1,3,4,6,7,8-hexahydro-4,6,6,7,8,8-hexamethylindeno[5,6-c]pyran

Bioaccumulation: aquatic / sediment

Description of key information

The aquatic BCF for HHCB 1584 l/kg is based on a WoE and the key value is derived from an aquatic OECD TG 305 test.

The aquatic BCF for HHCB-lactone is derived from the HHCB-dietary BMF, in which also the depuration of HHCB-lactone was measured, and this results in a BCF of 1293 l/kg 

HHCB-hydroxylated-carboxylic acid has a negligible aquatic BCF based on the log Kow of 0.6 based on the dissociated acid at environmental pH as calculated with SPARC.

Key value for chemical safety assessment

BCF (aquatic species):

1 584 L/kg ww

Additional information

Note: the complete WoE is presented in the file attached.

Bioaccumulation in aquatic environments A Weight of Evidence (WoE) – Endpoint Summary

 A wealth of information on the bioaccumulation potential of HHCB in both lab-based and field studies has been identified. Four lab-based studies have been selected as key studies as they satisfied REACH information requirements and utilized standard OECD guidelines for assessing metabolism and bioaccumulation in fish. The following studies are integrated in a WoE and are discussed further herein: An academic publication following the OECD TG 319B (Laue et al., 2020), one industry study report following the OECD TG 305-Aqueous Exposure (van Dijk, RCC, 1996), which is the key study in the EU RAR (2008), a poster from Butte and Ewald, 1999, who used the OECD TG 305-Aqueous Exposure and an industry report using the OECD TG 305-Dietary Exposure method (Schneider et al., Eurofins, 2020).

In the EU RAR (2008) for HHCB, a fish BCF value of 1584 derived from Van Dijk (RCC, 1996) was used as the main line of evidence in the assessment of HHCB bioaccumulation in fish among other information available. This key study, as described by the EU RAR, will also be presented here too in detail. In addition, the Butte and Ewald, 1999 study, also in the EU RAR, is included here as it presents a lipid corrected BCF and receives similar weight as the van Dijk (RCC, 1996) study. The whole review on bioaccumulation data presented at the time, however, will not be repeated here. A more recent review on the available studies relating to HHCB and its potential to bioaccumulate in various fish species has recently been submitted to the US-EPA as supplementary information for the PBT assessment of HHCB and is made available here for completeness (IFRA-ETF letter to US-EPA, 2020, attached to the Endpoint summary).

 Bioaccumulation endpoint – Key studies for WoE

For assessing the bioaccumulation potential of HHCB four key studies have been selected that demonstrate the potential of HHCB to undergo metabolic transformation in vitro (OECD 319B in vitro clearance using trout liver RT-S9), bioconcentrate from aqueous exposure in vivo (OECD 305-Aqueous Exposure) and biomagnify from food to fish in vivo (OECD 305-Dietary Exposure). The studies described here all receive Klimisch scores of 1 or 2 and can as such be used  in the bioaccumulation assessment of HHCB according to REACH. In addition to information provided on the parent compound HHCB, the bioaccumulation potential of two key metabolites of HHCB will be discussed.

The executive summaries of the four key studies are presented first in context of the parent compound and thereafter the bioaccumulation potential of the two primary metabolites. The two primary metabolites identified, HHCB-lactone and hydroxylated HHCB acid, have been found in both environmental monitoring studies (e.g. waste water monitoring studies) and in environmental fate studies (e.g. analogous to the OECD 314B) (Federle et al. 2002 and Itrich and Federle, 2003). These, or very similar metabolites are also expected in fish. Known standards of HHCB, its lactone and the acid were measured to have a log Kow of 5.9, 4.71 and 0.6, respectively.

HHCB (log Kow = 5.9)                Lactone (log Kow = 4.7)                 Hydroxy acid (log Kow = 0.6)

Fig. 1      HHCB and its anticipated metabolites in fish are HHCB-lactone and HHCB-hydroxylated acid (after cleavage of the lactone). The log Kow of HHCB and these metabolites were determined in an activated sludge-Die Away test (see reference in the text). The log Kows were 5.9, 4.7 and 0.6, respectively. The overall log Kow for HHCB in IUCLID is presented as 5.3 based on a standard HPLC-test.

The information presented herein (i.e. bullets 1-5) are integrated in a Weight of Evidence to derive a BCF and conclude on the bioaccumulation potential of HHCB and its metabolites (i.e. bullet 6). The Weight of Evidence is presented in the following order:

1)   Non-testing data aquatic bioaccumulation (Veith equation, ECHA guidance, 2017, R.7c.10.3.2);

2)    In vitro data on aquatic bioaccumulation and calculated BCF (Laue et al, 2020 and Laue et al. Suppl, 2020);

3)    In vivo tests for aquatic bioaccumulation; 3A OECD TG 305, van Dijk (RCC, 1996); and 3B OECD TG 305, Butte and Ewald  (1999)

4)   Fish dietary bioaccumulation test (OECD TG 305; Schneider et al., Eurofins, 2020);

5)    Other BCF information on HHCB-lactone and HHCB-hydroxylated acid (Federle, 2002 and Itrich and Federer, 2003); and

6)   Conclusion on BCF and Bioaccumulation

The following WoE for the BCF is applied:

1. Non-testing data aquatic bioaccumulation

For substances with log Kow < 6 and screening purposes, the Bioconcentration Factor (BCF) can be conservatively estimated (in absence of metabolization) using the following regression model by Veith et al., 1979, referenced in the ECHA guidance, 2017 (R.7c.10.3.2):

log BCF(wet weight) = 0.85.log Kow- 0.70. With log Kow = 5.3, the estimated BCF of HHCB is 6,383 l/kg.

The BCF estimated using a log Kow of 5.3 with BCF/BAF that includes metabolization is 1459 l/kg (EpiSuite). The Arnot and Gobas model for upper trophic level included in BCF/BAF program predicts a value of 893 l/kg also using a log Kow of 5.3. This indicates that if HHCB is metabolized by fish it would not bioaccumulate significantly and trophic dilution might be expected.

2.In vitro data on aquatic bioaccumulation – Calculated BCFs – Laue et al. (2020)

In vitro metabolism of HHCB (Galaxolide) has been studied using assays with trout S9 cell fractions and cryopreserved hepatocytes. Of the four studies that have reported the intrinsic clearance rate of HHCB in vitro; two studies suggest  that HHCB may be metabolized by fish (Weeks et al., 2020; and Laue et al., 2020) whereas two studies do not (WilResearch, 2013; and Weeks et al., 2019). All references are presented in the IFRA-ETF letter (2020).

Introduction: Of the four studies reporting in vitro intrinsic clearance of HHCB, the key study selected for the WoE is the study performed by Laue et al. (2020) as the HHCB exposure concentration was  0.05 umol (12.9 ug/L), which was 10-50x lower than the concentrations used in the other available studies. This study was a comprehensive evaluation of HHCB in vitro metabolism as it assessed the impact of four varying concentrations of HHCB to the in vitro intrinsic clearance rate in trout S9 cell fractions. Laue et al. (2020) showed that the lower concentrations resulted in the highest metabolic activity, indicating that high exposure concentrations are potentially oversaturating the metabolic pathway leading to an underestimation of the metabolic clearance rate (Laue et al., 2020, Nichols et al. 2013a). The study was performed following the OECD TG 319B and may be used in a WoE according to ECHA guidance (2017, R7.10.3). Method: In this study, S9 concentrations of 1mg/mL protein were used in accordance with OECD TG 319B. The incubation temperature was 12°C. The free concentration was calculated using the log Kow input for HHCB of 5.3 (OECD TG 123) and the method of Nichols et al. (2013b). Test concentrations of HHCB were 5, 1, 0.2 and 0.05 umol.

Results: In vitro intrinsic clearance rate of the lowest concentration tested or 0.05 umol was 0.8 (mL/h/mg/protein). The fraction-unbound (ƒu) was measured and calculated following the method of Nichols et al. (2013b) and resulted in fu 0.2. The measured value was 0.0093 (other calculation methods for ƒu varied between 0.0242, 0.2002, 0.3158, 0.0404, Table 4 in the Laue et al paper). The clearance depended on the concentration tested, as presented in the table below; the highest clearance rate will be used which corresponds with the lowest test concentration that is farthest below the water solubility of HHCB.

Study: Laue et al. (2020) and *Laue et al. Suppl 2020	Test Concentration (uM)	Liver S9 concentration (mg/mL)	in vitro intrinsic clearance rate (mL/h/mg/protein)	BCF fu is calculated according to Nichols et al. 2013*	BCF fu=1
	5	1	0.12	6272	833
	1	1	0.44	3937	413
	0.2	1	0.78	2859	341
	0.05	1	0.8	2825	339

Discussion: From the in vitro clearance rate a BCF can be calculated using a fu 0.2 and a calculated k1, resulting in a value of 2825 according to Nichols et al. (2013b, referenced in Laue et al, 2020). The ƒu calculated is considered to be the mechanistically correct way because substances bind to protein based on their log Kow and this is what is presented here, though it seems that for fragrance ingredients ƒu is 1, presents better results when compared to in vivo BCF (Laue et al., 2020).

For this BCF assessment we only use the clearance rate of 0.8 to show the HHCB significantly metabolises in fish though all concentrations indicated metabolic clearance.

3. In vivo tests for aquatic bioaccumulation (OECD TG 305: Aquatic exposure)

There are three reliable bioconcentration studies carried out with HHCB according to OECD TG 305. One is selected as key study because it is a study report in which a steady state BCF was measured (van Dijk, RCC, 1996). There is another, which also has derived a kinetic BCF. It is presented on a poster but has sufficient details to be used for assessment (Butte and Ewald, 1999, with the inclusion of some methodological data). Both  will be presented in detail below, because for the first one a study report is available and the Butte and Ewald study, is a lipid normalised BCF.

3A) Van Dijk, RCC (1996)

Introduction: The key study selected is the one with Blue gill sunfish which is using the standard guideline and total 14C-HHCB radioactivity which results in a BCF not only on HHCB but also its metabolites (van Dijk, RCC, 1996, which was published by Balk and Ford, 1999). Beside HHCB also the metabolites found are assessed for their retention times and log Kow values.

Method: An OECD TG 305 flow through study was performed, with nominal aquatic concentrations 1 and 10 ug/l 14C-HHCB and using DMF as a solvent. Bluegill sunfish (Lepomis macrochirus) of 1.2-1.4 gram were used. Uptake time was 28 days. Sampling times were -1, 0, 3,7, 14 and 28 days using 6 fish each time. Depuration time was also 28 days and sampling times were 1, 3, 7, 10, 14, 21 and 28 days.

Both steady state BCF and kinetic BCF were calculated. The steady state BCF was based on actual Cfish/Cwater. The k2 was based on depuration phase. Thin Layer Chromatography (TLC) was used to characterise HHCB and its metabolites. Non radiolabelled HHCB was detected with UV 254 nm. HPLC was used to additionally to characterise HHCB and its metabolites. For fish edibles, non-edibles and whole fish concentrations were measured. In this summary only whole fish values are reported.

Results: Measured concentrations in water were 0.91 and 8.94 ug/l 14C-HHCB during the test using LSC, TLC and HPLC. HHCB HPLC retention times for HHCB was 22.55 and 23.25 minutes. A metabolite was found with Retention time: 2.30 and 3.35 minutes. Steady state concentrations were reached within 3-7 days. 14C-HHCB Concentration in fish was 1.49 and 14.26 mg/kg bw fish for the low (0.91 ug/l) and high concentration (8.84 ug/l), respectively. The elimination k2 was -0.215 and -0.261/day for the low and high dose respectively.

BCF based on total radioactivity were 1635 and 1624 for the low and high dose, respectively. BCF based on the parent compound were 1613 and 1584. The latter is used as BCF for HHCB.

Discussion: The BCFs reported were not lipid and growth corrected. Growth correction is not needed when growth is similar in both uptake and depuration phase (Gobas and Lee, 2019). In view of absence of lipid correction, the study receives Kl2. The value of 1584 is in line with the Butte and Ewald BCF study, which resulted in a BCF of 1660 including lipid correction further supporting the results of the ‘van Dijk’ study. In the latter study growth correction was not performed and not needed because fish were not fed during treatment.

3B) Butte and Ewald (1999)

Introduction: This key study is selected to further support the BCF value of van Dijk (RCC, 1996) study. It is performed with Zebrafish (Brachidanio rerio) using the OECD TG 305 guideline.

Method: HHCB was dissolved in MeOH at a concentration of 7.3 ug/l and the concentration was maintained under flow-through conditions (250 ml / min) in a 200-litre basin at 20°C. in tap water. The fish weighed approx. 0.25 g, were approx. 3 cm long, 3 - 6 months old and had a corpulence factor of approx. 1 g / cm3. They were kept under flow conditions.  Fish were sampled 20 times and 5 fish were used per sample (no replicates). They were not fed during the experiment and therefore growth of the fish will not have occurred.

In the uptake phase (14 days) water and fish were sampled to follow the kinetics of bioaccumulation. After the uptake phase the fish were transferred to clean water for 28 days and fish samples were taken to follow clearance. The fish samples were homogenized with a mixture of hexane and isopropanol and was extracted with toluene. Concentrations in fish were measured with GC-MS/MS. BCF was calculated using the two-compartment model.

Results: The steady state BCF was measured as 624 l/kg HHCB. The kinetic BCF is 750 (k1/k2= 560/0.75) The 100% lipid based BCF is 33200 l/kg. When the latter value is lipid corrected to 5% the BCF would be 1660 l/kg. The steady state was reached within 5 days and half-lives were shorter than 3 days.

Discussion: The steady state and kinetic BCFs are well in line with each other (624 and 750 l/kg. Using the lipid BCF and correction for 5% lipid the BCF is 1660 l/kg.  Growth correction is not needed because fish were not fed during treatment. The uptake rate, the time to reach steady state and the half-life time all agree well with the van Dijk (RCC, 1996) study despite the latter not being lipid corrected.

4. In vivo HHCB BMF using dietary exposure (OECD TG 305)

Introduction: HHCB was tested in a dietary BMF (Biomagnification Factor) study according to OECDTG 305 using Bluegill sunfish (Lepomis macrochirus).

Method: Radio-labelled HHCB and the reference substance HCB (Hexachlorobenzene) was submitted in feed at nominal concentrations of 500 ug/g food/day. The concentrations were measured using LSC. The weight of the fish at the start was 1.87 g. The test duration was 42 days: 14 and 28 uptake and depuration phase, respectively. BMF is calculated as the concentration in organism/concentration in food. The BMFk is food intake*assimilation factor/k2

Results: The actual HHCB concentration was 548 ug/g food/day. The fish concentration at the end of the uptake phase was calculated to 14.6 ug/g/fish including gut content. The food was administered as 0.02 ug/g fish/day, the assimilation factor was 0.1648 and the depuration rate constant k2 was -0.1468, resulting in a BMF of was 0.022. The lipid content in the fish was averaged to 4.82% and was used for lipid normalization to 5%, resulting in BMFnormalised is therefore 0.023. The lipid food/fish correction value was 0.27, therefore the BMFlipid-corrected value resulted in 0.082. The BMFss was reached after 20 days (extrapolated value) and the DT50 is 5 days. Growth correction was not applied because the growth rate during uptake phase was not different from the depuration phase in which case growth correction does not need to be considered (Gobas and Lee, 2019). All validity criteria according to the OECD 305 Dietary Exposure method were met.

The reference substance HCB was dosed 41ug/g food/day (nominal 40 ug/g good/day). The food was administered 0.02 ug/kg fish/day, the assimilation factor was 0.1036 and the depuration constant k2 was -0.0125, resulting in BMF of 0.166. The BMFnormalised to 5% is almost the same because the lipid content of the fish was 4.99%. The lipid food/fish correction was 0.27 and therefore the BMFlipid-corrected value is 0.615. The BMFss was reached after 241 days (extrapolated value) and the DT50 was 90 days.

Discussion: The calculated kinetic lipid-corrected value BMFK of 0.082 for HHCB is below 1 and therefore not biomagnifying from food to fish (i.e. the concentration of HHCB in fish is less than that of its diet). The kinetic lipid-corrected BMF of 0.082 is also below the threshold of 0.1 and is considered indicative for a non-bioaccumulative substance with an extrapolated aqueous BCF of 1530 (ECHA guidance on PBT 12, page 73, Inoui et al., 2012.)

Conclusion: The kinetic lipid-corrected BMF is 0.082 and the extrapolated BMF to aqueous BCF is 1530 l/kg.

5. BCF information on HHCB-lactone and HHCB-hydroxylated acid

In the aqueous BCF study (van Dijk, RCC, 1996), a highly water-soluble metabolite of HHCB was characterized but not structurally elucidated (see BCF, aqueous phase). In comparing the HPLC retention time of the metabolite to that of the parent the retention time was 10x lower for the metabolite. Due to its high polarity, as indicated by HPLC, the metabolite is not considered to be HHCB-lactone (see BCF below) since Lactone is considered non-polar with a calculated log Kow of 4.7 (Federle et al, 2002 and Itrich and Federle, 2003, see biodegradation section). The key metabolite found in this BCF study is likely the HHCB-hydroxylated acid, which has a calculated log Kow of 0.6 (Fig. 1) and has negligible bioaccumulation potential. The data indicates that HHCB will be transformed to a HHCB-hydroxylated acid derivative that has a much lower BCF than HHCB or its Lactone.  

Supporting information on the bioaccumulation potential of HHCB-lactone in the juvenile Bluegill fish can be found in the Pilot BMF 305-Dietary exposure study (page 147 of the report). Fish weight was not presented but estimated to be similar to the main study (2.4 gram at day 14 of uptake). The depuration of this metabolite was followed on day 1, 3 and 7 of depuration. The calculated depuration rate constant (k2) was 0.2925 (Fig. 2) and the estimated depuration half-life was 2.4 days. To extrapolate a BCF from the kinetic BMF 305 pilot study, several equations can be used to estimate k1 one of them being the Sijm equation, also mentioned in OECD TG 305. From the equation below, k1 is calculated to be 393:  

k1= 520*(fish weight^-0.32))

Using the estimated k1 and the k2 derived from the BMF pilot study, the kinetic BCF is calculated to be 1293 where BCF = k1/k2.

Day	Log transformed average fish values in ug/kg
Depuration t=1	2.577492
Depuration t=3	1.944483
Depuration t=7	0.748

 

 

Fig. 2     Depuration of HHCB-lactone as measured in Pilot study of HHCB BMF (OECD TG 305-Dietary Exposure)

Conclusion on BCF

The final HHCB BCF value selected will be the BCF via aqueous exposure of van Dijk (RCC, 1996) of 1584 l/kg. This value is supported with the BCF of 1660 l/kg of Butte and Ewald (1999) which is lipid corrected and fish not being fed, growth correction is not applicable. The in vitro data support the metabolization in fish (Laue et al., 2020 and Suppl) and the BMF study support the absence of biomagnification and supports the BCF value via aqueous exposure: conversion of the BMF results in a BCF of 1530. Though the BCF value of van Dijk (RCC, 1996) is not lipid normalised, this is not needed because in the BCF study of Butte and Ewald (1998) shows a very similar BCF of 1660, which is lipid corrected. Growth correction is also not needed based on a paper of Gobas and Lee (2019).  

The BCF-Lactone is based on depuration in the HHCB dietary OECD 305 and is 1293 l/kg.

The BCF-hydroxylated carboxylic acid is negligible based on the log Kow of 0.6 based on the dissociated acid at environmental pH.

References:

Butte, W., and Ewald, F., Butte W, F Ewald (1999). Kinetics of accumulation and clearance of the polycyclic musk compounds Galaxolide (HHCB) and Tonalide (AHTN); Poster University Oldenburg, Germany (attached to study record).

ECHA guidance on Guidance on Information Requirements and Chemical Safety Assessment Chapter R.7c: Endpoint specific guidance, 2017, https://www.echa.europa.eu/documents/10162/13632/information_requirements_r7c_en.pdf

EU RAR  (2008), European Union Risk Assessment Report 1,3,4,6,7,8-HEXAHYDRO-4,6,6,7,8,8-HEXAMETHYLCYCLOPENTA-γ-2- BENZOPYRAN (1,3,4,6,7,8-HEXAHYDRO-4,6,6,7,8,8-HEXAMETHYLIN-DENO[5,6- C]PYRAN - HHCB) CAS No: 1222-05-5 EINECS No: 214-946-9 RISK ASSESSMENT https://echa.europa.eu/documents/10162/947def3b-bbbf-473b-bc19-3bda7a8da910

Gobas., F.A.P.C. and Lee, Y-S, 2019, Growth correcting the bioaccumulation factor in bioaccumulation assessments, Env. Toxicol. Chem, 38, 2065-2072.

Federle, T.W., Itrich, N.R., Lee, D.M., Langworthy, d., 2002, Recent Advances in the

Environmental Fate of Fragrance Ingredients, Setac presentation.

IFRA-ETF, 2020, Report on Bioaccumulation and Tropic Magnification Potential in the Aquatic Environment of 1,3,4,6,7,8-Hexahydro-4,6,6,7,8,8-hexamethylcyclopenta[g]-2-benzopyran (HHCB)

CAS RN 1222–05–5 (paper attached to Endpoint summary)

Itrich, N.W. and Federle, T.W., 2003, Biodegradation of Galaxolide in Activated sludge, P&Greport, E97-006, Central Products Safety Division/Product Safety and Regulatory Affairs, Environmental Science Department.

Inoue, Y, Hashizume, N. Tomohiko, Y., Hidekazu, Y, Murakami, H., Suzuki, Y., Koga, Y., , Ryoko Takeshige, R., Kikushima, E., Yakata, N., Otsuka, M., 2012, Comparison of bioconcentration and biomagnification factors for poorly water-soluble chemicals using common carp (Cyprinus carpio L.), Arch Environ Contam Toxicol, 63, 241-248.

Laue, H., Hostettler, H., Badertscher, R.P., Jenner, K.J., Sanders, G., Arnot, J.A., & Natsch, A., (2020 and Suppl). Examining Uncertainty in In Vitro−In Vivo Extrapolation Applied in Fish Bioconcentration Models. Environmental Science & Technology, 54 (15), 9483-9494; paper attached in the study record.

Nichols, J.W., Hoffman, A.D., ter Laak, T.L., Fitzsimmons, P.N., 2013a, Hepatic Clearance of 6 Polycyclic Aromatic Hydrocarbons by Isolated Perfused Trout Livers: Prediction From In Vitro Clearance by Liver S9 Fractions, Toxicological Sciences, Volume 136, 359-372, https://doi.org/10.1093/toxsci/kft219

Nichols, J. W.; Huggett, D. B.; Arnot, J. A.; Fitzsimmons, P. N.; Cowan-Ellsberry, C. E., 2013b, Towards improved models for predicting bioconcentration of well-metabolized compounds by rainbow trout using measured rates of in vitro intrinsic clearance. Environ. Toxicol.Chem., 32, 1611−1622.