Bisphenol A (4,4'-Isopropylidenediphenol)

Bisphenol A (BPA) is used primarily in the manufacture of epoxy resins and polycarbonate plastics. It can be found in interior coatings for cans and drums, reinforced pipe, adhesives, flooring, water main filters, baby bottles, artificial teeth, nail polish, and food packaging materials. Although insoluble in water, unreacted BPA can migrate from the resins used in food packaging to food surfaces. Therefore, the primary source of exposure to BPA for most people is through the diet. In 2004, the estimated production of BPA in the United States was approximately 2.3 billion pounds. Biomonitoring studies show that human exposure to BPA is widespread. Bisphenol A passed the animal data screen, underwent a preliminary toxicological evaluation, and is being brought to the Carcinogen Identification Committee for consultation. This is a compilation of the relevant studies identified during the preliminary toxicological evaluation. Epidemiological data No cancer epidemiology studies were identified. Animal carcinogenicity data • Long-term feeding studies o 103 week studies in male and female Fisher 344 rats: NTP (1982)  Increases (marginal) in leukemia in females  Increase in leukemia (marginal), testicular interstitial cell tumors (by pairwise comparison and trend), and mammary gland fibroadenomas (by trend) in males o 103 week studies in male and female B6C3F 1 mice: NTP (1982)  No treatment-related tumor findings in females  Increase in pituitary chromophobe carcinoma (by trend), lymphoma (by pairwise comparison) in males • Prenatal exposure studies o Mammary tumor study in female Wistar-Furth rats; evaluation at age 110 days: Murray et al. (2007)  Increase in mammary gland carcinoma in situ (by pairwise comparison) in female offspring o Reproductive system tumor study in female CD-1 mice; evaluation at age 18 months: Newbold et al. (2009)  Increase (non-significant) in reproductive system tumors

Bisphenol A (BPA) is an intermediate used in the production of polycarbonate, epoxide resins and styrene polyester resins. It is found in many everyday articles. A MAK Documentation (Greim 1996, translated;Greim 2000, translated), an EU Risk Assessment (EU 2003) and a BUA substance report with a supplementary report (BUA 1997(BUA , 1999 are available on the substance.

Metabolism and Toxicokinetics
Orally administered BPA is absorbed from the gastrointestinal tract and is completely eliminated. In urine, BPA is mainly present in the form of a glucuronic acid conjugate. In a feeding study with rats, 80% of the substance was excreted with the urine and faeces within two days after administration of a dose of 800 mg/kg body weight ( 14 C-labelled at position C2 in the propyl group, diluted with unlabelled substance at a ratio of 1:10, in the form of a 10% solution in propylene glycol) (Knaak and Sullivan 1966). Elimination was completed after 8 days; approximately 2/3 of the dose was excreted with the faeces and 1/3 with the urine. The biological half-life in urine was given as one day.
In a study with individuals, the kinetics of BPA was investigated after oral administration of 5 mg d 16 -BPA (Völkel et al. 2002). Free d 16 -BPA could not be found in the plasma, though the BPA glucuronide appeared rapidly in the plasma, attaining its maximum already after 33 minutes. Subsequently, the time course of the concentration changes showed a bi-exponential curve with half-lives of 89 and 202 minutes (see Figure 1). The data measured indicate a rapid and effective glucur-2 Figure 1 Concentration changes over time of glucuronide in the plasma of individuals after oral administration of 5 mg d 16 -BPA (Völkel et al. 2002) onidation in the liver and the absence of an enterohepatic recirculation. From these data, assuming that the absorbed BPA is immediately converted to glucuronide, a two-compartment model was developed (Völkel et al. 2002).
As the clearance of the elimination from the plasma compartment (0.12 ml/min) agreed well with creatinine clearance and with the kidney clearance calculated from the elimination data (0.13 ± 0.04 ml/min), the authors concluded that the BPA glucuronide does not bind to serum proteins and the elimination clearance from the plasma compartment corresponds to the kidney clearance (Völkel et al. 2002).
No data are available on bioavailability after dermal or inhalation exposure. According to a study by Schäfer et al. (2000) BPA does not accumulate in the endometrium or fatty tissue.

Critical Toxicity
Comprehensive literature on the toxicology of BPA in animal studies and humans has been compiled in the occupational-medical toxicological documentation of the MAK value of BPA (4.4′-isopropylidenediphenol) (Greim 1996, translated;Greim 2000, translated). In addition, a BUA substance report with supplementary report ("Bisphenol A (2,2′-bis-(4-hydroxyphenyl)propane)" reports no. 203 and no. 215) are available (BUA 1997(BUA , 1999. The most recent state of knowledge from animal study findings and in vitro investigations including the weak oestrogenic effect (weaker than that of oestradiol by a factor of 1 000) is described comprehensively in the occupational-medical toxicological documentation of the MAK value (Greim 1996, translated;Greim 2000, translated). The MAK value is based on the results of a 90-day inhalation study in rats resulting in a NOAEL of 10 mg/m 3 . Apart from non-specific effects without histological correlates, 50 or 150 mg/m 3 caused inflammatory hyperplastic changes in the frontal nose parts which were reversible (Greim 1996, translated).
The MAK documentation also provides data in humans as regards eye, nose and throat irritation at 5-15 mg/m 3 as well as data on skin and photosensitization.
Critical effects from the animal studies are hyperplastic changes in the frontal nose parts, a decrease in body weight and in a number of organ weights, as well as a weak oestrogenic effect. In humans the critical effects are eye, nose and throat irritation.

Exposure and Effects
Every exposure that can result in absorption either orally or by inhalation is to be assessed as external exposure to BPA. Although no data are available on the bioavailability of the substance after inhalation exposure, the MAK value is defined as concentration of the inhalable dust fraction, thus meaning that the contribution of exposure from inhalation -assuming complete bioavailability -should be calculated at a respiratory volume of 10 m 3 per working day. The internal exposure is determinable in the form of the BPA glucuronide in blood; free BPA was below the detection limit in a study in humans (Völkel et al. 2002).

Relationship between internal and external exposure
There are no studies on relationships between external and internal exposure in humans. Nevertheless, the requirements for a modeling are given.
The toxicokinetic model of Völkel et al. (2002) was modified in order to describe the concentration changes over time of the BPA glucuronide in plasma and its elimination in urine after 8-hour inhalation of 5 mg BPA/m 3 . In these calculations, a pulmonary respiratory activity of 1.76 m 3 /h with a work load of 50 W and an inhalative absorption of 100% was assumed for the time course of absorption (Csanády and Filser 2001). This resulted in absorption of about 70 mg BPA. Figure 2 shows that, according to the model calculations, the plasma concentration of BPA glucuronide attains about 87% of the value expected at the end of exposure already after three hours of exposure. However, measurement of the BPA glucuronide concentration in plasma does not appear to be suitable for the determination of the exposure, because the plasma glucuronide concentration decreases very rapidly when exposure ends. Based on the model calculations, it is expected that about 78% of the absorbed BPA is excreted with the urine in the form of 0.24 mmol BPA glucuronide (corresponding to 55 mg released bisphenol A) at the end of exposure (see Figure 3).

Relationships between internal exposure and effects
No studies with humans are available.

Selection of Indicators
Taking the pharmacokinetics of BPA into consideration, the determination of BPA glucuronide in blood and urine are possible methods. However, due to the rapid elimination from the blood, determination of BPA glucuronide in urine is the only practical alternative.

Methods
Total BPA in urine can be determined after incubation with β-glucuronidase using GC-MS (Arakawa et al. 2004).
The Working Group "Analyses of Hazardous Substances in Biological Material" has developed a method for this. After enzymatic hydrolysis using β-glucuronidase the analyte is enriched on a solid phase. This is followed by derivatization with pentafluorobenzoyl chloride. Determination is by GC/MS. The quantification limit is given as 0.1 μg/l (Leng et al. 2008, translated).

Background Exposure
In the environment, BPA is ubiquitous. It has been found in surface waters, sediments and in the ground water (EU 2003). Foods, especially those enclosed in packing material containing BPA, are secondary sources of exposure.
In a study with 36 non-exposed male individuals, the excretion of BPA in urine was determined over 5 days and found to be between < 0.21 and 14 μg/day (median 1.2 μg/day), corresponding to < 0.003-0.23 μg/kg body weight and day (median 0.02 μg/kg body weight and day). The intra-individual and inter-individual variation of BPA elimination in the urine in 5 individuals over 5 days is given as 91% and 84%, respectively (Arakawa et al. 2004).

Evaluation of the "Biologischer Leitwert" (BLW)
As no field studies on external and internal exposure are available, the derivation of a BAT value is not possible. As alternative, toxicokinetic data were considered, according to which the BPA concentration in urine amounts to 0.34 mmol BPA glucuronide ≙ 78 mg free BPA/l urine once the steady state has been reached after exposure to BPA at 5 mg/m 3 . A urine volume of 0.7 l excreted during an 8-hour shift was assumed and a work load of 50 W taken as basis.
Under these conditions, a BLW of 80 mg bisphenol A (after hydrolysis)/l urine is established. Samples have to be taken at the end of exposure or end of shift.

Interpretation of the Analytical Data
Owing to the kinetic circumstances (see Figure 2 and Figure 3), the conditions immediately at the end of the shift were taken as basis to derive the BLW. Application of the BLW is very closely bound to this sampling time.
At present no documentation is available as to the extent of occupational medical check-ups according to a guideline of the Berufsgenossenschaft (Employers' Liability Insurance Association).
The BLW relates to normally concentrated urine, in which the creatinine content should be in the range of 0.5-2.5 g/l. In the case of highly concentrated or greatly diluted urine, the physician must assess whether the BLW can be considered as being adhered to or exceeded. As a rule, where urine samples are outside the limits given above, repetition of the measurement in normally hydrated persons is recommended. A similar mode of renal elimination of BPA and creatinine has at present not been demonstrated, for which reason creatinine correction of the measured value is not considered necessary. Analytical determination of the urinary creatinine concentration additionally involves measurement inaccuracy, as reflected in intercomparison programmes on creatinine analysis in urine. Only 60-70% of the participating laboratories could obtain values within the tolerance range (Schaller et al. , 2004.