Csaba Leranth, Klara Szigeti-Buck, Neil J. MacLusky, and Tibor Hajszan

Exposure measurement data from several developed countries indicate that human beings are widely exposed to low levels of the synthetic xenoestrogen, bisphenol A. We reported previously that bisphenol A, even at doses below the reference safe daily limit for human exposure, recommended by the U.S. Environmental Protection Agency, impairs the synaptogenic response to 17β-estradiol in the hippocampus of ovariectomized rats. Recent experiments revealed that bisphenol A also interferes with androgen receptor-mediated transcriptional activities. Thus, to investigate whether bisphenol A impairs synaptogenesis in the medial prefrontal cortex (mPFC) and hippocampus of adult male rats, castrated and sham-operated animals were treated with different combinations of bisphenol A (300 μg/kg), testosterone propionate (1.5 mg/kg), and sesame oil vehicle. The brains were processed for electron microscopic stereology, and the number of asymmetric spine synapses in the mPFC and CA1 hippocampal area was estimated. In both regions analyzed, bisphenol A reduced the number of spine synapses in sham-operated, gonadally intact animals, which was accompanied by a compensatory increase in astroglia process density. In addition, bisphenol A prevented both the prefrontal and hippocampal synaptogenic response to testosterone supplementation in castrated males. These results demonstrate that bisphenol A interferes with the synaptogenic response to testosterone in the mPFC and hippocampus of adult male rats. Because the hippocampal synaptogenic action of androgens seems to be independent of androgen and estrogen receptors in males, the potential mechanisms that underlie these negative effects of bisphenol A remain the subject of further investigation.

SINCE THE 1950s, the synthetic xenoestrogen, bisphenol A, has been used in the manufacture of plastics that have a broad range of uses including dental prostheses and sealants (1), the polycarbonate lining of metal cans used to preserve foods (2), and such items as baby bottles (3) and clear plastic cages used in many research institutions to house laboratory animals (4). Bisphenol A is also used as an additive in many products, with a global production rate of more than 6 billion pounds per year. Whereas exposure measurement data from several developed countries, including the United States, consistently indicate that human beings are widely exposed to low levels of bisphenol A, probably on a continuous basis (5), there is considerable debate whether this exposure represents an environmental problem.

The relatively low affinity of bisphenol A for the nuclear estrogen receptors (ERs) and its weak bioactivity in standard tests of estrogenicity (6) initially led to the conclusion that exposure to bisphenol A has negligible biological effects in humans (7). However, recent findings suggest that bisphenol A may interfere with the development, function, and morphology of the brain. We reported earlier that bisphenol A, even at doses below the reference safe daily limit for human exposure, recommended by the U.S. Environmental Protection Agency (EPA), impairs the synaptogenic response to 17β-estradiol in the hippocampus of ovariectomized rats (8,9). Because remodeling of spine synapses on the dendrites of hippocampal pyramidal cells may contribute to the beneficial effects of estrogens on cognition (10), disruption of estrogen-induced spine synapse formation by bisphenol A may result in cognitive impairments, particularly in ages when estrogen levels are naturally low, such as in postmenopausal women.

Whereas earlier studies focused on the estrogenic properties of bisphenol A, recent experiments revealed that bisphenol A antagonizes androgen receptor (AR)-mediated transcriptional activities (11,12,13,14). Because androgens are just as critical in the cognitive functions (15) and synaptogenesis (16,17,18) in males as estrogens are in females, the potential exists that bisphenol A also interferes with the physiology and morphology of the adult male brain. Thus, the following experiments were performed to investigate whether bisphenol A impairs spine synapse formation in the medial prefrontal cortex (mPFC) and hippocampus of adult male rats, brain areas with crucial influence on cognitive functions.

Materials and Methods

Experimental animals

Male Sprague Dawley rats (280–300 g; Charles River Laboratories, Wilmington, MA) were kept under standard laboratory conditions in a 12-h light,12-h dark cycle, with tap water and regular rat chow available ad libitum. Experiments conformed to international guidelines on the ethical use of animals, and experimental protocols were approved by the Institutional Animal Care and Use Committee of Yale University School of Medicine.

Surgery, treatments, and tissue processing

Castration or sham operation was performed on d 1 under deep anesthesia with a ketamine-xylazine mixture (containing 25 mg/ml ketamine, 1.2 mg/ml xylazine, and 0.03 mg/ml acepromazine dissolved in saline; 3 ml/kg, im). One week later (on d 8), treatments were initiated, consisting of daily sc injections of testosterone propionate (TP), bisphenol A, or the sesame oil vehicle. On d 11, rats were killed under deep ether anesthesia by transcardial perfusion of heparinized saline, followed by a fixative containing 4% paraformaldehyde and 0.1% glutaraldehyde in 0.1 m phosphate buffer (pH 7.35). Brains were removed and postfixed overnight in the same fixative without glutaraldehyde. Tissue blocks containing the prefrontal region and the hippocampus were dissected out, and 100-μm-thick coronal vibratome sections were cut and sorted into several groups. Different groups of sections were then further processed for electron microscopy or glial fibrillary acidic protein (GFAP) immunostaining.

Electron microscopic stereology

The total number of asymmetric spine synapses in layer II/III of mPFC as well as the stratum radiatum of the CA1 hippocampal subfield was calculated as published previously (19,20). Due to the labor-intensive nature of electron microscopic stereology, the analysis was focused on these particular regions because our earlier studies demonstrated strong synaptogenic response to androgens in these areas of the adult male brain (16,18). First, using embedded sections, the volume of the sampling areas was estimated using the Cavalieri Estimator module of the Stereo Investigator system (MicroBrightField Inc., Villiston, VT) mounted on an Axioplan 2 light microscope (Zeiss, New York, NY). Because the mPFC and its neighboring cortical regions show very limited cytoarchitectonic differences in rodents, the precise anatomical borders of mPFC in rats remain the subject of intensive debate. To address this problem, we determined a sampling area with artificial borders that are related to easily identifiable macroanatomical structures. These borders are described in detail elsewhere (19). Thereafter 20 sampling sites for electron microscopic analysis were localized in both the mPFC and CA1 using a systematic-random approach, as published previously (19,20), and approximately four 75-nm-thick consecutive ultrasections were cut from each of these sampling sites. Digitized electron micrographs (Fig. 1​1)) were taken from neighboring ultrasections for the physical disector by a person, who was blind to the treatment of individual animals. The micrographs were taken in a transmission electron microscope (Tecnai 12; FEI Co., Hillsboro, OR) furnished with an HR/HR-B charge-coupled device camera system (Hamamatsu Photonics, Hamamatsu, Japan); and the pictures were coded for blind analysis. This sampling technique produced 20 dissectors for the mPFC and another 20 dissectors for CA1 from each brain. Asymmetric spine synapses were counted according to the rules of the disector technique (21), and the volumetric density of these synapses (synapse per square micrometer) was determined. Thereafter, the volumetric density was multiplied by the volume of the sampling area to arrive at the total number of spine synapses. The number of spine synapses was determined independently by two different investigators, and the results were cross-checked to preclude systematic analytical errors.