Registration Dossier

Environmental fate & pathways

Bioaccumulation: aquatic / sediment

Currently viewing:

Administrative data

Link to relevant study record(s)

Referenceopen allclose all

Endpoint:
bioaccumulation: aquatic / sediment
Type of information:
experimental study
Adequacy of study:
weight of evidence
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Acceptable, well documented publication which meets basic scientific principles.
Qualifier:
no guideline followed
Principles of method if other than guideline:
The main goal of this study is to describe the effect of molecular size on bioconcentration by accounting for conformational flexibility of molecules.
GLP compliance:
no
Radiolabelling:
not specified
Vehicle:
not specified
Test organisms (species):
other: fish
Route of exposure:
other: review article
Test type:
other: review article
Water / sediment media type:
not specified
Remarks on result:
not measured/tested

Results

The complexity of the effect of molecular geometry and flexibility on membrane permeation and subsequently on bioconcentration can be illustrated with n-pentadecane. The length of the n-pentadecane molecule, known also as the maximum diameter, is Dmax= 2.14 nm. The other two related dimensions, the effective and minimum cross-sections for this conformer, are Deff= 0.499 nm and Dmin= 0.492 nm, respectively. Due to the high hydrophobicity of n-pentadecane [log(Kow) = 7.71] and small effective cross-section of its lowest energy conformer, which is far below the critical 0.95 nm, one anticipates high bioconcentration for this chemical. The bioconcentration is in the range of 3.2 to 4.3 log units. Two other energetically reasonable conformers of n-pentadecane have got a heat formation of H°= –71.8 kcal/mol and H°= –68.5 kcal/mol, respectively. Based on the effective cross-section of the third conformer, Deff= 1.15 nm, a loss of membrane permeation may be expected, resulting in low bioconcentration. The experimentally measured BCF of n-pentadecane in carp, expressed as log (BCF), is in the range of 1.12 to 1.29. Two conclusions could be drawn from this simple example. First, molecular cross-sectional diameters appear to be strongly dependent on molecular flexibility Secondly, it is not clear which of the molecular dimensions controls the membrane permeation of chemicals.

The complex effect of molecular geometry and flexibility on the chemicals’ permeability is confirmed by superposition of the molecular effective cross-sectional diameters of the studied lipophilic chemicals with their experimentally measured BCFs.

The absence of any correlation indicates that this geometric characteristic is not related to the variation of BCF for highly hydrophobic chemicals. The hypothesis that the effective diameter controls permeability of chemicals assumes a strict spatial orientation of the molecules toward the cell membrane surface in a way that the molecular projection over the membrane does not exceed a certain threshold (anticipated to be around 0.95 nm). The appropriate orientation, however, is pre-vented by entropy (i.e., by chaotic movement of the molecules), which can explain the insufficiency of the effective cross-section of molecules to explain their permeability.

Two interesting features were displayed firstly, there is a well-outlined tendency of decreasing bioconcentration with the increase of the maximum cross-sectional areas of molecular conformers. The higher the molecular length (i.e., its Dmax value), the smaller are the chances of the molecule reaching the cell membrane at an appropriate angle. Secondly, most chemicals with Dmax under ~1.5 nm achieve high log (BCF)—in the range of 3 to 6, while the chemicals with Dmax greater than this threshold accumulate up to 3.3 units at most. The existence of such a transition point can be explained by a change in the mechanism of uptake of chemicals from passive diffusion through the phospholipid bilayer of the membrane to the more conservative passing of the membrane by the mechanism of exocytosis and endocytosis. Interestingly, the critical value of 1.5 nm for the threshold is comparable with the cell membrane architecture. The threshold of maximum diameter is comparable with the half thickness of leaflet constituting the lipid bilayer of the cell membrane.

From the present results, one could conclude that diffusion through the cell membrane is limited to molecules having a length not exceeding the threshold of about 1.5 nm. The latter could be assumed as the maximum tolerance of the cell membrane.

Conclusion

Analysis of the BCF data for narcotics with log (Kow) greater than 5.5 revealed that the maximum cross-sectional diameter can be used to explain the significant scatter around the maximum of the log (BCF)/log (Kow) curve. The chaotic collision of molecules with the cell membrane surface at different angles could explain the significance of this geometric characteristic, instead of generally accepted effective diameter. The drop in bioconcentration of chemicals at a maximum cross-sectional diameter of about 1.5 nm is an indication of a switch of the mechanism of uptake of chemicals into cells above this threshold. The value of this transition point can be used as an additional parameter of hydrophobicity for regression modeling of the BCF variation. Conformational flexibility tends to further increase the significance of entropy to cell permeability, which leads to additional decreases of BCF. The effect of this structural characteristic, however, needs to be further evaluated in order to be used for quantifying the bioconcentration of chemicals.

Endpoint:
bioaccumulation: aquatic / sediment
Type of information:
experimental study
Adequacy of study:
weight of evidence
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Literature study
Qualifier:
no guideline followed
Principles of method if other than guideline:
Literature study addressing the effects of molecular size and lipid solubility on the bioaccumulation potential of environmental contaminants.
GLP compliance:
no
Test organisms (species):
other: Aquatic organisms
Route of exposure:
other: not applicable, review
Test type:
other: not applicable, review
Water / sediment media type:
not specified
Remarks on result:
not measured/tested

BCF measured values plotted against log Kow values

- Estimating bioconcentration factors (BCF) from octanol/water partition coefficients (log KOW) is well established and essentially valid for neutral organics of intermediate lipophilicity (0 < log Kow < 6).

- On the other hand, chemicals with log Kow > 6 often have measured BCFs lower than calculated from linear QSARs. Apparently, BCFs no longer increase in correspondence with log Kow. A maximum range in log BCF of approximately 6–7 for compounds with log Kow 6–8 is observed, followed by a plateau or a gradual decrease with further increase in log Kow. The maximum BCF

associated with a given lipophilicity can be described by a bilinear worst-case function:

log BCF = 0.99 log Kow - 1.47 log (4.97 x 10-8 KOW + 1) + 0.0135

The bilinear curve resumes a linearly increasing part between log Kow 0 and 6, where the empirically postulated coincidence of log Kow and log BCF is reflected by a near-unity slope (0.99) for the 1st-order log Kow term and the intercept of about 0. Maximum log BCF values of approximately 7 are obtained for compounds with log Kow between 7 and 8. Compounds that are more lipophilic are observed to be less accumulating, which corresponds to the negative slope derived for the second log Kow term of the bilinear function.

- The apparent loss in linear relationships for superlipophilic compounds has been attributed – in part – to experimental artefacts. Theoretical considerations substantiate curvilinear relationships for chemicals with log Kow > 6: - Aqueous phase diffusion control of water to lipid transfer - Differences in phase (solvent) properties of natural lipids and octanol - Influence of steric conformations - Differences in thermodynamic properties of partitioning, e.g. enthalpy changes To test established QSARs, BCF data were provided by Umweltbundesamt for 31 plant protection agents and for 18 new chemicals. These chemicals were selected by two criteria: molecular weight > 300 g/mol and bioaccumulation data available. Since this- 34 -data set includes no compounds with log BCF > 4 (range in BCF data: 1.5 to 14600), additional data for compounds with very high BCF were taken from the literature. Most BCF data from Umweltbundesamt were qualified as ‘valid data'. However, for two compounds Umweltbundesamt classified the measurements as ‘invalid’. Critical inspection of the available BCF data revealed major deficiencies in data quality of several superlipophilic compounds. For nine compounds, the accumulation experiments have been conducted at concentrations above the water solubility of the test compounds. Consequently, the resulting BCF values are too low artefacts due to invalid experiments. These data were excluded from further analyses. The remaining data set from Umweltbundesamt is considered not sufficiently representative, because it includes no compounds with log BCF > 4. It is too limited with regard to activity domain as to provide new insights to relationships between BCF and log Kow. New insights to relationships between BCF and log Kow could not be provided. The currently available database is insufficient to conclusively substantiate the effects of molecular size and lipid solubility on the bioaccumulation potential of environmental contaminants.

Endpoint:
bioaccumulation: aquatic / sediment
Type of information:
experimental study
Adequacy of study:
weight of evidence
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Acceptable, well documented publication which meets basic scientific principles
Qualifier:
no guideline followed
Principles of method if other than guideline:
In this report, the application of selected physico-chemical and molecular attributes according to Lipinski's 'Rule of 5'* in the estimation of fish bioconcentration values is investigated.

*Lipinski et al. (1997). Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliv. Rev. 23, pp. 961-965).
GLP compliance:
no
Remarks:
not applicable
Test organisms (species):
other: Aquatic organisms
Route of exposure:
other: Not applicable, review
Test type:
other: Not applicable, review
Water / sediment media type:
not specified
Type:
BCF
Value:
< 2 000
Calculation basis:
other: Review literature dataset
Remarks on result:
other: Substances log Kow < 3 and log Kow > 10

Molecular weight or size-related cut-off criteria

- No robust evidence was found for cut-offs for bioconcentration related to molecular size. A criterion in molecular weight of 650 mg/mol seems to be more appropriate that the 500 g/mol stated by Lipinski et al. (1997) as it was only exceeded by a few non-B compounds (BCF < 2000) in the dataset. Cut-offs in molecular weight are pragmatic due to readily available metrics, but they lack a mechanistic rationale. As often, the problem is multivariate in nature and there exists a multitude of parameters than jointly and simultaneously influence each other. The correlation between molecular weight and size is rather weak (r square= 0.36 -0.45). Other properties related to molecular weight, such as solubilities in various media, absorption kinetics and bioavailability are also influential.

Lipophility related cut-off (log Kow)

The results from the analysis of log BCF values plotted against log Kow values confirm the implication of non-linear log Kow-based QSARs that chemicals with either low (log Kow < 3) of very high lipophilicity (logKow > 10) may have rather low BCF (non B compounds). The maximum BCF associated with a given lipophilicity has been described by a bilinear worst-case function. The curve was constructed (without regression and thus without calibration statistics) to represent the upper border of the data distribution of log BCF versus log Kow:

logBCFmax = 0.99 log Kow – 1.47 log (4.97x10-08 Kow + 1) + 0.0135    (1)

Chemicals with log Kow > 6 often have measured BCFs lower than calculated from QSARs. Apparently, these BCFs no longer increase in correspondence with log Kow, but plateau or gradually decrease with further increase in log Kow. Arnot and Gobas identified sorption as the main reason that accumulation decreases with increasing log Kow for highly lipophilic chemicals: if accumulation were quantified as the ratio of the concentrations in the organisms and the freely dissolved chemical concentration in water the BCF of high Kow chemicals would rise to 107and the level off with a further increase Kow. The rationale is that for more hydrophobic chemicals (log Kow > 6) membrane permeation rates become increasingly controlled and ultimately dominated by aqueous boundary layer transport rather than phospholipid bilayer permeation.

Upward deviations relative to equation (1) only concern hydrophilic molecules (log Kow < 2), but their BCFs remain below 100. In the log Kow range between 3 and 10, many BCF values are lower than calculated by the QSAR, but the higher than the B (BCF > 2000) of vB (BCF > 5000) criteria. Above log Kow 10, none of the few available data in the test and validation datasets, exceeds the B criterion. Downward scatter relative to Equation (1) over several orders of magnitude increases for more lipophilic compounds, and many be attributed to mitigating factors such as ionization, degradation or metabolism of the test substances as well as to major errors particularly in very high log Kow values.

In the confirmation dataset, one bioaccumulative compound has a log Kow slightly above 10, octabromiddiphenyl ether, and two compounds bioaccumulating by different mechanisms have a log Kow < 3, methyl mercury and tetraethyl lead. The consequence of these observations is that in addition to polybrominated compounds, also alkylated heavy metals are excluded from the applicability domain of the classification scheme.

Description of key information

The potential for bioaccumulation of Octadecanoic acid, ester with 2,2'-[oxybis(methylene)]bis[2- (hydroxymethyl)-1,3-propanediol] is assumed to be low based on all available data.

Key value for chemical safety assessment

Additional information

Experimental bioaccumulation data are not available for Octadecanoic acid, ester with 2,2'-[oxybis(methylene)]bis[2-(hydroxymethyl)-1,3-propanediol] (CAS 70969-57-2). The high log Kow (> 10), as an intrinsic chemical property of the substance, indicates a potential for bioaccumulation. However, the information gathered on environmental behaviour and metabolism, in combination with QSAR-estimated BCF values, provide enough evidence (in accordance to the Regulation (EC) No 1907/2006, Annex XI General rules for adaptation of the standard testing regime set out in Annexes VII to X, 1.2), to cover the data requirements of Regulation (EC) No 1907/2006, Annex IX and to state that the substance is likely to show negligible bioaccumulation potential.

Environmental behaviour

Due to the high adsorption potential an effective removal of the substance in conventional sewage treatment plants (STPs) by sorption to biomass is likely. The low water solubility (< 1.7E-4 mg/L) and high estimated log Kow (>10) indicate that the substance is highly lipophilic. If released into the aquatic environment, the substance will extensively adsorb to organic matter. Thus, the bioavailability in the water column will be rapidly reduced. The relevant uptake route in aquatic organisms is expected to be predominantly by ingestion of particle bound substance. 

Uptake/Absorption

If the substance is taken up by ingestion, absorption is expected to be low based on the molecular weight, size and structural complexity of the substance. For these large and complex structures a high degree of conformational flexibility is assumed. Dimitrov et al. (2002) revealed a tendency of decreasing log BCF in combination with an increase in conformational flexibility of molecules. The authors assume that this effect is related to an enhancement of the entropy factor on membrane permeability of chemicals. Thus, there is a high probability that the substance exists in a conformation which does not allow permeation of a biological membrane. Furthermore, the substance has a high molecular weight 1853.05 g/mol. Thus, the substance is unlikely that it is readily absorbed, due to the steric hindrance of crossing biological membranes. Following the ‘rule of 5’ (Lipinski et al., 2001), developed to identify drug candidates with poor oral absorption based on criteria regarding partitioning (log Kow > 5) and molecular weight (> 500 g/mol), the substance is considered to be poorly absorbed after oral uptake (also see Hsieh & Perkins, 1976).

This interaction between lipophilicity, bioavailability and membrane permeability is considered to be the main reason why the relationship between the bioaccumulation potential of a substance and its hydrophobicity is commonly described by a relatively steep Gaussian curve with the bioaccumulation peak approximately at log Kow of 6-7 (e.g., see Dimitrov et al., 2002; Nendza & Müller, 2007; Arnot and Gobas 2003). Substances with log Kow values above 10, which has been calculated for the test substance, are considered to have a low bioaccumulation potential (e.g., Nendza & Müller, 2007; 2010). Furthermore, for those substances with a log Kow value > 10 it is unlikely that they reach thepass level of being bioaccumulative according to OECD criteria for the PBT assessment (BCF > 2000; ECHA, 2017).

Conclusion

Octadecanoic acid, ester with 2,2'-[oxybis(methylene)]bis[2-(hydroxymethyl)-1,3-propanediol] (CAS 70969-57-2) is characterized by a low water solubility (< 1.7E-4 mg/L), high log Kow (> 10) and high molecular weight (1853.05 g/mol). Based on the physico/chemical properties such as low water solubility and high potential for adsorption a reduced availability in water is expected. The high molecular weight of the substance significantly reduces the absorption due to steric hindrance to cross biological membranes. It can be concluded that the bioaccumulation potential of Octadecanoic acid, ester with 2,2'-[oxybis(methylene)]bis[2- (hydroxymethyl)-1,3-propanediol] is negligible.

Taking all these information into account, it can be concluded that bioaccumulation of Octadecanoic acid, ester with 2,2'-[oxybis(methylene)]bis[2-(hydroxymethyl)-1,3-propanediol] unlikely to occur.