The Committee evaluated a group of 31 flavouring agents¹ that included aliphatic acyclic diols, triols, and related substances (see Table 1) using the Procedure for the Safety Evaluation of Flavouring Agents (see Figure 1, Introduction). All members of this group are aliphatic acyclic primary alcohols, aldehydes, acids, or related esters with one or more additional oxygenated functional groups. The group contains four subgroups: glycerol (No. 909) and 15 related glycerol esters and acetals (Nos 910, 911, 912, 913, 914, 915, 916, 917, 918, 919, 920, 921, 922, 923, and 924); propylene glycol (No. 925) and four related esters, acetals, and ketals (Nos 926, 927, 928, and 929); lactic acid (No. 930) and four lactate esters (Nos 931, 932, 934, and 935); and pyruvic acid (No. 936), its corresponding aldehyde (No. 937), two pyruvate esters (Nos 938 and 939) and one acetal of pyruvic acid (No. 933).

The Committee previously evaluated three members of the group. Glycerol (No. 909) was considered at the twentieth meeting, when an ADI "not specified" was established (Annex 1, reference 41). Propylene glycol (No. 925) was considered at the seventh meeting, when an ADI of 0–20 mg/kg bw was established (Annex 1, reference 7); it was further considered at the seventeenth meeting, when the ADI was increased to 0–25 mg/kg bw (Annex 1, reference 32). Ethyl lactate (No. 931) was considered at the eleventh (Annex 1, reference 14), twenty-third (Annex 1, reference 50), twenty-fourth (Annex 1, reference 53), and twenty-sixth meetings (Annex 1, reference 59). At its twenty-sixth meeting, the Committee included ethyl lactate in the group ADI ‘not specified’² with lactic acid.

Nine of the 31 substances (Nos 909, 929, 930, 931, 932, 934, 936, 937, and 938) have been detected as natural components of foods in cocoa, milk, cider, cognac, asparagus, tomato, and mushrooms (Maarse et al., 1999).

The total annual production of the 31 substances in this group of flavouring agents for use in food was reported to be 140 000 kg in Europe (International Organization of the Flavor Industry, 1995) and 21 000 000 kg in the USA (Lucas et al., 1999). These values are equivalent to a total daily per capita intake of 20 000 µg in Europe and 2 800 000 µg in the USA. The large difference in the annual volume of production between Europe and the USA is due to the inclusion in the USA of figures on the use of glycerol, triacetin, and propylene glycol as solvents in the preparation of compound flavour mixtures.

In Europe, three flavouring agents, namely, glycerol (17 000 µg/day), ethyl lactate (1900 µg/day), and butyl lactate (380 µg/day), accounted for approximately 97% of the total per capita intake. In the USA, three substances, namely glycerol (220 000 µg/day), triacetin (83 000 µg/day), and propylene glycol (2 400 000 µg/day), accounted for 96% of the total annual daily per capita intake. The per capita intakes of individual substances are shown in Table 2.

The aliphatic esters of propylene glycol, lactic acid and pyruvic acid and their parent compounds would all be expected to be readily absorbed from the gastrointestinal tract. Hydrolysis of the aliphatic esters is catalysed largely by hepatic esterases to give the component alcohol and carboxylic acid or aldehyde. After hydrolysis of glycerol esters in the intestine, glycerol is also readily absorbed. Glycerol, pyruvic acid, and lactic acid are endogenous in humans. Glycerol and pyruvic acid are metabolized completely and are not excreted. Lactic acid is also mainly metabolized, although urinary excretion may occur if the blood concentration is high. Propylene glycol can be metabolized, but high doses are likely to be excreted largely unchanged in the urine.

Glycerol is metabolized via the glycolytic pathway after it has been converted in the liver to glycerol-3-phosphate. Glycerol-3-phosphate is then oxidized to yield dihydroxyacetone phosphate, which is isomerized to glyceraldehyde-3-phosphate, eventually yielding pyruvic acid.

Pyruvic acid follows two primary routes of metabolism. Under aerobic conditions, it is converted to acetyl coenzyme A and enters the citric acid cycle. Under anaerobic conditions, primarily in muscles as a result of strenuous physical activity, pyruvic acid is reduced by lactic dehydrogenase to lactic acid.

Lactic acid diffuses through muscle tissue and is transported to the liver in the bloodstream. In the liver, it is converted to glucose by gluconeogenesis. Lactic acid can also be further catabolized in the lactic acid cycle (also known as the Cori cycle).

Propylene gycol can be oxidized to lactic acid via two biochemical pathways. If propylene glycol is phosphorylated, it can be converted to acetol phosphate, lactaldehyde phosphate, lactyl phosphate, and then lactic acid. If it is not phosphorylated, propylene glycol is successively oxidized to lactaldehyde, methylgloyoxal, and lactic acid.

Step 1 Twenty-eight of the 31 flavouring agents in this group are linear, simple branched aliphatic compounds. Twenty-two of these are in structural class I (Cramer et al., 1978) because they contain fewer than three different types of functional group (Nos 909, 913, 918–927, and 930–939). Six of these 28 substances are in structural class III because they contain three or more different types of functional group (Nos 910, 911, 914–917). The three remaining substances in the group are in structural class III because they are cyclic acetals and ketals (Nos 912, 928, and 929).

Step 2 The data on the metabolism of individual members of the group were sufficient to allow conclusions about their probable metabolic fate. The aliphatic esters of propylene glycol (Nos 926, 927, 928, and 929), lactic acid (Nos 931, 932, 934, and 935), and pyruvic acid (Nos 938 and 939) are expected to be hydrolysed to their component alcohol and carboxylic acid. Glycerol esters (Nos 910, 911, 912, 913, 914, 915, 916, 917, 918, 919, 920, 921, 922, 923, and 924) are expected to be hydrolysed to glycerol and carboxylic acids. Esters of propylene glycol are expected to be hydrolysed to propylene glycol and component acid. Esters of lactic acid and pyruvic acid are expected to be hydrolysed to lactic acid and pyruvic acid, respectively, and the corresponding alcohol. Acetals (Nos 927 and 933) are expected to be hydrolysed to their component alcohols and aldehydes, while ketals (Nos 928 and 929) are expected to be hydrolysed to their component ketones and alcohols. Glycerol (No. 909), lactic acid (No. 930), and pyruvic acid (No. 936) are endogeneous and are metabolized through the glycolytic and citric acid pathways. Propylene glycol (No. 925) is oxidized to lactic acid. For all substances in this group, therefore, the evaluation proceeded via the A side of the scheme.

Step A3 The daily per capita intakes of 22 of the substances in this group are below the threshold of concern for their respective structural classes (class I, 1800 µg; class III, 90 µg). These substances would not be expected to be of safety concern. The daily per capita intakes of the remaining nine substances (Nos 909, 910, 914, 916, 920, 925, 926, 930, and 931) exceed the threshold of concern for their respective structural classes. Evaluation of these substances therefore proceeds to step A4.

Step A4 Glycerol (No. 909), lactic acid (No. 930), and ethyl lactate (No. 931) are endogenous in humans and are therefore not expected to be a safety concern. Triacetin (No. 920), 3-oxohexanoic acid glyceride (No. 910), 3-oxodecanoic acid glyceride (No. 914), and 3-oxotetradecanoic acid glyceride (No. 916) are glycerol esters and are hydrolysed to glycerol. Propylene glycol (No. 925) and propylene glycol stearate (No. 926) are not endogenous in humans; however, the ester is expected to be hydrolysed to propylene glycol and stearic acid. Propylene glycol is known to be oxidized to lactic acid in mammals. These substances would therefore not be expected to be a safety concern.

In the unlikely event that all 23 substances in structural class I were to be consumed concurrently on a daily basis, the estimated per capita consumption in Europe and the USA would exceed the human intake threshold for class I. The estimated per capita consumption in Europe and the USA for combined intake of the eight flavouring agents in structural class III would also exceed the human intake threshold for class III. Given that the substances are expected to be efficiently metabolized by known metabolic pathways, the Committee considered that the combined intake would not give rise to concerns about safety.

On the basis of the predicted metabolism, the Committee concluded that the 31 aliphatic acyclic diols, triols, and related substances in this group would not raise safety concerns at the current levels of intake when used as flavouring agents. In applying the Procedure, the Committee noted that all of the available data on toxicity are consistent with the results of the safety evaluation.

This monograph summarizes the key data relevant to the evaluation of the 31 flavouring agents in this group. All members of this group are aliphatic acyclic primary alcohols, aldehydes, acids, or related esters with one or more additional oxygenated functional groups. The group consists of four subgroups: glycerol (No. 909) and 15 related glycerol esters and acetals (Nos 910, 911, 912, 913, 914, 915, 916, 917, 918, 919, 920, 921, 922, 923, and 924); propylene glycol (No. 925) and four related esters, acetals, and ketals (Nos 926, 927, 928, and 929); lactic acid (No. 930) and four lactate esters (Nos 931, 932, 934, and 935); and pyruvic acid (No. 936), its corresponding aldehyde (No. 937), two pyruvate esters (Nos 938 and 939) and one acetal of pyruvic acid (No. 933).

Quantitative data on natural occurrence and consumption ratios have been reported for seven flavouring agents in the group, which indicate that they are consumed predominantly from traditional foods (i.e., consumption ratio > 1) (Stofberg & Kirschman, 1985; Stofberg & Grundschober, 1987) (Table 2).

The daily per capita intake of this group of flavouring substances is 20 000 µg/day in Europe (330 µg/kg bw per day) and 2 500 000 µg/day in the USA (14 000 µg/kg bw per day). Glycerol accounted for approximately 85% (17 000 µg/day) of the total per capita intake in Europe; in the USA, propylene glycol accounted for 96% (2 400 000 µg/day) of the total per capita intake.

2.3.1 Biochemical data

Little specific information was available on the absorption and transformation of individual members of this group of flavouring substances. The esters, acetals, and ketals of glycerol, lactic acid, and pyruvic acid would be expected to be readily absorbed, as would the parent compounds. After hydrolysis of glycerol esters in the intestine, glycerol is readily absorbed. Glycerol and pyruvic acid are metabolized completely and are not excreted. Lactic acid is also largely metabolized, although urinary excretion may occur if the blood concentration is high. Propylene glycol can be metabolized, but at high concentrations is likely to be largely excreted unchanged in the urine.

Propylene glycol given orally to three persons at a dose of 1038 mg (0.017 g/kg bw) was rapidly absorbed and eliminated in the urine and saliva (Hanzlik et al., 1939). In a study of the pharmacokinetics of propylene glycol in humans, multiple oral doses were rapidly absorbed, and its rate of clearance from blood was dose-dependent (Yu et al., 1985).

In studies of the minor pathways of metabolism of propylene glycol, administration to rats in drinking-water resulted in excretion unchanged in the urine (Van Winkle, 1941). Propylene glycol given orally to rabbits was conjugated with glucuronic acid and excreted in the urine (Miura, 1911; Fellows et al., 1947).

In general, aliphatic esters of propylene glycol, lactic acid, and pyruvic acid are expected to be hydrolysed to their component alcohol and carboxylic acids. The hydrolysis is catalysed by classes of enzymes recognized as carboxylesterases or esterases (Heymann, 1980), the most important of which are the B-esterases, which, in mammals, predominate in hepatocytes (Heymann, 1980; Anders, 1989). The rates of hydrolysis follow first-order kinetics, with hydrolysis of the straight-chain esters occurring approximately 100 times more rapidly than that of branched-chain esters (Butterworth et al., 1975; Longland et al., 1977; Grundschober, 1977; Leegwater & van Straten, 1979).

Glycerol esters are hydrolysed to glycerol and the corresponding carboxylic acids (see Figure 1). The hydrolysis is catalysed by intestinal lipase (Tietz, 1986), which attacks the ester bonds at carbons 1 and 3. The ester bond at carbon 2 is more resistant to hydrolysis, possibly because of its stereochemistry and steric hindrance. The beta-monoglyceride can, however, spontaneously isomerise to the alpha-form (3-acylglycerol), permitting further hydrolysis to yield glycerol.

The rate of hydrolysis of glycerol esters depends on the surface area of the lipid–water interfaces, which increases greatly with the churning peristaltic movements of the intestine and the emulsifying action of bile acids. Lipase is rapidly denatured at these interfaces; however, colipase, a pancreatic protein that forms a 1:1 complex with lipase, inhibits the surface denaturation of lipase and anchors it to the lipid–water interface (Voet & Voet, 1990).

Studies of the hydrolysis of the glycerol fatty acid esters (tributyrin (No. 922) (Pilz, 1959; Pilz & Johann, 1967), glycerol 5-hydroxydecanoate (No. 923) (Als, 1975), and glycerol 5-hydroxydodecanoate (No. 924) (Als, 1975) showed complete hydrolysis to glycerol and the corresponding fatty acids, butyric acid, 5-hydroxy-decanoic acid, and 5-hydroxydodecanoic acid, respectively.

Hydroxylated and keto acids formed by hydrolysis of glycerol esters such as 5-hydroxydecanoic acid and 5-hydroxydodecanoic acid may form lactones by acid-catalysed intramolecular cyclization to yield five-member rings (see Figure 2). In aqueous media, equilibrium is established between the open-chain hydroxy-carboxylic acid and the lactone. At basic pH, the equilibrium favours the open-chain hydroxycarboxylate anion, but the lactone predominates at acidic pH. 5-Hydroxy-decanoic acid and 5-hydroxydodecanoic acid may form the delta-lactones delta-decalactone and delta-dodecalactone, respectively. Their metabolic fate can be predicted on the basis of an analogy with the known biotransformation of structurally related aliphatic lactones previously considered by the Committee (Annex 1, reference 132). Linear saturated 5-hydroxycarboxylic acids, which are formed from delta-lactones, are converted, via acetyl coenzyme A (CoA) to hydroxythioesters, which then undergo beta-oxidation and cleavage to yield an acetyl CoA fragment and a new beta-hydroxy-thioester reduced by two carbons. Even-numbered carbon acids continue to be oxidized and cleaved to yield acetyl CoA, while odd-numbered carbon acids yield acetyl CoA and propinyl CoA. Acetyl CoA enters the citric acid cycle directly, while propionyl CoA is transformed into succinyl CoA, which then enters the citric acid cycle.

Esters of propylene glycol are hydrolysed to propylene glycol and their component acid. In the presence of pancreatic lipase, propylene glycol stearate (No. 926) was hydrolysed to propylene glycol and stearic acid (Balls & Matlock, 1938).

Esters of lactic acid are hydrolysed to lactic acid and the corresponding alcohol. In rat plasma, ethyl lactate (No. 931) was hydrolysed to ethyl alcohol and lactic acid (Falke et al., 1981).

Esters of pyruvic acid are expected to be hydrolysed to pyruvic acid and the corresponding alcohol.

In general, acetals are hydrolysed to their component alcohols and aldehydes. Studies on the hydrolysis of 1,2,3-tris[(1´-ethoxy)ethoxy] propane (No. 913), which is readily hydrolysed to yield acetaldehyde and glycerol (DeSimone, 1976), support this conclusion.

Acetals of propylene glycol have also been shown to be hydrolysed to their component alcohol and aldehyde. In vitro, 1,2-di[(1´-ethoxy)ethoxy]propane (No. 927) was completely hydrolysed to acetaldehyde and propylene glycol (DeSimone, 1976). Potassium 2-(1´-ethoxy) ethoxypropanoate (No. 933), an acetal of lactic acid, was completely hydrolysed to lactic acid, acetaldehyde, and ethanol in simulated stomach fluids (Moreno et al., 1984). Aldehydes are oxidized to their corresponding carboxylic acids, which are subsequently metabolized through known biochemical pathways (Voet & Voet, 1990). For example, pyruvaldehyde (No. 937), the aldehyde of pyruvic acid, was metabolized to pyruvic acid when incubated with rat liver homogenate (Bonsignore et al., 1968).

Ketals are hydrolysed to their component ketones and alcohols. The related compound, benzaldehyde propylene glycol acetal, was hydrolysed in simulated gastric fluid and, to a lesser extent, in intestinal fluid (Morgareidge, 1962). Similarly, 2,2,4-trimethyl-1,3-oxacyclopentane (No. 929) would be expected to be hydrolysed in humans to yield acetone and propylene glycol.

Glycerol is endogenous in the human body. It enters the glycolytic pathway after its conversion in the liver to glycerol-3-phosphate by glycerol kinase. Glycerol-3-phosphate is then oxidized by glycerol-3-phosphate dehydrogenase to yield dihydroxyacetone phosphate (see Figure 3), which is then isomerized to glyceral-dehyde-3-phosphate, eventually yielding pyruvic acid.

Pyruvic acid is endogenous in the human body. It is a critical metabolic intermediate, and its fate depends on the oxidation state of the cell (see Figure 4). Under aerobic conditions, pyruvic acid is converted to acetyl CoA and enters the citric acid cycle, where it is completely metabolized. Under anaerobic conditions, lactate dehydrogenase catalyses the reduction of pyruvic acid to lactic acid and the oxidation of NADH to NAD⁺, primarily in muscles.

Lactic acid diffuses from the muscles and is transported through the bloodstream to oxygen-rich tissues such as the heart and liver, where it is catabolized further through the lactic acid cycle (also known as the Cori cycle) (see Figure 5), or converted to glucose via gluconeogenesis. Even in fully oxygenated muscle tissue, as much as 50% of the metabolized glucose is converted to lactic acid by way of pyruvic acid (Voet & Voet, 1990).

In resting women who received intravenous injections of [2-¹⁴C]pyruvate, analysis of blood glucose 1 h later showed 96% conversion of pyruvic acid to glucose (Hostetler et al., 1969). When [2-¹⁴C]pyruvate was incubated with liver slices from fasted normal rats, 86% had been used after 90 min of incubation. Of the radiolabel associated with metabolized pyruvic acid, 23% was associated with glycogen and glucose, 16% with CO₂, and 16% with lactic acid. In the presence of glycerol, the use of pyruvic acid was increased to 95%, accompanied by a decrease in conversion to glycogen (16%) and CO₂ (8.6%) and an increase in the production of lactic acid (Teng et al., 1953).

(v) Metabolism of propylene glycol (No. 925)

Propylene glycol can be oxidized to lactic acid via one of two pathways, depending on whether the glycol is phosphorylated (Rudney, 1954; Miller & Bazzano, 1965). In studies in vitro with rat liver, the free glycol was successively oxidized to lactaldehyde, methylglyoxal (pyuvaldehyde), and lactic acid (see pathway 1, Figure 6) (Ting et al., 1964; Miller & Bazzano, 1965), while the phosphorylated glycol followed the pathway of acetyl phosphate, lactaldehyde phosphate, lactyl phosphate, and lactic acid (Ruddick, 1972; see pathway 2, Figure 6). Lactate is subsequently converted to pyruvate, which enters the citric acid cycle and/or the gluconeogenesis pathway (Ruddick, 1972; Wittman & Bawin, 1974).

2.3.2 Toxicological studies

LD₅₀ values after oral administration were available for 12 of the 31 substances in this group. In rats, the values ranged from 2000 to 31 000 mg/kg bw (Nos 909, 920, 922, 925, 930–933, 912, and 929), indicating little acute toxicity of this group by the oral route (Smyth et al., 1941; Fassett & Roudabush, 1952; Dominguez-Gil & Cadorniga, 1971; deGroot et al., 1974; Bailey, 1976; Bartsch et al., 1976; Moreno, 1976, 1977, 1978; Moreno et al., 1984; Clary et al., 1998). The available values for mice ranged from 1100 to 5000 mg/kg bw (Nos 922, 935, and 928) (Gast, 1963; Moran et al., 1980; Moreno, 1980).

The results of short-term and long-term studies of toxicity conducted with substances in this group are summarized in Table 3.

Groups of five young rats of each sex were fed a diet containing glycerol at a concentration of 0, 1, 3, 6, 10, 15, 20, 30, 40, 50, or 60% (equivalent to 0, 1000, 3000, 6000, 10 000, 15 000, 20 000, 30 000, 40 000, 50 000, or 60 000 mg/kg bw per day) for 20 weeks. There was no significant difference in the body-weight gain at concentrations of glycerol ­ 30%, but reduced body-weight gain was observed at > 40%. Histological examination revealed no treatment-related changes at < 10%. The pathological changes observed at concentrations > 10% were marked hydropic and fatty degeneration of liver parenchymal cells. The NOEL was 5% glycerol in the diet, equivalent to 5000 mg/kg bw per day (Guerrant et al., 1947).

Ten men and four women were given glycerol orally at a dose calculated to result in an average daily intake of 24 000 mg/kg bw per day, for 50 days. No toxic effects were reported. The only effect was a slight tendency towards an increase in body weight (Johnson et al., 1933).

Groups of five male and five female Fischer 344 rats were given diets containing either 3-oxooctanoic acid or 3-oxotetradecanoic acid as esters of hydrogenated palm oil at a concentration calculated to provide a dose of 10 mg/kg bw per day, for 14 days. Detailed clinical examinations were conducted daily, and food consumption was measured on days 7 and 14. No physical signs of toxicity, abnormal body-weight gain, abnormal food consumption, or treatment-related effects were observed at necropsy. The absolute and relative weights of the liver and kidney were increased by 10% in female rats, but this effect was not considered to be biologically significant as no histological changes were found in these tissues and there were no other observed toxic effects (Gill & van Miller, 1987).

Studies were available on the lactones, delta-decalactone and delta-dodecalactone, which are formed from the hydrolysis of their respective glycerol esters glycerol 5-hydroxydecanoic acid and glycerol 5-hydroxydodecanoic acid. Groups of rats were fed a mixture of 30% delta-decalactone, 60% delta-dodecalactone, and 10% butyric acid at a concentration of 0.01% or 1% delta-decalactone or delta-dodecalactone in the diet for 49 weeks. These concentrations were calculated to result in average daily intakes of 1.5 or 150 mg/kg bw delta-decalactone and 3 or 300 mg/kg bw delta-dodecalactone. Histological examination revealed no adverse effects in any group. Haematology, blood chemistry, and urinary analysis showed no significant difference between test and control groups (Wilson, 1961).

In rats given 2,2,4-trimethyl-1,3-oxacyclopentane at a dose of 3.8 or 38 mg/kg bw per day for 14 days, there were no signs of toxicity at either dose (deGroot et al., 1974).

In a study of the tumour promoting potential of glycerol, groups of male and female C3H mice, 6–8 weeks old, were given various carcinogens followed by 0, 0.5, or 1% (v/v) glycerol solution until they were 1 year old. Animals in the control group received either 5% (v/v) glycerol (equivalent to 5000 mg/kg bw per day) or water. The animals were killed, and the incidences of liver and lung tumours were recorded. Among males, the incidence of liver tumours was 23% in those given glycerol and 39% in those given water. The tumour incidence in the lung was 21% with glycerol abd 41% with water. Similar results were obtained for female mice. Thus, lower incidences of liver and lung tumours were seen after glycerol treatment. No treatment-related adverse effects were reported (Witschi et al., 1989).

A study in which Sprague-Dawley rats were given glycerol in the diet at a concentration of 0, 5, 10, or 20% (equivalent to 0, 5000, 10 000 or 20 000 mg/kg bw per day) for 50 weeks was evaluated previously by the Committee (Annex 1, reference 41). No significant treatment-related effects were found on growth rate or gross or histological appearance. The NOEL was 20 000 mg/kg bw per day (Atlas Chemical Co., 1969).

A study in which Sprague-Dawley rats were given glycerol in the diet at a concentration of 0, 5, 10, or 20% (equivalent to 0, 2500, 5000, or 10 000 mg/kg bw per day) for 2 years was evaluated previously by the Committee (Annex 1, reference 41). No significant treatment-related effects were found on growth rate or gross or histological appearance. Changes observed in relative kidney weights were not accompanied by histopathological changes. The NOEL was 10 000 mg/kg bw per day (Atlas Chemical Co., 1969).

A study in which Long-Evan rats were given glycerol in the diet at a concentration of 0, 5, 10, or 20% glycerol (equivalent to 0, 2500, 5000, or 10 000 mg/kg bw per day) for 2 years was evaluated previously by the Committee (Annex 1, reference 41). There were no significant treatment-related effects. The NOEL was 10 000 mg/kg bw per day (Hine et al., 1953).

Groups of rats were fed a diet containing tributyrin, butyric acid, or ethyl butyrate to examine the occurrence of gastric lesions. Tributyrin was given at a concentration of 15 or 25% (equivalent to 7500 and 12 500 mg/kg bw per day) for 3–35 weeks. The animals has severely reduced body-weight gain, which represented approximately one-third of that of the control group. The 66 rats receiving tributyrin that were necropsied showed greatly enlarged stomachs with numerous irregular protuberances on the external surface. Microscopic examination revealed hyperplasia, hyperkeratosis, and occasional ulceration. The mucosa of the forestomach was covered in papillomas, resulting in a significant thickening of the forestomach wall (Salmon & Copeland, 1949).

A study in which rats were given propylene glycol in the diet at a concentration of 2.45% or 4.9% (equivalent to 900 and 1800 mg/kg bw per day) for 2 years was evaluated previously by the Committee (Annex 1, reference 33). No treatment-related adverse effects were found on growth, and histological examination revealed no treatment-related effects (Morris et al., 1942).

A study in which rats received propylene glycol in the diet at a concentration of 0, 310, 630, 1300, or 2500 mg/kg bw per day for 2 years was evaluated previously by the Committee (Annex 1, reference 33). No treatment-related adverse effects on body-weight gain, haematological, urinary, or clinical chemical end-points, or organ weights were found. The NOEL was 1300 mg/kg bw per day (Gaunt et al., 1972).

A study in which dogs received propylene glycol in the diet at a concentration of 0, 2000, or 5000 mg/kg bw per day for 2 years was evaluated previously by the Committee (Annex 1, reference 33). Increased erythrocyte destruction was found at the higher dose. No significant treatment-related effects on haematological, clinical chemical, or urinary end-points, or on gross or histological appearance were found (Weil et al., 1971).

As pyruvic acid is reduced to lactic acid in vivo, data on lactic acid can be used to evaluate the safety of pyruvic acid.

Groups of male and female Fischer 344 rats were fed diets containing the calcium salt of lactic acid at a concentration of 0, 2.5, or 5% for 2 years, calculated to provide a dose of 0, 2500, or 5000 mg/kg bw per day, respectively. No adverse effects were observed, and no evidence was found of a significant dose-related increase in the incidence of tumours in any organ or tissue of treated animals. No specific dose-related changes were observed in any of the haematological and biochemical parameters measured (Maekawa et al., 1991).

In a study to examine the tumour promoting potential of pyruvaldehyde in a two-stage model of stomach carcinogenesis, groups of male Wistar rats were given drinking-water containing 0.25% pyruvaldehyde for 32 weeks, alone or after 8 weeks’ treatment with a known tumour initiator, N-methyl-N-nitro-N-nitrosoguanidine (MNNG). Pyruvaldehyde alone caused no increase in the incidence of stomach hyperplasia or tumours. In rats pretreated with MNNG, pyruvaldehyde did not enhance the development of adenocarcinomas in the pylorus of the glandular stomach, but it significantly increased the incidence of hyperplasia (Takahashi et al., 1989).

The results of studies of genotoxicity with these substances are shown in Table 4.

    See Also:
       Toxicological Abbreviations