BPA Toxicokinetics

IFC Comments on Recent Mathematical Modeling study of BPA Toxicokinetics by Edginton and Ritter

Edington and Ritter present a mathematical model of the toxicokinetics of BPA and its glucuronidated metabolite, BPA-Glu, and estimate that the average steady state BPA plasma concentration in newborns is eleven times that in adults, when given the same weight-normalized dose.1  Additionally, they estimate by simulating typical feeding exposures, infants 3 and 6 months old show a five times greater steady state BPA plasma concentration when compared to adults.  However, this article contains no actual data regarding the metabolism of BPA in infants and young children.  Rather, it incorporates data on BPA metabolism from other studies, especially a study in adults by Volkel et al.2 (The Volkel et al. study employed a bolus administration of a 5-mg dose of fully deuterated BPA (roughly 70 mcg/kg bw for a 70-kg subject) and then tracked the excretion profile of BPA in the free and glucuronidated forms.)  Edington and Ritter use computer modeling to scale the information on BPA absorption, tissue update, glucuronidation, and excretion to a young child.  A concern is that there is tremendous amount of uncertainty in trying to apply the results of such a study, conducted in adults, to children.  The study of Volkel et al., while representing the most detailed information on BPA metabolism in humans, is still an unphysiological administration of BPA (roughly 35-70 times a typical adult exposure), given as a bolus.  In studies like this, it is important to remember that for every time that an assumption or extrapolation is made, the probability of significant error increases.  Regardless of how sophisticated a modeling system may be, there is no substitute for representative exposure data in the population of interest.

Further, in the Edginton and Ritter study, there were no good data, at the time this article was published, regarding the identity of the UDP-glucuronsyltransferase (UGT) enzyme(s) responsible for BPA glucuronidation in human infants; also there was little information on the kinetics of the activity of the enzyme(s).   Based on some data from previous studies, Edginton and Ritter assumed that UGT2B7 was the predominant isoform that accomplished BPA glucuronidation in humans and, because its activity in term neonates is only 5% that of adults, the further assumption was made that glucuronidation of BPA is dramatically reduced in neonates versus adults, especially in the early neonatal period.  However, a recent study of pooled adult human liver microsomes strongly suggests that the UGT2B15 isoform is the main UGT in humans.3  A recombinant preparation of UGT2B15 demonstrated a Michaelis-Menten constant (Km) that was very similar to that of pooled liver microsomes. Further, the Km values of the purified UGT2B15 preparation and the liver microsomes were 8.68 and 6.39 micromolar, respectively.  A BPA concentration of 8.68 micromolar (roughly 1913 ppb) suggests the hepatic microsomal UGT activity is functioning at only half its maximal velocity.   Given that the liver is exposed to plasma concentrations of BPA in portal blood that are much lower than these Km values, it is highly unlikely that this particular UGT could be saturated under conditions of normal dietary exposure.  Hanioka et al. also concluded that BPA is likely to broadly interact with a number of hepatic UGTs, including UGT2B7.  Thus, the odds are even greater that full glucuronidation of environmentally relevant doses of BPA will occur.

Edginton and Ritter propose that the average steady-state BPA plasma concentration in newborns is estimated to be 11 times greater than that in adults when given the same weight-normalized dose.  However, given the uncertainties involved in extrapolating results of adult BPA metabolism studies to children and new information on the identity of the UGT(s) responsible for BPA metabolism in humans, it is likely that the calculations of Edginton and Ritter represent an overestimate of circulating free BPA levels.  In concordance, the European Food Safety Authority (EFSA) in 2008 estimated that, based on infant studies of the glucuronidation of structurally similar molecules such as paracetamol, BPA doses of up to 1 mg/kg bw/day (300-500 times a typical infant exposure) could be glucuronidated by infants.4  While there is no question that the UGT activity of infants is typically lower than it is for adults, the more important question is whether the infant level of UGT activity is sufficient to cover the minute doses of BPA to which infants are exposed.  Based on the new data presented by EFSA and Hanioka et al., it appears more likely that even human infants will possess sufficient UGT activity to glucuronidate BPA at the levels to which they are exposed.

###

References Cited:

1. Edginton AN, Ritter L.  Predicting plasma concentrations of bisphenol A in young children (<2 years) following typical feeding schedules using a physiologically-based toxicokinetic model.  Environ Health Persp 2008.  doi: 10.1289/ehp.0800073 (available at http://dx.doi.org)
2. Volkel W, Colnot T, Csanady GA, Filser JG, Dekant W.  2002.  Metabolism and kinetics of bisphenol A in humans at low doses following oral administration.  Chem Res Toxicol 15:1281-1298.
3. Hanioka N, Naito T, Narimatsu S.  2008.  Human UDP-glucuronosyltransferase isoforms involved in bisphenol A glucuronidation.  Chemosphere.  doi: 10.1016/j.chemosphere.2008.09.053.
4. European Food Safety Authority.  2008.  Toxicokinetics of Bisphenol A:  Scientific Opinion of the Panel on Food Additives, Flavourings, Processing Aids, and Materials in Contact with Food (AFC).  Question no. EFSA-Q-2008-382.  EFSA Journal 759:1-10.