Registration Dossier

Data platform availability banner - registered substances factsheets

Please be aware that this old REACH registration data factsheet is no longer maintained; it remains frozen as of 19th May 2023.

The new ECHA CHEM database has been released by ECHA, and it now contains all REACH registration data. There are more details on the transition of ECHA's published data to ECHA CHEM here.

Diss Factsheets

Administrative data

Link to relevant study record(s)

Description of key information

As described in detail in Section 1 of the registration substance, Turpentine Oil from Pulping Processes (TOPP) is a complex UVCB substance containing a range of constituent types. The principal constituents are bicyclic terpenes (e.g. α-pinene, β-pinene, δ-3-carene), with smaller proportions of monoterpenes, sesquiterpenes and their hydrogenated and oxygenated terpenoid derivatives. The other constituents of note are low molecular weight sulfur-containing constituents, dimethyl sulfide, dimethyl disulfide and methyl mercaptan, The Boundary Composition for the registration substance limits the total amount of sulfur-containing constituents to 6% by weight expressed as sulfur.

Where available, key endpoint data are reported for ‘whole substance’ and these are also used for classification conclusions for the registered substance. It is also important to consider the possible impact of individual constituents on the overall toxicity of the substance. Information on constituents has therefore been gathered from published sources and where possible original study reports have also been obtained and summarised.

For endpoints where measured data for whole substance are not available, a weight of evidence approach has been followed based on constituent data. Constituent data for all endpoints are relevant to support read-across for the weight of evidence approach.

For the purposes of environmental fate and exposure assessment, a ‘block’ approach is used, whereby constituents with similar physicochemical and environmental properties are grouped together (Section 4). For worker exposure, the approach used is to consider whole substance (based on pinene data) and sulfides (based on dimethyl disulfide data) (Section 9.0). However, for exposure of humans via the environment, separate DNELs are calculated for individual blocks using the best available data for each. The blocks are also referred to for convenience in the toxicokinetics discussion below.

Key value for chemical safety assessment

Additional information

Turpentine Oil from Pulping Processes (TOPP) is a UVCB substance, composed mainly of bicyclic terpene hydrocarbons (C10H16), with small proportions of other terpenes and sulfur-containing species. A detailed description of the composition of TOPP has been discussed in Section 1. No in vitro or in vivo toxicokinetic studies are available for the whole substance and it would not be scientifically feasible to conduct such a test.

The terpene constituents of TOPP are all naturally occurring secondary metabolites of pine trees. Metabolism and other biological behaviours of these constituents have been extensively reported in the published literature and this is discussed further below.


No quantitative data are available for absorption of constituents of TOPP. However, there is clear evidence from clinical and systemic effects reported in acute and repeated dose toxicity studies that constituents of TOPP are absorbed via both the oral and inhalation routes. There was no evidence of systemic effects in an acute dermal toxicity study with turpentine oil, although local skin irritation effects were observed. Positive results obtained in skin sensitisation tests (Local Lymph Node Assay) with individual constituents of TOPP indicate that these do penetrate the epidermis and reach the lymph nodes via Langerhans cells. Pathological examination in the available acute and repeated dose tests did not, however, indicate any specific target organ toxicity for the terpenes.

Available studies via the oral, dermal and inhalation routes for sulfur containing constituents also show clear evidence of systemic uptake.

There is a QSPR to estimate the blood:air partition coefficient for human subjects as published by Meulenberg and Vijverberg (2000). The resulting algorithm uses the dimensionless Henry’s Law coefficient and the octanol:air partition coefficient (Koct:air) as independent variables.

Using these values for the constituent blocks of TOPP results in a high blood:air partition coefficient for the terpene alcohols (approximately 147:1) and moderate values (2.7:1 to 5.1:1) for sulfur-containing constituents. Uptake into the systemic circulation following inhalatory exposure to these constituents is therefore likely.


Most studies describe the partition of turpentine rather than individual constituents such as the studies by Sperling et al., (1967) (rats and mice) and Savolainen and Pfäffi (1978) (rats), which showed that in these animals, terpenes accumulate in the peripheral fat, kidneys and brain. 


For blood:tissue partitioning a QSPR algorithm has been developed by De Jongh et al. (1997) in which the distribution of compounds between blood and human body tissues as a function of water and lipid content of tissues and then-octanol:water partition coefficient (Kow) is described. Using this value for the constituent blocks of TOPP predicts that, should systemic exposure occur, distribution of terpenoid constituents wouldprimarily be into fat, with potential distribution into liver, muscle, brain and kidney but to a much lesser extent, whereas for sulfur-containing constituents partitioning into fat and organs is much lower.

Table 5.2.1 Partition coefficients for constituent blocks of TOPP


Log Kow

Water solubility (mg/L)

Molecular weight (g/mol)

Vapour pressure (Pa)







Soluble fraction in blood

1 Pinene












2 δ-3-carene












3 Dipentene












4 Dimethyl sulfide












5 Methyl mercaptan












6 Sesquiterpenes












7 Terpene alcohols












8 Camphene












9 Dimethyl disulfide














Available published information on the metabolic pathways of constituents of TOPP has been reviewed and is summarised in a report attached in Section 13 and in the Chemical Safety Report. Please note that the metabolic pathways (figures) cited below are attached in Section 13 and CSR Section 5.2.3.


The chirality of α-pinene varies with the species of pine tree. In European pines it is of the (-) 1S,5S configuration and North American pines it is (+) 1R,5R. Metabolic studies often do not state the isomer composition of the α-pinene used; however, this can influence the nature of the metabolites formed (e.g. Ishida et al., 1981). The metabolic transformations of alpha pinene have been determined by White et al(1979) in rat liver microsomes, Ishida et al. (1981) in rabbits, and in humans by Eriksson and Levin, (1990, 1996) and Schmidt et al. (2013). Alpha-pinene is an inducer of cytochrome P450 enzymes in mammals and its principal route of phase I metabolic transformation is the production of various hydroxylated terpenols,shown in the metabolic pathway scheme below, catalysed by cytochromes P450.

An in vitro study using rat liver microsomes identified α-pinene oxide ([1] in Figure 5.1.1) as a principal metabolite (White et al., 1979). Epoxides such as [1] are frequent intermediates in the hydroxylation of alkenes. In a series of studies in which urine from sawmill workers exposed to alpha-pinene, beta-pinene and d-3-carene via their work was collected and hydrolysed with beta-glucuronidase, and the resultant aglycones analysed by GC/EI-MS, the principal metabolites were identified exclusively as those derived from alpha pinene; no metabolites of beta pinene and d-3-carene were identified. Note, although environmental monitoring of a-pinene, b-pinene and d-3-carene was carried out, the concentrations of these materials to which the workers were exposed in the experiment were not reported. The metabolites identified were cis- [2] and trans-verbenol [3] (see in Figure 5.1.1). The recoveries of the verbenols from hydrolysed urine were in the range of 85 to 94%. These metabolites were not detectable in urine that was not treated with glucuronidase, indicating that the two verbenols are excreted in urine as theglucuronides (Eriksson and Levin, 1990). In a follow-up study the same authors analysed the urine of similarly exposed workers using EI and CI MS post enzymatic hydrolysis and GC separation following TMS derivatisation (Eriksson and Levin, 1996). In addition to the cis-and trans-verbenol identified in the earlier study, they showed evidence for the production of a trans diol derivative of trans-verbenol [metabolite 4 in Figure 5.1.1] and its alcohol-aldehyde oxidation product [metabolite 5 in Figure 5.1.1]. The diol derived from cis-verbenol [metabolite 6 in Figure 5.1.1] was also identified.

The metabolism of α-pinene in laboratory rabbits was reported by Ishida (Ishida et al., 1981). Rabbits were orally dosed with (+)- (-)- and (±)-a-pinene and the urinary metabolites subjected to glucuronidase and arylsulfatase digestion. Metabolites were purified by silica gel chromatography, GC and TLC. Pure metabolites were identified by EI MS and 1H-NMR. Myrtenol [7] and its oxidation product myrtenic acid ([8] in Figure 5.1.1) were identified. The yields of myrtenol [7] were highly dependent of the nature of the chirality of the α-pinene used: (+) 14.9% (-) 0.9% (±) 9.3%. Myrtenol ([7] in Figure 5.1.1) has also been identified in the urine of human volunteers orally exposed to a cough medication containing monoterpenes, including α-pinene, as the active medicinal ingredients (Meesters et al., 2008).

The metabolites of α-pinene isolated from the urine of humans who had not been occupationally exposed to monoterpenes using a highly sensitive GC MS/MS method following TMS derivatisation were (1S,2S,5S)-cis-verbenol, (1R,2S,5R)-trans-verbenol and (1S,5R)-(+)-myrtenol [7] (Schmidt et al., 2013). Borneol, bornylacetate and myrtenol [7] have been identified as metabolites from α-pinene in human urine after acute poisoning with pine oil (Koppel et al.,1981). The biosynthesis of borneol structures from α-pinene seems unlikely considering the uncertainty in identification of borneol and lack of mechanistic explanation for their biosynthesis. Therefore, these structures are not shown in Figure 5.1.1.

A composite pathway for α-pinene metabolism in mammals is shown in Figure 5.1.1.


In contrast to α-pinene, the isomer of β-pinene in different conifer species is always (-)-b-pinene. Ishida et al. (1981) considered the principal route of (-)-β-pinene metabolism in rabbits to be via the allylic oxidation of the alkene moiety to form an epoxide intermediate [metabolite 1, Figure 5.1.2]. This was further ring-opened and hydrolysed to yield (-)-a-terpineol [2] and (-)-1-p-menthene 7,8 diol [3] in 39% and 30% yield respectively, the magnitude of the yield implying that this is the major catabolic route. (+)-Trans-pinocarveol [4] and (-)-trans-10-pinanol [5] formed by ring and alkene hydroxylation respectively of β-pinene were relatively minor metabolites. In the brushtail possum (Trichosurus vulpecula) myrtenic acid was identified as a non-conjugated metabolite, which was excreted in the urine after oral administration of b-pinene (Southwell et al.,1980).

A composite pathway for β-pinene metabolism in mammals is shown in Figure 5.1.2.


The isomeric composition of d-3-carene varies with the source species of coniferous tree: in the European Pinus sylvestris it is (-)-d-3-carene, whereas in Pinus palustris, a native of the south-eastern USA, it is (+)-d-3-carene. As with α-pinene the isomeric composition will affect its metabolism and this is not always reported. The in vitro metabolism of (-)-d-3-carene was investigated using human liver microsomes and with recombinant human cytochrome P450 enzymes expressed in E. coli (Duisken et al., 2005) and the metabolites identified by GC/MS. Two metabolites were detected: δ-3-carene epoxide [1] and δ-3-carene-10-ol [2]. The epoxide was produced exclusively by the CYP1A2 P450 isozyme, whereas 10-hydroxylation was effected by CYP2B6, CYP2C19, and CYP2D6. The metabolism of (+)-d-3-carene by rabbits produced three methyl-hydroxylated and carboxylated metabolites as their glucoronide and sulfate conjugates in urine (Ishida et al., 1981). The compounds identified were (+)-d-3-carene-8-ol [3], (+)-d-3-carene-9-carboxylic [4] acid and (+)-d-3-carene-9,10-dicarboxylic acid [5]. The only metabolite of (+)-d-3-carene detected in the urine of environmentally exposed humans wasd-3-carene-10-ol [2] (Schmidt et al., 2013).

A composite pathway for δ-3-carene metabolism in mammals is shown in Figure 5.1.3.

Dipentene (Limonene)

Limonene is the 4-R-(+) isomer of dipentene (the racemic (±) form). The limonene of commerce is mostly extracted from the peel of Citrus fruit and the material found in most turpentine extracts is usually the racemate, although some Pinaceae such as Abies spp. also contain (+)-limonene. This summary of metabolic pathways contains studies on all isomeric forms of dipentene, indeed in some studies the isomer is not stated although it may influence both the rate of metabolism and the nature of the metabolites.

R-(+)-limonene is a farnesyl transferase inhibitor that has shown anti-tumour properties, consequently its metabolism in humans, especially cancer patients, has been extensively studied. In a study by Poon et al. (1996) R-(+) limonene was given orally to cancer patients. Plasma and urine samples collected from the patients were examined by reversed-phase HPLC-CI and electrospray ionization MS. The drug underwent rapid conversion to hydroxylated and carboxylated derivatives. Characterisation and structural elucidation of the metabolites were achieved by LC/MS and NMR. Five major metabolites were detected in the plasma extracts, namely limonene-1,2-diol [1], limonene-8,9-diol (uroterpinol) [2], perillic acid [3], an isomer of perillic acid [4], and dihydroperillic acid [5]. Urinary metabolites comprised the glucuronides of the two isomers of perillic acid [3] [4], dihydroperillic acid [5], limonene-8,9-diol [2], limonene-10-ol [13] and a mono-hydroxylated limonene, which was not characterised further.

In a similar study, a group of 32 cancer patients were administered 99 courses of R-(+)-limonene at doses from 0.5 to 12 g/kg per day administered orally in 21-day cycles (Virgushin et al., 1998). Predominant circulating metabolites were perillic acid [3], dihydroperillic acid [5], limonene-1,2-diol [1], limonene-8,9-diol (uroterpinol) [2] and an isomer of perillic acid [4]. Both cis and trans perillic acid [3], and cis-[6] and trans-[7] dihydroperillic acid were detected in urine hydrolysates. In two other similar studies the following metabolites were identified: dihydroperillic acid [3] and perillic acid [5] (major metabolites). Two minor metabolites were found to be the respective methyl esters of these acids (Crowell et al.,1992, 1994). In these human in vivo studies the metabolites in urine were exclusively O‑glucuronides for the hydroxylated derivatives of limonene. In the case of perillic acid and its derivatives, they are esterified at the carboxyl group by glucuronic acid to form glucuronyl ester conjugates.

In an in vitro study using human liver microsomes and recombinant human cytochrome P450 enzymes, the metabolic transformations of both (-)- and (+)-limonene was studied by Miyazawa et al., (2002). The oxidation rate by human liver microsomes was similar for both (-)- and (+)-limonene, indicating that little enantioselectivity was shown by the cytochrome P450 enzymes responsible. Two types of metabolites, (+)- and (-)-trans-carveol [8] (a product of 6-hydroxylation) and (+)- and (-)-perillyl alcohol [9] (a product of 7-hydroxylation), were identified, and the latter metabolites were found to be formed more extensively than the former ones. The rate of limonene oxidation activities correlated well with amount of CYP2C9 P450 isozyme. Different P450 enzymes were responsible for oxidising (+)- and (-)-limonene, with CYP2C8, CYP2C9, CYP2C18 and CYP2C19 responsible for the (+) isomer and CYP3A4 (-)-limonene. The authors state that species-related differences exist in the oxidations of limonenes in the CYP2B subfamily of P450 in rats and humans, thus suggesting that animal models may not always be suitable for assessing the human catabolism (and hence toxicity) of these compounds.

These differences were further explored by Shimada et al.(2002) who performed a comprehensive in vitro study on the metabolism of (-)- and (+) limonene P450 enzymes in liver microsomes of mice, rats, guinea pigs, rabbits, dogs, monkeys and humans. This study was initiated by the finding that (+)-limonene causes renal toxicity (kidney and bladder tumours) in male rats, but not in female rats and other animal species including mice, guinea pigs, rabbits, and dogs. Both (+) and (-) limonene enantiomers were converted to respective carveols [8], perillyl alcohols [9], and carvones [10] (oxidative metabolites of carveols) by liver microsomes of dogs, rabbits, and guinea pigs. However, mice, rats and humans produced carveols [8] and perilly alcohols [9], but not carvones [10]. Additionally, when (+)-carveol [8] and (+)-carvone [10] were used as substrates, dogs, rabbits, and guinea pigs metabolised them to (+)-carvone [10] and (+)-carveol [8] respectively, but humans, rats, and mice did not convert (+)-carveol [8] to (+)-carvone, [10], but metabolised (+)-carvone [10] to (+)-carveol [8] with male rats having the highest rates. Consequently, these results suggest that there are species-related differences in the metabolism of limonenes by P450 enzymes, particularly in the route from (+)-carveol to (+)-carvone that will affect the metabolic profile of limonene catabolism in different species.

Watabe et al. (1980) reported the isolation of 1,2-[11] and 8,9-epoxides [12] of (+)-limonene and their conversion to 1,2- [1] and 8,9-limonene diols [2] by rat liver microsomes. Epoxides are usually only isolated in in vitro metabolism studies, as they are rapidly converted to the diols, which is catalysed by epoxide hydratase. Interestingly, neither epoxide was positive in the Ames test, which is atypical for this class of compounds (cf. the epoxide formed from trans-anethole (see below)).

A composite pathway for limonene metabolism in mammals is shown in Figure 5.1.4. Conjugates have been omitted to simplify the scheme.



Terpinolene is found in relatively low concentrations (0.9% typical concentration in TOPP), however, there are data showing low acute oral toxicity (rat LD50 3740 mg/kg) and a very low repeat dose feeding toxicity. No specific data on the catabolic metabolism of terpinolene in mammals or other vertebrates could be found in the open scientific literature; however, the compound is a positional isomer of dipentene (limonene). Consequently, compounds formed by degradative metabolism of terpinolene would be expected to form a similar pattern to that compound, viz: formation of an epoxide formed from the endocyclic carbon bond and also possibly the exocyclic one, followed by formation of diols. The free ring methyl group would also be expected to be oxidised via cytochromes P450 to the alcohols and carboxylic acid substituents. An analogous beta-oxidation of the ring double bond to afford product analogous to carveol and carvone formed from dipentene would also be expected.

A composite pathway for terpinolene metabolism in mammals is shown in Figure 5.1.5.


α-Terpineol is found in low concentrations in TOPP (0.51% w/w on average). In an early study in which α-terpineol was administered orally to humans, the oxidation product p-menthane-1,2,8-triol [1] was identified as a urinary metabolite (Horning et al., 1976).

In a systematic study on the metabolism of α-terpineol in rats, Madyastha and Srivatsan (1988) studied both its urinary metabolites and its in vitro transformation by hepatic P450 preparations. Note: the metabolites shown are those after hydrolysis of the glucuronide conjugates. After oral administration to rats (600 mg/kg body weight), α-terpineol was metabolised to p-menthane-1,2,8-triol [1] probably formed from the epoxide intermediate (1,2-epoxy-α-terpineol [2] shown in bracket in the metabolic scheme). This epoxide intermediate was demonstrated in the P450 in vitro experiment, but not in vivo. The major routes for the biotransformation was allylic (beta) methyl oxidation and the reduction of the l,2-double bond of α-terpineol in the rat. Although allylic oxidation of C-l methyl to the alcohol seems to be the major pathway, the alcohol p-menth-l-ene-7,8-diol (8-hydroxy α-terpineol) [3] (shown in brackets in the metabolic scheme) was not isolated from the urine samples. Presumably, this compound is rapidly further oxidised to oleuropeic acid [4], which is in turn reduced to the dihydro compound (dihydroolopeic acid) [5].

A composite pathway for alpha terpineol metabolism in mammals is shown in Figure 5.1.6.


1,8-Cineole (Eucalyptol) is found in very low concentrations in TOPP, indeed in average samples of TOPP it is undetectable. The metabolism of 1,8-cineole has been investigated in koalas (in vivo) (Boyle et al.,2001), in rat and human liver microsomes (in vitro) (Miyazawa et al., 200l) and two species of rabbit, one of which (Brachylagus idahoensis) is a species that feeds on sagebrush, which contains high levels of 1,8-cineole (Shipley et al,2012). It would be expected that the koala and Brachylagus idahoensis, both of which feed on plants containing large amounts of 1,8-cineole, would have adaptive specialist metabolism for the compound and consequently would be poor models for human metabolism and this seems to be true, as both species metabolise 1,8-cineole to more highly oxidised metabolites such as 7-cineolic acid [1] and 9-cineolic acid [2]. These are further oxidised to hydroxylated cineoleic acids: 7-hydroxy-9-cineolic acid [3] and 9-hydroxy-7-cineolic acid [4] and eventually to cineole dicarboxylic acid (5). These polar metabolites were excreted in the urine, and also faeces of the cineole “specialists” as the aglycones rather than glucuronide conjugates. This atypical xenobiotic metabolism is unlikely to be a valid model for the metabolites produced by humans and these metabolites are shown in red in the metabolic pathway scheme below. In contrast, the metabolism in vitro by rat and human liver microsomes and in the non-specialist rabbit species, the cottontail rabbit (Sylvilagus nuttalli), were the two ring-hydroxylated metabolites 2-hydroxy-1,8-cineole [6] and 3-hydroxy-1,8-cineole [7] and in addition, the methyl-hydroxylated compounds 7-hydroxy-1,8-cineole [8] and 9-hydroxy-1,8-cineole [9] were also identified. In the in vivo studies in the non-specialist cottontail rabbit metabolites 6, 7, 8 and 9 were excreted in the urine as the glucuronide conjugates. Those metabolites considered to be relevant to human metabolism are depicted in black in the scheme in Figure 5.1.7.


In contrast to the monoterpenes discussed above, trans-anethole is an aromatic compound found in a number of plants and is derived via a completely different biosynthetic pathway. Its concentration in turpentine is low (0.1% in a typical sample). Its extensive use as a food-flavouring additive, data suggesting it is a carcinogen in rats and evidence that a genetoxic propene oxide derivative is formed by in vivo metabolism has led to several metabolic studies on the compound. Metabolism studies have been conducted in rats (oral administration) (Sangster et al., 1984a, 1984b, Solheim and Scheline, 1973), mice (IP injection administration) (Sangster et al., 1984a, 1984b) and humans (oral administration) (Sangster et al, 1987).


The earliest paper (Solheim and Scheline, 1973), which investigated the metabolism of trans anethole in the rat and identified the urinary metabolites by GC and GC/MS concluded that epoxidation of the side chain to form trans-anethole oxide [1] was a minor metabolic reaction. Instead, the major reaction was O-demethylation, which gave rise to 4’‑hydroxypropenylbenzene [2] and 4’-hydroxycinnamic acid [3] (Note: metabolite [3] was not identified in the Sangster et al. papers, see below). The major urinary metabolite identified was 4’-methoxy hippuric acid [4], which is the glycine conjugate of 4’-methoxybenzoic acid [5], which in turn is derived from 4'-methoxy-trans-cinnamic acid [6] formed from the beta-oxidation of trans-anethole. Essentially the entire applied dose of trans-anethole was recovered as metabolites [2], [3] and [4] in the urine, with [2] and [3] being found as their glucuronides.

In the Sangster et al. (1984a) paper, 4’-(14C)-methoxytrans-anethole was orally administered to Wistar rats and via IP injection to mice. Eleven labelled metabolites were identified in rat urine and ten in mouse urine, with the principal metabolites being two isomers of 1-(4'-methoxyphenyl)propane-1,2-diol [7], 2-hydroxy-1-methylthio-1-(4'-methoxyphenyl)propane [8] and 4’-methoxyhippuric acid [4].Metabolites [7] and [8], which are necessarily formed via trans-anethole oxide [1] were isolated as their glucuronides and their identification is in contrast to the findings of Solheim and Scheline, (1973), who considered that epoxidation was a minor route (see above). 4’-Hydroxypropenylbenzene [2], was excreted extensively in urine as the glucuronide.

Note: trans-anethole oxide [1] has shown to be both positive in the Ames test and carcinogenic to mice (Kim et al., 1999). Nevertheless, both an Ames test and an in vitro chromosome aberration study with trans-anethole itself, reported in the disseminated REACH dossier, gave negative results with and without metabolic activation.

In their second paper (Sangster et al., 1984b), the influence of trans-anethole dose (range 0.05-1500 mg/kg) on metabolism and species differences between rats and mice were investigated. The major route of metabolism of trans-anethole in rodents was via oxidative O-demethylation to give 4'-hydroxypropenylbenzene [2], though the importance of this route was found to decrease with increasing dose. The rat favoured the epoxidation route, resulting in the elimination in the urine of two 1-(4'-methoxyphenyl)propane-1,2-diol isomers [7], excretion of which rose from 2 to 15% of the dose over the dose range studied. In contrast the mouse favoured beta-oxidation of the propenyl moiety, yielding cinnamyl compounds, 4-methoxybenzoic [5] and 4’-methoxyhippuric acids [4], the elimination of the latter two rising from 10 to 42% of the dose over the dose range studied. Note: neither of Sangster’s 1984 papers discuss the potential influence on the metabolic pathway of the method of administration (oral or IP injection) which was different in rats and mice.

Human metabolism was studied by Sangster et al. (1987), where human volunteers were administered 100 mg or 1 mg of [14C]-methoxy-trans-anethole and their urinary metabolites identified. At both doses, the major metabolite was 4’-methoxyhippuric acid [4] (56% of the dose), accompanied by much smaller amounts of the two isomers of 1-(4'-methoxyphenyl)propane-1,2-diol [2] (together 3%). A composite metabolic pathway for trans-anethole metabolism in rats, mice and humans is shown in Figure 5.1.8.

Dimethyldisulfide, Dimethylsulfide, Methanethiol

Maintenance of an adequate GuSH/GuSSGu ratio is essential for cell survival and function. Under normal conditions, the cellular concentrations of endogenous thiols (reduced and oxidised glutathione) are controlled by disulfide exchange reactions.

Dimethyldisulfide, dimethylsulfide, and methanethiol (methyl mercaptan) are considered together, as they are metabolically interchangeable via sulfide exchange: animal studies that have examined the metabolites of any one of these materials are generally able demonstrate the presence of all three: dimethyl disulfide [1], dimethyl sulfide [2], methanethiol [3], generally in addition to dimethyl sulfoxide [4], dimethyl sulfone [5], and sulfate [6], the latter three of which are excreted in the urine (Blom et al. 1990) (rats and dogs), (Williams et al., 1966) (rabbits).The labile nature of the S-S bond results in a variety of metabolic routes for detoxification. The disulfide bond may be reduced to give the corresponding thiol in a reversible reaction in vivo. Therefore, the metabolic transformations of thiols are also applicable to disulfides. Thiol-disulfide exchange reactions are common in vivo and result from nucleophilic substitution by sulfur. In addition to the interconversion of dimethyl disulfide [1], dimethyl sulfide [2], methanethiol [3], thio-disulfide exchange reactions with endogenous cellular thiols (reduced glutathione, GuSH) or disulfides (oxidised glutathione, GuSSGu) will produce mixed disulfides that may also undergo reduction. (Cotgreave et al., 1989; Brigelius, 1985; Sies et al., 1987). Further metabolism of any free thiols produced will generate thiosulfinates, thiosulfones and eventually sulfate, which are generally excreted. Oxidation of thiols is catalysed by cytochromes P450 and flavin mono-oxygenases, generally in the liver. Thiols may also be methylated via S-adenosylmethionine (SAM)-dependent thiol methylation to yield a thio-ether, which is usually oxidised to the sulfoxide (major) and sulfone (minor) polar metabolites, which are then excreted. The composite pathway for the inter-conversion of dimethyldisulfide, dimethylsulfide and methanethiol in mammals is shown in Figure 5.1.9.


Summary and conclusions on metabolism

The process of catabolic degradation of xenobiotics such as the monoterpehavnes described above takes place mainly in the liver and is principally catalysed by cytochromes P450. It leads to increasingly polar molecules such as primary alcohols (formed by hydroxylation of methyl groups) or secondary alcohols (hydroxylation of ring methylene groups). The primary alcohols may then also be further oxidised to carboxylic acids, which are even more polar. This type of phase I metabolism, is usually considered a detoxification mechanism and if sufficiently polar (such as dicarboxylic acids), the products may be excreted in the urine. Much more common is the formation of conjugates by coupling the alcohols to very polar molecules such as glucuronic acid or sulfate (phase II metabolism), and these are excreted in the urine. 

Of more toxicological significance is the formation of electrophilic, potentially genotoxic metabolic intermediates such as epoxides (oxiranes). The potential for producing epoxides has been shown for all the monoterpenes, which have an endocyclic carbon-carbon double bond) (α-pinene, -3-carene, α-terpineol), exocyclic carbo-carbon double bond (β-pinene) or both (dipentene, terpinolene). The one exception is 1,8-cineole. There is evidence that trans-anethole also forms an epoxide, however, its formation appears to be species-dependent (see above). Epoxides are often not detected in in vivo metabolism studies, as they are reactive and usually have short biological half-lives, being mainly hydrolysed to cis-diols. This reaction is catalysed by the enzyme epoxide hydrolase, and the identification of cis-diols may be used to infer the existence an epoxide intermediate. Epoxides are also detoxified by reaction with glutathione, and the conjugates produced are further processed to yield acetylcysteine derivatives, which are sometimes detected in metabolism studies. However, in vitro studies using liver microsomes do frequently demonstrate the production of epoxides produced by P450 catalysed oxidation of alkenes or benzene rings. Their toxicological significance, however, is dependent on several factors, including the biological half-life of the epoxide and its inherent nucleophilicity, and their existence may not necessarily be used to infer inherent chronic toxicity. However, in vitro genotoxicity studies reported in REACH disseminated dossiers for the constituents all gave negative results (see Section 7.6), so there is no evidence of epoxide formation causing damage to DNA or chromosomes.    

It is often difficult to predict the effects of degradative metabolism on toxicity across different chemical structures and there are examples where even the chirality at a single carbon centre can fundamentally affect the course of the metabolic pathway, which in turn can affect long-term toxicity. There are also major differences between different mammalian species, and even sexes, which means that animal “models” such as rats and mice may not in fact predict the toxicological outcome of the administration of a particular xenobiotic in humans very well. This also means that such studies may either over- or under-estimate human toxicity. There are numerous examples of these effects in the literature and some can be seen in the examples described above.

Notwithstanding the above caveats, there are some general observations that can be drawn from the metabolism of turpentine constituents, particularly the monoterpenes: 

·        Beta-oxidation of methyl and ring methylene, whereby the carbon next to an alkene group is oxidised to a primary or secondary (ring) alcohol is very common and this applies to all the monoterpenes andtrans-anethole. It may be considered a detoxification step.

·        Primary alcohols may be further oxidised to carboxylic acids. This reaction is also catalysed by P450 enzymes (demonstrated in all compounds) (detoxification).

·        All the primary and secondary alcohols, and phenols (trans-anethole) metabolites are principally excreted as the glucuronide (or in some species sulfate) conjugates in mammals. This also applies to manocarboxylic acids, although dicarboxylic acids may be excreted unconjugated (dipentene, 1,8-cineole). (detoxification)

·        Exo-cyclic carbon-carbon double bonds may isomerise (b-pinene) (of uncertain toxicological significance)

·        Carbon-carbon double bonds may be reduced (alkene to alkane) (dipentene, α-terpineol) (potentially a detoxification step).

·        All the compounds that possess a carbon-carbon double bond show the potential to generate epoxides during metabolism (enhancing toxicity). The exception is 1,8-cineole.

Although the chemical structures of metabolites differ according to the identity of the starting point, there is no evidence from the literature that any of these constituents produces a metabolite of specific concern for toxicity most of the metabolic transformations may be considered to be detoxification. The exception is; the formation of potentially genotoxic epoxide metabolites, but the available genotoxicity data do not indicate a hazard, most likely because these metabolites are short-lived and rapidly converted to diols in a futher detoxification step. It is therefore considered to be likely that the metabolism of constituents results in detoxification. It is also concluded that is is likely that a given constituent would demonstrate a similar toxicological hazard profile to other constituents of similar chemical structure and physicochemical properties.



The monoterpenes discussed in above are all of a hydrophobic nature, and metabolism in mammals is principally by oxidations catalysed by cytochrome P450 enzymes, mainly in the liver. This process yields various hydoxylated metabolites, diols via epoxide hydratase and further oxidised metabolites such as peryllic acid, all of which are of higher polarity, although still not polar enough to be excreted via the kidneys.

Most animal studies on the urine metabolites show that almost all the excreted metabolites are conjugates, principally glucuronides, which are highly polar. Toxicokinetic measurements on the uptake and clearance or excretion of turpentine or its constituents are few. However, in a study in which eight male volunteers were exposed to turpentine vapour (constituent ratios not stated), Filipsson (1996) showed that the mean relative uptakes of alpha pinene, beta pinene, and delta-3-carene were 62%, 66%, and 68% respectively of the amount applied. Between 2% and 5 % of the net uptake was excreted unchanged in the expired air after the end of exposure. After the end of the two-hour exposure initial blood clearance was rapid but the half-lives for blood clearance of a-pinene, β-pinene, and d-3-carene were 32, 25, and 42 hours respectively for the later slow elimination phase. Peak blood concentrations of the three monoterpene components were in the order d-3-carene > α-pinene >> β-pinene. The concentrations of urinary metabolites were not measured.

In a more thorough study on the uptake and clearance of (-)- and (+)- α-pinene administered by inhalation, the peak blood concentration of α-pinene was shown to be directly proportional to the administered dose and the clearance rate of α-pinene was similar to the experiment in which the administered material was turpentine. The long clearance time of a-pinene from the blood indicates that it would take two days for it to be completely cleared. The elimination of unchanged a-pinene via the lungs was about 7.6% of the dose. The renal excretion of unchanged a-pinene was very low, with <0.001% of the administered dose being detected in the urine. However, urinary metabolites of a-pinene were not identified or measured (Falk et al., 1990b). The renal elimination of verbenols after experimental exposure to (+)- and (-)- α-pinene was studied in humans following exposure to 10, 225, and 450 mg/m3 total terpenes in an exposure chamber. The pulmonary uptake was about 60%. About 8% was eliminated unchanged in exhaled air. Depending on the exposure level, about 1-4% of the total uptake was eliminated as conjugates of cis- and trans-verbenol (the major metabolites). Most of the verbenols were eliminated within 20 hours after a 2 hour exposure. The renal excretion of unchanged a-pinene was less than 0.001% as with the previous study (Levin et al. 1992); however, no study was found that gave a complete balanced pharmacokinetic study on the uptake, partitioning and elimination ofa-pinene or similar monoterpenes present in turpentine.

In summary, the relatively slow removal ofa-pinene when administered to humans, and its multiphase elimination profile from the blood, is characteristic of other similar hydrophobic compounds, whereby the initial fast elimination is due to first-pass hepatic metabolism. The slower elimination phase(s) is/are due to partitioning into lipid-rich organs such as the brain, kidney and peripheral fat as evidenced by the partitioning ofa-pinene into these organs in animal models. Further slow elimination is then due to either tissue oxidation to hydrophilic metabolites or re-partitioning into blood, followed by hepatic oxidation, conjugation and renal excretion.

The sulfur-containing constituents are of low n-octanol/water partition coefficient and high vapour pressure. It is therefore likely that a significant proportion of inhaled substance will be excreted unchanged in expired air.