1.1.1. Nomenclature

• Chem. Abstr. Serv. Reg. No.: 78-59-1
• EC/List No.: 201-126-0
• Chem. Abstr. Serv. name: isophorone
• IUPAC systematic name: 3,5,5-trimethylcyclohex-2-en-1-one
• Synonyms: 3,5,5-trimethylcyclohex-2-enone; 3,5,5-trimethyl-2-cyclohexene-1-one; 1,1,3-trimethyl-3-cyclohexene-5-none; 3,5,5-trimethyl-2-cyclohexen-1-one; isoacetophorone; isooctopherone; α-isophorone; 3,5,5-trimethyl-2-cyclohexenone; and other depositor-supplied synonyms and acronyms (NCBI, 2021).

1.1.2. Structural and molecular information

• Relative molecular mass: 138.21 (NCBI, 2021)
• Chemical structure:

  
• Molecular formula: C₉H₁₄O

1.1.3. Chemical and physical properties

• Description: colourless liquid with a peppermint-like odour (IFA, 2021a) or camphor-like odour (NCBI, 2021)
• Odour threshold: odour may be noticeable at concentrations of 2–5 ppm (NCBI, 2021)
• Boiling point: 215 °C (NCBI, 2021)
• Melting point: −8.1 °C (NCBI, 2021)
• Density: 0.92 g/cm³ at 20 °C (IFA, 2021a)
• Relative vapour density: 4.77 (air = 1) (IFA, 2021a)
• Vapour pressure: 0.59 hPa at 25 °C (IFA, 2021a)
• Auto-ignition temperature: 460–470 °C at 1013 hPa (ECHA, 2021a)
• Lower explosion limit: 0.87 vol.% (50 g/m³) (IFA, 2021a)
• Upper explosion limit: 3.8 vol.% (220 g/m³) (IFA, 2021a)
• Solubility: sparingly soluble (12 g/L at 20 °C) in water (IFA, 2021a); soluble in ether, acetone, and alcohol; high solvent power for vinyl resins, and cellulose esters (NCBI, 2021)
• Flash point: 84 °C (NCBI, 2021)
• Stability and reactivity: the combustible substance can react dangerously with air, and the formation of peroxides is possible (IFA, 2021a); exposure to sunlight in aqueous solutions can result in the formation of photodimers by 2+2 photocycloaddition (Gonçalves et al., 1998)
• Octanol/water partition coefficient (P): log K_ow = 1.70 (NCBI, 2021)
• Conversion factor: 1 ppm is equivalent to 5.74 mg/m³ at 1013 mbar [101.3 kPa] and 20 °C (IFA, 2021a)
• Dynamic viscosity: 2.62 cP at 20 °C [2.62 × 10⁻⁵ hPa.s] (NCBI, 2021).

1.1.4. Impurities

Impurities include up to 4% isomeric β-isophorone (3,5,5-trimethyl-3-cyclohexen-1-one) and traces (< 1%) of 1,3,5-trimethylbenzene, mesityl oxide (2-methyl-2 pentene-4-one), phorone (2,6-dimethyl-2, 5-heptadiene-4-one), and isoxylitones (NCBI, 2021).

1.2.1. Production process

Isophorone is produced by the aldol condensation of acetone at high temperature (200 °C) and pressure (3.6 MPa) in the presence of aqueous potassium hydroxide. Alternatively, isophorone can be manufactured using calcium oxide, hydroxide, or carbide, or mixtures thereof, at atmospheric pressure and high temperature (350 °C) (NCBI, 2021).

1.2.2. Production volume

Isophorone is listed as a High Production Volume chemical by the Organisation for Economic Co-operation and Development (OECD) (OECD, 2004, 2009). Worldwide production capacity has been estimated at 50 000 tons [45 400 tonnes] in 1990 (NCBI, 2021). In 2016, the United States Environmental Protection Agency (US EPA) estimated an aggregated production volume of 10 000 000–50 000 000 pounds [~4536–22 680 tonnes] in the USA (US EPA, 2016). Isophorone is registered under the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) Regulation, and more than 100 tonnes per year are manufactured in and/or imported to the European Economic Area (ECHA, 2021a).

1.2.3. Uses

Isophorone is a widely used solvent and chemical intermediate used at industrial sites and in the manufacture of lacquers and vinyl/acetate-based polymers, inks and paints, nitrocellulose finishes, and washing and cleaning products (US EPA, 2000; ECHA, 2021a; NCBI, 2021). Isophorone is used in the manufacture of agrochemicals and is a constituent of certain pesticides. For example, in the USA, isophorone is exempt from the requirement of a tolerance when used as an inert ingredient in pesticide formulations applied to beet, ginseng, rice, spinach, sugar beet, and Swiss chard (Federal Register, 2006). Isophorone is also used as an intermediate for the manufacture of other chemicals, such as 3,5-xylenol, 3,3,5-trimethylcyclohexanol, and trimethylcyclohexanone (US EPA, 2000).

1.3.1. Air

Established methods exist for the measurement of isophorone in indoor workplace air, with protocols dated as early as 1955 (Kacy & Cope, 1955). Modern methods commonly involve the use of sampling tubes containing a solid adsorbent matrix through which a volume of air is passed and can be used in conjunction with personal air samplers. Early approaches to this method employed adsorption on a charcoal matrix (White et al., 1970), e.g. National Institute for Occupational Safety and Health (NIOSH) Method 2508 (NIOSH, 2003). Owing to the decomposition of isophorone desorbed on charcoal, polymer-based collection matrices are now favoured (Brown & Purnell, 1979; Levin & Carleborg, 1987). These developments formed the basis of (superseding) NIOSH Method 2556, which uses a XAD-4 resin mesh as the adsorptive collection matrix with desorption using diethyl ether and analysis by gas chromatography (GC) with flame ionization detection (FID). This method has a limit of detection (LOD) of 1 µg per sample and has achieved average recoveries of isophorone of 94.1%, in the range of 55–831 µg (Levin & Carleborg, 1987; NIOSH, 2003). One study made inline measurements of emissions of isophorone and other volatile organic compounds (VOCs) from headspace chambers in which electronic components were heated to 75–200 °C. Analysis was performed using gas chromatography-mass spectrometry (GC-MS) (Paz et al., 2012). [The Working Group noted that this study was conducted in the context of detecting the onset of equipment overheating and electrical fires in on-board instruments found in aircraft, submarines, and other vessels. While determining human exposure was not the primary rationale for this study, such electrical emissions may still be a relevant potential source of isophorone.]

Methods for the determination of isophorone in workplace air have also been developed in China. These methods are based on the earlier principle of adsorption to a charcoal matrix but report good recoveries (> 90%) if the analysis is carried out within 10 days of sampling (Kang et al., 2006; Chen & He, 2009).

1.3.2. Water

Several methods have been used to measure isophorone in various aqueous samples, including drinking-water (US EPA, 2012), natural waters (Sheldon & Hites, 1978; Horng & Huang, 1994; Ma et al., 2009), industrial effluents (US EPA, 1984), and landfill leachate (Ghassemi et al., 1984). Solid-phase or solvent extraction techniques are capable of separating isophorone from an aqueous matrix before chromatographic analysis. The US EPA has developed several approved methods for the determination of isophorone in water. US EPA Method 525.3 (US EPA, 2012) is the most recent issue of a series of methods used to detect isophorone and more than 100 other organic chemicals in drinking-water. In brief, 1 L of sample is passed through a solid-phase extraction (SPE) device. Analytes are eluted from the SPE device with organic solvents. The eluent is then dried by passing through an anhydrous sodium sulfate column, concentrated by evaporation with nitrogen gas, and made up to a volume of 1 mL with ethyl acetate and internal standard solutions. Samples are then analysed by GC-MS, with reported LODs of 0.004–0.014 µg/L. US EPA Method 609 (US EPA, 1984) covers the determination of isophorone and selected nitroaromatics in industrial wastewater using solvent extraction with methylene chloride [dichloromethane]. Extracts are dried, concentrated to a volume of 10 mL in hexane and can be analysed by GC with FID, or GC with electron capture detection (ECD), although the former is recommended since the LOD is lower (LOD of 5.7 µg/L with GC-FID compared with 15.7 µg/L with GC-ECD). The abovementioned analytes were also analysed using an SPE approach in lake water (Horng & Huang, 1994). Low concentrations of isophorone (LODs, 1.4–3.2 ng/L) have also been measured in sea water by a method using GC coupled with headspace solid-phase microextraction (HSPME) with activated carbon fibre (Ma et al., 2009).

1.3.3. Soil and sediment

No methods were identified for the measurement of isophorone in natural soils.

Isophorone has been measured in lake sediments using both solvent (diethyl ether) (McFall et al., 1985) and Soxhlet (Wang et al., 2002) extraction methods followed by chromatographic clean-up of extracts and analysis by GC-MS. Two studies have reported isophorone measurements made in soils subjected to various aerobic and anaerobic conditions in vitro (Brown et al., 2021) and spiked with known quantities of isophorone (Singh et al., 1998). These studies used HSPME followed by gas chromatography quadrupole time-of-flight mass spectrometry (GC-Q-TOF-MS), and florisil column extraction followed by GC-MS, respectively. [The Working Group noted that although the soils analysed in these two studies were subjected to experimental conditions, the sample preparation, extraction, and analysis steps are probably applicable to environmentally sampled or natural soils.]

1.3.4. Food and consumer products

Methods have been described for the determination of isophorone in various edible samples, including fish and shellfish sampled in their natural habitat (McFall et al., 1985; Camanzo et al., 1987) and shop-bought food items (Kataoka et al., 2007). Isophorone was measured in samples of oyster and clam using solvent extraction with diethyl ether, homogenization, and centrifugation, followed by fractionation by gel-permeation chromatography and analysis by GC-MS (McFall et al., 1985). A similar approach, using Soxhlet extraction with GC-MS, was used to analyse isophorone in several fish species (Camanzo et al., 1987). Kataoka et al. (2007) collected samples of 32 food items and determined isophorone content using HSPME and GC-MS, with a reported LOD of 0.5 pg/mL [0.0005 µg/L] and recoveries of > 84% [the Working Group noted the versatility of this method across the wide variety of food items analysed].

Additionally, several reports have described methods for the chemical profiling of a variety of natural products cultivated for human consumption (see Section 1.4.1). These methods have assessed the VOC composition of different samples of which isophorone is a major constituent. The rationales for undertaking such analyses include the VOC profiling of honeys for identification and for description of antifungal, medicinal and aromatic properties (Alissandrakis et al., 2007; El-Sayed et al., 2018; Anand et al., 2019); the VOC profiling of plant extracts (e.g. Clerodendrum infortunatum L.; Gera et al., 2020) to determine taxonomy and aromatherapeutic and pharmaceutical applications; the analysis of saffron for the purposes of geographical and commercial discrimination (Masi et al., 2016; Liu et al., 2018); the analysis of the intermediate distillate of the Chinese traditional medicine, Xingnaojing injection, to optimize its extraction process (Fang et al., 2017); and the analysis of Jiashi muskmelon juice for odour profiling – a major determinant of consumer acceptance (Pang et al., 2012). [The Working Group noted that, while the rationales for the abovementioned analyses were not related to human exposure, the methods used would be applicable to determinations of isophorone made in this context.] Selected methods for measure of isophorone in these items are summarized in Table 1.1 and mostly involved HSPME techniques or direct headspace analysis by GC-MS.

Other consumer products for which methods of isophorone determination have been reported in the context of human exposure include clofibrate (a pharmaceutical used to control high blood levels of cholesterol and triglyceride), which was analysed directly by GC-MS after dissolution in chloroform (Johansson & Ryhage, 1976). Children’s inflatable aquatic toys and learning devices (armbands, beach balls and bathing rings) were also analysed by GC-MS after solvent extraction with dichloromethane and filtration (Wiedmer et al., 2017; Wiedmer & Buettner, 2018). [The Working Group noted that the two studies quantifying isophorone in aquatic toys used a combination of sensory analyses (i.e. a panel of trained assessors) and instrumental analyses to identify the principal chemical signatures of odours, a technique termed “gas chromatography-olfactometry”. The instrumental aspects of this approach are relevant to exposure assessment more broadly.]

1.3.5. Biological specimens

Although isophorone is mostly excreted via the urine in animal systems, data on excretion in humans are sparse (see Section 4.1 of the present monograph). There are no standardized protocols for the measurement of biomarkers of isophorone exposure in humans, and only one study reporting the determination of isophorone in human urine was available. In this study with a case–control design (Hanai et al., 2012), isophorone and other urinary VOCs were investigated as potential non-invasive diagnostic markers of lung cancer. Urine samples were frozen at −80 °C until use and, after thawing, underwent centrifugation and filtration. In an approach similar to that used by Ma et al. (2009) for sea water samples, isophorone and other VOCs were extracted from urine samples by HSPME using divinylbenzene/carboxen/polydimethylsiloxane fibres. Fibres with adsorbed compounds were analysed by GC-TOF-MS [limits of detection were not reported]. [The Working Group noted that although there was a lack of studies reporting methods for isophorone determination in human urine, methods used for other aqueous samples and in animals (e.g. hexane extraction of rabbit urine followed by GC analysis; Dutertre Catella et al., 1978) may be used for the analysis of human urine.]

Finally, an experimental study reported isophorone measurements on spiked plasma samples that underwent florisil column extraction and analysis by GC-MS (Singh et al., 1998). [The Working Group noted that the authors did not specify the origin (i.e. human or animal) of the plasma used in this experiment, and that samples were spiked with isophorone. Nevertheless, the study provided an example of isophorone determination in a plasma matrix, which may be relevant to human biomonitoring.]

1.4.1. Occurrence in the environment, food, and consumer products

The wide use of isophorone as a chemical intermediate and solvent for lacquers, inks, vinyl resins, herbicides, copolymers, coatings, and other products in a variety of industrial settings permits its entry into the environment from urban centres and industrial sites via atmospheric emissions due to volatilization; and via water and soil contamination due to waste disposal, industrial effluents and runoff. While isophorone is rapidly removed from the air by photochemical breakdown, and to a lesser extent washout, it may persist in natural waters and soil for longer periods. In water, volatilization and sorption to sediments and particulates are not expected to be significant removal mechanisms of isophorone and, in soils, microbial degradation is expected to occur (ATSDR, 2018). Isophorone is also present in food items, and in products whose manufacture involves its application, including food packaging (Sasaki et al., 2005; Skjevrak et al., 2005) and children’s aquatic toys (Wiedmer et al., 2017; Wiedmer & Buettner, 2018). While the dominant sources of isophorone in the environment appear to be anthropogenic in nature, it has been found to occur naturally, including in several botanical specimens, such as cranberries (NCBI, 2021) and saffron, and in honey (see Section 1.3.4), and in the defensive froth or secretions of grasshoppers (Eisner et al., 1971). [The Working Group noted that the precise origin of isophorone in these natural specimens is a subject of inquiry.] Concentrations of isophorone reported in different environmental media, including food and other consumer products, are summarized in Table 1.2 and described throughout the following sections. [The Working Group noted that many of the measurements reviewed throughout Section 1.4.1 were made for method development and validation purposes and do not necessarily reflect the actual distribution of isophorone in the environment.]

Table 1.2

Occurrence of isophorone in environmental samples, food, consumer products, and biological specimens.

(a) Environmental occurrence

There is a notable scarcity of ambient air measurements of isophorone in the literature, despite its volatility and known sources of atmospheric emissions. The US EPA publishes national estimates of isophorone emissions via its National Emissions Inventory (NEI) on the basis of data provided by state, local, and tribal air agencies, and supplemented by data collected by the US EPA. Estimated isophorone emissions by sector in the USA in 2017 are presented in Table 1.3. This suggests that the five sectors with the highest emissions of isophorone are coal-powered electricity generation, waste disposal, industrial surface coating and solvent use, industrial processes not elsewhere classified, and chemical manufacturing, which contribute 38%, 29%, 19%, 6.5%, and 3.4%, respectively, of total emissions (US EPA, 2017). Atmospheric emissions of isophorone may be produced by coal combustion; isophorone was measured at a concentration of 490 ppb [0.49 mg/kg] in coal fly ash from a power station in the USA (Harrison et al., 1985). Overheating electrical components have also been shown to be a source of atmospheric emission of isophorone. Under experimental conditions, a resistor heated at a constant temperature of 200 °C for 5 hours emitted isophorone at 128 ng/g of component per hour (Paz et al., 2012). [The Working Group noted that these data indicated that isophorone exposure may occur as a result of electrical and other fires involving the combustion of isophorone-containing materials. The former is of relevance to those involved in the burning of electronic waste, an activity that is particularly prevalent in low- and middle-income countries such as those in West Africa and from where occurrence and exposure data for isophorone appear to be absent.]

Table 1.3

Estimated isophorone emissions in the USA in 2017 by emission sector^a .

Data on the occurrence of isophorone in surface waters (excluding effluents from industrial sites) are sparse, with the few measurements available from the USA not detecting the compound or detecting trace amounts (< 2 µg/L) (Sheldon & Hites, 1978; US EPA, 1982; Hall Jr et al., 1987). Lake sediments in both the USA and China have been found to contain isophorone. Mean concentrations at three sites at Lake Pontchartrain, Louisiana, USA, between May and June 1980, were between 0.9 and 12 ng/g [0.0009–0.012 mg/kg] (McFall et al., 1985). Much higher concentrations (1.01–17.21 mg/kg) were measured in five sediment samples collected from Donghu Lake, Wuhan, China, in November 2000 (Wang et al., 2002). A small number of studies (US EPA, 1974, 1975; Keith et al., 1976; Feng et al., 2020) have measured isophorone in drinking-water. A range of 1.5–2.9 µg/L was reported for an unknown number of samples collected in New Orleans, USA in 1974 (US EPA, 1974). Isophorone was detected in 3 out of 11 samples collected in Philadelphia, USA, between 1975 and 1977 (Suffet et al., 1980), but the concentrations were not reported. A study conducted in China comparing methods for the removal of isophorone from reservoir-sourced drinking-water reported concentrations that were all below 5.18 ng/L [0.005 µg/L] (Feng et al., 2020).

The highest environmental concentrations of isophorone have been measured in industrial effluents (Table 1.2): concentrations were 120 µg/L in sewage effluent, 185 µg/L in paint and ink effluent, 237 µg/L in pharmaceutical effluent, 753 µg/L in pulp and paper effluent, and as high as 1380 µg/L in effluent from petroleum refining (US EPA, 1982, 1983). Isophorone has also been found in the turf crumb rubber of synthetic sports pitches, which is made from recycled tyres (Perkins et al., 2019).

1.4.2. Occupational exposure

Few estimates are available of the number of workers exposed to isophorone worldwide. NIOSH estimated in 1978 that about 1.5 million workers were potentially exposed to isophorone in the USA (NIOSH, 1978a). However, the NIOSH National Occupational Exposure Survey estimated that just 47 097 workers (10 353 of whom were women) were exposed to isophorone in 1981–1983, with the highest numbers seen among operators of printing machines, painting and paint spraying machines, textile machines, and miscellaneous machines, as well as hand packers and packagers, assemblers, non-construction labourers, unspecified mechanics and repairers. The industries with the highest representation among exposed workers were rubber and miscellaneous plastics products, printing and publishing, fabricated metal, chemicals and allied products, and miscellaneous manufacturing industries (NIOSH, 1990). [The Working Group noted that it is unclear how representative these estimates are of current exposure prevalence.]

No studies on occupational exposure of workers exposed to isophorone during its manufacture were available to the Working Group. Isophorone has been measured in air in several occupational settings where isophorone is used as an ink in screen-printing and other types of printing and coating, and in plastics manufacture. Although isophorone is a constituent of some herbicides, e.g. Betanal (Bayer Crop Science, 2010) and Satunil (Arysta LifeScience, 2013), no information was available on exposures during manufacture or use of these herbicides. [The Working Group noted that the available data were sparse and were collected in only a few countries.]

In an epidemiological study by Rodrigues et al. (2020), the authors reported isophorone exposures in three facilities, located in East Fishkill in New York, Burlington in Vermont, and San José in California, USA, in which the following operations were carried out: semiconductor manufacture, masking, and module manufacture; and the manufacture of printers, hard disk drives, tape drives, and Winchester disks. [The Working Group noted that, although isophorone was present in at least one of these facilities, no information was provided as to in which operation the isophorone occurred.]

An Institut national de recherche et de sécurité (INRS) database called Solvex (INRS, 2021) is derived from a French occupational exposure database (COLCHIC) of measurements taken by French prevention authorities for risk assessment purposes since 1987 (Mater et al., 2016). Solvex provides summary statistics for personal measurements by industry, occupation, or task. The industry categories that showed sufficient data on isophorone exposure (more than 50 samples) that could be used to calculate statistics were printing, and reproduction of documents (100 results; mean, 0.99 mg/m³; range, 0.15–12 mg/m³), metallurgy (83 results; mean, 0.65 mg/m³; range, 0.05–21 mg/m³), manufacture of metal products (62 results; mean, 0.65 mg/m³; range, 0.05–10 mg/m³), and manufacture of electrical equipment (57 results; mean, 0.62 mg/m³; range, 0.04–3 mg/m³). Of these measurements, 88% had been taken before 2001.

Compliance measurements are also available from the United States Occupational Safety and Health Administration (OSHA) (OSHA, 2021) and discussed in Lavoué et al. (2013). Between 1984 and 2020, 755 personal isophorone measurements varying in duration between 11 and 880 minutes were collected. Of the 8% of measurements that were made after 2000, isophorone was not detected in ~30%. The isophorone measurements made between 1984 and 2020 ranged from < 0.015 ppm [< 0.09 mg/m³] (the smallest reported detected value) to 40 ppm [230 mg/m³] (interquartile interval, < 0.015–0.6 ppm [< 0.09–3.4 mg/m³]). The three most visited industries were: commercial printing, not elsewhere classified (n = 239; median < 0.015 ppm [< 0.09 mg/m³]; 90th percentile, 2.3 ppm [13 mg/m³]); plastics products, not elsewhere classified (n = 150; median < 0.015 ppm; 90th percentile, 1.1 ppm [6.3 mg/m³]); and blank books and loose-leaf binders (n = 41; median, < 0.015 ppm [< 0.09 mg/m³]; 90th percentile, 1.9 ppm [11 mg/m³]).

Table 1.4 summarizes the results of the identified literature on occupational exposure measurements in specific workplaces.

Table 1.4

Occupational exposure to isophorone measured in workplace air.

A NIOSH health hazard evaluation was conducted during two visits in 1977 to a metals coating company near Chicago, Illinois, USA (NIOSH, 1978b). Personal breathing zone (PBZ) and area-sample air measurements were collected in three coating lines and two reclaimed-solvent areas using charcoal sampling tubes and analysed using GC. Isophorone was below the LOD in several air measurements collected in one reclaimed-solvent area and 1.0 ppm [5.74 mg/m³] in the second area. Isophorone and several other solvents were measured for workers carrying out different tasks on three coating lines during the first visit and on one coating line during the second visit. [Ventilation was present, but its effectiveness was unclear.] Most of the general area samples in the coating lines contained non-detectable concentrations of isophorone. A PBZ sample for the finish coater collected over 4 hours contained non-detectable concentrations of isophorone. Time-weighted average (TWA) concentrations for exposure to isophorone were estimated at 1.5 ppm [8.61 mg/m³] for the prime coater, and 0.75 [4.30 mg/m³] and 0.97 ppm [5.57 mg/m³] for the finish coater, in short-term samples collected over 1–1.5 hours. PBZ samples collected over 5.5–6 hours showed non-detectable concentrations of isophorone for the prime coat operator and, in three out of four samples, for the finish coat operator (the fourth sample contained isophorone at a concentration of 0.74 ppm [4.25 mg/m³]). Short-term samples collected over a period of less than 35 minutes for the unwind operator, rewind operator, and finish coat operator all contained non-detectable concentrations of isophorone in the PBZ samples. The coating line for a different product was evaluated on the second visit and showed a mean isophorone concentration of 1.0 ppm [5.74 mg/m³] in PBZ samples across the various tasks. [The Working Group noted that sampling times were generally longer during the second visit, which may have improved the ability to detect isophorone in the PBZ samples.]

NIOSH measured occupational exposures to isophorone and other solvents in several screen-printing operations in the USA between 1978 and 1984. [The Working Group noted that these investigations were generally triggered by workers’ reports of nausea, headache, or eye and nose irritation.] Measurements were made in these studies using PBZ or area air sampling onto charcoal tubes, which were then analysed using GC.

Several PBZ samples were collected in 1977 at a small specialty screen-printing operation in Pittsburgh, Pennsylvania, USA (NIOSH, 1979). Of seven samples, all except two had non-detectable concentrations of isophorone (0.30 ppm [1.72 mg/m³] during screen printing in an unventilated area and 25.7 ppm [148 mg/m³] in a very short-term sample while cleaning the screens [ventilation effectiveness was unclear]; the LOD was 0.3 mg per sample.) [The Working Group noted this was more than five times the short-term ceiling limit value in effect at that time, 5 ppm [28.7 mg/m³].]

In 1980, NIOSH (1981) measured isophorone exposures in a company in Ridgefield, New Jersey, USA, employing 54 workers to screen-print, cut, laminate, and sew decals. Isophorone was a component of the printing ink and was also used directly in spray bottles as a “reducer” and on the printing screens as an anti-static coating. [The Working Group noted that employees using these sprays reported acute respiratory and neurological symptoms.] Isophorone was detected in the PBZ air of only the screen printers: their full-shift TWA (8 hours) concentrations were 0.7 and 14 ppm [4.0 and 80.4 mg/m³]. It was noted that the ventilation was poorly designed.

In a small specialty printing company in Augusta, Georgia, USA, PBZ and area air samples were collected in 1982 for two screen-printing workers using gloss vinyl inks containing 35–40% isophorone (NIOSH, 1983). Ventilation was considered poorly designed. The median air concentration across eight PBZ and area samples was 1.3 ppm [7.5 mg/m³], with the highest concentrations noted near the drying racks (area sample, 2.5 ppm [14 mg/m³]) and while printing decals (PBZ sample, 3.4 ppm [20 mg/m³]).

Lastly, NIOSH (1984) investigated exposures to isophorone and other solvents at a silk-screen printing vinyl-wallcovering manufacturer in Chicago, USA, in 1984. Isophorone (a solvent in the “retarder”) was not detected during screen printing operations at this facility. [The Working Group noted that the full-shift air sampling rates were lower than those used in previous NIOSH studies, which could have affected the detection limit, which was not given for isophorone.]

A study focusing on isophorone exposures among screen-printing workers was carried out in a 34 000 ft² [3160 m²] facility in the USA [location unspecified] (Samimi, 1982). Isophorone was a main constituent of the inks and ink thinners, ranging from 10% to 75% among the various products, which were used to screen-print plastic, paper, or metal sheets at this mostly unventilated workplace. The product was dried in a ventilated dryer or hung to dry in a [presumably unventilated] room. It was noted that isophorone exposures were highest for workers involved in press operations, drying operations, ink formulation, and screen cleaning. The authors collected 78 short-term (50–90 minutes) PBZ air samples using charcoal tubes among workers expected to have highest solvent exposures. [The Working Group noted that the sampling and analytical methods used in this study were similar to those used in the above series of NIOSH studies.] TWA PBZ exposure concentrations were generally highest among printing press operators (mean, 23 ppm [132 mg/m³] and workers involved in paint mixing (mean, 17.8 ppm [102 mg/m³]), but mean concentrations for all workers were above 8 ppm [46 mg/m³]. It was also noted that PBZ concentrations were higher than corresponding area samples. [The Working Group noted that this is a typical finding for many solvents, aerosols, and particulates studied in occupational settings.]

[The Working Group noted that it was unclear how representative of current exposure conditions these 40-year-old studies of screen-printing workers are. However, it is notable that a recent publication identified isophorone as “the most widely used screen-printing ink solvent (comprising 75% of the total solvent)” (Kiurski et al., 2016).]

A small study was carried out of indoor air in a paper and cardboard printing company in Slovakia in 2015 (Vilcekova & Meciarova, 2016). Quantitative measurements were made for total VOCs in a location with a floor area of 144.3 m². Although no quantitative exposure levels were measured for individual VOCs, qualitative analysis was done for isophorone and 20 other VOCs by sampling with a zNose 4300 electronic nose in four cycles [“cycle” was not defined] (Meciarova et al., 2014). Total VOC concentrations fluctuated throughout the day and were typically above 40 mg/m³, with the highest levels (spiking above 120 mg/m³) seen in the latest part of the 8-hour day. Isophorone was one of the three most commonly occurring individual VOCs, appearing in all four cycles.

Cai et al. (2019) measured isophorone and several other ketones and aldehydes in workplace air of the production workspaces of 10 large-scale plastic-product manufacturers in China. Air concentrations for seven of the plants were reported as 0.0065 mg/m³ [the Working Group interpreted this to be the LOD], and the highest concentration measured at a plant was 2.1 mg/m³. [The Working Group noted that few details were provided about the sampling and analytical methods used or about the facilities themselves in this study. All measured values were noted by the authors to be well below occupational exposure limits in China.]

In addition to the printing and coating operations described above, isophorone has occasionally been reported in office settings. A recent study by Davis et al. (2019) found isophorone to be emitted in 8% of 3D printers tested. A NIOSH investigation (NIOSH, 2014) of a large government office complex in the USA in 2011 noted that grab samples had been historically collected for isophorone, but no information was provided on whether isophorone was detected.

1.4.3. Exposure of the general population

Only one study reporting quantitative measurements of isophorone in the general population was available. A case–control study (Hanai et al., 2012; Table 1.2) measured isophorone in the headspace of urine samples from 20 lung cancer cases and 20 healthy controls for the purpose of identifying novel diagnostic markers of lung cancer. A mean concentration of 130 nM [18 µg/L] was reported among control participants, and concentrations ranged from 39 to 1412 nM [5.4–195 µg/L]. It has been reported that the most likely routes of isophorone exposure in the general population are the inhalation of contaminated air and the ingestion of contaminated drinking-water (ATSDR, 2018), as well as direct contact with lacquers, paints, inks, and adhesives (US EPA, 2000), and exposure as a bystander during professional spraying and agrochemical use (ECHA, 2022). [The Working Group noted that while exposure to chemicals and products containing high concentrations of isophorone is probably higher in occupational settings, exposure to such products also occurs among hobbyists in the general population. However, there are substantial gaps in the published data on environmental exposures. The relative importance of exposure to other sources of isophorone described in this section (food, food packaging, and inflatables) remains to be quantified. It was further noted that, in many populations, isophorone exposure via drinking-water may be higher among people drinking bottled water due to migration from plastic bottles.]

1.5.1. Exposure limits and guidelines

(a) Occupational exposure limits

The American Conference of Governmental Industrial Hygienists (ACGIH, 2001) recommended a 15-minute ceiling limit of 28 mg/m³ for isophorone. There is no occupational exposure limit for the European Union, although occupational exposure limits have been defined by some European Union countries. This is the case for Germany, which has a MAK value (maximum workplace concentration) of 11 mg/m³ (IFA, 2021b). European Union Council Directives state that pregnant or breast-feeding workers and persons under age 18 years may not be occupationally exposed to isophorone (Council directive 92/85/EEC; Council Directive 94/33/EC; European Council, 1992, 1994). Table 1.5 summarizes the current occupational exposure limits for isophorone in some countries with an 8-hour time-weighted average (TWA), or 15-minute short-term or ceiling limits (IFA, 2021b). In the USA, NIOSH has established an “immediately dangerous to life and health” (IDLH) limit for isophorone of 200 ppm on the basis of acute inhalation toxicity data in humans (NIOSH, 1994).

Table 1.5

Occupational exposure limits for isophorone in various countries.

1.5.2. Reference values for biological monitoring of exposure

No reference values related to isophorone biological monitoring were available.

1.6.1. Exposure assessment methods

The exposure assessment for the study by Rodrigues et al. (2020) was completed by conducting site visits at the three facilities and compiling more than 700 000 documents with site-specific job, task, and process information, and an industrial hygiene database with more than 10 000 samples across 31 chemicals and dusts of interest (including isophorone). Exposure was assigned by group exposure matrices and time period (“manufacturing era” to record periods where processes remained relatively stable) using division, department, and job title coupled with the sampling data, which were used to assign a mean exposure.

1.6.2. Critical review of exposure assessment methods

The exposure assessment for the study by Rodrigues et al. (2020) was well described for the cohort overall, and it was a key strength that such detailed and extensive site-specific information and a large database of industrial hygiene measurements were available to assign mean exposure by work group and over manufacturing eras. In the earlier work, the authors identified a confidence for each exposure estimate (Rodrigues et al., 2019). [The Working Group noted that although confidence estimates were apparently available, they were not used in the later analysis (Rodrigues et al., 2020).] The number of samples used for isophorone in particular was not reported (although availability of at least 20 samples across all three facilities was a criterion for inclusion) and so the strength of the exposure assessment for isophorone exposure with respect to hygiene data was difficult to evaluate. In addition, all 126 836 employees were categorized into 1 of 10 exposure groups per manufacturing era (three eras per facility), so there was likely to be substantial heterogeneity within each group. Job histories were only available for the parts of workers’ careers that were spent at these facilities; this may not be a weakness if workers typically spent most or all of their career in these facilities, but if they spent significant time periods employed in other jobs, there could be an issue with exposure misclassification. An additional weakness was that industrial hygiene measurements were not randomly collected, such that their generalizability across the combination of chemical, facility, exposure group, and era was unclear.

Oral administration (gavage)

In a well-conducted study that complied with Good Laboratory Practice (GLP), groups of 50 male and 50 female B6C3F₁ mice (age, 6–8 weeks) were given isophorone (purity, ≥ 94%) at a dose of 0, 250, or 500 mg/kg body weight (bw) in corn oil, for the control group and the groups at the lower and higher dose, respectively, by gavage, 5 days per week, for 103 weeks (NTP, 1986). At study termination, survival was 16/50, 16/50, and 19/50 in males, and 26/50, 35/50, and 34/50 in females, for the control group and the groups at the lower and higher dose, respectively. In females, there was a significant positive trend in survival in females, and the survival rates at both doses were significantly higher than that of the controls. In males, the survival rates of the treated groups were similar to that of the control group. A significant decrease in body-weight gain was observed in females at the higher dose (about 5% lower at the end of the exposure period compared with controls). No significant difference in body weight was observed in females at the lower dose or in males at either dose. All mice (except one missing male mouse in the control group) underwent complete necropsy, and histopathological examination was performed on all gross lesions, and main tissues and organs.

In male mice, there was a significant positive trend in the incidence of hepatocellular adenoma or carcinoma (combined) (P = 0.027, incidental tumour trend test; P = 0.025, Cochran–Armitage trend test), with incidence being significantly increased at the higher dose: control, 18/48 (37%); lower dose, 18/50 (36%); and higher dose, 29/50 (58%); P = 0.033, Fisher exact test. The incidence of hepatocellular adenoma was 6/48, 7/50, and 13/50, and the incidence of hepatocellular carcinoma was 14/48 (29%), 13/50 (26%), and 22/50 (44%), for the control group and the groups at the lower and higher dose, respectively. The incidence of hepatocellular carcinoma and of hepatocellular adenoma or carcinoma (combined) in males at the higher dose exceeded the upper bound of the ranges observed in historical controls in this laboratory – hepatocellular carcinoma, 218/1034 (mean ± standard deviation, 21.1 ± 7.6%); range, 8.3–36%; and hepatocellular adenoma or carcinoma (combined), 335/1034 (32.4 ± 9.4%); range, 14–50%. There was a significant positive trend in the incidence of fibrosarcoma of the subcutis (P = 0.044, life-table trend test; P = 0.019, incidental tumour trend test; P = 0.023, Cochran–Armitage trend test), with the increase in incidence being significant at the higher dose: control, 3/48; lower dose, 4/50; higher dose, 10/50; P = 0.042, Fisher exact test; P = 0.009, incidental tumour test. The incidence of fibroma of the subcutis was 0/48, 2/50, and 3/50, for the control group and the groups at the lower and higher dose, respectively. In the subcutis, one sarcoma was observed at the higher dose, and one neurofibrosarcoma was observed in controls. In the skin, the incidence of fibroma was 2/48, 1/50, and 0/50, for the control group and the groups at the lower and higher dose, respectively; and one neurofibrosarcoma was observed at the lower dose. There was a significant positive trend (P = 0.034, incidental tumour trend test; P = 0.033, Cochran–Armitage trend test) in the incidence of mesenchymal tumours (fibroma, fibrosarcoma, neurofibrosarcoma or sarcoma, combined) of the integumentary system (skin and subcutis, combined) with the incidence being significantly increased at the higher dose: control, 6/48 (12%); lower dose, 8/50 (16%); and higher dose, 14/50 (28%); P = 0.048, Fisher exact test; P = 0.050, incidental tumour test. In addition, the incidence of mesenchymal tumours (fibroma, fibrosarcoma, neurofibrosarcoma, or sarcoma, combined) of the integumentary system in male mice at the higher dose exceeded the upper bound of the range observed in historical controls in this laboratory – 70/1040 (6.7 ± 6.5%); range, 0–22%. There was a significant increase in the incidence of malignant lymphoma of the lymphohaematopoietic system at the lower dose: control, 7/48 (15%); lower dose, 18/50 (36%); higher dose, 5/50 (10%); P = 0.013, Fisher exact test. The incidence of malignant lymphoma of the lymphohaematopoietic system at the lower dose exceeded the upper bound of the range observed in historical controls in this laboratory – 126/1040 (12.1 ± 5.1%); range, 2.1–22.0%.

In female mice, there was no significant increase in tumour incidence in any of the treated groups compared with controls (NTP, 1986).

[The Working Group noted this was a well-conducted GLP study that used multiple doses, an adequate number of animals per group, an adequate duration of exposure and observation, and males and females.]

3.2.1. Oral administration (gavage)

In a well-conducted GLP study, groups of 50 male and 50 female F344/N rats (age, 6–7 weeks) were given isophorone (purity, ≥ 94%) at a dose of 0, 250, or 500 mg/kg bw in corn oil, for the control group and the groups at the lower and higher dose, respectively, by gavage, 5 days per week, for 103 weeks (NTP, 1986). There was a significant negative trend in survival in males; survival at study termination was: 33/50, 33/50, and 14/50 in males, and 30/50, 23/50, and 20/50 in females, for the control group and the groups at the lower and higher dose, respectively. The survival rate was significantly decreased in males at the higher dose. Gavage errors accounted for all of the 36 accidental deaths of male and female rats. Deaths related to gavage error increased with dose in females. A decrease in body weight was observed in males and females at the higher dose (7% lower at the end of the exposure period). All rats underwent complete necropsy, and histopathological examination was performed on all gross lesions, and main tissues and organs.

In male rats, there was a significant positive trend in the incidence of tubular cell adenoma or tubular cell adenocarcinoma (combined) of the kidney (P = 0.014, life-table trend test; P = 0.034, incidental tumour trend test), with incidence being significantly increased at the higher dose: control 0/50, lower dose, 3/50 (6%); higher dose, 3/50 (6%); P = 0.025, life-table test. The incidence of tubular cell adenoma or carcinoma (combined) of the kidney in both treated groups (3/50, 6%) was higher by 15-fold than the incidence reported for historical controls in this laboratory (4/1091, 0.4%). [The range of incidence of kidney tubular cell tumours in the historical control groups was not reported.] The incidence of tubular cell adenoma of the kidney was 0/50, 0/50, and 2/50, and the incidence of tubular cell adenocarcinoma of the kidney was 0/50, 3/50, and 1/50, for the control group and the groups at the lower and higher dose, respectively. There was a significant positive trend in the incidence of carcinoma of the preputial gland (P = 0.002, life-table trend test; P = 0.019, incidental tumour trend test; P = 0.006, Cochran–Armitage trend test), with the incidence being significantly increased at the higher dose – control, 0/50; lower dose, 0/50; and higher dose, 5/50 (10%); P = 0.028 Fisher exact test; P = 0.012, life-table test. The incidence of preputial cell carcinoma in historical controls was 19/1094 (1.7%). [The range of incidence of preputial gland carcinoma in historical control groups was not reported.]

Regarding non-neoplastic lesions in male rats, the incidence of tubular cell hyperplasia of the kidney was 0/50, 1/50, and 4/50, for the control group and the groups at the lower and higher dose, respectively.

In female rats, there was no significant increase in tumour incidence in any of the treated groups compared with controls (NTP, 1986).

[The Working Group noted that this was a well-conducted GLP study that used multiple doses, an adequate number of animals per group, an adequate duration of exposure and observation, and males and females. The Working Group also noted the high number of accidental gavage-related deaths in males and females in this study.]

3.2.2. Inhalation

Two groups of 10 male and 10 female Wistar rats [age not reported; body weight, approximately 140 g at 2 weeks before exposure] were exposed by inhalation (whole-body exposure) to isophorone at a concentration of 0 (control) or 250 ppm [1413 mg/m³] for 6 hours per day, 5 days per week, for 18 months. In treated rats, slight conjunctivitis and irritation of the nasal mucosa with a bloody discharge were observed. Frequent haemorrhages were found with oedema in the alveoli of the lungs, and microvacuolization was found in the liver. Tumour incidence was not reported (Dutertre-Catella, 1976). [The Working Group noted that this study was inadequate for the evaluation of the carcinogenicity of isophorone in experimental animals due to incomplete reporting, and lack of details regarding the study design, postmortem evaluation, and tumour incidence.]

4.1.1. Humans

Only one study was performed to study dermal absorption of isophorone in human skin. Low permeability (< 20%) was demonstrated, and no potential to damage the skin was observed after dermal application of isophorone (25 mg/mL) for 60 minutes (Fasano & McDougal, 2008). Urinary isophorone was detected at higher concentrations in 20 lung cancer patients than in 20 healthy control volunteers (Hanai et al., 2012). [The Working Group noted that there was pharmacokinetic disposition, but the significance of this study was not clear.]

4.1.2. Experimental systems

(b) Metabolism and excretion

Metabolites of isophorone have been identified in the urine of animals exposed to isophorone by oral administration (Dutertre-Catella et al., 1970, 1978; Dutertre-Catella, 1976). In rabbits and rats treated with isophorone (1 g/kg bw) by gavage, the following metabolites were identified in the urine: 3,5,5-trimethyl-2-cyclohexene-1-ol (isophorol) and its glucuronic conjugate; 3-carboxy-5,5-dimethyl-2-cyclohexene-l-one; 3,5,5-trimethylcyclohexan-1-one (dihydroisophorone); and cis- and trans-3,5,5-trimethylcyclohexan-1-ol (Dutertre-Catella, 1976; Dutertre-Catella et al., 1978). After isophorone ingestion, more dihydroisophorone and less isophorol was found in rat urine than in rabbit urine. Truhaut et al. (1973) identified isophorone in the urine of rats and rabbits given 3,5,5-trimethylcyclohexan-1-one (dihydroisophorone) (1 g/kg bw) by gavage (Truhaut et al., 1973). Dutertre-Catella (1976) and Dutertre-Catella et al. (1978) proposed that the metabolism of isophorone involves methyl oxidation to 3-carboxy-5,5-dimethyl-2-cyclohexene-l-one, reduction of the ketone group to isophorol, reduction of the ring double bond to dihydroisophorone, and dismutation of dihydroisophorone to cis- and trans-3,5,5-trimethylcyclohexan-1-ol (Dutertre-Catella, 1976; Dutertre-Catella et al., 1978). The metabolic pathways of isophorone are presented in Fig. 4.1. [The Working Group noted that the enzymes implicated in the metabolism of isophorone in rodents have not been identified.]

Fig. 4.1

Metabolic scheme for isophorone.

Studies in rats and rabbits suggest that urine is the predominant route of elimination of isophorone (Dutertre-Catella et al., 1970, 1978; Truhaut et al., 1973; Dutertre-Catella, 1976). After oral administration of isophorone, rats and rabbits excreted unchanged isophorone and metabolites in the urine, and unchanged isophorone in the expired air. Forty-eight hours after ingestion of dihydroisophorone (a metabolite of isophorone) by rats or rabbits, an estimated 50–70% of the administered dose was present as glucuronic conjugates in the urine (Dutertre-Catella, 1976; Dutertre-Catella et al., 1978). [The Working Group noted that the rate and extent of excretion were not reported.]

4.2.1. Is genotoxic

(b) Experimental systems

(i) Non-human mammals in vivo

See Table 4.1.

Table 4.1

Genetic and related effects of isophorone in non-human mammals in vivo.

Intraperitoneal injection of isophorone did not result in micronucleus formation in bone marrow cells (polychromatic erythrocytes) of male and female CD-1 mice (O’Donoghue et al., 1988).

(ii) Non-human mammalian cells in vitro

See Table 4.2.

Table 4.2

Genetic and related effects of isophorone in non-human mammalian cells in vitro.

Tests for unscheduled DNA synthesis in rat primary hepatocytes treated with isophorone were reported to give weakly positive results in two studies (Selden et al., 1994) [the Working Group noted the use of a non-standard protocol in one of the studies], although the same test system gave a negative result in a third study (O’Donoghue et al., 1988). Gene mutation studies in mouse L5178Y/Tk^+/− lymphoma cells treated with isophorone (at concentrations ranging from 400 to 1500 µg/mL) gave one positive result without metabolic activation at non-cytotoxic concentrations (McGregor et al., 1988), three additional positive results without metabolic activation (NTP, 1986; Tennant et al., 1987; Honma et al., 1999b) [but the Working Group noted concerns over cytotoxicity and/or study design], and three studies gave negative or equivocal results with and without metabolic activation (O’Donoghue et al., 1988; Sofuni et al., 1996; Honma et al., 1999a). Tests for chromosome aberrations were reported to give positive results with and without metabolic activation in one study in Chinese hamster lung fibroblast cells (Matsuoka et al., 1996) [the Working Group noted that a non-standard protocol was used and no indication of cytotoxicity was provided], but negative results were reported in three additional studies in Chinese hamster ovary cells (with isophorone at concentrations up to 1600 µg/mL) with and without metabolic activation (NTP, 1986; Tennant et al., 1987; Gulati et al., 1989). In Chinese hamster ovary cells, tests for sister-chromatid exchange were reported to give positive results without metabolic activation in two studies (Tennant et al., 1987; Gulati et al., 1989) [the Working Group noted that no indication of cytotoxicity was provided], but results were negative with or without metabolic activation (with isophorone at concentrations of up to 1000 µg/mL) in a third study (NTP, 1986).

(iii) Non-mammalian experimental systems

See Table 4.3.

Table 4.3

Genetic and related effects of isophorone in non-mammalian experimental systems.

Isophorone did not induce micronucleus formation in a hen’s egg test (Greywe et al., 2012). Isophorone exposure by feeding or injection gave negative results in the sex-linked recessive lethal test in Drosophila melanogaster (Foureman et al., 1994). Isophorone did not induce mutation in Salmonella typhimurium in the presence or absence of metabolic activation (Mortelmans et al., 1986; NTP, 1986; Tennant et al., 1987; Kubo et al., 2002).

4.2.2. Evidence relevant to other key characteristics

4.2.3. High-throughput in vitro toxicity screening data evaluation

The analysis of the in vitro bioactivity of the agents reviewed in IARC Monographs Volume 130 was informed by data from high-throughput screening assays generated by the Toxicity Testing in t^he 21st Century (Tox21) and Toxicity Forecaster (ToxCast) research programmes of the government of the USA (Thomas et al., 2018). Isophorone was one of thousands of chemicals tested across the large assay battery of the Tox21 and ToxCast research programmes of the US EPA and the United States National Institutes of Health. Detailed information about the chemicals tested, assays used, and associated procedures for data analysis is publicly available (US EPA, 2021). A supplementary table (Annex 2, Supplementary material for Section 4, Mechanistic Evidence, web only; available from: https://publications.iarc.fr/611) provides a summary of the findings including the assay name, the corresponding key characteristic, the resulting “hit calls” both positive and negative, and any reported caution flags for isophorone. The results were generated with the software “kc-hits” (key characteristics of carcinogens – high-throughput screening discovery tool) (available from: https://gitlab.com/i1650/kc-hits) using the US EPA ToxCast and Tox21 assay data and the curated mapping of key characteristics to assays available at the time of the evaluations performed for the present monograph. Findings and interpretations from these high-throughput assays for isophorone are discussed below.

After mapping against the key characteristics of carcinogens, the ToxCast/Tox21 database contained 291 assays in which isophorone was tested. Of these, it was found to be active and without caution flags in seven assays relevant to the key characteristics of carcinogens. [The Working Group noted that the cytotoxic limit for isophorone is 14.8 µM.]

Isophorone was active in five assays mapped to key characteristic 8 (KC8), “modulates receptor-mediated effects”. It was active in one assay related to G protein-coupled receptor (GPCR) binding activity in human platelets with a half-maximal activity concentration (AC₅₀) of 22.7 µM. In HepG2 cells, isophorone activated the nuclear receptor subfamily 1, group I, member 2 (NR1I2) (AC₅₀, 48.7 µM); the nuclear receptor subfamily 1, group H, member 3 (NR1H3) (AC₅₀, 36 µM); the nuclear receptor subfamily 1, group H, member 2 (NR1H2) (AC₅₀, 63.3 µM); and the retinoid X receptor, β (RXRB) (AC₅₀, 51.4 µM).

In addition, isophorone was active in two assays mapped to KC10, “alters cell proliferation, cell death, or nutrient supply”, in HEK293 cells. The assay measurements were performed 32 and 40 hours after exposure, with AC₅₀s of 40.93 and 36.74 µM, respectively.