Studies in experimental animals

Study

Key findings/claims

FSANZ response

'Bisphenol A affects early bovine embryo development and metabolism that is negated by an oestrogen receptor inhibitor'

Choi et al (2016) science Reports, 6:29318.

• In vitro incubation of bovine embryos with BPA resulted in (i) a decreased percentage (by up to 20%) in embryos considered to be suitable for subsequent transfer into a host, and (ii) an up to 50% increase in glucose consumption rate by the embryos.

• This is an in vitro study where the embryos were cultured in solutions of BPA.
• The metabolic detoxification of BPA that occurs following oral ingestion is therefore bypassed.
• Oral studies in laboratory animals have addressed the potential embryofetal toxicity of BPA, and these studies have previously been taken into account in risk assessments and establishment of the tolerable daily intake (TDI).

'Disruption of adult expression of sexually selected traits by developmental exposure to bisphenol A.'

Jašarevic et al (2011) Proceedings of the National Academy of sciences USA, published online ahead of print, 27 June 2011.

• Male deer mice exposed to BPA through maternal diet exhibited compromised learning abilities and exploratory behaviours compared to control males.
• Female mice spent more time in nose-to-nose contact with control males than with males exposed to BPA. The authors' concluded that female deer mice have a reduced sexual preference for males exposed to BPA.

• The level of BPA in feed was 50 mg per kg of feed (i.e. 50 ppm), however feed consumption and estimated doses of BPA received by the deer mice were not reported.
• Male deer mice exposed to BPA showed no changes in appearance, body weight, sensory development, or adult circulating concentrations of testosterone and corticosterone. A mechanistic explanation for the reported findings is lacking.
• Parameters relevant to mating were not investigated. Reduced time in nose-to-nose contact may not be indicative of reduced sexual preference.
• This is the first study of BPA in deer mice. It is unknown whether the findings in deer mice may be applicable to other rodents let alone humans.

'Similarity of BPA pharmacokinetics in rhesus monkeys and mice: relevance for human exposure'

Taylor et al (2011)Environmental Health Perspectives119(4):422-430

• BPA administered as a single oral dose exhibited similar concentration vs time profiles of free BPA measured in blood serum of mice and monkeys.
• This finding adds weight to the potential relevance to humans of low dose findings reported in some mouse studies.

• Close scrutiny of the data shows important differences in the BPA serum profiles in mice and monkeys. In mice, the serum concentration of free BPA 24 hours after dosing was approximately 12% of the maximum serum concentration observed at 1 hour after dosing. In monkeys, the corresponding value was only 2.5%.
• This difference results in a 3-fold greater systemic exposure to free BPA in mice compared to monkeys and could result in accumulation of free BPA in mice following repeated dosing. This result did not receive any comment in the paper.
• The conclusion that serum profiles of free BPA are similar in monkeys and mice is therefore not considered to be generally valid.
• Other studies have shown that mice and rats are less efficient than humans with regard to BPA detoxification. This diminishes the human relevance of low dose findings reported in some rodent studies.

'Oral exposure to BPA increases dimethylbenzanthracene-induced mammary cancer in rats'

Jenkins et al (2009) Environmental Health Perspectives117:910-915.

• In rats, combination treatment with a carcinogenic chemical and lactational exposure to BPA resulted in increased numbers of mammary tumours per animal and reduced time to first tumour.

• Studies have shown that a minimal fraction of BPA administered to dams is transferred to breast milk. BPA has shown no carcinogenic potential in chronic studies in mice and rats at relatively high dose levels.
• Histopathologic evaluation revealed no changes in the carcinoma score or tumour burden (expressed as percentage of bodyweight).
• With regard to tumour latency, uncertainty in the measurement method (palpation) is likely to be large and was not addressed in the publication.
• The shortcomings in this study and a lack of concordance with other studies indicate that this study is not useful for hazard assessment.

'No effect of route of exposure on plasma BPA throughout 24h after administration in neonatal female mice'

Taylor et al (2008) ReproductiveToxicology25(2):169-176

• In neonatal mice, the route of administration (oral vs subcutaneous injection) gave no significant difference in plasma levels of BPA.
• Studies that use non-oral administration of BPA during the neonatal period should not be dismissed as unsuitable for hazard assessment.

• BPA concentrations were determined only in ether extracts of blood samples and only a small fraction of the administered dose was recovered (
• Well designed human studies show that BPA is rapidly and extensively detoxified following oral exposure.
• BPA studies using non-oral routes of administration, for which detoxification pathways are largely bypassed, should be given little weight in the hazard assessment of BPA.

'Estrogenic chemicals in plastic and oral contraceptives disrupt development of the fetal mouse prostate and urethra'

Timms et al (2005) Proceedings of the National Academy of  sciences USA 102(19):7014-7019

• Pregnant mice receiving oral BPA (10 µg/kg bodyweight/day) produced male offspring with increased number and size of prostate ducts.
• Urethral constriction was also reported and it was stated that this could contribute to urine flow disorders.

• The study used a single dose level and small animal numbers.
• Prostate weights were not reported.
• No effects on the prostate have been observed in more robust studies with multiple dose levels and larger group sizes.
• No disorders which could be attributed to altered urine flow (e.g. adverse bladder or kidney effects, altered clinical chemistry parameters) have been observed in well designed animal studies on BPA.

'Exposure to a low dose of BPA during fetal life or in adulthood alters maternal behavior in mice'

Palanza et al (2002)

Environmental Health Perspectives110: 415-422

• Female mice exposed to BPA (10 µg/kg bodyweight/day) either as foetuses or in adulthood spent less time nursing their offspring and more time out of the nest compared with the control group.

• Statistically significant effects were reported for prenatal exposure or postnatal exposure. However, no effects were observed following combined pre- and postnatal exposure and there was no explanation for this apparent anomaly.
• No adverse effects were reported for the offspring.
• Use of only a single dose level is a deficiency (see above).
• The small reported differences in maternal behaviour are not considered to be adverse effects.

'Exposure to BPA advances puberty'

Howdeshell et al (1999) Nature401: 763-764

• A low oral dose of BPA (2.4 µg/kg bodyweight/day) administered to pregnant mice advanced puberty in the female offspring.

• A non-standard endpoint was used for assessing puberty in mice and is of questionable biological significance.
• Use of only a single dose level prevents dose-response considerations which are integral to toxicological hazard assessment.
• Statistical analysis was not adequately documented.

'A physiologically based approach to the study of BPA and other estrogenic chemicals on the size of reproductive organs, daily sperm production, and behavior'

Vom Saal et al (1998) Toxicology and Industrial Health14: 239-260

• Mice given low oral doses of BPA during pregnancy (2 or 20 µg/kg bodyweight/day) gave offspring with reduced sperm production at the higher dose, higher preputial weights and lower seminal vesicle weights at the lower dose, and lower epididymal weights at both dose levels.

• This study shares weaknesses with the above study (Nagel et al 1997) such as the use of: (i) small animal numbers per group, (ii) an obsolete mouse strain, and (iii) data from only one randomly selected male per litter.
• An additional weakness is the unusual/unexplained findings of low dose only effect on weights.
• The US National Toxicology Program stated that it was not able to confirm any of the statistically significant findings in this paper and concluded that the data were inadequate for hazard assessment.
• No effects on these parameters have been observed in more robust studies employing multiple dose levels and larger group sizes.

'Relative binding affinity-serum modified access (RBA-SMA) assay predicts the relative in vivo bioactivity of the xenoestrogens BPA and octylphenol'

Nagel et al (1997) Environmental Health Perspectives105: 70-76

• Mice given low oral doses of BPA during pregnancy (2 or 20 µg/kg bodyweight/day) gave offspring with larger prostates at both dose levels.

• The study used an 'in-house' mouse strain which was subsequently destroyed thus preventing replication of the findings by other researchers.
• The use of small animal numbers per group raises questions regarding statistical validity.
• Reproductive organs in male mice can vary in size depending on social status (e.g. dominant males usually have larger prostates). The reported increase in average prostate weight (approx 30%) may be an artefact of the study design in which one male per litter was randomly selected for analysis.
• Microscopic analyses of prostates were not conducted.
• No effects on the prostate have been observed in several subsequent studies in mice and rats employing multiple dose levels and larger group sizes.