Methyl Methacrylate - PSL1

2.0 Summary of Information Critical to Assessment of "Toxic"

2.1 Identity, Properties, Production, and Uses

Methyl methacrylate (C₅H₈0₂) is a colourless, volatile liquid with an acrid, fruity odour. It has a relatively high vapour pressure (4000 Pa at 20°C ), moderate water solubility (15 800 mg/L), and a low log octanol:water partition coefficient (K_ow ) (1.38) (ACGIH, 1986; Dean, 1985; Tanii and Hashimoto, 1982). Methyl methacrylate is manufactured commercially using the acetone cyanohydrin process. Common analytical methods used to quantify and qualify acrylic compounds include gas chromatography (GC) (Shen and Woo, 1988); mass spectrometry (MS) (Gjos et al., 1983); GC/MS (Horna et al., 1986); nuclear magnetic resonance (NMR) (Shen and Woo, 1988); and infrared spectroscopy (O'Neill and Christensen, 1975).

Methyl methacrylate is not known to occur naturally and is not produced in Canada. Methyl methacrylate used in Canada is imported primarily from the United States. In 1988, Canada imported 20.11 kilotonnes from the United States and minor amounts from the United Kingdom (0.14 kilotonnes) and West Germany (0.04 kilotonnes). The amount of MMA forecast to be imported in 1993 is 24 kilotonnes (CPI, 1989).

Methyl methacrylate polymerizes easily, especially when heated or in the presence of hydrochloric acid. The polymer forms clear, ceramic-like resins and plastics, commonly known as Plexiglas and Lucite. In Canada, MMA is used captively in the production of cast acrylic sheet, acrylic emulsions, and molding and extrusion resins. Demand for these products increased sharply in the early 1980s but has moderated recently. Its use is expected to increase by 3 to 4% by 1993 (CPI, 1989).

2.2 Entry into the Environment

Methyl methacrylate can enter the environment during transport, bulk storage, and use. Data on emissions of MMA in Canada have not been identified. Estimates from the United States Toxic Chemical Release Inventory on the emissions to air, water, and soil from plants in the United States correspond to about 0.46% of production (TRI, 1989). Most emissions, i.e., 98%, were estimated to be to air, with very small amounts to water and soil. Assuming the same level of emissions in Canada and Canadian imports of 22 kilotonnes, it is estimated that 100 000 kg/yr or 11.4 kg/h are emitted into the Canadian environment. The major importers of MMA in Canada are located in Morrisburg and Niagara Falls, Ontario; therefore, releases from its use in manufacturing cast acrylic sheet, and molding and extrusion resins are expected to occur in southern Ontario.

2.3 Exposure-related Information

2.3.1 Fate

The environmental fate of MMA is determined by the rates of photolysis, hydrolysis, volatilization, and biotic degradation, as well as adsorption to sediment and soil, and bioconcentration by aquatic organisms. The significance of each of these factors is discussed in this subsection.

No data were available on the photolysis rates of MMA; however, because the ultraviolet/visible absorption maximum is 231 nm, MMA should not absorb radiation >290 nm (the radiation reaching the earth's surface) and photolyze. Free radicals formed in natural waters by the action of light might react with MMA; however, environmentally pertinent data are limited in this area. Because MMA is highly reactive with hydroxyl radicals, its lifetime in the atmosphere is short. The estimated half-life of MMA in the troposphere at a latitude such as that of Toronto will vary from <5 hours in summer to a few days in winter (Bunce, 1992). The reported photo-oxidation half-life of MMA is 1.1 to 9.7 hours (Howard et al., 1991). Methyl methacrylate is readily polymerized by light and heat (Hawley, 1981).

Hydrolysis, which is base-catalyzed, is not significant at neutral and acidic pH. Based on the measured second order hydrolysis rate constant of 200/(M · h) at 25°C and pH 11 (Ellington et al., 1987), the hydrolysis half-life of MMA is estimated to be 3.9 years at pH 7 and 14.4 days at pH 9 (Howard, 1989).

No field or laboratory data were identified on the rate of volatilization. Based on the Henry's Law constant, the half-life for evaporation from a river 1-metre deep with a 1 m/s current and 3 m/s wind is calculated to be 6.3 hours. Because of the high vapour pressure of MMA and its weak adsorption to soil, evaporation from soil is expected to be rapid (Howard, 1989).

Several studies have shown that MMA can be biodegraded. The aqueous aerobic degradation half-life is estimated to be 1 to 4 weeks and the anaerobic degradation half-life to be 4 to 16 weeks (Howard et al., 1991).

No data were identified on the adsorption of MMA to soil or sediment. From the log K_ow , the log K_oc is calculated to range from 1.17 to 2.13 (depending on the relationship to log K_ow chosen) which indicates that little adsorption to soil or sediment should occur. Although no studies have been conducted to measure the bioconcentration factor (BCF) for MMA, based on the equation by Veith et al. (1979) (which estimates BCF from the log K_ow ), the BCF is about 3. Methyl methacrylate is therefore not expected to bioconcentrate or biomagnify in food chains.

2.3.2 Concentrations

No information was found in the literature on MMA levels in any environmental medium in Canada. In a study conducted in Atlantic Canada to detect organic and inorganic contaminants in edible shellfish, MMA was not detected [detection limit (D.L.) =0.01 µg/g wet weight (w.w.)] in any of the 30 assayed samples from various locations (Environment Canada, 1989).

In a database containing 5700 entries on the frequency of organic compounds identified in water in the United States, MMA was listed four times, once in river water and three times in drinking water (Shackelford and Keith, 1976). One of the drinking water listings was a result of analyses of 204 water samples collected from 14 heavily industrialized river basins in the United States (Ewing and Chian, 1977). Methyl methacrylate was detected (D.L. = 1 µg/L) only once in the survey (10 µg/L in final tap water after chlorination in Chicago). It is not known how valid this number is as no additional information was provided.

Due to the lack of identified data, the environmental fate and concentrations of MMA were predicted using a Level III Fugacity Model (Mackay and Paterson, 1991; 1981; 1982; and Mackay et al., 1985) based on the physical and chemical properties of the substance, its transformation half-lives, and emission rates (Mackay et al., 1992). The following assumptions were included:

• MMA is used and emitted only in southern Ontario, in an area of approximately 170 000 km²;
• 95%, 4.5%, and 0.5% of emissions were to air, water, and soil, respectively; and
• total emissions were 100 000 kg/yr or 11.4 kg/h (see Subsection 2.2).

The following environmental concentrations were predicted:

• 2.44 x 10^-4 µg/m³ in air;
• 0.13 ng/L in surface water;
• 1.2 x 10^-6 µg/g in soil;
• 8.7 x 10^-8 µg/g in sediment; and
• 1.5 x 10^-7 µg/g in fish.

The environmental persistence was estimated to be about one day.

The MMA monomer may be present in food as a result of migration from food wrap made from polymethyl methacrylate. However, quantitative data on concentrations of MMA in food have not been identified. Migration of MMA from bone cement into prepared tissue media has also been reported in a limited study (IARC, 1979).

2.4 Effects-related Information

2.4.1 Experimental Animals and In Vitro

In all species studied, the acute toxicity of MMA was low, though in some cases, effects have been observed at the site of entry (i.e., the lungs after inhalation) after short periods of exposure to relatively low concentrations. [Indeed, intra-alveolar congestion/hemorrhage, pulmonary vasodilation, and edema were observed in rats exposed to concentrations as low as 100 ppm of MMA for 2, 3, or 4 hours (Raje et al., 1985)]. Inhalation of high concentrations of MMA for brief periods resulted in significant changes in pulmonary function, edema, hemorrhage, and congestion in the lung, irritation of the eye and mucous membranes, prostration and narcosis with death in 24 hours (Deichmann, 1941; Spealman et al., 1945; McLaughlin et al., 1973; Mir et al., 1974; Tansy et al., 1980a;

Raje et al., 1985; U.S. EPA, 1985). The 4-hour LC₅₀ s for MMA in rats ranged from 3750 to 7093 ppm (15 375 to 29 080 mg/m³) (Kennedy and Graepel, 1991; Tansy et al., 1980a) andthe 8-hour LC₅₀ s were 4634 ppm (19 000 mg/m³) in rats, rabbits, and guinea pigs (U.S. EPA, 1985). The oral LD₅₀ s ranged from 5.0 m/kg (4.7 g/kg) in dogs to 10.0 mL/kg (9.44g/kg) in rats (Spealman et al., 1945). The signs of toxicity following oral administration included decreased respiration, loss of reflexes, coma, corrosion of the stomach walls, and liver and kidney degeneration (Deichmann, 1941; Spealman et al., 1945).

In short-term, repeated-dose studies, death, decreases in body weight, cardiovascular effects, changes in respiration rate, increases in level of blood urea nitrogen, and pulmonary damage were observed after exposure to high concentrations of MMA. Mice were found to be more susceptible than rats; effects on the respiratory tract of mice were observed at the lowest tested concentration of 500 ppm (2050 mg/m³) for 10 days [IBT for NTP (1986)]. Histopathological effects were limited to those at the site of entry (i.e., the lung in inhalation studies). Renal effects were also reported in rats which were administered MMA by subcutaneous injection for 34 days, though available data in the published account of this study were inadequate for evaluation (Miller et al., 1982).

Gross or microscopic pathological effects reported in long-term, repeated-dose studies in rats, mice, and dogs are limited. In most studies conducted to date, animals have been exposed to MMA by inhalation; the effects most commonly observed in these investigations were decreases in body weight gain and irritation of the skin, nasal cavity, and eye at high concentrations [generally greater than or equal to 500 ppm (2050 mg/m³)] [IBT for Rohm and Haas (1977a); IBT and Batelle Northwest for NTP (1986)]. Other effects on the kidney, such as renal cortical necrosis and tubular degeneration, and liver necrosis have also been reported (Tansy et al., 1980b; NTP, 1986; Deichmann-Gruebler and Read, 1989). On the basis of decreases in final mean body weight and squamous metaplasia at the site of entry (i.e., the lung), the lowest reported "no-observed-effect-levels"(NOELs) and "lowest-observed-effect-levels (LOELs) in a subchronic inhalation bioassay in which several dose levels were administered were 250 and 500 ppm (1025 and 2050 mg/m³) in mice, respectively [IBT for Rohm and Haas (1977a); IBT and Batelle Northwest for NTP (1986)]. With the exception of effects at the site of entry, histopathological effects have not been observed in the two most extensive bioassays in rats at concentrations less than or equal to 1000 ppm (4100 mg/m³) [IBT for Rohm and Haas (1977a); IBT and Batelle Northwest for NTP (1986)]. In less extensive and well documented studies conducted by Tansy et al. (1976; 1980b; 1980c), effects on the trachea and some indications of liver damage in mice were recorded at the lowest concentration of 116 ppm, administered for 7 hours/day for 6 months (though the statistical significance of the pulmonary changes was not specified and similar effects were observed in some of the sham-exposed control animals). In a supplementary study, there was weak evidence of an effect on liver function (barbiturate sleeping time) in groups of 20 male mice administered "intermittent daily exposures" of 100 ppm for a total of 160 hours (Tansy et al., 1980c). Initial reports of reduced fat deposits after exposure for 3 months were not confirmed in repeat studies (Tansy et al., 1980b; 1980c).

In the few studies identified in which the chronic toxicity and carcinogenicity of MMA were investigated, the observed effects were generally similar to those reported in short- and long-term studies, including inflammation and epithelial hyperplasia of the nasal cavity and degeneration of the olfactory sensory epithelium. Based on the results of a well documented inhalation study in F344/N rats and B6C3F₁ mice conducted at Batelle Northwest Laboratories and reported by the NTP (1986) and Chan et al. (1988), it was concluded that there was no evidence of carcinogenicity of MMA for male F344/N rats and male and female B6C3F₁ mice exposed to 500 or 1000 ppm (2050 or 4100 mg/m³) and female rats exposed to 250 or 500 ppm (1025 or 2050 mg/m³). From observations of inflammation and degeneration of the olfactory epithelium and minimal increases in the numbers of alveolar macrophages in the nasal cavity in rats at all dose levels, the LOEL for rats was considered to be 250 ppm (1025 mg/m³). In mice, the LOEL was considered to be 500 ppm (2050 mg/m³) on the basis of lower mean body weights in exposed mice and localized histopathological effects at the site of entry (including inflammation and degeneration of the olfactory epithelium).

In earlier studies conducted by Hazelton Laboratories for Rohm and Haas (1977b; 1979), no treatment-related increases in tumour incidence occurred in either golden hamsters or albino rats (strain not reported) exposed to 0, 25, 100, or 400 ppm (0, 102.5, 410, or 1640 mg/m³) MMA 6 hours/day, 5 days/week for 18 months and 2 years, respectively. At the highest concentration, body weight decreased significantly in both species, mortality increased in hamsters, and the incidence of mild rhinitis increased slightly in the nasal mucosa in rats. The NOELs and LOELs were considered to be 100 ppm and 400 ppm (410 and 1640 mg/m³), respectively, in both species based on the observed effects on body weights (both species) and respiratory tract (rats) in the group exposed to the highest concentration.

In an early study (Borzelleca et al., 1964), the ratio of kidney weight to body weight was increased in a small group of female rats exposed to 2000 ppm of MMA in drinking water for 2 years. The LOEL was, therefore, considered to be 2000 ppm [about 146 mg/(kg b.w. · day) for females]; the NOEL was 60 ppm [about 5 mg/(kg b.w. · day) for females], though it should be noted that the administered doses varied considerably in this study. In another study of extremely small groups of beagle dogs (n=2) exposed to doses of up to 1500 ppm MMA [about 38 mg/(kg b.w. · day)] in their feed for 2 years, there were no treatment-related effects upon gross or histopathological examination, urinalysis, or hematology (Borzelleca et al., 1964).

Based on the limited available information, MMA has not been mutagenic in a number of standard in vitro studies in Salmonella typhimurium (Lijinsky and Andrews, 1980; Waegemaekers and Bensink, 1984). However, MMA has been mutagenic and clastogenic in mammalian cells in culture (Moore et al., 1988; Doerr et al., 1989; Ishidate et al., 1981; Galloway et al., 1985; NTP, 1986). At high atmospheric concentrations (1000 ppm), MMA induced chromosome damage in mice after a single, but not after multiple, exposures (Anderson and Richardson, 1976; U.S. EPA, 1985). Results of a dominant lethal study in male mice administered similarly high concentrations, however, were negative (Anderson and Hodge, 1976). The results of a bone marrow micronucleus test were also negative (Hachitani et al., 1981).

Based on the available studies on the developmental effects of MMA in different species, no significant differences were found in the number of dead or live fetuses, and litter size after inhalation or intraperitoneal (i.p.) exposure. Gross abnormalities in rats were observed only following intraperitoneal injection or inhalation during pregnancy of doses only slightly less than acute lethal doses [(125 mg/kg b.w.) by i.p. injection (Singh et al., 1972); 110 mg/L (110 000 mg/m³) by inhalation (Nicholas et al., 1979)]. In the former study, in which group sizes were small, Singh et al.,(1972) reported a dose-dependent increase in hemangiomas; effects observed in the mothers were not addressed. Nicholas et al.,(1979) reported decreases in maternal body weight gain associated with decreases in food consumption. Ossification was delayed, and there were early deaths, hematomas, and a decrease in fetal weight in the offspring of rats exposed to doses at which decreases in body weight gain in the mothers were observed (probably associated with decreases in food consumption).

Slight decreases in maternal weight gain and increased incidence of early resorptions were reported in rats exposed during pregnancy to 1000 ppm MMA (4100 mg/m³) (Hodge and Palmer, 1977). In a well documented study in rats, there was no embryo- or feto- toxicity and no increase in the incidence of malformations or variations following exposure during pregnancy to concentrations that ranged from 99 to 2028 ppm (406 to 8315 mg/m³; NOEL = 8315 mg/m³), though there was transient overt maternal toxicity (Solomon et al., 1991). Slight fetotoxicity manifested as a decrease in fetal weight in offspring of pregnant mice administered a single dose level of 116 ppm (476 mg/m³) (Tansy, 1975) or delayed ossification in rats at higher concentrations in the absence of maternal toxicity, has been reported (Hodge and Palmer, 1977; Luo et al., 1986). No effects were observed on reproductive performance in males in the only reproductive study identified (Anderson and Hodge, 1976).

Very few studies of the neurotoxicity of MMA have been identified. In an inhalation study by Innes and Tansy (1981) in which chloralose-urethanized mature male rats were exposed to 400 ppm (1640 mg/m³) of MMA for 60 minutes, depression of multiple-unit electrical activity in the lateral hypothalamus and ventral hippocampus was observed. Methyl methacrylate markedly impaired locomotor activity and learning, while significantly increasing aggressive behaviour in male rats orally administered 500 mg/kg b.w. for 21 days (Husain et al., 1985). In a separate study under the same experimental conditions (Husain et al., 1989), a significant increase in cholesterol (26%) and triglycerides (65%) and a slight decrease in the total phospholipid content of the sciatic nerve were noted.

2.4.2 Humans

Available clinical studies on human volunteers are restricted to investigations of effects on the skin following patch testing (Spealman et al., 1945; van Joost et al., 1988; Kaaber et al., 1979). In cross-sectional studies of populations exposed occupationally to MMA, effects on the skin (Rajaniemi and Tola, 1985), lungs (Jedrychowski, 1982; Andrews et al., 1979), nervous system (Seppalainen and Rajaniemi, 1984; Schwartz et al., 1989) and blood (NIOSH, 1976; Lang et al., 1986) have been examined. There are few quantitative data on exposure of these workers to MMA, however, and interpretation of several of these studies is complicated by concomitant exposure of the examined populations to other compounds.

Several historical cohort studies have been conducted to examine the mortality rate from cancer of the colon or rectum among male workers at two of the Rohm and Haas Company's plastics manufacturing plants, in Bristol, Pennsylvania and Knoxville, Tennessee (DeFonso and Maher, 1981; 1986; Maher and DeFonso, 1987a; 1987b; Walker et al., 1991). Earlier investigations are subsumed, however, by the most recent study by Walker et al., (1991), in which data were re-analyzed by the period of employment of the workers. In this investigation, the two cohorts were comprised of 6548 white men hired from January 1, 1946 to December 31, 1982 in the Bristol plant and 3381 white men hired between January 1, 1943 and December 31, 1982 in the Knoxville plant, followed either to death or to December 31, 1986. The population of workers at the Bristol plant was further divided into an early cohort (men employed at some time between 1933 and 1945 inclusive) and a late cohort (1946 to 1982, inclusive).

In the two cohorts with workers first hired at later dates (Knoxville and the late cohort at Bristol), there was no excess mortality due to cancer of the colon or rectum. In the early cohort, there was an apparent excess of deaths due to colon cancer. This cohort worked in conditions that likely involved high exposures to the vapour phase of ethyl acetate (EA) and MMA monomer, as well as to a variety of volatile by-products of the EA/MMA polymerization process. While the risk was highest in the subgroup of workers having the greatest cumulative exposure, there was no trend of increasing risk with increasing exposure after allowing for a long latency period. There was no systematic pattern of excess risk of cancer at any other site. For respiratory cancer, however, there was a significantly high Standardized Mortality Ratio (SMR = 1.44) in the Knoxville cohort, with no excess in either of the Bristol cohorts. In view of the large number of statistical estimates in this study, and the absence of clear dose-response trends, the evidence of an association of MMA with respiratory cancer is far from convincing. The apparent excess may have been due to statistical fluctuation or to confounding by other occupational exposures in the environment at the time (e.g., lead, ethylene dichloride, methylene chloride, and acrylonitrile). If the apparent excess reflects a true risk, it is noteworthy that the cause of the risk appears to have disappeared for workers hired after 1946, and this excess risk has not been confirmed in other studies of MMA-exposed workers.

In a limited study of a much smaller cohort of workers exposed for considerably shorter periods to MMA at two American Cyanamid Company plants between 1951 and 1983, there was no excess mortality for any type of cancer examined (Collins et al., 1989). However, it should be noted that workers were not exposed during the early period in which the excess mortality due to colon cancer was observed by Walker et al., (1991). Though there was a very weak indication of an excess of rectal cancer, and weak-to-moderate indication of an excess risk of lung cancer in a population-based (as opposed to cohort-based), case-control study of workers in Montreal exposed to MMA, these results are inconclusive due to the small numbers of subjects involved, i.e., 21 (Siemiatycki, 1991).

There was no increase in the number of sister chromatid exchange (SCE) in the peripheral lymphocytes in 31 male workers occupationally exposed to MMA [mean value/8 h ranged from 0.70 to 21.6 ppm (3.0 to 90 mg/m³)] in four factories as compared with the control group (7.85 + 2.66 vs. 7.49 + 2.33) (Marez et al., 1991). The distribution of SCE, however, was significantly higher in the group exposed to MMA at peak concentrations ranging from 114 to 400 ppm (467 to 1640 mg/m³), although the number of individuals in this exposure subgroup was small (n = 6). The control group consisted of 31 healthy male workers with similar mean age and smoking habits.

2.4.3 Ecotoxicology

The only information on aquatic toxicity of MMA includes acute toxicity tests on fish, water flea (Daphnia magna), and algae. Because MMA is a volatile compound, it requires aquatic testing procedures that eliminate potential losses through volatilization, i.e., maintenance of closed systems, with measured concentrations (not nominal) throughout the test period. As few of the aquatic toxicity studies met these criteria, actual concentrations may have been less than nominal, and the values presented may have been questionable.

Bailey et al., (1985) studied the toxicity of MMA to juvenile bluegill sunfish (Lepomis macrochirus) at 22°C under static and flow-through conditions of various durations (1 to 96 hours). The lowest value was achieved in the 96-hour flow-through assay, the LC ₅₀ being 191 mg/L. In shorter exposure flow-through assays, the LC₅₀ s were:420, 373, 367, 360, 360, and 356 mg/L for 1, 2, 4, 8, 16, and 24 hours exposure, respectively. The 96-hour LC ₅₀ for rainbow trout (Oncorhynchus mykiss) under flow-through conditions was >79 mg/L, the highest concentration to which fish were exposed. Sublethal/behavioural responses were noted among the fish in the 40 and 79 mg/L dose groups (Bowman, 1990).

The 24-hour EC₅₀ for immobilization of Daphnia magna was 720 mg/L, with the extrapolated EC₀ and EC₁₀₀ being 502 and 1042 mg/L, respectively (Bringmann and Kuhn, 1982). The 24-hour LC₅₀ for Daphnia magna was 1760 mg/L and for the LC₀ and LC₁₀₀ , 875 and 2500 mg/L, respectively (Bringmann and Kuhn, 1977). In a similar study, the toxicity threshold for onset of inhibition of cell multiplication was 447 mg/L for the flagellate protozoan Entosiphon sulcatum after 72 hours of exposure (Bringmann, 1978). These studies were done in static, open systems and only the nominal concentrations were reported.

Toxicity thresholds for onset of inhibition of cell multiplication following 8 days of exposure to MMA were 120 mg/L for the blue green alga Microcystis aeruginosa and 37 mg/L for the green alga Scenedesmus quadricauda, at pH 7 (Bringmann and Kuhn, 1976; 1978a; 1978b). The 96-hour LC₅₀ for the green alga Selenastrum capricornutum was 170 mg/L with a NOEL of 100 mg/L (Forbis, 1990). No studies on effects on aquatic vascular plants were identified.

Data on the acute or chronic effects of MMA on terrestrial organisms are sparse; field or laboratory studies on birds, terrestrial invertebrates, or terrestrial plants have not been identified. Although there were no field data on the effects on wild mammals, toxicity studies have been conducted on laboratory mammals and can be used for extrapolation to effects on wild mammals (see Subsection 2.4.1).