1 : Concentrations of CAS#6731 -36 -8, CAS#873 -94 -9 and t-butanol in fish tissue samples *

Sample number	Spike level, time (min.) and replicate	CAS#6731-36-8 (µg/g)	CAS#6731-36-8 % recovery	CAS#873-94-9 (µg/g)	t-butanol (µg/g)
1 2 5 6	Control-0-1 Control-0-2 Control-60-1 Control-60-2	<LLQ** <LLQ <LLQ <LLQ	--*** -- -- --	-- -- -- --	<LLQ <LLQ <LLQ <LLQ
Samples spiked with CAS#6731-36-8
7 8 9 10 11 12 13 14 15 16   17 18 19 20 21 22 23 24 25 26	10 µg/g 0-1 10 µg/g 0-2 10 µg/g 10-1 10 µg/g 10-2 10 µg/g 30-1 10 µg/g 30-2 10 µg/g 45-1 10 µg/g 45-2 10 µg/g 60-1 10 µg/g 60-2   1 µg/g 0-1 1 µg/g 0-2 1 µg/g 10-1 1 µg/g 10-2 1 µg/g 30-1 1 µg/g 30-2 1 µg/g 45-1 1 µg/g 45-2 1 µg/g 60-1 1 µg/g 60-2	10.56 10.08 8.64 9.44 9.60 8.96 6.41 10.00 9.92 10.24   0.88 1.14 1.06 0.87 0.57 0.54 0.67 0.90 0.49 0.96	105.6 100.8 86.4 93.6 96.0 88.8 63.9 99.2 98.4 102.4   88.0 114.4 105.6 86.4 57.0 54.2 66.6 88.8 49.0 96.0	0.23 0.17 <LLQ 0.23 0.20 0.16 0.13 0.30 0.29 0.29   0.02 0.03 0.03 0.02 <LLQ <LLQ <LLQ <LLQ <LLQ <LLQ	2.77 <LLQ <LLQ 1.07 <LLQ <LLQ 1.85 <LLQ <LLQ 0.97   <LLQ <LLQ <LLQ <LLQ <LLQ <LLQ <LLQ <LLQ <LLQ <LLQ
Samples spiked with t-butanol
27 28 29 30 31 32	5 µg/g 0-1 5 µg/g 0-2 5 µg/g 30-1 5 µg/g 30-2 5 µg/g 60-1 5 µg/g 60-2	-- -- -- -- -- --	N/A**** N/A N/A N/A N/A N/A	-- -- -- -- -- --	1.29 <LLQ <LLQ <LLQ <LLQ <LLQ

* tert-butanol (purity of 99.5%) and CAS#873 -94 -9 / 3,3,5 -trimethylcyclohexane (purity of 96.4%) were used for identification and quantification of degradation products anticipated from the degradation of CAS#6731 -36 -8.

** <LLQ = <10.4 ng/mL for CAS#6731 -36 -8, <1.08 ng/mL for CAS#873 -94 -9 and <10.7 ng/mL for t-butanol

*** -- = not analyzed for specific compound

**** N/A = not applicable

Validity criteria fulfilled:

not applicable

Conclusions:

When spiking freshly prepared fish homogenate with the organic peroxide CAS#6731-36-8 at residue levels, care should be taken to extract the tissues as quickly as possible. The lower extraction efficiencies could be attributed to the metabolism of CAS#6731-36-8 by peroxidases still active in the tissues. The detection of t-butanol may also be limited as a result of metabolism as well as tissues, especially those associated with the liver have elevated levels of alcohol-dehydrogenase.

Executive summary:

The objective of this study was to evaluate the impact of contact time on the recovery of the organic peroxide CAS#6731 -36 -8 from freshly prepared and potentially metabolically active fish tissue. Samples of freshly homogenized rainbow trout were spiked with the organic peroxide CAS#6731 -36 -8 at either 10 µg/g or 1.0 µg/g. In addition, samples were also spiked at 5 µg/g with t-butanol, a known metabolite of CAS#6731 -36 -8. For comparison tissue samples were prepared with only carrier solvent (100% ethanol) to evaluate for matrix interferences. The samples were allowed to sit on the bench top with samples removed in duplicate over the course of 1 hour. At each sampling interval, 20 mL of acetonitrile was added to the tissue. The tissue were thoroughly homogenized with the solvent and centrifuged to pellet the tissue. Aliquots were removed from the extraction solvent concentrated to dryness and re-suspended in either ethyl acetate for analysis of the peroxide or hexane for analysis of the t-butanol. The concentration of CAS#6731 -36 -8 in samples spiked at 10 µg/g nominal ranged from a maximum of 10.56 µg/g at T0 to a miminum of 6.41 µg/g at 45 minutes. When normalized to T0 recovery varied from 100% to a minimum of 79.5%. Samples spiked at 1.0 µg/g (nominal) showed more variability and ranged from 1.14 µg/g to a minimum of 0.49 µg/g at 0 and 60 minutes, respectively. When normalized to the average concentration at T0, sample recovery ranged from an average of 100% at T0 to a minimum of 49% at 60 minutes. In general, samples spiked at 1.0 µg/g showed more extreme losses of recovery with incubation time, which were likely due to the presence of metabolically active tissues in the freshly prepared fish homogenate. This study demonstrates the importance when working with metabolically labile materials at trace levels, to work quickly with the tissues to minimize the potential impact of metabolism on the organic peroxide. Samples were also analyzed for t-butanol, a proposed primary degradation product. When spiked at 5 µg/g in fish tissue, all samples had less than the limit of quantification other than a single sample at T0. Four out of twelve samples spiked with 10 µg/g CAS#6731 -36 -8 had detectable quantities of t-butanol while no samples spiked at 1 µg/g CAS#6731 -36 -8 had detectable quantites. Liver and other tissues are known to have significant levels of alcohol dehydrogenase, which would quickly metabolize the alcohol to its concomitant aldehyde.

Description of key information

The existing in vivo cannot currently be used for risk assessment due to major methodological deficiencies. Primarily the lack of depuration phase the fact that the data is pre GLP and pre guideline. The test material also meets all of the criteria for a feeding study being more appropriate. A new study meeting the guideline reccommendations or read across to a valid study for a close analogue has been proposed to the authorities. The most appropriate existing study is the Trout S9 in vitro metabolic study used in conjunction with a BCF metabolism correction model. The study is conducted to GLP in triplicate and is based on the current OECD 319 methodology. This study is considered at present as the best evidence of non B available at this time. This provides BCF values of 433 to 766 L/Kg.

Key value for chemical safety assessment

BCF (aquatic species):

766 L/kg ww

Additional information

The bioaccumulation test in CarpCyprinus carpio was conducted on the reference substance in a NITE study. The test substance concentrations were 0.2 and 0.02 mg/L maintained in a solvent. The measured concentrations within the test water was an established value of 60% or greater. The duration of exposure was 8 wk, with continuous water flow-through method. A Bioconcentration factor (BCF) at 6510 -6200 for the 1st concentration section (0.2 mg/L) and 4960-5970 for the 2nd concentration section (0.02 mg/L), based on steady state values. No kinetics were provided, nor tissue concentrations. This in vivo study is thought to be fundamentally flawed as the test substance appears to diminish relatively rapidly in water and more so in fish tissues. This can be attributed to the fact that the test was performed at water concentrations 10 and 100 times the estimated water solubility of the test substance. Moreover, critical information on the relevance of time related recovery factors and analytical methodology data are missing. The study cannot be considered valid. An in vitro study was performed using the trout S9 in vitro metabolism assay (Erhardt 2008) and the metabolism results were combined with a BCF uptake depuration model (Arnott & Gobas, 2003, 2004). When metabolic rate (Kmet) is set to 0, the model calculates a BCF of 46097. When Kmet is experimentally determined and the values obtained from the two methods incorporated into the equation (arterial hepatic and arterial hepatic and portal blood flow) further to a trout hepatocyte in vitro study, BCFs are calculated as 766 and 443 L/Kg respectively. In the absence of a valid in vivo study the BCF value of 766 L/Kg will be used for risk assessment as a worst case estimate.