Distribution of bismuth in the brain after intraperitoneal dosing of bismuth subnitrate in mice: implications for routes of entry of xenobiotic metals into the brain.

Bismuth (Bi) can produce neurotoxic effects in both humans and animals under certain dosing conditions, but little else is known about the effects of Bi in the brain. In the present study we determined the distribution of Bi in the brains of adult female Swiss-Webster mice 4, 7, 14, 21 and 28 days after a single 2500 mg/kg i.p. injection of Bi subnitrate (BSN), which establishes a depot of absorbable Bi and produces morphological signs of neurotoxicity. Sections of brains were processed by autometallographic (AMG) procedures that produced silver grains at the site of Bi localization (AMGBi). Ventricular dilation was observed in all BSN-dosed mice. Among treated mice there were marked interanimal differences in the absolute amount of AMGBi, but consistent regional and cellular patterns of AMGBi were observed. AMGBi was observed in many cell types in brain regions adjacent to fenestrated blood vessels of the circumventricular organs (CVOs) and olfactory epithelium. Prominent intrasomal AMGBi was observed in nuclei containing large cell bodies, including cranial motor neurons innervating somatic muscle, lateral vestibular and red nucleus and pontine/medullary reticular nuclei. In the hypothalamus, the supraoptic and paraventricular nuclei demonstrated the densest AMGBi. In the cerebellum, Purkinje and granule cell layers with the densest AMGBi were in folia adjacent to the fourth ventricle. In the hippocampus, AMGBi was densest in the fasciola cinerum, polymorph cells of the dentate gyrus, and pyramidal cell layer of the CA3 regions. Neuropil of subcortical auditory nuclei (cochlear nucleus, trapezoid body, lateral lemniscus and nucleus of lateral lemniscus, medial geniculate nucleus and inferior colliculus) had a high density of AMGBi. Among nonneuronal cells, ependyma and meninges lining the ventricular and subarachnoid spaces were labeled extensively. Glial labeling was prominent adjacent to CVOs, in subependymal regions, and in fiber tracts. Presumptive perivascular cells lining large blood vessels had extremely dense AMGBi as early as 4 days after dosing. Smaller blood vessels had moderate AMGBi. However, in regions (e.g. cerebral cortex, striatum) known to have low brain Bi levels after i.p. dosing, vascular deposits accounted for most of the AMGBi. Several animals had foci of AMGBi which suggested that vascular or perivascular aberrations may have contributed to the unusually dense accumulations. The results of the present studies indicate that Bi accumulates predictably in certain regions and cell types. The pattern of regions and cells with the highest AMGBi accumulations is very similar to pattern reported for other xenobiotic metals (i.e. mercury, silver, gold), and supports the hypothesis that these metals may share some mechanisms for entry, distribution and storage in the brain.

Highest brain bismuth levels and neuropathology are adjacent to fenestrated blood vessels in mouse brain after intraperitoneal dosing of bismuth subnitrate.

A small fraction of humans ingesting bismuth (Bi)-containing medications develops neurotoxicity in which neuropsychiatric signs precede motor dysfunction. Large ip doses of Bi subnitrate (BSN) produce similar signs in mice, but little is known about the pathogenesis of neurotoxicity in either species. Adult female Swiss-Webster mice received a neurotoxic dose (2500 mg/kg ip) of BSN. Bi distribution and neuropathology were determined as follows: (1) Regions of central and peripheral nervous system were assayed for Bi by atomic absorption spectrometry (AAS) 28 days after dosing, (2) regional brain Bi distribution was demonstrated in histologic sections by autometallography 28 days after dosing, and (3) blood/brain barrier status and neuropathologic effects were evaluated by light and electron microscopic techniques 1, 3, and 7 days and 2, 3, 4, and 5 weeks after dosing. By AAS, Bi levels were highest in olfactory bulb (approximately 7 ppm), hypothalamus (approximately 7 ppm), septum (approximately 3 ppm), and brain stem (approximately 3 ppm). Striatum and cerebral cortex had the least Bi (approximately 1 ppm). Regional distribution by autometallography showed that high Bi levels were associated with diffusion of Bi from fenestrated blood vessels of circumventricular organs and olfactory epithelium. All treated mice had hydrocephalus, but no other pathology was demonstrable by light microscopy. By electron microscopy, dramatic expansion of the extracellular space between choroid plexus epithelial cells was observed. Dendrites in the neuropil of the hypothalamus and septum exhibited vacuoles and membranous debris. Based on the Bi distribution and lesions, we propose that diffusion of Bi from fenestrated blood vessels contributes to pathogenesis of neurotoxicity in mice. This proposal is consistent with the clinical features of Bi-related neurotoxicity in humans.

Differential complementary localization of metabolic enzymes for quinolinic acid in olfactory bulb astrocytes.

The cellular localizations of the synthetic [3-hydroxyanthranilic acid oxygenase (3HAO)] and degradative [quinolinic acid phosphoribosyltransferase (QPRT)] enzymes of the endogenous excitotoxin quinolinic acid were studied in the adult rat main olfactory bulb by immunohistochemical techniques. 3HAO and QPRT were expressed only in astrocytes. The two enzymes were differentially expressed by astrocytes in a complementary pattern: 3HAO staining was strongest at the glomerular-external plexiform layer junction; QPRT staining was strongest at the glomerular-olfactory nerve layer junction. The complementary distributions of these metabolic enzymes suggests that there could be a gradient of quinolinic acid across the glomerular layer of the main olfactory bulb. Such a gradient could function to restrict the ingrowth of new olfactory axons to the glomeruli and/or to stabilize the formation of new synapses.