Aluminium chloride

Neurotoxicity

Administrative data

Endpoint:

neurotoxicity

Remarks:

other: Single exposure with 14 days observation before termination of study

Type of information:

experimental study

Adequacy of study:

disregarded due to major methodological deficiencies

Reliability:

3 (not reliable)

Rationale for reliability incl. deficiencies:

other: basic information given, deficiencies in experimental design (animal numbers)

Data source

Reference

Reference Type:

publication

Title:

Aluminium-induced model of motor neuron degeneration: subperineurial injection of aluminium in rabbits.

Author:

Kihira, T., Yoshida, S., Komoto, J., Wakayama, I., Yase, Y.

Year:

1995

Bibliographic source:

NeuroToxicol. 16 (3), 413-424

Materials and methods

Principles of method if other than guideline:

The present study was designed to morphologically evaluate the neurotoxic effects of Al following its injection into the peripheral nervous system on spinal motor neurons and on the permeability of the bloodnerve-barrier.

GLP compliance:

not specified

Test material

Test material information

Constituent 1

Reference substance name:

Reference substance 002

Details on test material:

no data on purity of the compound

Test animals

Species:

rabbit

Strain:

other: Japanese White

Sex:

not specified

Administration / exposure

Route of administration:

other: subperineurial injection

Details on exposure:

Thirty-two rabbits were divided into five groups and inoculated with the following agents: Group 1 (n = 4); non-treated; Group 2 (n = 9), 0 .9% NaCl (50 ul to 500 ul, pH 4 .5); Group 3 (n = 4), maltol (100mM, 50 ul) ; Group 4 (n = 11), AlCl3 solution (83 mM, 50-500 ul, pH 4.0); and Group 5 (n = 4), Al-maltol (100 mM, 50 u1, pH 4.5). Each agent was injected into the subperineurial space of the right sciatic nerve 1 cm proximal to the angle of the knee with a microsyringe (27- or 33-gauge) over a 10-minute period to prevent direct injury to the nerve fibers. After two weeks, the in-life phase was terminated.

Analytical verification of doses or concentrations:

not specified

Duration of treatment / exposure:

2 weeks

Frequency of treatment:

Single treatment

No. of animals per sex per dose:

total: thirty-two rabbits
Group 4 animals (treatment with 83 mM AlCl3 solution): 11 in total (6 receiving 50 ul, 2 receiving 100 ul, 1 receiving 200 ul, 2 receiving 500 ul)

Control animals:

yes, concurrent no treatment

Details on study design:

Two weeks after treatment, the rabbits were anesthetized with sodium pentobarbital (Nembutal), and perfused through the ascending aorta with saline, 2.5-5% phosphate buffered glutaraldehyde (pH 7.4), or 10% phosphate buffered formalin. The lumbar spinal cord (at the level of the fourth and fifth lumbar spinal cord), sciatic nerves, and ganglia were removed, embedded in paraffin, and 6-um-thick paraffin sections prepared. The sections were stained with hematoxylin-eosin, Bodian silver, and Kluver Barrera stains and with a modified Bielshowski silver stain. Parts of the spinal cord and sciatic nerve of rabbits from Groups 1, 2-1 (50 ul, saline injected group), and 4-1 (50 ul, AlCl3 injected group), which were removed before fixation, were used for metal analysis by particle-induced X-ray emission spectrometry.
For examination of the permeability of the blood brain barrier and blood-nerve barrier, 0.5 ml of saline containing horseradish peroxidase was injected intravenously via ear vein two weeks after subperineurial injection in one rabbit each from Groups 1, 2, and 3 and in two rabbits from Group 5. Rabbits were anesthetized and perfused with 2.5% glutaraldehyde 2.5 hours after HRP injection. Sections 50 um thick from the 5th lumbar spinal cord, sciatic nerve, and ganglion were cut with a vibratome and processed to demonstrate peroxidase activity. The vibratome sections were incubated with 10% 2,2'-diaminobenzidene tetrahydrochloride sucrose and 0.01% H202) for 20 minutes, and mounted in glycerin for light microscopic examination, or postfixed with 1% osmic tetroxide and embedded for electron microscopic examination.
For morphometric evaluation, 25 serial sections, 6-um thick, were obtained from the 5th lumbar spinal cord, and every third section was stained by HE, KB, and Bodian silver methods, respectively. These sections were divided into the conventional anterior and posterior areas with the vertical line passing through the central canal to the line drawn between the outer edge of the posterior median septum and anterior median fissure, and the number of spheroids / globules in Bodian stained tissue sections was counted. The numbers of central/ peripheral chromatolytic neurons, and of degenerated neurons in the anterior area and posterior area of 25 sections was counted and recorded using nucleoli as a marker.
The avidin-biotin immunohistochemical technique for phosphorylated and non-phosphorylated neurofilament, tau, microtubule-associated protein 2 and ubiquitin was used on spinal cord sections. The following mouse monoclonal antibodies were used in this study: phosphorylated and non-phosphorylated high- and intermediate-weight neurofilament proteins, microtubule-associated protein tau, MAP2 and PHF/ubiquitin.
For electron microscopic examination, tissue specimens from the sciatic nerve and the 5th lumbar spinal cord were postfixed with 1% osmic tetroxide, dehydrated and embedded in Epon. Semi-thin sections were stained with toluidine blue, and ultrathin sections were stained with uranyl acetate and lead citrate.

Results and discussion

Results of examinations

Details on results:

Argyrophilic round structures (spheroids/globules), which were faintly stained with HE, were observed using the modified Bielschowski silver stain in the neurophil of the spinal anterior horn in Groups 3, 4 and 5, and, to a lesser degree, in Groups 1 or 2. Immunohistochemically, the spheroids / globules were positive for SMI 31, the monoclonal antibody recognizing phosphorylated neurofilaments, but were negative for monoclonal antibodies to non-phosphorylated neurofilaments, tau, MAP-2, or ubiquitin. Electron microscopy disclosed some swollen axons that were filled with interwoven, small bundles of 10-mm neurofilaments and membranous structures. Most of these structures were distributed in the anterior horn of the spinal cord, but a small number in Groups 3 and 5 were located in the posterior horn. The number ot spheroids/globules in Group 5 was significantly increased compared with that in Group 0 or 3. The neurons in the lumbar anterior horn in Groups 4 and 5 and, to a lesser degree, in Group 3, showed central and peripheral chromatolytic change. The numbers of central and peripheral chromatolytic neurons in Groups 3, 4-1 (Al chloride 50 ul injected group) and 5 were significantly increased compared with those in Groups 1(uninjected control) or 2-1 (50 ul saline control). The number of degenerative neurons in the anterior area of the 5th lumbar spinal cord was significantly increased in Group 5 compared with that in any of the other groups. The frequency of spheroids/globules and that of neuronal degeneration showed a positive correlation with the dose of Al in Group 4 , but not with the dose of saline in Group 2.
Electron microscopically, the neuronal soma and dendrites in the anterior horn at the 5th lumbar spinal cord in the Al-treated rabbits showed marked edematous change. The Al-treated rabbits also showed more marked changes than those in the saline/maltol-treated rabbits or controls in terms of the fragmentation of granular endoplasmic reticulum, slight increase of neurofilaments, accumulation of free ribosomes, inward folding of the nuclear membrane, and deposition of lipid droplet-like structures in anterior horn neurons. Lipid droplet-like structures, sometimes surrounded by free ribosomes, showed variable electron density patterns including homogenous, double ringlike, or irregularly lucent areas framed with residual fibrous structures.
The sciatic nerve at the injected site showed mild to moderate edematous change in the subepineurial space, but the portion 2 cm proximal to the injected site showed mild edematous change in the subepineurial space. There were no differences in the morphological changes, among the AlCl3 / maltol, maltol, and saline injected rabbit groups, although these changes were more marked in the groups injected with 500 ul and 200 ul than in those injected with 100 ul and 50 ul. However, the axons and myelin sheath were well preserved in both the injected and proximal portion in every group. HRP reactive product was seen in the axons and cytoplasm of the Schwann cells in the proximal part of the right sciatic nerve, as well as epithelial cells and pericytes of the spinal vessels in Group 5 rabbits, but not in Groups 1, 2, or 3 rabbits. The axons that contained HRP reactive product showed disarrangement of neurofilaments.

Effect levels

Basis for effect level:

other: see 'Remark'

Remarks on result:

not measured/tested

Remarks:

Effect level not specified

Applicant's summary and conclusion

Conclusions:

The authors concluded, that Al, subperineurially injected, was absorbed into the spinal cord and induced degeneration of spinal motor neurons in these rabbits. They further concluded that the retrograde transport of A1 into spinal motor neurons via the peripheral nervous system may exacerbate neuronal degeneration in ALS.
The results of the study are not considered reliable for the following reasons:
- Only very few animals were used.
- Test substance concentrations were not analytically verified.
- The different treatment groups are not comparable (i.e. only one Al-maltol dose level was tested whereas 4 AlCl3 doses were applied). Furthermore, the substances were applied in different volumes.
- Comprehensive histopathological data are not provided.

Therefore, the study is disregarded and not used for assessment.

Executive summary:

The present study was designed to morphologically evaluate the neurotoxic effects of Al following its injection into the peripheral nervous system on spinal motor neurons and on the permeability of the bloodnerve-barrier.

A total of 32 rabbits was divided into five groups and inoculated with saline (Group 2), maltol (Group 3), AlCl3 (Group 4), Al-maltol (Group 5), or nothing (Group 1). Each agent was injected into the subepineurial space of the right sciatic nerve. For examination of the permeability of the blood brain barrier and blood-nerve barrier, 0.5 ml of saline containing horseradish peroxidase was injected intravenously via ear vein two weeks after subperineurial injection in one rabbit each from Groups 1, 2, and 3 and in two rabbits from Group 5.

After two weeks, the rabbits were killed, the lumbar spinal cord, sciatic nerves and ganglia were fixed, removed and embedded in paraffin.

For morphometric evaluation, 25 serial sections, 6-um thick, were obtained from the 5th lumbar spinal cord, and every third section was stained by HE, KB, and Bodian silver methods, respectively. These sections were divided into the conventional anterior and posterior areas with the vertical line passing through the central canal to the line drawn between the outer edge of the posterior median septum and anterior median fissure, and the number of spheroids / globules in Bodian stained tissue sections was counted. The numbers of central/ peripheral chromatolytic neurons, and of degenerated neurons in the anterior area and posterior area of 25 sections was counted and recorded using nucleoli as a marker.

The avidin-biotin immunohistochemical technique for phosphorylated and non-phosphorylated neurofilament, tau, microtubule-associated protein 2 and ubiquitin was used on spinal cord sections. The following mouse monoclonal antibodies were used in this study: phosphorylated and non-phosphorylated high- and intermediate-weight neurofilament proteins, microtubule-associated protein tau, MAP2 and PHF/ubiquitin.

For electron microscopic examination, tissue specimens from the sciatic nerve and the 5th lumbar spinal cord were postfixed with 1% osmic tetroxide, dehydrated and embedded in Epon. Semi-thin sections were stained with toluidine blue, and ultrathin sections were stained with uranyl acetate and lead citrate.

Argyrophilic round structures (spheroids/globules), were observed in the neurophil of the spinal anterior horn in Groups 3, 4 and 5, and, to a lesser degree, in Groups 1 or 2. However, these changes only reached statistical significance in Group 5, as compared to Groups 1 and 3. Neurons in the lumbar anterior horn in Groups 4 and 5 and, to a lesser degree, in Group 3, showed central and peripheral chromatolytic change. However, in the case of Group 4, these changes were only statistically significant in the low-dose group (50 ul AlCl3 solution) as compared to Group 1 and 2 (50 ul saline), and a dose-response relationship was not observed.

The sciatic nerve at the injected site showed mild to moderate edematous change in the subepineurial space, but the portion 2 cm proximal to the injected site showed mild edematous change in the subepineurial space. There were no differences in the morphological changes, among the AlCl3 / maltol, maltol, and saline injected rabbit groups, although these changes were more marked in the groups injected with 500 ul and 200 ul than in those injected with 100 ul and 50 ul. However, the axons and myelin sheath were well preserved in both the injected and proximal portion in every group. HRP reactive product was seen in the axons and cytoplasm of the Schwann cells in the proximal part of the right sciatic nerve, as well as epithelial cells and pericytes of the spinal vessels in Group 5 rabbits, but not in Groups 1, 2, or 3 rabbits. The axons that contained HRP reactive product showed disarrangement of neurofilaments.

The present study has a number of deficiencies:

- Only very few animals were used.

- Test substance concentrations were not analytically verified.

- The different treatment groups are not comparable (i.e. only one Al-maltol dose level was tested whereas 4 AlCl3 doses were applied). Furthermore, the substances were applied in different volumes.

- Comprehensive histopathological data are not provided.

Therefore, the study is disregarded and not used for assessment.