Certain chlorinated propanols occur as contaminants in hydrolysed vegetable proteins. Processing of defatted vegetable proteins by traditional hydrochloric acid hydrolysis leads to the formation of 3-chloro-1,2-propanediol and 1,3-dichloro-2-propanol. These two compounds were evaluated by the Committee at its forty-first meeting (Annex 1, reference 107), when it concluded that 3-chloro-1,2-propanediol is an undesirable contaminant in food and considered that its concentration in hydrolysed proteins should be reduced to the lowest level technically achievable. Since that time, new data have become available, and the Codex Committee on Food Additives and Contaminants asked the Expert Committee to re-evaluate 3-chloro-1,2-propanediol.

2.1.1 Absorption, distribution, and excretion

3-Chloro-1,2-propanediol crosses the blood–testis barrier and the blood–brain barrier and is distributed widely in the body fluids (Edwards et al., 1975). The compound has been found to accumulate in the cauda epididymis of rats and to a lesser extent in that of mice, as observed by autoradiography (Crabo & Appelgren, 1972). This finding was disputed by Jones et al. (1978), who did not observe tissue-specific retention of radiolabel in rats given [³⁶Cl]3-chloro-1,2-propanediol intraperitoneally at a dose of 100 mg/kg bw. Neither 3-chloro-1,2-propanediol nor its metabolite beta-chlorolactate accumulated in tissues.

Male Wistar rats given a single intraperitoneal injection of [¹⁴C]3-chloro-1,2-propanediol at a dose of 100 mg/kg bw exhaled 30% of the dose as ¹⁴CO₂ and excreted 8.5% unchanged in the urine after 24 h (Jones, 1978). In another study, in which rats received a single intraperitoneal injection of[³⁶Cl]3-chloro-1,2-propanediol at a dose of 100 mg/kg bw, 23% of the radiolabel was recovered in the urine as beta-chlorolactate (Jones et al., 1978).

2.1.2 Biotransformation

3-Chloro-1,2-propanediol is detoxified by conjugation with glutathione, yielding S-(2,3-dihydroxypropyl)cysteine and the corresponding mercapturic acid, N-acetyl-S-(2,3-dihydroxypropyl)cysteine (Jones, 1975). The compound also undergoes oxidation to beta-chlorolactic acid and further to oxalic acid (Jones & Murcott, 1976). An intermediate metabolite, beta-chlorolactaldehyde, may also be formed, as traces have been found in the urine in rats (Jones et al., 1978). The intermediate formation of an epoxide has been postulated but not proven (Jones, 1975).

There is ample evidence that the microbial enzymes halohydrin dehalogenases can dehalogenate haloalcohols to produce glycidol (van den Wijngaard et al., 1989), a known mutagen in Salmonella typhimurium tester strains (Zeiger et al., 1988) and a genotoxin in vitro and in vivo (National Toxicology Program, 1990). In a review of the metabolism of 3-chloro-1,2-propanediol, Lynch et al. (1998) concluded that the main metabolic route in mammals is formation of beta-chlorolactate and oxalic acid, while many bacteria metabolize 3-chloro-1,2-propanediol primarily via glycidol.

2.1.3 Effects on enzymes and other biochemical parameters

The activity of all glycolytic enzymes in the epididymal and testicular tissue of rats was reduced when they were given a daily subcutaneous injection of 3-chloro-1,2-propanediol at a dose of 6.5 mg/kg bw for 9 days (Kaur & Guraya, 1981a). It has been suggested that the mechanism is inhibition of glyceraldehyde 3-phosphate dehydrogenase and triosephosphate isomerase by the beta-chlorolactate metabolite (Mohri et al., 1975; Suter et al., 1975; Jones & Porter, 1995; Lynch et al., 1998).

Incubation of ram sperm with 3-chloro-1,2-propanediol in vitro inhibited the glycolysis of spermatozoa (Brown-Woodman et al., 1975), possibly by indirect inhibition of glyceraldehyde-3-phosphate dehydrogenase (Mohri et al., 1975; Suter et al., 1975). The decrease in the activity of spermatozoan glycolytic enzymes was suggested to be a result of an altered epididymal milieu (Kaur & Guraya, 1981b).

The inhibition of spermatozoan glycolysis by 3-chloro-1,2-propanediol (and/or its metabolites) resulted in reduced sperm motility (Jones, 1983). The inhibition was reversible and has been attributed to the S-enantiomer of the substance (Porter & Jones, 1982; Stevenson & Jones, 1984; Dobbie et al., 1988; Jones & Porter, 1995; Jones & Cooper, 1999). In addition, 3-chloro-1,2-propanediol decreased testosterone secretion in cultured Leydig cells from rats (Paz et al., 1985).

Rats receiving 3-chloro-1,2-propanediol at a dose of 6.5 mg/kg bw per day for 9 days had significantly decreased (p < 0.05) levels of RNA and protein in the testis and epididymis, and these changes were parallelled by increases in the concentrations of proteinase and ribonuclease. The DNA content was unchanged (Kaur & Guraya, 1981c).

Increased blood urea nitrogen and serum creatinine concentrations, chronic progressive nephropathy, and renal tubule-cell lesions—all indicative of overt nephrotoxicity—were generally seen at doses somewhat higher than those that caused testicular and epidydimal effects. The nephrotoxicity was associated with the R-enantiomer of 3-chloro-1,2-propanediol (Porter & Jones, 1982; Dobbie et al., 1988; Jones & Cooper, 1999).

2.2.1 Acute toxicity

The LD₅₀ of 3-chloro-1,2-propanediol in rats treated orally was reported to be 150 mg/kg bw (Ericsson & Baker, 1970).

2.2.2 Short-term studies of toxicity

Groups of eight male Fischer 344 rats were given a single subcutaneous injection of 3-chloro-1,2-propanediol at a dose of 75 mg/kg bw and killed after 24 h or 3, 8, 25, or 75 days. A slight but significant (p < 0.05) increase in liver weight was observed after 24 h but not at later sacrifices. Histologically, the hepatocytes showed mild-to-moderate cytoplasmatic swelling in the periportal area (Kluwe et al., 1983).

Intraperitoneal injection of 3-chloro-1,2-propanediol to male Sprague-Dawley rats at a single dose of 100 mg/kg bw caused increased diuresis for up to 15 days. Higher doses (not reported) caused aneuresis and death, and histological examination of the kidney showed acute glomerular nephritis. The type of kidney lesion was characteristic of oxalic acid poisoning, and crystals of calcium oxalate were seen in urine by microscopy. Oral treatment with 3-chloro-1,2-propanediol at 10 mg/kg bw per day for 5 days did not increase diuresis (Jones et al., 1978).

In another study, intraperitoneal injection of 3-chloro-1,2-propanediol at a dose of 100 or 120 mg/kg bw caused severe proteinuria and glucosuria in male Wistar rats. Oliguria and anuria were observed, and four of nine animals died. The five surviving animals had decreased appetite and body weight, proteinuria, dose-related diuresis, and increased water intake (Morris & Williams, 1980).

Testing of the (R)- and (S)-isomers of 3-chloro-1,2-propanediol, synthesized under laboratory conditions, showed that only the (R)-isomer induced diuresis and gluco-suria in rats (Porter & Jones, 1982; Dobbie et al., 1988; Jones & Cooper, 1999).

Oxalic acid, a metabolite of 3-chloro-1,2-propanediol, appeared to play a important role in the development of kidney damage (Jones et al., 1979). Birefringent crystals characteristic of calcium oxalate seen in tubules at the cortico-medullary junction of rats 1 day after treatment with 3-chloro-1,2-propanediol at a single subcutaneous dose of 75 mg/kg bw were considered to be early morphological changes. On day 75, focal tubule necrosis, regeneration, and tubule dilatation were observed in the kidney (Kluwe et al., 1983).

Groups of 20 Sprague-Dawley rats of each sex were given 3-chloro-1,2-propanediol at a dose of 0, 30, or 60 mg/kg bw per day by gavage on 5 days per week for 4 weeks. Ten animals of each sex from each group were killed on day 2, and their blood was examined for clinical chemical parameters. On day 2, rats at the higher dose had increased serum alanine aminotransferase activity (males, p < 0.05; females, p < 0.001) and increased concentrations of creatinine (females, p < 0.001), urea, and glucose (females, p < 0.05). On day 25, the treated rats had increased alanine aminotransferase activity (males at the higher dose, p < 0.001; females at both doses, p < 0.001) and increased urea concentrations (males at the higher dose, p < 0.001; females at the higher dose, p < 0.05). Statistically significantly (p < 0.05) decreased values for haemoglobin concentration and erythrocyte volume fraction were found for both male and female treated rats. Female rats at the higher dose had a decreased erythrocyte count (p < 0.001). Treated rats had lower body-weight gain, which was statistically significant (statistics not reported) at termination of the study. After 2 days of treatment, the relative weight of the kidney was increased (p < 0.001) in males at the higher dose and in females at both doses, and on day 25, the relative weights of the kidney, liver, and testis (males at the higher dose) were significantly increased (p < 0.01 or 0.001) in treated rats. Histopathological examination revealed chronic progressive nephropathy in eight females at the higher dose and mild tubule dilatation in the testis of three males at the lower dose and seven at the higher dose. One male at the higher dose had severe atrophy of both testes (Marchesini & Stalder, 1983).

Groups of 20 Fischer 344 rats of each sex were given drinking-water containing 3-chloro-1,2-propanediol at a concentration of 0, 100, 300, or 500 mg/L for 90 days, corresponding to average daily intakes of 9, 27, and 43 mg/kg bw for males and 11, 31, and 46 mg/kg bw for females. Ten animals of each sex per group were killed after 30 days of treatment. Clinical chemical and haematological parameters were determined, and histopathological examinations were carried out on controls and rats at the highest dose.

Slight anaemia (p < 0.05 or 0.001) was seen in females at the two higher doses after 30 days and in rats of each sex after 90 days of treatment (p < 0.05 or 0.01); however, there was no morphological evidence of impaired haematopoiesis or increased degradation of erythrocytes. A dose-dependent decrease (p < 0.01) in plasma creatinine concentration was seen in rats of each sex at the two higher doses after 30 days of treatment and in all treated groups at terminal sacrifice (p < 0.05 or 0.01). Serum phosphate concentrations were increased in male rats at the highest dose at the interim (p < 0.01) and terminal sacrifices (p < 0.05). A statistically significant (p < 0.01), dose-dependent increase in relative weights was found for the kidney and liver, and the increase in the relative kidney weight was significant at the lowest dose. Histopathological examination showed a lower incidence of crystalline precipitation in the kidneys of animals at the highest dose than in the controls. The livers of about half of the treated males were found to contain single hepatocytes with two or three nuclei after 90 days of treatment, and the epididymides of treated rats had an increased number of exfoliated spermatozoids (Marchesini et al., 1989).

Three of six male macaque monkeys (Macaca mulatta) given 3-chloro-1,2-propanediol orally at a dose of 30 mg/kg bw per day for 6 weeks showed haematological abnormalities including anaemia, leukopenia, and severe thrombocytopenia (Kirton et al., 1970). The Committee noted that two of the affected monkeys died during the study due to bone-marrow depression.

2.2.3 Long-term studies of toxicity and carcinogenicity

Fifty female CHR/Ha Swiss mice received 3-chloro-1,2-propanediol subcuta-neously at a dose of 1 mg/week for 580 days, and a second group of 50 mice received the compound dissolved in acetone by topical application at a dose of 2 mg three times per week. No changes were observed in the group treated by dermal application, but local sarcomas were found at the site of injection in one dosed and one control mouse treated subcutaneously (Van Duuren et al., 1974).

Three groups of 26 male and female Charles River CD rats received 3-chloro-1,2-propanediol at a dose of 0, 30, or 60 mg/kg bw by gavage twice weekly. After 10 weeks, the doses were increased to 35 and 70 mg/kg bw. The animals were treated for 72 weeks, and the study was terminated after 2 years. Three parathyroid adenomas were found in male rats at the higher dose, but this finding was not statistically significant with respect to the control group. While the females showed no signs of toxicity, male rats had a higher mortality rate than controls, and all treated males had severe testicular degeneration and atrophy (Weisburger et al., 1981).

Four groups of 50 specific pathogen-free Fischer 344 rats of each sex, 5–6 weeks old at the start of the study, underwent an 11-day acclimatization period and then received drinking-water containing 3-chloro-1,2-propanediol (purity, 98%) at a concentration of 0, 20, 100, or 500 mg/L, equivalent to mean daily intakes of 0, 1.1, 5.2, and 28 mg/kg bw per day for males and 0, 1.4, 7.0, and 35 mg/kg bw per day for females, for 104 weeks. Feed and water were provided ad libitum. The feed was certified laboratory chow, with contaminants within an acceptable range according to the Environmental Protection Agency of the USA. The test substance was stable in water for > 4 days, and it was prepared twice a week and tested once per group per week. The water contained a mean of 2.7 mg/L of 3-chloro-1,2-propanediol, equivalent to an intake of 0.15 mg/kg bw per day for males and 0.19 mg/kg bw per day for females, determined once per week.

The animals were examined daily for changes in health or behaviour. Food consumption was recorded weekly up to week 19 and body weight weekly up to week 20, and then both were recorded monthly; from week 88 to the end of the study, body weight was again recorded weekly. Water consumption was recorded weekly from start to week 20 of the study and fortnightly thereafter. Ophthalmological examinations were performed regularly. Haematological parameters and blood chemistry were evaluated in blood samples taken at weeks 103–105 from all surviving animals. All animals found dead, animals killed in extremis, and those killed at the end of the experiment were subjected to complete necropsy and histopathological examination. The liver, kidneys, spleen, pancreas, heart, adrenals, testis, epididy-mides, and brain were weighed.

The body weights of rats at the highest dose were significantly (p < 0.05) reduced after the first week of treatment. At termination, the body weights of animals at the two higher doses were significantly reduced (p < 0.05), with reductions of 33% in males and 35% in females at the highest dose. However, the mortality rate was unaffected by treatment: at terminal sacrifice, more than 42% of this group were still alive. The food and water intake of rats at the highest dose were significantly (p < 0.05) reduced. No treatment-related clinical signs were noted. The haematological and blood clinical chemical parameters varied considerably within groups, but no consistent, significant, dose-related effects were observed. The reduced body weights of rats at the two higher doses obviated a conclusion about an effect on organ weights. However, the body weights of animals at the lowest dose were unaffected, and males showed a significant (p < 0.05) increase in absolute kidney weight. Treatment-related pathological, hyperplastic and neoplastic findings are listed in Table 1.

Chronic progressive nephropathy was found in all groups, and the incidence increased with dose, being significant at the two higher doses (p < 0.05). Female rats were more severely affected than males. Significant correlations (p < 0.01) were found between the severity of nephropathy and the increase in incidence of renal tubule hyperplasia and renal adenoma (see below). Advanced chronic progressive nephropathy accounted for a significant (males) and dose-dependent (both sexes) rate of premature deaths (p < 0.005 for males and p < 0.10 for females). The treatment-related distribution of advanced chronic progressive nephropathy in rats of each sex was reflected in significant, dose-dependent increases in kidney weight (p < 0.05), serum creatinine concentration (p < 0.01), and blood urea nitrogen concentration (p < 0.01). Papilliform hyperplasia of the urothelium covering the renal papilla was seen almost exclusively in animals at the two higher doses. Both the incidence and the severity of the lesions increased in a dose-dependent pattern. The incidence of papillary urothelial hyperplasia correlated to the severity of chronic progressive nephropathy. A dose-dependent increase in the frequency of epithelial single-cell degeneration was observed in the epididymides, which was significant at the two higher doses (p < 0.001).

Dose-related alterations in the incidence of hyperplasia and/or tumours were observed in all groups, with increases in the kidney (tubule hyperplasia and adenomas), the testis (Leydig-cell hyperplasia, adenomas and adenocarcinomas), mammary gland (males: fibroadenomas, adenomas, and adenocarcinomas), and preputial gland (adenomas and carcinomas), and decreases in the pancreas (males: hyperplasia, adenomas, and carcinomas) (see Table 1). The increased incidence of tubule hyperplasia in the kidneys of animals of each sex was the most sensitive end-point, as it was seen even at the lowest dose. Although it did not reach statistical significance at this dose (p = 0.073 for males and 0.099 for females), the Committee considered that it reflected a treatment-related, dose-dependent increase, which was highly significant (p < 0.0001) in a trend analysis. Nodular Leydig-cell hyperplasia was present in a high proportion of controls, and the incidence in treated animals decreased significantly in a dose-dependent pattern.

When the incidences of nodular Leydig-cell hyperplasia, adenomas, and carcino-mas were combined for statistical analysis, no significant difference was found between groups. The Committee noted that the decreased frequency of hyperplasia might be associated with the concomitant increases in the incidences of adenomas and carcinomas, so that the effect would not be significant when all three were combined. When the incidences of preputial gland adenomas and carcinomas were combined for statistical analysis, the resulting increased incidence was significant at the two higher doses. The Committee noted that the report clearly stated that the preputial gland was not included in the standard protocol and was examined only when it was removed incidentally with other tissues or organs. Thus, few were investigated. When pancreatic hyperplasia and neoplastic lesions were combined for statistical analysis, the decrease in incidence was significant at all doses (p < 0.05).

The authors concluded that treatment with 3-chloro-1,2-propanediol increased the incidences of renal and testicular Leydig-cell tumours. The renal tumours developed in a dose-dependent fashion in animals of each sex and were considered to be secondary to the treatment-related increase in the incidence of chronic progressive nephropathy. The treatment-related increase in the incidence and frequency of Leydig-cell tumours was considered to represent hormone-mediated promotion and was suggested to be associated with the treatment-related decrease in testosterone concentration and the increase in those of estradiol, prolactin, progesterone, and follicle-stimulating and luteinizing hormones. 3-Chloro-1,2-propanediol also caused a dose-related increase in the incidence of mammary tumours in males, and this effect was considered to be secondary to the hormonal activity of functionally active Leydig-cell tumours, which were suggested to produce less androgen and more estrogen or progesterone. The Committee noted that the hormones mentioned were not measured in the study. In addition, the treatment caused a dose-related increase in the incidence of preputial gland tumours, which was suggested to be secondary to the disturbed endocrine balance of treated animals with large Leydig-cell tumours, analogous to the induction of mammary tumours (Sunahara et al., 1993).

2.2.4 Genotoxicity

The results of studies of genotoxicity with 3-chloro-1,2-propanediol are summarized in Table 2.

A test for micronucleus formation in bone marrow in vivo was performed in Crl:HanWistBR rats according to a protocol conforming to OECD 474. The highest dose was determined from a range-finding study in which single oral doses of 20–100 mg/kg bw were administered once daily for 2 days to groups of male and female rats; doses > 60 mg/kg bw per day were severely toxic and caused some deaths. Male animals were used in the main study, as no substantial sex difference in toxicity was seen. Groups of six males were given 3-chloro-1,2-propanediol orally at a dose of 15, 30, or 60 mg/kg bw per day for 2 days. Piloerection was seen at the highest dose and was associated with a clear reduction in the ratio of polychromatic to normochromatic erythrocytes, indicating bone-marrow cytotoxicity and hence indicating that the substance and/or its metabolites had reached the bone marrow. There was no increase in the number of micronucleated polychromatic erythrocyte stem cells at any dose (2000 polychromatic erythrocytes scored per animal). Cyclophosphamide, used as the positive control, caused a clear increase in the number of micronuclei (Marshall, 2000).

A test for unscheduled DNA synthesis was performed in male Han Wistar rats according to a protocol that conformed to OECD 486. The highest dose of 100 mg/kg bw was chosen on the basis of a study that had shown severe toxicity at an oral dose of 150 mg/kg bw. In the main study, a single oral dose of 40 or 100 mg/kg bw was administered to the animals, and hepatocytes were recovered for analysis of unscheduled DNA synthesis by autoradiography after 12–24 h (four animals per dose) and 2–4 h (five animals per dose). No signs of toxicity were seen at either dose, and no increase in unscheduled DNA synthesis was seen. The two positive control compounds, N-2-fluorenylacetamide and N-nitrosodimethylamine, both gave clear positive results (Fellows, 2000).

2.2.5 Reproductive toxicity

3-Chloro-1,2-propanediol has been reported to inhibit male fertility (Gunn et al., 1969; Helal, 1982), although the effect was reversible (Ericsson & Youngdale, 1970; Jones, 1983). The mechanism of this activity of 3-chloro-1,2-propanediol is unknown, but its metabolites inhibit enzymes involved in spermatozoan glycolysis, reducing the motility of the spermatozoa (Jones, 1983). The inhibition of spermatozoan motility by 3-chloro-1,2-propanediol was suggested to be due in part to alkylation of cysteine (Kalla & Bansal, 1977). The compound also affects several enzymes of epithelial cells in the testis and caput epididymis, resulting in decreased glycolysis (Gill & Guraya, 1980). Only the (S)-isomer, synthesized under laboratory conditions, specifically inhibited glycolysis in boar sperm (Stevenson & Jones, 1984).

3-Chloro-1,2-propanediol has two specific effects on the reproductive tract of male rats. These effects are dose-dependent and have been termed the ‘high-dose effect’ and the ‘low-dose effect’. The ‘high-dose effect’, seen after a single intra-peritoneal injection of 75 mg/kg bw 3-chloro-1,2-propanediol, consisted of bilateral retention cysts or spermatocoele of the caput epididymis 5–7 days after treatment (Cooper & Jackson, 1973). Use of electron microscopy showed that administration of 3-chloro-1,2-propanediol by gavage at a dose of 140 mg/kg bw specifically affected the epithelia in the initial segment of the epididymis 2 h later. The cellular lesions were characterized by sloughing of the epithelium, which led to obstruction of the epididymal tract (Hoffer et al., 1973). The back-pressure of the testicular fluid caused oedema, inhibition of spermatogenesis, and atrophy of the testis (Jones, 1983). Histological examination of testes from rats given injections of 40 mg/kg bw per day for 20 days revealed total inhibition of spermiogenesis due to degeneration and disappearance of the spermiogonia from the tubules. Proliferation of the epithelial cells of the ducts in the cauda epididymis was observed, and several blood vessels showed thickened walls (Samojlik & Chang, 1970).

The ‘low-dose effect’, which was evident a few days after an oral dose of 5–10 mg/kg bw per day, was directed towards mature sperm contained in the cauda epididymis. The spermatozoa were rendered incapable of fertilization but showed no visible change in morphology (Jones, 1983). Male rats given 3-chloro-1,2-propanediol by subcutaneous injection at 15 or 40 mg/kg bw per day became infertile 6 and 3 days after commencement of treatment, respectively. Treatment with 15 mg/kg bw per day for 30 days, followed by a recovery period of 18 days, resulted in recovery of fertility (Samojlik & Chang, 1970). The lowest daily oral doses shown to cause infertility in male rats, as determined by mating studies, were 5 mg/kg bw for 14 days (Coppola, 1969), 6.5 mg/kg bw for 10 days (Gunn et al., 1969), 2.5 mg/kg bw with ‘continuous’ treatment (Erickson & Bennett, 1971), 8 mg/kg bw for 4 days (Turner, 1971), and 8 mg/kg bw (by subcutaneous injection) for 3 days (Black et al., 1975).

Groups of five albino male rats given 3-chloro-1,2-propanediol orally at a dose of 0.5, 1, 2, 4, or 6 mg/kg bw per day for 10–12 days showed 2.5%, 20%, 45%, 85%, and 100% sterility (on the basis of histological degree of spermiogenesis), respectively (Helal, 1982).

As reported in a summary, groups of five male Wistar rats were given 3-chloro-1,2-propanediol in distilled water at a dose of 0, 0.1, 0.5, 1, 2, 3, 4, 5, or 10 mg/kg bw per day by gavage for 7 days before and during mating. Each male was mated with five virgin females, which were killed on day 14 of gestation and examined for pregnancy status. 3-Chloro-1,2-propanediol had no adverse effect on male fertility at doses < 3 mg/kg bw per day, with respect to pregnancy rate and total numbers of implantations and live embryos; however, the pre-implantation loss was significantly greater (p = 0.05) for female rats mated with males given 3 mg/kg bw per day than in controls. The NOEL was 2 mg/kg bw per day (Parish, 1989).

3-Chloro-1,2-propanediol has also been reported to affect the fertility of male hamsters (30–100 mg/kg bw per day orally for 7 days), gerbils (20 mg/kg bw per day orally for 50 days), guinea-pigs (50–70 mg/kg bw per day orally or subcutaneously for 45 days), dogs (8 mg/kg bw per day subcutaneously for 30 days), rams (25 mg/kg bw per day intramuscularly for 4 days), and rhesus monkeys (30 mg/kg bw per day orally for 42 days) (Jones, 1978, 1983; Jones & Cooper, 1999). The compound was reported to have no effect on fertility in mice, quail, or rabbits (Jones, 1978).

Groups of 10 female rats were given 3-chloro-1,2-propanediol subcutaneously at a dose of 0 or 10 mg (approximately 25 mg/kg bw) every second day for 30 days. Significant (p < 0.01) decreases were noted in the relative weights of the ovary, uterus, and vagina when compared with controls. Histological examination ahowed that the ovary was small and had widespread follicular atresia and degeneration of corpora lutea; the uterus was regressed, and the lumen was lined with columnar epithelium; atrophic changes were observed in the vaginal epithelium. The protein and RNA contents of the uterus and vagina were significantly (p < 0.01) reduced when compared with controls. The authors suggested a luteolytic and possibly antiestrogenic effect of 3-chloro-1,2-propanediol in female rats (Lohika & Arya, 1979).

In a computer-assisted analysis of sperm motion to determine the relationship between dose and effect on sperm glycolysis as evidenced by impaired sperm mobility in the epididymis, male Long-Evans rats were given drinking-water containing 3-chloro-1,2-propanediol at concentrations providing a dose of 5, 10, or 20 mg/kg bw per day for 8 days. The percentage of motile sperm was significantly reduced at the two higher doses. Multivariate analysis of end-points of motion, including curvi-linear velocity, linearity of swim path, velocity, and lateral head displacement, showed significant differences from controls at the two higher doses (Toth et al., 1992).

3-Chloro-1,2-propanediol was administered by gavage to groups of 10 adult male CD rats at a dose of 0, 1, 5, or 25 mg/kg bw per day for 14 days. The animals were killed on day 15 or 29. At necropsy, testis weight, distribution of DNA ploidy in testicular cell suspensions, testicular and epididymal histological appearance, and epididymal sperm concentration, motility, morphology, and breakage were determined. Before the kill at day 15, the males were cohabited with untreated females in a 1:2 ratio. The females were killed on presumed gestational day 13 and examined for pregnancy status. At the highest dose, minor decreases in body weight and relative food consumption were reported, and testicular and epididymal lesions were observed. The distribution of DNA ploidy was found to be predictive of testicular damage; however, the effects were more pronounced on day 29 (in the group allowed a 2-week recovery) than on day 15. Sperm motion was altered, and the percentage of motile sperm was reduced on day 15 at the two higher doses. At the highest dose, sperm velocity, amplitude of lateral head displacement, and epididymal sperm concentrations were reduced, and the incidence of sperm breakage was increased. The NOEL was 1 mg/kg bw per day (Hoyt et al., 1994).

3-Chloro-1,2-propanediol was given orally to male Sprague-Dawley rats at a dose of 0, 2, or 8 mg/kg bw per day for 2 weeks. An additional group was given the higher dose for 4 weeks. At the end of dosing, the males were mated with untreated females. No treatment-related effects were found on body weight, food consumption, or the weights of the testis, epididymis, or prostate, and no significant effects were found on sperm number, viability, or maturation rate. At the higher dose, sperm motility was decreased after 2 h of incubation, and sperm activity was decreased both at the time of initial collection and after 2 h of incubation. At the lower dose, sperm activity was decreased only after 2 h of incubation. The group given the higher dose and allowed to recover showed no effects on sperm motility or activity. None of the females mated with males at the higher dose became pregnant. The lower dose had no effect on fertility. After recovery, males at the higher dose copulated with and successfully impregnated females (Yamada et al., 1995).

3-Chloro-1,2-propanediol administered orally to male Han rats at a dose of 20 mg/kg bw per day for at least 5 days caused lesions in the testes and epididymides. At doses of 5, 10, and 20 mg/kg bw per day, sperm motility was significantly depressed. Females mated with males at any dose failed to become pregnant. No effects of treatment were found on sperm morphology (Woods & Garside, 1996).

Male rats received 3-chloro-1,2-propanediol by oral gavage at a dose of 1, 3, or 10 mg/kg bw per day for 9 days and were then mated with untreated females. At the highest dose, no pregnancies resulted from the matings. A decreased pregnancy rate and number of implants were reported at 3 mg/kg bw per day. Analysis of sperm motility revealed treatment-related decreases in the percentage of motile sperm, sperm velocity, and amplitude of lateral head displacement at the highest dose and decreased sperm velocity and amplitude of lateral head displacement at the intermediate dose. Sperm from males treated at the highest dose did not reach the oviducts of females, and few sperm from males treated at the next lowest dose reached this location. Similarly, the percentage of fertilized eggs in the oviducts of mated females was decreased in a dose-dependent manner. The NOEL was 1 mg/kg bw per day (Ban et al., 1999).

2.2.6 Special studies: Neurotoxicity

Groups of three male BALB/c mice were given racemic 3-chloro-1,2-propanediol intraperitoneally at a dose of 25, 50, or 100 mg/kg bw per day for up to 5 days. The mice given three daily doses of 100 mg/kg bw died and were found to have discrete widespread lesions in the gray matter, from the cortex to the spinal cord. No signs of lesion haemorrhage were observed. Administration of 50 mg/kg bw per day for 5 days did not cause deaths; however, mice killed on day 6 showed small vacuolated lesions in many brainstem centres. Another group of mice treated with an additional four daily doses developed more severe and widespread lesions. Only one spinal cord lesion developed in one mouse at the lowest dose administered on day 5. However, as with the intermediate dose, additional dosing during the second week resulted in severe lesions. The authors attributed the development of central nervous system lesions partly to inhibition of glycolysis by the (S) enantiomer of 3-chloro-1,2-propane-diol (Cavanagh & Nolan, 1993; Cavanagh et al., 1993).

Single intraperitoneal injections of 3-chloro-1,2-propanediol to female Wistar rats at doses of 250–1000 mg/kg bw resulted in deaths associated with the development of neurological lesions within 24 h. Histologically, the lesions appeared as watery cytoplasmic swelling of astrocytes, mainly in the brain stem. Lower doses did not cause brain lesions. Administration of 100 mg/kg bw per day for 2 days was overtly toxic and resulted in death, with the formation of large vacuolated lesions at many sites within the brain. Treatment at 50 mg/kg bw per day for up to 5 days resulted in cumulative central nervous system lesions. For several days after the completion of treatment, regenerative processes rather than necrosis of astrocytes predominated. As in the case of mice, the authors attributed the development of central nervous system lesions after a high dose of 3-chloro-1,2-propanediol in part to inhibition of glycolysis by the (S) enantiomer and energy deprivation of the affected brain regions (Cavanagh & Nolan, 1993; Cavanagh et al., 1993).

A synergistic effect of 3-chloro-1,2-propanediol and copper ions in decreasing the motility of human spermatozoa was observed in vitro (Kalla & Singh, 1981). When the compound was incubated with ejaculated human sperm, the motility of the spermatozoa was inhibited and their metabolic activity was reduced, as measured by glucose and oxygen uptake and lactate production (Homonnai et al., 1975).

In its pure state, 3-chloro-1,2-propanediol is a relatively non-volatile liquid with a boiling-point of 114–120 °C at the reduced pressure of 14 mm Hg (1.9 kPa). It is soluble in a range of solvents, including water, ethanol, and ether. (Windholz, 1976). 3-Chloro-1,2-propanediol is also referred to in the literature as 3-monochloro-1,2-propanediol and alpha-chlorohydrin. It is an asymmetric molecule that can exist in two optically active forms, (R) and (S).

During the 1980s, 3-chloro-1,2-propanediol was quantified in acid-hydrolysed vegetable protein by gas chromatographic techniques that typically allow quantification of concentrations > 10 mg/kg. The analytical approach involves extraction and concentration, followed by chemical derivatization with a reagent such as heptafluorobutyrylimidazole to provide a derivative that is suitably volatile and can be readily detected. The application of mass spectrometric detection has increased the sensitivity of the analysis, so that more recent reports quote a limit of quantification of 0.01 mg/kg.

As interest has spread from acid-hydrolysed vegetable protein to a wider range of foods and ingredients, extraction methods have been developed to deal with a wide range of matrices.

The most widely used analytical method is based on gas chromatography with mass spectrometric detection. Samples are prepared by extraction onto a column packed with diatomaceous earth and subsequent elution of 3-chloro-1,2-propanediol with diethyl ether. The concentrated extract is taken for derivatization and analysis by gas chromatography. Quantification is achieved by comparing the intensity of a characteristic peak with that of the internal standard.

This method, which allows measurement of 3-chloro-1,2-propanediol at concentrations down to 0.01 mg/kg, has been validated with a range of foods and food ingredients, including acid-hydrolysed vegetable protein, in an international collaborative trial. It has been accepted as an official method by AOAC International (2000).

3-Chloro-1,2-propanediol was originally identified as a contaminant of the savoury food ingredient acid-hydrolysed vegetable protein, which is produced by treating protein-rich extracts from soya beans or other vegetable sources with concentrated hydrochloric acid at high temperature (Velisek et al., 1980). The protein extracts contain small amounts of fats and oils, which react with the hot acid to form 3-chloro-1,2-propanediol and smaller amounts of other chlorinated propanols.

The compound has since been found to be present at low concentrations in several other foods and food ingredients, including various cereal products whose manufacture involves high temperatures (i.e. roasting or toasting). The origin and formation of 3-chloro-1,2-propanediol in these products is not yet fully understood.

4.1.1 Acid-hydrolysed vegetable protein

The concentrations of 3-chloro-1,2-propanediol initially reported, in 1980, in acid-hydrolysed vegetable protein were as high as 100 mg/kg (Velisek et al., 1980). It has subsequently been shown that the compound occurs in this food as a racemic mixture of (R) and (S) isomers (Velisek & Dilezal, 2000). After the initial concern about this contaminant, many manufacturers refined their production processes to minimize its presence. Surveys of acid-hydrolysed vegetable protein products on the market in the United Kingdom, conducted in 1990, 1992, and 1998, showed that the median concentration had fallen from > 10 mg/kg in 1990 to 0.01–0.02 mg/kg in 1998 (Ministry of Agriculture, Fisheries, and Food, 1999a).

4.1.2 Soya sauce

Soya sauce and similar products such as oyster sauce can be manufactured by various processes, including treatment of crude soya extracts with strong acid as well as the traditional fermentation route. After reports of high concentrations of 3-chloro-1,2-propanediol, surveys of soya sauce were conducted in several countries in 1999–2000. These consistently showed that some products contain significant concentrations, while others contain concentrations close to or below the limit of detection.

For example, a survey conducted in the United Kingdom in mid-1999 showed that nearly one-fourth of the 40 samples contained 3-chloro-1,2-propanediol at a concentration > 1 mg/kg, the highest being 30 mg/kg (Ministry of Agriculture, Fisheries, and Food, 1999b). Unpublished analyses carried out in other European countries had found concentrations in the range 6–124 mg/kg in some samples (reported in Ministry of Agriculture, Fisheries, and Food, 1999b). Analysis of 90 products in Canada showed concentrations ranging from not detectable (< 0.01 mg/kg) to 330 mg/kg (Health Canada, 2000). A survey of 21 samples of soya sauce purchased in the USA showed that the concentration in nine exceeded 1 mg/kg, the highest being 85 mg/kg (Food & Drug Administration, 2000).

4.1.3 Food additives

By analogy with acid-hydrolysed vegetable protein, use of acid treatment in the manufacture of food additives could lead to the generation of 3-chloro-1,2-propanediol from fats or oils in the starting materials. In a recent survey in the United Kingdom (Food Standards Agency, 2001a), 3-chloro-1,2-propanediol was found in two of seven samples of modified starch, the highest concentration being 0.49 mg/kg in a sample of yellow dextrin. In the same survey, the compound was not detected (< 0.01 mg/kg) in samples of caramel colours.

4.1.4 Other food ingredients

3-Chloro-1,2-propanediol can be detected at low concentrations in certain cereal products, including malted cereals and malt extracts that are produced with high temperature treatment—so-called ‘dark’ malts and malt extracts that have a pronounced flavour and colour. These malts and malt extracts are used in a wide range of foods, such as bread, biscuits, breakfast cereals, beers, sauces, and gravies. 3-Chloro-1,2-propanediol has been found at concentrations ­ 0.5 mg/kg in these ingredients (Food Standards Agency, 2001a). Similar concentrations are found in roasted barley, which is a characteristic ingredient of certain beers. As these products typically make up 1–10% of the final food, the concentration of 3-chloro-1,2-propanediol will be one or two orders of magnitude lower.

4.1.5 Food contact materials

Some migration of 3-chloro-1,2-propanediol may occur from food contact materials that contain epichlorohydrin-based ‘wet strength’ resins, e.g. sausage casings, tea bags. and coffee filter paper. Similar resins are also sometimes used in the treatment of drinking-water, e.g. as flocculants. However, the significance of these sources is likely to decrease with the increasing availability and use of higher-grade resins which contain much lower concentrations of 3-chloro-1,2-propanediol (Commission of the European Union, 2000).

4.1.6 Foods

A range of foods was analysed for the presence of 3-chloro-1,2-propanediol in a recent survey in the United Kingdom (Food Standards Agency, 2001b), but most of the products tested were those in which fat and chloride are present in acidic conditions and which are processed at high temperatures and/or have a long shelf-life at ambient temperature. The results of the survey, covering 300 food products, are shown in Table 3. When quantifiable concentrations were found, they were low; the concentration in only one sample in this survey exceeded 0.1 mg/kg.

In surveys of acid-hydrolysed vegetable protein and soya sauce, the distribution of concentrations of 3-chloro-1,2-propanediol was markedly skewed, a large number of products having low or undetectable concentrations and a relatively small number having very high concentrations. The results of three surveys in the United Kingdom are shown in Figure 1 (Ministry of Agriculture, Fisheries, and Food, 1999a). However, the results of these surveys should be interpreted with caution, as sampling depended on provision of samples by selected food manufacturers and may not have been representative of all the products currently on the market.

The Canadian survey of 90 samples of soya sauce mentioned above (Health Canada, 2000) also showed that the distribution of contamination was markedly skewed, the median concentration being 0.5 mg/kg and the mean, 18 mg/kg.

3-Chloro-1,2-propanediol appears to be formed by the action of heat and/or acid during the processing of raw ingredients. It seems unlikely that there would be significant annual variation from natural sources.

3-Chloro-1,2-propanediol and 1,3-dichloro-2-propanol are formed when chloride ions react with triglycerides in foods under a variety of conditions, including food processing, cooking, and storage. These compounds have been found in a variety of foods and food ingredients, most notably in hydrolysed protein products and soya sauces. It has been shown that 3-chloro-1,2-propanediol is a precursor of 1,3-dichloro-2-propanol in protein hydrolysates (Collier et al., 1991). Their formation can be minimized by use of good manufacturing practices. When both compounds are found in soya sauces, the concentration of 1,3-dichloro-2-propanol is far lower than that of 3-chloro-1,2-propanediol.

Information on the concentrations of 3-chloro-1,2-propanediol in food, food ingredients, and protein hydrolysates was submitted by the United Kingdom, the USA, and the International Hydrolyzed Protein Council. A submission from the Czech Republic contained information on the reactions of chlorinated propanols and other chlorinated compounds but did not include information useful for this assessment. Only the USA supplied a national estimate of intake of 3-chloro-1,2-propanediol.

5.2.1 Relevant period of Intake

At any level of intake that can reasonably be expected to be encountered, 3-chloro-1,2-propanediol would not show acute toxic effects. This analysis therefore addresses only the probable long-term consumption of 3-chloro-1,2-propanediol from its presence in foods.

5.2.2 National estimates of Intake

The data submitted by the United Kingdom showed that 3-chloro-1,2-propanediol occurs in savoury foods, at above the limit of detection of 0.01 mg/kg in 30% of samples; only two samples had a residual concentration > 0.1 mg/kg. While the distribution of representative residual concentrations in all food groups could not be determined, the mean residual concentration in these savoury foods was 0.012 mg/kg. In soya sauces, however, concentrations > 300 mg/kg have been found, with a mean of 18 mg/kg in a survey of commercially available soya sauces in Canada. As the concentrations of residual 3-chloro-1,2-propanediol in the soya sauces surveyed are so much higher than the background concentration, the intake of this compound would be dominated by consumption of contaminated soya sauces.

Information was received from Australia on the intake of soya sauce and hydrolysed vegetable protein. The Committee sought data on soya sauce consumption in individual countries from a worldwide producer, and data on per-capita consumption of soya sauce in Japan and the United States were received.

The intake estimate prepared in the USA focused on the presence of 3-chloro-1,2-propanediol in soya sauces and in hydrolysed vegetable protein; low background concentrations in other foods were not considered relevant to the overall intake. As hydrolysed vegetable protein is used as a flavour and flavour enhancer in many food products, a large proportion of the American public consumes it daily, although many, if not most, people are unaware of its presence in some or all their foods. It was deemed appropriate to use ‘food disappearance poundage’ (the amount of a substance reported to be used, or disappear, in the food supply) in order to estimate the per-capita intake of hydrolysed vegetable protein, and thus the total poundage of reported to be used in food in 1 year was divided by the population of the USA and 365 days to determine the daily per-capita intake. The data used were made available to the Food & Drug Administration by industry. The reported poundage of hydrolysed vegetable protein in 1990 was 40 million pounds (about 18 million kg). The International Hydrolyzed Protein Council subsequently reported to the Food & Drug Administration that 63 million pounds (about 29 million kg) were produced in North America in 1994. Assuming that 63 million pounds of hydrolysed vegetable protein were used by the food industry in 1994 and that the population of the USA was 260 million (US Census Bureau, 1999), the per-capita daily disappearance of hydrolysed vegetable protein was estimated to be approximately 300 mg/person per day. An upper-bound residual concentration of 3-chloro-1,2-propanediol of 1 mg/kg (the current voluntary ‘specification’ for industry in the USA) was then used with the estimated intake of hydrolysed vegetable protein to yield an estimated intake of 3-chloro-1,2-propanediol of 0.3 µg/person per day. The report notes that if the hydrolysed vegetable protein industry were to adopt a limit for 3-chloro-1,2-propanediol of 1 mg/kg, the actual intake would be lower than that estimated.

Soya sauces are produced either by natural fermentation of soya or by manufacture of hydrolysed soya proteins and flavourings. The analysis from the USA noted that some soya sauces surveyed in Canada (Health Canada, 2000) and England (Ministry of Agriculture, Fisheries, and Food, 1999a) contained 3-chloro-1,2-propanediol at concentrations > 300 mg/kg. A survey of one brand of soya sauce used in Sweden (Backstrom, 1999) showed concentrations ranging from the limit of detection (not specified, but probably 0.01 mg/kg) to > 120 mg/kg. In a survey of 90 commercially available soya sauces analysed in Canada, 50 samples contained < 1 mg/kg. The average concentration of 3-chloro-1,2-propanediol in the 90 samples was 18 mg/kg. The average concentrations in samples from only three countries exceeded this. The mean intake of soya sauces in the USA that was used in the intake assessment was 8 g/person per day. The estimated intakes of 3-chloro-1,2-propanediol derived by combining the mean residual concentration of the substance in soya sauce and the consumption of soya sauce were 140 µg/person per day at the mean and 290 µg/person per day at the 90th percentile of consumption.

The results for the 90 samples were highly skewed (median, 0.51 mg/kg; mean, 18 mg/kg; mode, 0.01 mg/kg). Elimination of the two highest concentrations of 3-chloro-1,2-propanediol (180 and 330 mg/kg) would leave an average concentration of 13 mg/kg; elimination of the highest 10% would leave an average concentration of < 5 mg/kg (range, 0.01–46 mg/kg). The report stated that the estimates were highly conservative and that only regular consumers of soya sauces would be expected to ingest 3-chloro-1,2-propanediol at the concentrations noted.

5.2.3 Consumption of soya sauce in Australia, Japan, and the USA

Australia submitted information on the intake of soya sauces. The mean intake of consumers of soya sauce was approximately 11 g/person per day, with a 95th percentile intake of 35 g/person per day. The worldwide producer of soya sauces reported that the per-capita consumption of soya sauce in Japan is approximately 10 L/year, or 27 mL/day, or 30 g/person per day. In contrast, the per-capita consumption in the USA is 10 ounces per year or 0.7 mL/day, approximately one-thirtieth of the Japanese intake. The difference between the per-capita intake of soya sauce in the USA described here and the value of 8 g/person per day reported in the national estimate from the USA is due to use only of data from a food consumption survey in the latter. It was noted that the value was likely to be valid only for regular consumers of soya sauces.

In the absence of information on residual concentrations of 3-chloro-1,2-propanediol in soya sauces in Australia and Japan, the intake of this compound in those countries could not be estimated. However, if it is assumed that the residual concentrations of 3-chloro-1,2-propanediol in all soya sauces are similar to those in the 90 soya sauces surveyed in the USA for the national estimate, a rough estimate can be made. This assumption is not unreasonable, as the data used in the USA were for soya sauces produced domestically in Canada and the USA as well as soya sauces imported from Asia. The Australian intake of 3-chloro-1,2-propanediol would be 200 µg/person per day at the mean and 400 µg/person per day at the 90th percentile. The Japanese intake would be 540 µg/person per day at the mean and 1100 µg/person per day at the 90th percentile, assuming consumption of soya sauce at twice the mean.

5.2.4 International estimates of exposure

The descriptions of the GEMS/Food regional diets contain limited information on consumption of soya sauce: the Far Eastern diet includes soya sauce, at a level of 11 g/person per day, and the Middle Eastern diet at 0.1 g/person per day; none is reported in the other regional diets, representing consumption of < 0.1 g/person per day. The background intake of 3-chloro-1,2-propanediol can be estimated roughly from the data submitted by the United Kingdom on residual concentrations in savoury foods. If it is assumed that approximately one-eighth of the diet, 180 g, consists of savoury foods that might contain 3-chloro-1,2-propanediol (out of 1500 g/day solid food, used as the default level in the USA, and that those foods contain a mean residual concentration of 3-chloro-1,2-propanediol of 0.012 mg/kg, the background intake is approximately 2 µg/person per day.

As 3-chloro-1,2-propanediol is found infrequently in foods, it is unlikely that a regulatory limit would have much effect on overall intake. For example. an enforced regulatory limit of 20 µg/kg would lower the mean concentration in savoury foods only from 0.012 to 0.006 µg/kg. However, because the distribution of residual concentrations of 3-chloro-1,2-propanediol in soya sauces is highly skewed and, additionally, brand loyalty might result in regular consumption of contaminated brands, a regulatory limit on the concentration of 3-chloro-1,2-propanediol in soya sauces could have a large effect in reducing the intake of this contaminant. For example, in the national estimate of intake in the USA, removal of the two highest values (out of 90) lowered the average concentration of 3-chloro-1,2-propanediol from 18 to 13 mg/kg, while rejecting the highest 10% (equivalent to rejecting all samples containing > 50 mg/kg) reduced the mean to 5 mg/kg. A regulatory limit for 3-chloro-1,2-propanediol of 1 mg/kg would lower the mean of the distribution of residual concentrations (and the resulting estimated intake) to 0.13 mg/kg.

The mechanism of formation of 3-chloro-1,2-propanediol in acid-hydrolysed vegetable protein is relatively well understood. The protein is manufactured by treating protein-rich extracts from soya beans and other vegetable sources with concentrated hydrochloric acid at high temperature. The treated product is neutralized and supplied either as a liquid suspension containing about 40% solids, as a paste (about 85% solids), or as a dry powder. However, the starting materials also contain small amounts of fats and oils that react with the hot acid to form 3-chloro-1,2-propanediol and, to a lesser extent, other chlorinated propanols.

There are thus several strategies for limiting contamination of the final product:

• reducing the concentrations of fats and oils in the starting materials,
• carefully controlling the acid hydrolysis, and
• removing any chlorohydrins formed during acid hydrolysis, e.g. by alkali treatment.

Research in the Czech Republic has shown that 3-chloro-1,2-propanediol formed by the action of acid in the production of acid-hydrolysed vegetable protein is a racemic mixture of (R) and (S) isomers and that the decomposition rates of the two isomers when treated with alkali are equal (Velisek & Dilezal, 2000).

The dramatic reduction in the concentrations of 3-chloro-1,2-propanediol in acid-hydrolysed vegetable protein on the market in the United Kingdom over the past 10 years is due to use of revised manufacturing processes. In a survey conducted in 1998, 3-chloro-1,2-propanediol could not be quantified (< 0.01 mg/kg) in 21 of 50 samples; a further 17 samples contained < 0.05 mg/kg, and and only two had > 1 mg/kg. In contrast, a similar survey in 1990 showed that the concentration in 31 of 37 samples exceeded 1 mg/kg.

The International Hydrolyzed Protein Council (2001), which represents manufac-turers of acid-hydrolysed vegetable proteins sold in the USA, confirmed that it is technically possible to product acid-hydrolysed vegetable proteins with 3-chloro-1,2-propanediol at concentrations < 0.1 mg/kg. However, the Council’s member companies stated that the organoleptic properties of the products would be adversely affected if the concentration of 3-chloro-1,2-propanediol was reduced below 0.1 mg/kg. Similarly, German producers have stated that they are unable to produce an acceptable product for use containing < 0.1 mg/kg.

No studies have been reported on the possibility of making similar reductions in the concentration of 3-chloro-1,2-propanediol in soya sauce.

The generation of low concentrations of 3-chloro-1,2-propanediol in other foods is less well understood. Trials were conducted by the brewing and malting industries in the United Kingdom to modify their processes for the production of roasted malts and dark, highly flavoured, speciality malts. However, they have been unable to make significant reductions in the concentrations of 3-chloro-1,2-propanediol in these products without compromising the flavour, and hence the quality, of the malt. The concentrations in foods resulting from use of these ingredients are < 0.01 mg/kg (Commission of the European Union, 2000).

7.1.1 Biotransformation

Although the metabolism of 3-chloro-1,2-propanediol has been studied in a variety of test systems, comparisons with human tissue are limited, limiting the possibility of making valid extrapolations in such a way that would allow for quantitative adjustments in risk estimates or extrapolations of safety.

Differences in metabolism between mammals and bacteria form the basis for deciding the relevance of genotoxicity in bacterial test systems to mammals, particularly humans. Certain authors have concluded that there are likely to be important differences between the bacterial and mammalian pathways (Lynch et al., 1998). The issue is complicated by the use of mammalian microsomal fractions in the bacterial test systems. Moreover, conflicting reports have been made concerning the proportion of biotransformation that proceeds by each of the two major metabolic pathways in bacterial and mammalian systems (Jones, 1975; Jones et al., 1978; Jones & O’Brien,1980). Hence, no conclusion can be reached from studies of metabolism about the relevance of bacterial systems. The differences in metabolism are likely to be quantitative and not in kind.

7.1.2 Toxicological studies

The study by Sunahara et al. (1993), in which 3-chloro-1,2-propanediol was administered to Fischer 344 rats for 2 years, is the most appropriate study for assessing the tumorigenic risk for humans. In this study, the incidences of renal tubule adenomas and benign and malignant tumours of the mammary gland in male rats and renal-cell adenomas in female rats were associated with treatment. Increased incidences of interstitial-cell tumours of the testis were also observed in male rats. These tumours are common, variable in incidence, and associated with old age in Fischer 344 rats. This thus represents an equivocal finding and should not be used for risk assessment in this case. The incidence of renal tubule hyperplasia was also increased with treatment in both male and female rats. This lesion is not associated with the chronic progressive nephritis observed in these animals but represents a unique, treatment-related lesion thought to be part of a proliferative process and a precursor to tubule neoplasia. Hence, this observation of treatment-related, dose-dependent tubule hyperplasia strongly corroborates the treatment-related, dose-dependent renal tubule neoplasia in this study.

3-Chloro-1,2-propanediol was mutagenic to bacteria in the standard plate incorporation assay (Stolzenberg & Hine, 1979, 1980; Silhankova et al., 1982) and the preincubation assay (Zeiger et al., 1988; Ohkubo et al., 1995). However, these observations were made only at extremely high doses which would not be recommended today for standard mutagenicity testing. The results are therefore of questionable value. Tests with yeast and mammalian cells in culture also resulted in positive findings (Rossi et al.,1983; Henderson et al., 1987; Görlitz, 1991; May, 1991), but these studies were compromised by conditions such as excessive doses, which seriously detracted from their validity.

In an attempt to clarify the issue and perhaps to provide data that are more relevant to risk assessment, two studies of genotoxicity in mammals were conducted in vivo: an assay for micronucleus formation in rat bone marrow and an assay for unscheduled DNA synthesis (Fellows, 2000; Marshall, 2000). Both studies were well performed, and neither showed genotoxicity.

The most important information from these studies for risk assessment is on the mechanism by which 3-chloro-1,2-propanediol induces carcinogenicity and toxicity. Thus, it can cause irreversible changes that will be ‘remembered’, such as gene alterations. The potential to induce irreversible effects in carcinogenesis has given rise to the hypothesis that there is no practical biological threshold for such a process. This has a profound effect on the procedures used to assess risk or to predict safety.

What is important in the present context is the insight these studies give to the induction of renal and mammary gland neoplasia seen in Fischer 344 rats treated with 3-chloro-1,2-propanediol. Assays of genotoxicity in vivo may be intrinsically less sensitive than those conducted in vitro. Moreover, mouse bone marrow does not provide a good metabolic or chromosomal surrogate for rat kidney or mammary gland. The rat liver used in the assay of unscheduled DNA synthesis may be a somewhat more relevant tissue because of its metabolic capability, but it is not the target tissue. Nevertheless, the negative results in these assays of genotoxicity in vivo indicate that 3-chloro-1,2-propanediol induces neoplasia in rats by a non-genotoxic mechanism. Moreover, the studies conducted in vitro, although with positive results, are so significantly flawed that extrapolation of results for the DNA of these organisms to that of kidney and mammary gland tissue in rats would be ill-advised. Thus, there is no credible evidence that 3-chloro-1,2-propanediol induces neoplasia by a mechanism involving direct interaction with DNA and that there is a practical threshold for this induction.

7.1.3 Human data

No clinical or epidemiological studies or studies with biomarkers that are relevant for establishing dose–response relationships or for risk assessment were available.

The doses administered and the responses in respect of mammary gland tumours, renal tubule-cell adenomas, and hyperplasia in male rats and renal tubule-cell adenomas and hyperplasia in female rats in the study of Sunahara et. al. (1993) are summarized in Table 1.

The most appropriate approach for including an assumption of a biological threshold is the familiar determination of a NOEL for a surrogate biological end-point and the application of factors to ensure safety. The effect used in this approach is not necessarily tumour incidence: if precursor lesions such as preneoplastic hyperplasia are relevant to the process, they should be used.

Health Canada (1999) arrived at a provisional TDI for 3-chloro-1,2-propanediol of 1.1 µg/kg bw per day by applying an uncertainty factor of 1000 to the LOEL of 1.1 mg/kg bw per day found in the long-term study of toxicity and carcinogenicity of Sunahara et al. (1993).

The Quantitative Risk Assessment Committee (2000) of the Food & Drug Administration of the USA determined a unit cancer risk for 3-chloro-1,2-propanediol of 2.1 x 10^–3 (mg/kg bw per day)^–1 on the basis of the same study, citing the occurrence of testis carcinoma in 3/50 rats at a dose of 28 mg/kg bw per day and 0/50 controls. This unit risk is equivalent to a dose of 0.5 µg/kg bw per day for an excess lifetime cancer risk of 1/10⁶. The Committee did not state whether it considered 3-chloro-1,2-propanediol to be a genotoxic carcinogen in humans.

The extrapolation of results in rodents to humans is an exercise in allometry, requiring scaling based on some measure of body size or other relevant physiological or anatomical feature. Cross-species extrapolations tend to be complex, and there are usually no data for achieving the most appropriate result for a specific case. Fortunately, dose per body mass has been used extensively for this purpose and often serves as a default dose metric. This is generally a successful approach for the extrapolation of toxic phenomena. However, other dose metrics have been used for pharmacological extrapolations, such as dose per body surface area, and these are favoured by some. The prediction of risk for humans even with these two metrics can vary by a factor of 5–10 or more when the test species are rodents. This uncertainty in cross-species extrapolation should be kept in mind in the overall assessment process and before excessive demands are made on the accuracy of other parameters.

The study in Fischer 344 rats reported by Sunahara et al. (1993) provides the only appropriate data for a risk assessment based on carcinogenicity for 3-chloro-1,2-propanediol. In this study, the NOEL for tumours of the mammary gland and kidney in male rats corresponds to a mean daily intake of 1.1 mg/kg bw per day. The value for the precursor, renal tubule hyperplasia, appears to be lower than the lowest dose tested, 1.1 mg/kg bw per day. If a strict interpretation were made, this would not be considered to be a NOEL but a dose better characterized as a LOEL, even though it may be close to a NOEL. A larger uncertainty factor than usual should perhaps be used in this case. Similar results were found for female rats, but the value is somewhat higher. The factor chosen to account for the uncertainty associated with cross-species (not scaling) sensitivity and human variation determines the ‘safe’ dose at which no risk exists for humans. Use of factors of 100 to 500, the latter to take into account the uncertainty in the NOEL for tubule hyperplasia, leads to ‘tolerable’ concentrations of 2–10 µg/kg bw per day. An important consideration is that these values represent intake of a chemical that is toxic only after essentially a lifetime. Individual daily doses are not of much consequence; it is cumulative intake that is important, and this should be kept in mind in making recommendations for allowable concentrations in foods.

Absorption, distribution, metabolism, and excretion

3-Chloro-1,2-propanediol crosses the blood–testis barrier and the blood–brain barrier and is widely distributed in body fluids. The parent compound is partly detoxified by conjugation with glutathione, resulting in excretion of the corresponding mercapturic acid, and is partly oxidized to beta-chlorolactic acid and further to oxalic acid. Approximately 30% is broken down to and exhaled as CO₂. In these studies, however, much of the administered dose was not accounted for. Intermediate formation of an epoxide has been postulated but not proven. There is some indication that microbial enzymes can dehalogenate haloalcohols to produce glycidol (a known genotoxin in vitro and in vivo).

Toxicological studies

The oral LD₅₀ of 3-chloro-1,2-propanediol in rats is 150 mg/kg bw.

In several studies in which the compound was given to rats at repeated doses in excess of 1 mg/kg bw per day, it decreased sperm motility and impaired male fertility. At doses > 10–20 mg/kg bw per day, alterations in sperm morphology and epididymal lesions (spermatocoele) were found in rats. 3-Chloro-1,2-propanediol reduced fertility in males of several other mammalian species at slightly higher doses than in the rat.

In rats and mice, 3-chloro-1,2-propanediol at doses > 25 mg/kg bw per day was associated with the development of dose-related central nervous system lesions, particularly in the brain stem.

In several short-term studies in rats and mice, the kidney was the target organ for toxicity. In a 4-week study in rats treated by gavage at 30 mg/kg bw per day, 3-chloro-1,2-propanediol increased the relative kidney weights. In a 13-week study in rats given an oral dose of 9 mg/kg bw per day, a similar effect was seen.

In the pivotal long-term study in Fischer 344 rats, the absolute weight of the kidney was reported to be significantly increased by administration of 3-chloro-1,2-propanediol in drinking-water, at all doses tested. Also at all doses tested, the incidence of tubule hyperplasia in the kidneys of animals of each sex was higher than in controls. Although the incidence did not reach statistical significance at the lowest dose tested (1.1 mg/kg bw per day), the Committee concluded that it represented part of a compound-related dose–response relationship. Overt nephrotoxicity was seen at higher doses (5.2 and 28 mg/kg bw per day).

The results of most assays for mutagenicity in bacteria in vitro were reported to be positive, although negative results were obtained in the presence of an exogenous metabolic activation system from mammalian tissue. The results of assays in mammalian cells in vitro were also reported to be generally positive. It should be noted, however, that the concentrations used in all these assays were very high (0.1–9 mg/ml), raising serious questions about their relevance. The weight of the evidence indicates that 3-chloro-1,2-propanediol is not genotoxic in vitro at concentrations that are not toxic. The results of assays conducted in vivo, including a test for micronucleus formation in mouse bone marrow and an assay for unscheduled DNA synthesis in rats, were negative. The Committee concluded that 3-chloro-1,2-propanediol was not genotoxic in vivo.

Altogether, four long-term studies of toxicity and carcinogenicity were available; three (two with mice and one with rats) did not meet modern standards of quality. Nevertheless, none of the three studies indicated carcinogenic activity. In the fourth study, conducted in Fischer 344 rats, 3-chloro-1,2-propanediol was associated with increased incidences of benign tumours in some organs. These tumours occurred only at doses greater than those causing renal tubule hyperplasia, which was selected as the most sensitive end-point.

Occurrence

3-Chloro-1,2-propanediol has been detected at concentrations in excess of 1 mg/kg in only two food ingredients: acid-hydrolysed vegetable protein and soya sauce. In both ingredients, a range of concentrations has been reported, from below the limit of quantification (0.01 mg/kg with a method that has been validated in a range of foods and food ingredients) up to 100 mg/kg in some samples of acid-hydrolysed vegetable protein and more than 300 mg/kg in some samples of soya sauce.

Formation of 3-chloro-1,2-propanediol in acid-hydrolysed vegetable protein has been found to be related to production processes, and the concentration can be reduced markedly with suitable modifications. The source of 3-chloro-1,2-propanediol in soya sauce is being investigated; by analogy with hydrolysed vegetable protein, however, the contaminant may arise during acid hydrolysis in the manufacture of some products. Traditionally fermented soya sauces would not be contaminated with 3-chloro-1,2-propanediol.

3-Chloro-1,2-propanediol has also been quantified at low concentrations in a range of other foods and food ingredients, notably a number of cereal products that have been subjected to high temperatures, e.g., roasting or toasting. The concentrations are generally less than 0.1 mg/kg. Slightly higher concentrations (up to 0.5 mg/kg) have been found in food ingredients such as malt extracts, but the resulting concentrations in finished foods are below 0.01 mg/kg.

Estimates of dietary intake

Information on the concentrations of 3-chloro-1,2-propanediol in food, food ingredients, and protein hydrolysates was submitted by the United Kingdom, the USA, and the International Hydrolyzed Protein Council. The USA supplied a national estimate of the intake of 3-chloro-1,2-propanediol. Information on the consumption of soya sauce in Australia, Japan and the USA was also received.

At any level of intake that might reasonably be expected, 3-chloro-1,2-propanediol would not have acute effects. This analysis therefore addresses only long-term intake of 3-chloro-1,2-propanediol from its presence in foods.

The data submitted by the United Kingdom showed that 3-chloro-1,2-propanediol is found in some savoury foods, about 30% of samples containing concentrations above the limit of detection of 0.01 mg/kg. The mean residual concentration in these savoury foods was 0.012 mg/kg.

In a survey of 90 samples of commercially obtained soya sauces, 50 samples contained less than 1 mg/kg; the average concentration in the 90 samples was 18 mg/kg. The results of this survey were taken as representative of all soya sauces for the purpose of the intake assessment. Intake of 3-chloro-1,2-propanediol would be dominated by consumption of soya sauces contaminated with the compound.

When estimating the intake of 3-chloro-1,2-propanediol from food other than soya sauce, it was assumed that about one-eighth of the diet, 180 g (on the basis of 1500 g/day of solid food), consists of savoury foods that might contain 3-chloro-1,2-propanediol and that the mean residual concentration of the compound in those foods is 0.012 mg/kg. On this basis, the intake of 3-chloro-1,2-propanediol from foods other than soya sauces is approximately 2 µg/person per day.

The mean and 90th percentile consumption of soya sauce that were used in the intake assessment from the USA were 8 and 16 g/person per day, respectively (consumers only), and the resulting estimates of intake of 3-chloro-1,2-propanediol were 140 µg/person per day for mean consumption and 290 µg/person per day for consumption at the 90th percentile. The mean consumption of soya sauce in Australia (consumers only) was approximately 11 g/person per day, and for consumers at the 95th percentile it was approximately 35 g/person per day, resulting in intake of 3-chloro-1,2-propanediol of 200 µg/person per day for mean consumption of soya sauce and 630 µg/person per day at the 90^th percentile of consumption. Per-capita consumption of soya sauce in Japan (approximating consumption by consumers only) was approximately 30 g/person per day, resulting in an intake of 3-chloro-1,2-propanediol of approximately 540 µg/person per day for mean consumption of soya sauce. Intake at the 95th percentile in Japan would be 1100 µg/person per day assuming consumption of soya sauce that is twice the mean.

The Committee chose tubule hyperplasia in the kidney as the most sensitive end-point for deriving a tolerable intake. This effect was seen in the long-term study of toxicity and carcinogenicity in rats in a dose-related manner, although the effect did not reach statistical significance at the lowest dose. The Committee concluded that the LOEL was 1.1 mg/kg bw per day and that this was close to a NOEL.

The Committee established a provisional maximum tolerable daily intake (PMTDI) of 2 µg/kg bw for 3-chloro-1,2-propanediol on the basis of the LOEL of 1.1 mg/kg bw per day and a safety factor of 500, which included a factor of 5 for extrapolation from a LOEL to a NOEL. This factor was considered to be adequate to allow for the absence of a clear NOEL and to account for the effects on male fertility and for inadequacies in the studies of reproductive toxicity. Data available to the Committee indicated that the estimated mean intake of 3-chloro-1,2-propanediol by consumers of soya sauce would be at or above this PMTDI.

As 3-chloro-1,2-propanediol is found infrequently in foods, a regulatory limit would be unlikely to have much effect on the overall intake of non-consumers of soya sauces. However, because the distribution of residual 3-chloro-1,2-propanediol in soya sauce is highly skewed and because it is likely that brand loyalty could result in regular consumption of highly contaminated brands, a regulatory limit on the concentration of 3-chloro-1,2-propanediol in soya sauce could markedly reduce the intake by consumers of this condiment.

AOAC International (2000) AOAC Official Method 2000.01: Determination of 30-chloro-1,2-propanediol in food and food Ingredients.

Backstrom, K. (1999) Unpublished information dated 27 October 1999, provided from Livsmedels Verket, Sweden, to P. Wilson, Food & Drug Administration, Washington DC, USA.

Ban, Y., Asanabe, U., Ingaki, S., Sasaki, M., Nakatsuka, T. & Matsumoto, H. (1999) Effects of alpha-chlorohydrin on rat sperm motions in relation to male reproductive functions. J. Toxicol. Sci., 24, 407–413.

Black, D.J., Glover, T.D., Shenton, J.C. & Boyd, G.P. (1975) The effects of alpha-chlorohydrin on the composition of rat and rabbit epididymal plasma: a possible explanation of species difference. J. Reprod. Fertil., 45, 117–128.

Brown-Woodman, P.D., White, I.G. & Salamon, S. (1975) Effects of alpha-chlorohydrin on the fertility of rams and the metabolism of spermatozoa in vitro. J. Reprod. Fertil., 43, 381.

Cavanagh, J.B. & Nolan, C.C. (1993) The neurotoxicity of alpha-chlorohydrin in rats and mice: II. Lesion topography and factors in selective vulnerability in acute energy deprivation syndrome. Neuropathol. Appl. Neurbiol., 19, 471–479.

Cavanagh, J.B., Nolan, C.C. & Seville, M.P. (1993) The neurotoxicity of alpha-chlorohydrin in rats and mice: I. Evolution of the cellular changes. Neuropathol. Appl. Neurobiol., 19, 240–252.

Collier, P.D., Cromiue, D.D.O. & Davies, A.P. (1991) Mechanisms of formation of chloropropanols present in protein hydrolysates. J. Am. Oil Chem. Soc., 68, 785–790.

Commission of the European Communities (2000) Final report of Scientific Cooperation Task 3.2.6 ‘Provision of validated methods to support the Scientific Committee on Foods recommendation regarding 3-monochloropropane-1,2-diol in hydrolysed vegetable protein and other foods’, Brussels.

Cooper, E.R.A. & Jackson, H. (1973) Chemically induced sperm retention cysts in the rat. J. Reprod. Fertil., 34, 445–449.

Coppola, J.A. (1969) An extragonadal male antifertility agent. Life Sci., 8, 43–48.

Crabo, B. & Appelgren, L.E. (1972) Distribution of ¹⁴C-alpha-chlorohydrin in mice and rats. J. Reprod. Fertil., 30, 161–163.

Dobbie, M.S., Porter, K.E. & Jones, A.R. (1988) Is the nephrotoxicity of (R)-3-chlorolactate in the rat caused by 3-chloropyruvate? Xenobiotica, 18, 1389–1399.

Edwards, E.M., Jones, A.R. & Waites, G.M.H. (1975) The entry of alpha-chlorohydrin into body fluids of male rats and its effect upon incorporation of glycerol into lipids. J. Reprod. Fertil., 43, 225–232.

Epstein, S.S., Arnold, E., Andrea, J., Bass, W. & Bishop, Y. (1972) Detection of chemical mutagens by the dominant lethal assay in the mouse. Toxicol. Appl. Pharmacol., 23, 288–325.

Erickson, G.I. & Bennett, J.P. (1971) Mechanism of antifertility activity of minimal dose level of alpha-chlorohydrin in the male rat. Biol. Reprod., 5, 98.

Ericsson, R.J. & Baker, V.F. (1970) Male antifertility compounds: Biological properties of U-5897 and U-15,646. J. Reprod. Fertil., 21, 267–373.

Ericsson, R.J. & Youngdale., G.A. (1970) Male antifertility compounds: Structure and activity relationships of U-5897, U-15,646 and related substances. J. Reprod. Fertil., 21, 263–266.

Fellows, M. (2000) 3-MCPD: Measurement of unscheduled DNA synthesis in rat liver using an in vitro/in vivo procedure. Unpublished report No. 1863/1-D5140 from Covance Laboratories Ltd.

Food & Drug Administration (2000) Submission to FAO dated 14 December 2000.

Food Standards Agency (2001a) Survey of 3-monochloropropane-1,2-diol (3-MCPD) in food ingredients, Food Safety Information Sheet 11/01, February 2001.

Food Standards Agency (2001b) Survey of 3-monochloropropane-1,2-diol (3-MCPD) in selected food groups, Food Safety Information Sheet 12/01, February 2001.

Frei, H. & Würgler, F.E. (1997) The vicinal chloroalcohols 1,3-dichloro-2-propanol (DC2P), 3-chloro-1,2-propanediol (3CPD) and 2-chloro-1,3-propanediol (2CPD) are not genotoxic in vivo in the wing spot test of Drosophila melanogaster. Mutat. Res., 394, 59–68.

Gill, S.K. & Guraya, S.S. (1980) Effects of low doses of alpha-chlorohydrin on phosphatase, beta-glucosidase, beta-glucuronidase and hyaluronidase of rat testis and epididymis. Ind. J. Exp. Biol., 18, 1351–1352.

Gunn, S.A., Gould, T.C. & Anderson, W.A.D. (1969) Possible mechanism of posttesticular antifertility action of 3-chloro-1,2-propanediol. Proc. Soc. Exp. Biol. Med., 132, 656–659.

Görlitz, B.D. (1991) In vitro mammalian cell HPRT-test with 3-chloro-1,2-propanediol. Unpublished report No. G91/3 from Fraunhofer-Institute für Toxikologie und Aerosolforschung, Hanover, Germany.

Health Canada (1999) Memorandum on 3-chloro-1,2-propanediol dated 21 October 1999, Government of Canada, Ministry of Public Health, Food Additives and Contaminants Section, Ottawa.

Health Canada (2000) Survey of soy sauces and similar products, summary results reported to FAO by the Food and Drug Administration, memorandum dated 20 April 2000, Ottawa.

Helal, T.Y. (1982) Chemosterilant and rodenticidal effects of 3-chloro-1,2-propanediol (Epibloc) against the albino laboratory rat and the Nile field rat. Int. Pest. Control, 24, 20–23.

Henderson, L.M., Bosworth, H.J., Ransome, S.J., Banks, S.J., Brabbs, C.E. & Tinner, A.J. (1987) An assessment of the mutagenic potential of 1,3-dichloro-2-propanol, 3-chloro-1,2-propanediol and a cocktail of chloropropanols using the mouse lymphoma TK locus assay. Unpublished report No. ULR 130 ABC/861423 from Huntingdon Research Centre Ltd, Huntingdon, Cambridgeshire, United Kingdom.

Hoffer, A.P., Hamilton, D.W. & Fawcett, D.W. (1973) The ultrastructural pathology of the rat epididymis and after administration of alpha-chlohydrin (U-5897). 1. Effects of a single high dose. Anat. Rec., 175, 203–230.

Homonnai, Z.T., Paz, G., Sofer, A. Yedwab, G.A. & Kraicer, P.F. (1975) A direct effect of alpha-chlorohydrin on motility and metabolism of ejaculated human spermatozoa. Contraception, 12, 579–589.

Hoyt, J.A., Fisher, L.F., Hoffman, W.P., Swisher, D.K. & Seyler, D.E. (1994) Utilization of a short-term male reproductive toxicity study design to examine effects of alpha-chlorohydrin (3-chloro-1,2-propanediol). Reprod. Toxicol., 8, 237–250.

International Hydrolyzed Protein Council (2001) Submission to FAO dated 28 February 2001.

Jaccaud, E. & Aeschbacher, H.U. (1989) Evaluation of 3-chloro-1,2-propanediol (3MCPD) in the bone marrow and colonic micronucleus mutagenicity tests in mice. Unpublished report No. 1265 from Nestec Ltd Research Centre, Nestlé, Switzerland.

Jones, A.R. (1975) The metabolism of 3-chloro-, 3-bromo- and 3-iodopropan-1,2-diol in rats and mice. Xenobiotica, 5, 155–165.

Jones, A.R. (1978) The antifertility actions of alpha-chlorohydrin in the male. Life Sci., 23, 1625–1646.

Jones, A.R. (1983) Antifertility actions of alpha-chlorohydrin in the male. Aust. J. Biol. Sci., 36, 333–350.

Jones, A.R. & Cooper, T.G. (1999) A re-appraisal of the post-testicular action and toxicity of chlorinated antifertility compounds. Int. J. Androl., 22, 130–138.

Jones, P. & Jackson, H. (1976) Antifertility and dominant lethal mutation studies in male rats with dl-alpha-chlorohydrin and an amino-analogue. Contraception, 13, 639–646.

Jones, A.R. & Murcott, C. (1976) The oxidative metabolism of alpha-chlorohydrin and the chemical induction of spermatocele. Experientia, 32, 1135–1136.

Jones, A.R. & O’Brien, R.W. (1980) Metabolism of three active analogues of the male antifertility agent, chlorohydrin, in the rat. Xenobiotica, 10, 365–370.

Jones, A.R. & Porter, L.M. (1995) Inhibition of glycolysis in boar spermatozoa by alpha-chlorohydrin phosphate appears to be mediated by phosphatase activity. Reprod. Fertil. Dev., 7, 1089–1094.

Jones, A.R., Davies, P., Edwards, K. & Jackson, H. (1969) Antifertility effects and metabolism of alpha- and epi-chlorohydrins in the rat. Nature, 224, 83.

Jones, A.R., Milton, D.H. & Murcott, C. (1978) The oxidative metabolism of alpha-chlorohydrin in the male rat and the formation of spermatocele. Xenobiotica, 8, 573–582.

Jones, A.R., Gadel, P. & Murcott, C. (1979) The renal toxicity of the rodenticide alpha-chlorohydrin in the rat. Naturwissenschaften, 66, 425.

Kalla, N.R. & Bansal, M.P. (1977) In vivo and in vitro alkylation of testicular cysteine by alpha-chlorohydrin administration. Ind. J. Exp. Biol., 15, 232–233.

Kalla, N.R. & Singh, B. (1981) Synergistic effect of alpha-chlorohydrin on the influence of copper ions on human spermatozoa. Int. J. Fertil., 26, 65–57.

Kaur, S. & Guraya, S.S. (1981a) Effects of low doses of alpha-chlorohydrin on the enzymes of glycolytic and phosphogluconate pathways in the rat testis and epididymis. Int. J. Androl., 4, 196–207.

Kaur, S. & Guraya, S.S. (1981b) Effects of low doses of alpha-chlorohydrin on the dehydrogenase and oxidase of rat epididymal epithelium and sperms: A correlative histochemical and biochemical study. Andrologia, 13, 225–231.

Kaur, S. & Guraya, S.S. (1981c) Biochemical observations on the protein and nucleic acid metabolism of the rat testis and the epididymis after treatment with low doses of alpha-chlorohydrin. Int. J. Fertil., 26, 8–13.

Kirton, K.T., Erickson, R.J., Ray, J.A. & Porbes, A.D. (1970) Male antifertility compounds: Efficacy of N-5897 in primates (Macaca mulatta). J. Reprod. Fertil., 21, 275–278.

Kluwe, W.M., Gupta, B.N. & Lamb, J.C., IV (1983) The comparative effects of 1,2-dibromo-3-chloropropane (DBCP) and its metabolites, 3-chloro-1,2-propaneoxide (epichlorohydrin), 3-chloro-1,2-propanediol (alpha-chlorohydrin), and oxalic acid, on the urogenital system of male rats. Toxicol. Appl. Pharmacol., 70, 67–86.

Lohika, N.K. & Arya, M. (1979) Antifertility activity of alpha-chlorohydrin (3-chloro-1,2-propanediol, U-5897) on the female rats. Acta Eur. Fertil., 10, 23–28.

Lynch, B.S., Bryant, D.B., Hook, G.J., Nestmann, E.R. & Munro, I.C. (1998) Carcinogenicity of monochloro-1,2-propanediol (alpha-chlorohydrin, 3-MCPD). Int. J. Toxicol., 17, 47–76.

Majeska, J.B. & Matheson, D.W. (1983) Quantitative estimate of mutagenicity of tris-[1,3-dichloro-2-propyl]-phosphate (TCPP) and its possible metabolites in Salmonella (abstract). Environ. Mutag., 5, 478.

Marchesini, M. & Stalder, R. (1983) Toxicity of 3-chloro-1,2-propanediol in a 4 weeks gavage study on rats. Part I. Unpublished report No. LA 70/1082 from the Société d’Assistance Technique Pour Produits Nestlé SA, Switzerland.

Marchesini, M., Stalder, R. & Perrin, I. (1989) Subchronic toxicity of 3-chloro-1,2-propanediol, 90 days administration in drinking water of Fischer F344 rats. Unpublished report No. 1264 from Nestec Ltd Research Centre, Nestlé, Switzerland.

Marshall, R.M. (2000) 3-MCPD: Induction of micronuclei in the bone-marrow of treated rats. Unpublished report No. 1863/2-D5140 from Covance Laboratories Ltd.

May, C. (1991) In vitro sister chromatid exchange assay in mammalian cells. Unpublished report No. 91/4 CM from Fraunhofer-Institute für Toxikologie und Aerosolforschung, Hanover, Germany.

Ministry of Agriculture, Fisheries, and Food (1999a) Survey of 3-monochloropropane-1,2-diol (3-MCPD) in acid hydrolysed vegetable protein, Food Surveillance Information Sheet 181, London.

Ministry of Agriculture, Fisheries, and Food (1999b) Survey of 3-monochloropropane-1,2-diol (3-MCPD) in soy sauce and related products, Food Surveillance Information Sheet 187. London.

Mohri, H., Suter, D.A.J., Brown-Woodman, P.D.C., White, J.G. & Ridley, D.D. (1975) Identification of the biochemical lesion produced by alpha-chlorohydrin in spermatozoa. Nature, 255, 75–77.

Morris, J.D. & Williams, L.M. (1980) Some preliminary observations of the nephrotoxicity of the male antifertility drug (±) alpha-chlorohydrin. J. Pharm. Pharmacol., 32, 35–38.

National Toxicology Program (1990) Toxicology and carcinogenesis studies of glycidol (CAS No. 556-52-5) in F344/N rats and B6C3F1 mice (gavage studies), NTP Technical Report Series No. 374; NTP-TR-374; NIH 90-2829, Research Triangle Park, North Carolina, USA.

Ohkubo, T., Hayashi, T., Watanabe, E., Endo, H., Goto, S., Endo, O., Mizoguchi, T. & Mori, Y. (1995) Mutagenicity of chlorohydrins. Nippon Suisan Gakkaishi, 61, 596–601 (in Japanese).

Painter, R.B. & Howard, R (1982) The HeLa DNA synthesis inhibition test as a rapid screen for mutagenic carcinogens. Mutat. Res., 92, 427–437.

Parish, W.E. (1989) Effect of 3-chloropropane-1,2-diol on rat fertility. Unpublished summary report No. D 89/005 from Unilever Research, Sharnbrook, Bedford, United Kingdom.

Paz, G., Carmon, A. & Homonnai, Z.T. (1985) Effect of alpha-chlorohydrin on metabolism and testosterone secretion by rat testicular interstitial cells. Int. J. Androl., 8, 139–146.

Piasecki, A., Ruge, A. & Marquardt, H. (1990) Malignant transformation of mouse M2-fibroblasts by glycerol chlorohydrines contained in protein hydrolysate and commercial food. Arzneim.-Forsch./Drug Res., 40, 1054–1055.

Porter, K.G. & Jones, A.R. (1982) The effect of the isomers of alpha-chlorohydrin and racemic beta-chlorolactate on the rat kidney. Chem.-Biol. Interactions, 41, 95–104.

Quantitative Risk Assessment Committee (2000) Assessment of carcinogenic upper bound lifetime risk from 3-monochloro-1,2-propanediol exposure in soy and oyster sauces and related products. Report dated 17 November 2000, Washington DC, USA, Food & Drug Administration.

Rossi, A.M., Migliore, L., Lascialfari, D., Sbrana, I., Lorieno, N., Tortoreto, M., Bidioli, F. & Pantarotto, C. (1983) Genotoxicity, metabolism and blood kinetics of epichlorohydrin in mice. Mutat. Res., 118, 213–226.

Samojlik, E. & Chang, M.C. (1970) Antifertility activity of 3-chloro-1,2-propanediol (U-5897) on male rats. Biol. Reprod., 2, 299–304.

Silhankovà, L., Smid, F., Cernà, M., Davidek, J. & Velisek, J. (1982) Mutagenicity of glycerol chlorohydrines and of their esters with higher fatty acids present in protein hydrolysate. Mutat. Res., 103, 77–81.

Stevenson, D. & Jones, A.R. (1984) The action of (R)- and (S)-alpha-chlorohydrin and their metabolites on the metabolism of boar sperm. Int. J. Androl., 7, 79–86.

Stolzenberg, S.J. & Hine, C.H. (1979) Mutagenicity of halogenated and oxygenated three-carbon compounds. J. Toxicol. Environ. Health, 5, 1149–1158.

Stolzenberg, S.J. & Hine, C.H. (1980) Mutagenicity of 2- and 3-carbon halogenated compounds in the Salmonella/mammalian-microsome test. Environ. Mutag., 2, 59–66.

Sunahara, G., Perrin, I. & Marchesini, M. (1993) Carcinogenicity study on 3-monochloropropane-1,2-diol (3-MCPD) administered in drinking water to Fischer 344 rats. Unpublished report No. RE-SR93003 submitted to WHO by Nestec Ltd, Research & Development, Switzerland.

Suter, D.A.I., Brown-Woodman, P.D.C., Mohri, H. & White, G.I. (1975) The molecular site of action of the anti-fertility agent, alpha-chlorohydrin, in ram spermatozoa. J. Reprod. Fertil., 43, 382–383.

Toth, G.P., Wang, S.R., McCarthy, H., Tocco, D.R. & Smith, M.K. (1992) Effects of three male reproductive toxicants on rat cauda epididymal sperm motion. Reprod. Toxicol., 6, 507–515.

Turner, M.A. (1971) Effects of alpha-chlorohydrin upon the fertility of spermatozoa of the cauda epididymides of the rat. J. Reprod. Fertil., 24, 267–269.

US Census Bureau (1999) Statistical Abstract of the United States, Washington DC, Government Printing Office

Van Duuren, B.L., Goldschmidt, B.M., Katz, C., Seidman, J. & Paul, J.S. (1974) Carcinogenic activity of alkylating agents. J. Natl Cancer Inst., 53, 695–700.

Velisek, J.D. & Dilezal, M. (2000) Summary of research findings reported to JECFA by the United Kingdom Food Standards Agency.

Velisek, J.D., Davidek, J., Kubelka, V., Janicek, G., Svobodova, Z. & Simicova, Z. (1980) New chlorine containing organic compounds in protein hydrolysates. J. Agric. Food Chem., 28, 1142–1144.

Weisburger, E.K., Ulland, B.M., Nam, J., Gart, J.J. & Weisburger, J.H. (1981) Carcinogenicity tests of certain environmental and industrial chemicals. J. Natl Cancer Inst., 67, 75–88.

van den Wijngaard, A.J., Janssen, D. & Witholt, B. (1989) Degradation of epichlorohydrin and halohydrins by bacterial cultures isolated from freshwater sediments. J. Gen. Microbiol., 135, 2199–2208.

Windholz, M., ed. (1976) The Merck Index—An Encyclopedia of Chemicals and Drugs, 9th Ed., Rahway, New Jersey, Merck & Co.

Woods, J. & Garside, D.A. (1996) An in vivo and in vitro investigation into the effects of alpha-chlorohydrin on sperm motility and correlation with fertility in the Han Wistar rat. Reprod. Toxicol., 10, 199–207.

Yamada, T., Inoue, T., Sato, A., Yamagishi, K. & Sato, M. (1995) Effects of short-term administration of alpha-chlorohydrin on reproductive toxicity parameters in male Sprague-Dawley rats. J. Toxicol. Sci., 20, 195–205.

Zeiger, E., Anderson, B., Haworth, S., Lawlor, T. & Mortelmans, K. (1988) Salmonella mutagenicity tests: IV. Results from the testing of 300 chemicals. Environ. Mol. Mutag., 11 (Suppl. 12), 1–158.

    See Also:
       Toxicological Abbreviations