Essential oil of Spearmint obtained from the aerial part of Mentha spicata and/or Mentha cardiaca (Lamiaceae) obtained by distillation

Adsorption / desorption

Link to relevant study record(s)

Reference

Endpoint:

adsorption / desorption, other

Remarks:

other: QSAR

Type of information:

(Q)SAR

Adequacy of study:

key study

Study period:

Not relevant.

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

results derived from a valid (Q)SAR model and falling into its applicability domain, with adequate and reliable documentation / justification

Remarks:

Values for individual constituents of this natural complex substance (NCS) were calculated using a validated QSAR. All constituents fall within the applicability domain of the QSAR. Documents and references are provided.

Justification for type of information:

1. Substance
This section is aimed at defining the substance for which the (Q)SAR prediction is made.
1.1 CAS number: N/A. Related CAS numbers: 91770-24-0 (Mentha spicata, ext.) and 84696-51-5 (Spearmint, ext.). Common name: Spearmint oil. Spearmint oil is an UVCB/NCS consisting of the following constituents:
(-)-β-Bourbonene (CAS 5208-59-3) (EC -) (C15H24)
1,8-Cineole (CAS 470-82-6) (EC 207-431-5) (C10H18O)
3-Octanol (CAS 589-98-0) (EC 209-667-4) (C8H18O)
Germacrene D (CAS 37839-63-7) (EC -) (C15H24)
L-Carvone (CAS 6485-40-1) (EC 229-352-5) (C10H14O)
Linalool (CAS 78-70-6) (EC 201-134-4) (C10H18O)
L-Limonene (CAS 5989-54-8) (EC 227-815-6) (C10H16)
para-Cymene (CAS 99-87-6) (EC 202-796-7) (C10H14)
Sabinene hydrate (CAS 546-79-2) (EC 208-911-7) (C10H18O)
Terpinen-4-ol (CAS 562-74-3) (EC 209-235-5) (C10H18O)
trans-Dihydrocarvone (CAS 5524-05-0) (EC 226-872-4) (C10H16O)
α-Pinene (CAS 80-56-8) (EC 201-291-9) (C10H16)
β-Myrcene (CAS 123-35-3) (EC 204-622-5) (C10H16)
β-Pinene (CAS 127-91-3) (EC 204-872-5) (C10H16)
γ-Terpinene (CAS 99-85-4) (EC 202-794-6) (C10H16)

1.2 EC number: N/A. Related EC numbers: 294-809-8 (Mentha spicata, ext.) and 283-656-2 (Spearmint, ext.). For constituents, see 1.1.
1.3 Chemical name: Essential oil of Spearmint obtained from the aerial part of Mentha
spicata and/or Mentha cardiaca (Lamiaceae) obtained by distillation, Common name
Spearmint oil. The NCS/UVCB consists of the constituents listed in 1.1.
1.4 Structural formula: See 1.1
1.5 Structure codes:
a. SMILES:
(-)-β-Bourbonene:
SMILES: C=C1C2C(C3(C2C(C(C)C)CC3)C)CC1
InCHI: 1S/C15H24/c1-9(2)11-7-8-15(4)12-6-5-10(3)13(12)14(11)15/h9,11-14H,3,5-8H2,1-2,4H3/t11-,12+,13-,14+,15-/m0/s1

1,8-Cineole:
SMILES: O(C(CCC1C2)(C2)C)C1(C)C
InCHI: 1S/C10H18O/c1-9(2)8-4-6-10(3,11-9)7-5-8/h8H,4-7H2,1-3H3

3-Octanol:
SMILES: OC(CCCCC)CC
InCHI: 1/C8H18O/c1-3-5-6-7-8(9)4-2/h8-9H,3-7H2,1-2H3

Germacrene D
SMILES: CC(C)C1CCC(C)=CCCC(=C)C=C1
InCHI: 1S/C15H24/c1-12(2)15-10-8-13(3)6-5-7-14(4)9-11-15/h7-8,10,12,15H,3,5-6,9,11H2,1-2,4H3/b10-8+,14-7-

L-Carvone:
SMILES: O=C(C(=CCC1C(=C)C)C)C1
InCHI: 1S/C10H14O/c1-7(2)9-5-4-8(3)10(11)6-9/h4,9H,1,5-6H2,2-3H3

Linalool:
SMILES: OC(C=C)(CCC=C(C)C)C
InCHI: 1S/C10H18O/c1-5-10(4,11)8-6-7-9(2)3/h5,7,11H,1,6,8H2,2-4H3

L-Limonene
SMILES: C(=CCC(C(=C)C)C1)(C1)C
InCHI: 1S/C10H16/c1-8(2)10-6-4-9(3)5-7-10/h4,10H,1,5-7H2,2-3H3/t10-/m1/s1

para-Cymene
SMILES: c(ccc(c1)C)(c1)C(C)C
InCHI: 1S/C10H14/c1-8(2)10-6-4-9(3)5-7-10/h4-8H,1-3H3

Sabinene hydrate
SMILES: OC(C(C1(C2)C(C)C)C1)(C2)C
InCHI: 1S/C10H18O/c1-7(2)10-5-4-9(3,11)8(10)6-10/h7-8,11H,4-6H2,1-3H3

Terpinen-4-ol
SMILES: OC(CCC(=C1)C)(C1)C(C)C
InCHI: 1/C10H18O/c1-8(2)10(11)6-4-9(3)5-7-10/h4,8,11H,5-7H2,1-3H3

trans-Dihydrocarvone
SMILES: O=C1CC(C(=C)C)CCC1C
InCHI: 1/C10H16O/c1-7(2)9-5-4-8(3)10(11)6-9/h8-9H,1,4-6H2,2-3H3/t8-,9-/m1/s1

α-Pinene
SMILES: C(C(CC1C2)C1(C)C)(=C2)C
InCHI: 1S/C10H16/c1-7-4-5-8-6-9(7)10(8,2)3/h4,8-9H,5-6H2,1-3H3

β-Myrcene
SMILES: C(C=C)(=C)CCC=C(C)C
InCHI: 1S/C10H16/c1-5-10(4)8-6-7-9(2)3/h5,7H,1,4,6,8H2,2-3H3

β-Pinene
SMILES: C(C(CC1C2)C1(C)C)(C2)=C
InCHI: 1S/C10H16/c1-7-4-5-8-6-9(7)10(8,2)3/h8-9H,1,4-6H2,2-3H3

γ-Terpinene
SMILES: C=1(C(C)C)CC=C(C)CC1
InCHI: 1S/C10H16/c1-8(2)10-6-4-9(3)5-7-10/h4,7-8H,5-6H2,1-3H3

b. InChI: See 1.5a, not used for prediction
c. Other structural representation: Not available
d. Stereochemical features: not relevant for this endpoint
2. General information
General information about the compilation of the current QPRF is provided in this section.
2.1 Date of QPRF: February 11, 2016
2.2 QPRF author and contact details:
A.C. Belfroid (Angelique), P. Englebienne (Pablo)
HaskoningDHV Nederland B.V.
P.O. Box 151
6500 AD Nijmegen, The Netherlands
[e-mail]; [e-mail]
http://www.royalhaskoningdhv.com
3. Prediction
3.1 Endpoint (OECD Principle 1)
a. Endpoint: Soil adsorption Coefficient, also referred to as log Koc
b. Dependent variable: Log Koc (dimensionless)
3.2 Algorithm (OECD Principle 2)
a. Model or submodel name: KOCWIN
b. Model version: Version 2.00
c. Reference to QMRF: QSAR model for Soil adsorption Coefficient log Koc drafted 5.10.2015 by A.C. Belfroid (Angelique) and P. Englebienne (Pablo)
d. Predicted value (model result):
Log Koc:
(-)-β-Bourbonene (CAS 5208-59-3): 4.28
1,8-Cineole (CAS 470-82-6): 2.34
3-Octanol (CAS 589-98-0): 1.53
Germacrene D (CAS 37839-63-7): 4.3
L-Carvone (CAS 6485-40-1): 2.13
Linalool (CAS 78-70-6): 1.88
L-Limonene (CAS 5989-54-8): 3.05
para-Cymene (CAS 99-87-6): 3.03
Sabinene hydrate (CAS 546-79-2): 1.88
Terpinen-4-ol (CAS 562-74-3): 1.91
trans-Dihydrocarvone (CAS 5524-05-0): 2.13
α-Pinene (CAS 80-56-8): 3.01
β-Myrcene (CAS 123-35-3): 3.03
β-Pinene (CAS 127-91-3): 3.01
γ-Terpinene (CAS 99-85-4): 3.05

e. Predicted value (comments):No comments
f. Input for prediction: KOCWIN uses the molecular connectivity index (MCI) and series of statistically-derived fragment contribution factors for polar compounds.
g. Descriptor values:
The non-polar Koc was tuned with a training set of 69 chemicals, while coefficients for individual fragments in KOCWIN were derived by multiple regression of reliably measured log Koc values on 447 chemicals. In total 36 different types of fragments exist (see Appendix).
3.3 Applicability domain (OECD principle 3)
a. Domains: Discuss whether the query chemical falls in the applicability domain of the model as defined in the corresponding QMRF (section 5 of QMRF, Defining the applicability domain - OECD Principle 3). If additional software/methods were used to assess the applicability domain then they should also be documented in this section. Include a discussion about:
i. descriptor domain: The molecular weights of the substances for which the log Koc is predicted fall within the applicability domain (range of molecular weight for the training set substances: 32.04-665.02).
ii. structural fragment domain: All predicted substances are chemically similar to the training and validation set substances. All functional groups that are present in the substances for which predictions are made are covered by the model.
iii. mechanism domain: Not relevant
iv. metabolic domain: Not relevant
c. b. Structural analogues: A set containing 158 chemicals was used for validation of the QSAR algorithm, yielding a correlation coefficient (r2) of 0.85, and a maximum absolute residual of 1.74. For only one constituent of Spearmint oil was there experimental Koc data determined:

1,8-Cineole (CAS 470-82-6): logKoc (calc): 2.34 and logKoc (exp): 2.33

c. Considerations on structural analogues: The residual in the estimation of the constituent for which there is experimental data available (-0.01) is within what was observed for chemicals in the validation set.
3.4 The uncertainty of the prediction (OECD principle 4) The training and validation sets are not from one lab, but a collection. The sets are nevertheless considered robust as they are large: 516 and 158 chemicals respectively. For more information on the quality of these datasets we refer to the Help file of KOCWIN 2.00.
3.5 The chemical and biological mechanisms according to the model underpinning
the predicted result (OECD principle 5).
The model was developed by statistical approach. No mechanistic basis for this physico-chemical property was set a priori, but a mechanistic interpretation of molecular descriptors was provided a posteriori.

4. Adequacy (Optional)
4.1 Regulatory purpose: The predictions are used to gather the required data for filling the “Adsorption/Desorption in soil” data gap in the REACH dossier for a UVCB substance.
4.2 Approach for regulatory interpretation of the model result: REACH accepts and encourages the use of QSARs to fill data gaps. The requirements for REACH are met as:
- a robust study summary and endpoint summary are included in the dossier;
- a QMRF for the QSAR model is included in the dossier;
- a QPRF for the predictions is included in the dossier.
4.3 Outcome: The model results (log Koc) is compatible with the requirements of REACH for the Adsorption/Desorption in soil endpoint.
4.4 Conclusion: As all criteria for the application of QSARs under REACH are met, the predictions for the soil adsorption/desorption endpoint are considered to be adequate and fit-for-purpose.

Qualifier:

according to guideline

Guideline:

other: REACH Guidance on QSARs R.6

Deviations:

not applicable

Principles of method if other than guideline:

Principle of the method if other than guideline >> NCSs, consisting of a number of constituents, do not have one single sorption coefficient (Koc). The log Koc can be based on the range of the log Koc values from the calculated or measured values of the individual constituents. Calculated and measured data on the constituents are obtained from KOCWIN v2.00. The relevance and reliability of the used QSAR for these constituents are discussed in the attached QMRF and QPRF.

GLP compliance:

no

Remarks:

Not relevant.

Type of method:

other: Constituents' approach with QSAR.

Key result

Type:

log Koc

Value:

>= 1.53 - <= 4.3 dimensionless

Remarks on result:

other: Range of log Koc values of the individual constituents.

Substance	CAS	log Koc
L-Carvone	6485-40-1	2.13
L-Limonene	5989-54-8	3.05
β-Myrcene	123-35-3	3.03
Terpinen-4-ol	562-74-3	1.91
1,8-Cineole	470-82-6	2.34
(-)-β-Bourbonene	5208-59-3	4.28
trans-Dihydrocarvone	5524-05-0	2.13
germacrene D	23986-74-5	4.3
sabinene hydrate	546-79-2	1.88
3-Octanol	589-98-0	1.53
α-Pinene	80-56-8	3.01
γ-Terpinene	99-85-4	3.05
β-Pinene	127-91-3	3.01
Linalool	78-70-6	1.88
para-Cymene	99-87-6	3.03

KOCWIN v.2.0 model details

 

Reference to the type of model used

KOCWIN estimates the soil adsorption log Koc. In KOCWIN (version 2) estimates of Koc can be made with two separate estimation methodologies: (1) estimation using first-order Molecular Connectivity Index (MCI) or (2) estimation using log Kow (octanol-water partition coefficient). As the MCI approach was shown to have a better fit than the log Kow approach, this method was used.

 

The first-order molecular connectivity index (MCI) was successfully used to predict soil sorption coefficients (Koc) for nonpolar organics, but extension of the model to polar compounds has been problematic.  Therefore,two separate regressions were performed in the MCI methodology.  The first regression related log Koc of non-polar compounds (n = 69) to the first-order MCI. The second regression included 447 compounds having correction factors for specific chemical classes or structural fragments.  The regression coefficients were derived via multiple linear regression of the correction descriptors to the residual error of the prediction from the non-polar equation. In total there are36 types of fragments having correction factors.

 

Description of the applicability domain

The applicability domain is based on the maximum number of instances that a fragment can be used in a chemical (based on the training and validation sets, summarized in Appendix D in the help file of KOCWIN) and on molecular weight. The minimum and maximum values for molecular weight are the following:

Training Set Molecular Weights:

Minimum MW: 32.04

Maximum MW: 665.02

Average MW: 224.4

 

Currently there is no universally accepted definition of model domain. However, users may wish to consider the possibility that log Koc estimates are less accurate for compounds outside the MW range of the training set compounds, and/or that have more instances of a given fragment than the maximum for all training set compounds.  It is also possible that a compound may have a functional group(s) or other structural features not represented in the training set, and for which no fragment coefficient or correction factor was developed.  These points should be taken into consideration when interpreting model results.

 

Description and results of any possible structural analogues of the substance to assess reliability of the prediction

External validation with a dataset containing 158 substances resulted in a correlation coefficient (r2) of 0.850 and a standard deviation of 0.583. The external validation set includes a diverse selection of chemical structures that rigorously test the predictive accuracy of the model.

 

Uncertainty of the prediction

All constituents for which estimations were made fall within the applicability domain of the model.

 

Predictivity

The first-order molecular connectivity index (MCI) was successfully used to predict soil sorption coefficients (Koc) for nonpolar organics. The extension of the model with correction factors makes this model also applicable for polar substances. The QSARs have an r2 of 0.967 and 0.90 respectively with an average deviation of 0.199 and 0.273.  If the Training datasets are combined in to one dataset of 516 compounds (69 having no corrections plus 447 with corrections), the MCI methodology has an r², standard deviation and average deviation of 0.916, 0.330 and 0.263

Validity criteria fulfilled:

not applicable

Conclusions:

The log Koc range of the constituents of Spearmint oil is 1.53 - 4.30.

Executive summary:

As Spearmint oil is a naturally complex substance consisting of multiple constituents, the substance does not have a single sorption coefficient (log Koc). Therefore, in line with the NCS protocol, soil adsorption coefficients for the individual known constituents were calculated using the KOCWIN v2.00 QSAR by US-EPA.

The log Koc range for Spearmint oil constituents was found to be 1.53-4.30.

Description of key information

The range of log Koc values for the known constituents of Spearmint oil was found to be 1.53 - 4.30.

Additional information

Spearmint oil is a Natural Complex Substance (NCS) with L-Carvone and L-Limonene as its main constituents. As Spearmint oil consists of many constituents, there is no single value for adsorption (log Koc).

The log Koc of Spearmint oil was estimated by calculation. Log Koc values for the known constituents were estimated using the QSAR KOCWIN v2.0 according to the Molecular Connectivity Index method. The range of log Koc values for the known constituents of Spearmint oil was found to be 1.53 - 4.30.