Download Autometallographic Tracing of Bismuth in Human Brain Autopsies

Journal of Neuropathology and Experimental Neurology
Copyright q 2001 by the American Association of Neuropathologists
Vol. 60, No. 7
July, 2001
pp. 705 710
Autometallographic Tracing of Bismuth in Human Brain Autopsies
MEREDIN STOLTENBERG MD, PHD, JEAN-ANASTASE HOGENHUIS, MD, JEAN-JACQUES HAUW, MD,
AND GORM DANSCHER, DRMEDSC
Abstract. For decades, drugs containing bismuth have been used to treat gastrointestinal disorders. Although a variety of
adverse effects, including neurological syndromes, have been recorded, the biological/toxicological effects of bismuth ions
are far from disclosed. Until recently, only quantitative assessments were possible, but resent research has made histochemical
tracing of bismuth possible. The technique involves silver enhancement of bismuth crystallites by autometallography (AMG).
In the present study, the localization of bismuth was traced by AMG in sections of paraffin-embedded brain tissue obtained
by autopsy from 6 patients suffering from bismuth intoxication in a period ranging from 1975 through 1977. Tissue was
analyzed at light and electron microscopical levels, and the presence of bismuth further confirmed by proton-induced x-ray
emission (PIXE). Clinical data and bismuth concentrations in blood, cerebellum, and thalamus were measured by atomic
absorption spectrophotometry (AAS) and are reported here. Histochemical analyses demonstrate that bismuth accumulated in
neurons and glia cells in the brain regions examined (neocortex, cerebellum, thalamus, hippocampus). Cerebellar blood vessels
stained most intensely. The PIXE and AAS data correlated with the histochemical staining patterns and intensities. At the
ultrastructural level, bismuth was found to accumulate intracellularly in lysosomes and extracellularly in the basement membranes of some vessels.
Key Words:
Autometallography (AMG); Brain; Central Nervous System; Heavy metals.
INTRODUCTION
For more than 100 years, gastrointestinal disorders
have been treated with drugs containing bismuth. Modern
variants include a combination of bismuth salts and antibiotics (De-Nolt, Pyloridt) that are used to treat patients
suffering from Helicobacter pylori-associated peptic ulcers. Bismuth compounds are used in a wide variety of
products, including additives to dental root canal sealers
(1), catheters, in order to make them visible on x-ray
films (2), and as shotgun pellets (3, 4). Despite its relatively low toxicity, bismuth treatments occasionally are
accompanied by adverse side effects. The best-documented case of its neurotoxicity is the outbreak of bismuth encephalopathy among several hundred patients in
France (5, 6). The most widely used drug at that time
was bismuth subnitrate (BIS), which was given in doses
of 5 to 25 g daily for periods of 4 wk to 30 yr (5, 7–9).
The calamity was never fully understood and the involvement of unknown factors has not been excluded (6).
Neurons appear to be selectively sensitive to toxic effects of bismuth. Bismuth has been reported to cause neuronal degeneration in rat hippocampus slices (10, 11), and
the application of bismuth in organotypic cultures of rat
hippocampus resulted in a selective degeneration of CA1
From the Department of Neurobiology (MS, GD), Institute of Anatomy, University of Aarhus, Aarhus, Denmark; Laboratoire de Neuropathologie (JAH, JJH), Hôpital de la Salpêtrière, Paris, France.
Correspondence to: Meredin Stoltenberg, MD, PhD, Department of
Neurobiology, Institute of Anatomy, University of Aarhus, DK-8000
Aarhus C, Denmark.
This study was supported by the Aarhus University Research Foundation, the Danish Medical Association Research Fund, ‘‘Direktør E.
Danielsen og Hustrus Fond,’’ ‘‘Helga og Peters Kornings Fond’’ and
‘‘Direktør Jacob Madsen og Hustrus Fond.’’
pyramidal cells, while CA3 pyramidal cells, dentate granule cells, and subicular neurons appeared to be resistant
(11). Spinal cord neurons of rats and mice exposed to
metallic or compound bismuth accumulate substantial
amounts of bismuth in their motor neurons (4, 12, 13),
and intraperitoneal (IP) injections of bismuth subnitrate
result in the uptake of bismuth in large numbers of neurons and result in dilated ventricles (14).
The histochemical tracing of bismuth in brains of mice
exposed to bismuth subnitrate was first demonstrated by
Ross et al (14). The authors found that the autometallographic (AMG) technique revealed a highly organized
pattern of staining that was not present in control animals.
Detailed protocols for the detection of bismuth in tissues
at light and electron microscopic levels have been published (15). The AMG method is based on silver enhancement of bismuth sulfide/selenide clusters.
This study is the first in which AMG has been used to
detect bismuth in human brain sections. The objective of
the study was to describe the distribution and subcellular
localization of bismuth in the brains of patients suffering
from bismuth intoxication.
MATERIALS AND METHODS
The 6 patients from whom tissue was obtained in the present
study were all suffering from bismuth intoxication. In all cases
bismuth was administered for benign digestive illnesses such as
peptic ulcers and chronic colopathy. All patients were treated
with bismuth subnitrates (pharmaceutical specialties and dosage
remain unknown) for periods of 3 months to more than 20 yr.
Treatment was discontinued for all patients when clinical evidence of bismuth intoxication appeared. The clinical presentation of bismuth intoxication was, for all patients, a myoclonic
encephalopathy. Three patients (E1718, E1838, and E1962)
died rapidly, without improvement in the clinical symptoms,
705
706
STOLTENBERG ET AL
TABLE
Patient Data
Bismuth concentrations
Patient
code
Sex
E1718
E1719
E1723
E1798
E1838
E1962
f
f
f
m
f
f
Age
Duration of
bismuth intake*
Benign digestive
lesions
blood1
mg/1**
blood2
mg/1**
cerebellar
cortex
mg/kg**
68
57
62
77
80
80
3 months
4 years
.20 years
1.5 years
2 years
2 years
chr col
chr col
chr col
chr col
gdu
chr col
1.20
—
—
1.70
2.25
—
0.53
0.64
0.4
0.05
0.21
0.3
7
9.5
20
2.6
10.5
16
thalamus
mg/kg**
5.3
6
25
3.1
9.5
8
* in all cases, bismuth salts are subnitrates in micronized form. Pharmaceutical specialties and dosages remain unknown.
** Normal values (controls) 0.002–0.5 mg/1 or mg/kg.
1
Before admission.
2
After admission.
— Data not available.
Abbreviations: chr col 5 chronic colopathy; gdu 5 gastroduodenal ulcer.
within 2 wk after the discontinuation of bismuth treatment. The
elevated levels of bismuth before admisssion in 2 of the patients
(E1718 and E1838) later dropped significantly. The other patients (E1719, E1723, and E1798) had slight improvements in
their clinical symptoms 1 month after the discontinuation of
bismuth therapy. In patient E1798, bismuth levels also dropped
substantially after admission (Table).
Tissue Preparation for Light and Electron Microscopy
The brains were examined after several months of formalin
fixation. One-cm-thick coronal slices of the brain and horizontal
slices of the brainstem were performed. Selected areas were
dissected, dehydrated, kept in toluene for 6 hours (h), and immersed in paraffin wax for 1 h prior to embedding in paraffin.
In order to perform routine neuropathological examinations, 5mm sections were cut and stained with hematoxylin and eosin,
PAS, and bodian-luxolfast blue. For AMG development, 5-mm
sections were deparaffinated, rehydrated, and rinsed in distilled
water. Sections were dipped in a 0.1% gelatin solution and dried
prior to AMG. The autometallographic technique used in the
present study has previously been described in detail (15, 16).
Sections for electron microscopy were cut at a thickness of 10
mm, then AMG developed and epon re-embedded. Ultrathin
sections were cut and examined in a JEOL 100S electron microscope (15, 16).
Proton-Induced X-Ray Emission (PIXE) Analysis
Twenty-five-mm-thick sections were cut and placed on an
‘‘aerosol quality’’ Nuclepore filter, (a high-purity and approximately 1 mg/cm2-thick polycarbonate membrane). Each sample
was bombarded with both 2 and 3 MeV protons in order to
obtain the best detection limits for a wide range of elements
(15, 17). Because the exact weight of the dry tissue was unknown, the PIXE analysis yielded only relative concentrations
of the elements. Since the analysis was carried out to determine
whether or not the tissue contained bismuth and other AMGdetectable heavy metals, we elected not to measure absolute
levels.
J Neuropathol Exp Neurol, Vol 60, July, 2001
Controls
Controls included the following: 1) sections from brains of
patients (4 female and 2 male, 60–80 yr of age) with no known
history of bismuth treatment history; 2) bismuth-containing sections that were not subjected to AMG; and 3) PIXE analyses
of the preceding 2 categories.
RESULTS
For all subjects, routine neuropathological examination
revealed no significant lesions, with the exception of
small, nonspecific lymphocytic infiltrates observed
around a few small- to medium-sized venules in the white
matter. Neither gliosis nor neuronal loss was observed.
However, after analysis with the AMG method, neurons
as well as glia cells were found to have accumulated bismuth (Fig. 1a). Astrocytes, in particular, exhibited abundant staining (Fig. 1b). In all patients, cerebellum, thalamus, and neocortex contained the most intensely stained
neurons (Fig. 2a–c). In particular, the Purkinje cells were
found to be heavily loaded (Fig. 2d). On the other hand,
hippocampus was almost devoid of staining except in patient E1723 (see Discussion section). The AMG-Bistained neurons were of different sizes, but in general the
larger neurons contained more AMG grains than the
smaller neurons. In the electron microscope, AMG grains
were observed exclusively in the lysosomes of stained
neurons (Fig. 3a). Large inclusion bodies were not observed.
The AMG-stained glia cells were particularly numerous in the white matter and close to the neurons in the
gray matter. For example, the glia cells around the Purkinje cells were heavily loaded. From LM and EM criteria we estimated that the bismuth-containing glia cells
were, for the most part, astroglia; however, this was not
confirmed by immunohistochemistry. At ultrastructural
levels, bismuth observed in glia was found in lysosomes.
BISMUTH IN HUMAN BRAIN
707
Fig. 1. Sections from the temporal cortex, AMG-developed for 60 min and counterstained with toluidine blue, patient E1718.
a: Neurons and glia cells accumulate bismuth (3450). b: Astrocyte exhibiting intense staining, in close contact with a vessel
(31,100).
Some vessels were stained in all brain samples (Fig.
2e), and the 2 patients with the highest concentrations of
bismuth showed extensive dense staining of the vessels
(Fig. 2f). The most pronounced staining of vessels was
found in the cerebellum (Fig. 2f). At the EM level, the
AMG grains were found in the basal lamina of vessels
(Fig. 3b) and in the lysosomes of glia cells and neurons
(Fig. 3a).
There was a good correlation between the concentration of bismuth as measured by atomic absorption spectrophotometry (AAS) and the staining intensity (seen
with AMG). Only the 2 patients (E1962, E1723) with the
highest cerebellar bismuth concentrations showed a massive staining of the vessels (Table; Fig. 2a, f).
Comparable sections from subjects that had not been
exposed to bismuth were blank and neither AAS nor
PIXE analyses of the tissues indicated the presence of
bismuth. Sections from bismuth-exposed brain autopsies
that were not AMG-developed were devoid of staining.
However, PIXE analyses showed the presence of bismuth
and sulfur in the sections.
DISCUSSION
Because the AMG technique is a tool for revealing
endogenous and exogenous metals in tissue sections (Au,
Ag, Hg, Zn, and Bi), it is important to ensure that the
metal/metal-containing molecule clusters that are enhanced by AMG silver can be identified. Thus, it is important that the specificity of the technique be tested.
PIXE analysis of brain sections from the bismuth-intoxicated patients contained bismuth, but no gold, silver, or
mercury. Sections that were not AMG-developed showed
no staining, and brain sections from patients with no
known history of exposure to bismuth, mercury, silver or
gold were devoid of staining. The conclusion is that the
AMG grains shown in the sections developed around bismuth ions. Data proving that the catalytic bismuth compounds were bismuth sulfide/selenide clusters have been
published previously (15).
It has been demonstrated that bismuth can penetrate
the blood barrier (4, 5, 12, 14, 15, 18), and bismuth has
been shown to cause selective degeneration of CA1 neurons in hippocampal brain slices (11). Although some of
the patients analyzed in the present study had ingested
high doses of bismuth for many years, they did not show
significant AMG accumulations in the hippocampal region and we found no signs of degeneration in the CA1
area. Our observations do not support the idea that hippocampal neurons are the main ‘‘target’’ in bismuth intoxications. On the other hand, accumulations of bismuth
in large neurons of the motor cortex and cerebellum
might explain the frequent clinical presentation of the
myoclonic encephalopathy that is seen during bismuth
intoxication. In a recent study, a build-up of bismuth clusters in motor neurons in the spinal cord of mice was
observed, irrespective of the dose and chemical form of
the bismuth. This further supports the hypothesis that bismuth produces myoclonic encephalopathy as a consequence of the accumulation of bismuth in neurons that
affect human motor abilities.
One of the patients (E1723) had ingested bismuth subnitrate for more than 20 years. In this patient there was
massive staining in all parts of the brain, even in areas
otherwise devoid of bismuth traces such as hippocampus,
suggesting that all cell types and areas of the central nervous system can accumulate bismuth in extreme situations.
Retrograde axonal transport of bismuth has recently
been shown to take place in rat motor neurons and sensory ganglion cells after injections in the soleus muscle
J Neuropathol Exp Neurol, Vol 60, July, 2001
708
STOLTENBERG ET AL
Fig. 2. Sections AMG-developed for 60 min and counterstained with toluidine blue. a: Patient E1962. Cerebellum exhibits
high bismuth-AMG staining in neurons and glia cells. 3450. b: Patient E1719. Large, bismuth-containing neurons from the
thalamus (arrows); glia cells are stained as well (arrowheads) (31,100). c: Patient E1718. High bismuth-AMG staining seen in
neurons and glia cells from the temporal cortex (3450). d: Patient E1962. Section from the cerebellum. Purkinje cells, in particular,
are heavily loaded with bismuth (31,100). e: Patient E1838. Section from the motor cortex. Bismuth-AMG grains can be seen
in the vessel wall (arrows), neuron (*), and glia cell near the vessel (arrowhead) (31,100). f: Patient E1723. Massive and very
dense bismuth-AMG staining is seen in cerebellar vessels (3300).
J Neuropathol Exp Neurol, Vol 60, July, 2001
709
BISMUTH IN HUMAN BRAIN
Fig. 3. Electron micrograph of a neuron from the temporal cortex, AMG-developed for 60 min. The relatively poor histochemical quality is a consequence of the tissue being originally paraffin-embedded, patient 1718. a: Bismuth-AMG grains (arrows)
are seen in lysosome-like structures (arrows) near the nucleus of the neuron (nc) (38,500). b: Bismuth-AMG grains (arrows) are
seen in the basal lamina of a vessel (310,000).
(13). Axonal transport of silver and mercury has previously been shown to take place (19–22). This transport
mechanism of toxic metals is important from a toxicological point of view in that it represents a pathway of
entry into the CNS that circumvents the blood-brain barrier. However, bismuth uptake via vessels is undoubtedly
the most important path by which bismuth enters the
CNS (13). Just a few weeks after placing bismuth gunshot pellets intraperitoneally, bismuth can be found in the
brain and spinal cord (4).
The presence of AMG-Bi grains in the lysosomes and
in the basal lamina of some capillaries in the human
brains is in accordance with findings in different species
(4, 12–15). The lysosomal storage of bismuth, silver mercury, and gold is believed to represent the last step in a
detoxification process (23–29). However, alterations in
lysosomal activity might result from this storage. Recently, mercury has been shown to cause the death of
dorsal root ganglia cells in rats (30, 31). This subtle toxic
effect might reflect the influence of mercury clusters on
the lysosomal system. Whether or not bismuth affects the
human nervous system in a similar way remains to be
demonstrated. It is hypothesized that bismuth accumulates primarily in neurons involved in motor activities.
In conclusion, patients suffering from bismuth intoxication will accumulate bismuth sulfide/selenide clusters
in the lysosomes of both glia cells and neurons and in
basal membranes of certain capillaries.
ACKNOWLEDGMENT
We wish to thank Ms. H. Krunderup, Ms. D. Jensen, Ms. L. Munkøe,
Ms. K. Wiedemann, and Mr. A. Meier for their excellent technical assistance.
REFERENCES
1. Geurtsen W, Lehmann F, Spahl W, Leyhausen G. Cytotoxicity of
35 dental resin composite monomers/additives in permanent 3T3
and three human primary fibroblast cultures. J Biomed Mater Res
1998;41:474–80
2. Pariente JL, Bordenave L, Bareille R, Ohayon-Courtes C, Baquey
C, Le Guillou M. In vitro cytocompatibility of radio-opacifiers used
in ureteral endoprosthesis. Biomaterials 1999;20:523–27
3. Kraabel BJ, Miller MW, Getzy DM, Ringelman JK. Effects of embedded tungsten-bismuth-tin shot and steel shot on mallards (anas
platyrhynchos). J Wildl Dis 1996;32:1–8
4. Pamphlett R, Danscher G, Rungby J, Stoltenberg M. Tissue uptake
of bismuth from shotgun pellets. Environ Res 2000;82:258–62
5. Martin-Bouyer G. Intoxications par les sels de bismuth administrés
par voie orale Enquête épidémiologique. Thérapie 1976;31:683–702
6. Martin-Bouyer G, Foulon B, Guerbois H, Barin C. Aspects épidémiologiques des encéphalopathies après administration de bismuth
par voie orale. Thérapie 1980;35:307–13
7. Winship KA. Toxicity of bismuth salts. Adv Drug React Ac Pois
Rev 1983;2:103–21
8. Slikkerveer A, de Wolff FA. Pharmacokinetics and toxicity of bismuth compounds. Toxicology Management Review 1989;4:303–23
9. Supino-Viterbo V, Cicard C, Risvegliato M, Rancurel G, Buge A.
Toxic encephalopathy due to ingestion of bismuth salts: Clinical
and EEG studies of 45 patients. J Neurol Neurosurg Psychiatry
1977;40:748–52
10. Bruinink A, Reiser P, Müller M, Gähwiler BH, Zbinen G, Bès A.
Neurotoxic effects of bismuth in vitro. Toxicology in Vitro 1992;
6:285–93
11. Müller M, Rietschin L, Grogg F, Streit P, Gähwiler BH. Selective
degeneration of CA1 pyramidal cells by chronic application of bismuth. Hippocampus 1994;4:204–9
12. Pamphlett R, Stoltenberg M, Rungby J, Danscher G. Uptake of
bismuth in motor neurons of mice after single oral doses of bismuth
compounds. Neurotoxicol Teratol 2000;22:559–63
13. Stoltenberg M, Schiønning JD, Danscher G. Retrograde axonal
transport of bismuth.An ultrastructural autometallographic study.
Acta Neuropathol (in press)
J Neuropathol Exp Neurol, Vol 60, July, 2001
710
STOLTENBERG ET AL
14. Ross JF, Switzer RC, Poston MR, Lawhorn GT. 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. Brain Res 1996;725:137–54
15. Danscher G, Stoltenberg M, Kemp K, Pamphlett R. Bismuth autometallography. Protocol, specificity, and differentiation. J Histochem Cytochem 2000;48:1503–10
16. Danscher G. Histochemical demonstration of heavy metals. A revised version of the sulfide silver method suitable for both light
and electron microscopy. Histochemistry 1981;71:1–16
17. Kemp K, Danscher G. Multi-element analysis of the rat hippocampus by proton-induced X-ray emission spectroscopy (phosphorous,
sulphur, chlorine, potassium, calcium, iron, zinc, copper, lead, bromine and rubidium). Histochemistry 1979;59:167–76
18. Escourolle R, Bourbon R, Galli A, et al. Étude neuropathologique
et toxicologique de douze cas d’encéphalopathie bismuthique (EB).
Rev Neurol (Paris) 1977;133:153–63
19. Danscher G. Exogenous selenium in the brain. A histochemical
technique for light and electron microscopical localization of catalytic selenium bonds. Histochemistry 1982;76:281–93
20. Arvidson B. Retrograde axonal transport of metals. J Trace Elem
Exp Med 1989;2:343–47
21. Arvidson B. Inorganic mercury is transported from muscular nerve
terminals to spinal and brainstem motor neurons. Muscle Nerve
1992;15:1089–94
22. Schiønning JD. Retrograde axonal transport of mercury in rat sciatic nerve. Toxicol Appl Pharmacol 1993;121:43–49
23. Fowler BA, Brown HW, Lucier GW, Bread ME. Mercury uptake
by renal lysosomes of rats ingesting methyl mercury hydroxide.
Arch Pathol 1974;98:297–301
J Neuropathol Exp Neurol, Vol 60, July, 2001
24. Norseth T, Brendeford M. Intracellular distribution of inorganic and
organic mercury in rat liver after exposure to methylmercury salts.
Biochem Pharmacol 1971;20:1101–7
25. Danscher G. Localization of gold in biological tissue. A photochemical method for light and electronmicroscopy. Histochemistry
1981;71:81–88
26. Danscher G. Light and electron microscopic localisation of silver
in biological tissue. Histochemistry 1981;71:177–86
27. Danscher G, Møller-Madsen B. Silver amplification of mercury sulfide and selenide: A histochemical method for light and electron
microscopic localization of mercury in tissue. J Histochem Cytochem 1985;33:219–28
28. Møller-Madsen B, Danscher G. Localization of mercury in CNS of
the rat. I. Mercuric chloride (HgCl2) per os. Environ Res 1986;41:
29–43
29. Schiønning J, Møller-Madsen B. Autometallographic mapping of
mercury deposits in the spinal cord of rats treated with inorganic
mercury. Acta Neuropathol 1991;81:434–42
30. Schiønning JD, Larsen JO, Eide R. A stereological study of dorsal
root ganglion cells and nerve root fibers from rats exposed to mercury vapor. Acta Neuropathol 1998;96:185–90
31. Schiønning JD, Larsen JO, Tandrup T, Brændgaard H. Selective
degeneration of dorsal root ganglia and dorsal nerve roots in methyl
mercury-intoxicated rats: a stereological study. Acta Neuropathol
1998;96:191–201
Received February 5, 2001
Revision received April 6, 2001
Accepted April 9, 2001