- Email: [email protected]
Age-related changes in the neuropil in the rat inferior olive nucleus: A quantitative electron microscopic study
BrainResearchBulletin, Vol. 8, pp. 381-388, 1982. Printedin the U.S.A.
Age-Related Changes in the Neuropil in the Rat Inferior Olive Nucleus: A Quantitative Electron Microscopic Study KOICHI
IKARI,
MICHIHIKO
HAYASHI
AND RYO OGATA
Department of Neuropsychiatry, Faculty of Medicine, Kyushu University, I-I, Maidashi I-chome, Higashi-ku, Fukuoka 812, Japan Received
I July
198 I
IKARI, K., hf. HAYASHI AND R. OGATA. Age-related changes in the neurapil in the rat inferior olive nucleus: A quantitative electron microscopic study. BRAIN RES. BULL 8(4) 381-388, 1982.-Age-related ultrastructural changes in the neuropil in the rat inferior olive nucleus were examined at 3, 6. 12, 18, 24 and 30 months old. The profiles of axon terminals, dendrites and nstroglial processes from random samplings within the neuropif were traced. Subsequently, the percentages of these pro&d areas in relation to the area of neuropil (relative volume fraction) were examined using the image analyzer system. The relative volume fractions of both axon terminals and dendrites in relation to the neuropil were found to have decreased in the aged rats, while the relative volume fraction of astroglial processes had progressively increased with aging. Inferior olive nucleus
Aging
Quantitative electron microscopic study
THE inferior olive in the human brain, one of the li~p~~ic nuclei, contains lipofuscin granules (aged pigment) in the cell soma beginning in the fetal stage [3]. The intracellular accumulation of lipofuscin granules were reported to be correlated directly with aging in this nucleus [21]. Mann and Yates [20) proposed that cell death may be a result of lipofuscin pigment accum~ation. The inferior olive nucleus is also reported capable of mantling its rno~holo~~~ integrity into advanced age, without significant change in nerve cell number [23,24]. The physiological aging process in inferior olive must be, therefore, examined from another point of view. Zeman [35] reported that a stunting of the dendritic apparatus occurs when the rate of pigment production is higher than the rate of regenerative molecular synthesis. Several quantitative morphometrical studies have been carried out on the changes in cell processes in the neuropil with aging [6, 7, 9, 10, 19, 29, 30, 331. These studies reported atrophic or dystrophic changes in cell processes, or proliferation of astrogiiai cell processes. As to the inferior olive, quantitative uItr~t~ctur~ studies on the neuropil of aged mammals are very few. The present study was undertaken to assess the age-related ultrastructural changes in the neuropil in the rat inferior olive nucleus. METHOD
The animals used were male Wistar King rats. They were reared in closed colony with water and a standard laboratory rat chow. Four animals from each of the following ages were prepared: 3,6, 12,18,24 and 30 months old. Animals exhibiting any grossly detectable pathology at the time of perfusion
Copyright
Q 1982 ANKHO
Intemation~
Neuropil
were not used. The animals were anesthetized with pentobarbital sodium (35 mg/kg IP) and perfused through the heart with buffered saline followed by 4% paraformaldehyde in Millonig buffer at pH 7.3. Blocks of the brain stem were cut from the level of C, to about 0.3 cm rostral to the pontine-medulhuy junction, sliced and postfixed in 1% osmium tetraoxide, dehyd~ted and embedded in epon 812. Ultrathin sections were cut and stained with many1 acetate and lead citrate. The inferior olive nucleus was identified in 1 Frn thick toluidine blue stained sections (Fig. 1) from which areas were selected for ultrastructural examinations. An electron microscope (JEOL-100-B) was used to examine and photograph areas within the neuropil of the inferior olive at a direct rn~~~tion of 10,000~. Random samphngs within the neuropil were performed on two to three different tissue blocks from each rat producing 40 electron micrographs per subject animal, totalhng 960 electron micrographs. Photographic enlargement resulted in a final magnification of 30,000x on the prints. The profiles of axon terminals, dendrites and astrogliai processes in neuropil of each rni~r~~phs were traced. The percentages of the profiled areas in relation to the area of neuropil (named relative volume fraction) were then calculated using the image analyzer system. The “neuropil” in this context means the remainder of the inferior olive nucleus after exclusion of cell bodies, myelinated fibers and blood vessels. The criteria for identification of axon terminals were the presence of synaptic vesicles and/or of synaptic membrane differentiation. The criteria for identification of dendrites were as follows: any round or cylindrical profiles of cell processes which showed postsynaptic membrane differentiation on the surface and/or
Inc .-0361-9230/82/~381-~$03.~/0
382
IKARl,
HAYASNl
AND OGATA
any round or cylindrical profiles of greater than 0.3 pm in diameter which contained evenly spaced microtubl~s~ or any round or cylindrical profiles of smaller than 0.3 pm in diameter which contained no microtubles or glio~l~ent (the smallest dendrite and spine) 1181 and showed rather clear cytoplasmic matrix. The criteria for identification of astroglial cell processes were as follows: round, cylindri~ai or irregumrly shaped profibs which filled up the space between neuronal cell processes. These showed clear cytoplasmic matrix and contained contents, glycogen particles, and were poor in other organelIes [2, 12, IS]. The profIles which could not be ident~ed by the above mentioned criteria remained as unclassified. All of the jdent~able profiles except for unmyel~ted axons were traced and the percentage of the area of each profile for the area of neuropil (named relative volume fraction) was measured (Figs. 2,3,4). Figure 2 is a line drawing of the profiles of neuronal and astroglial processes in the ~e~ropil ~hoto~aphed in Fig. 4. Additional electron microscopic observations were made on the lipofuscin granules in the cell soma for one or two animals at 5 weeks, 3, 6, 18 and 30 months of age. FIG. I. inferior olive nucleus of a 3 month old rat. Epon 1 t.lm, toluidine blue stained. P: pyramis meduliae oblongatae. medz n. olivaris accesorius medialis. inf: n. olivaris inferior. dors: n. olivaris accessorius dorsalis. 50x
r
RESULTS
TabIe 1 shows the relative volume fractions of cell processes within the neuropil at various ages. Mean values (percentage) for each animal were derived from analyses of 40 electron micrographs. Mean values for groups of each age were calcuiated from ~dividual mean values and treated statistica~y with the ~de~ndent f-test for multiple comparison test (Dunnet’s test). The volume fraction of axon termi-
(B)
FIG.2.An example of profiles tracing the (A) presynaptic terminals and dendrites, (B) astroglial processes ofthe same rn~c~~~h
as in Fig. 4.
NEUROPIL
IN THE RAT INFERIOR
383
OLIVE TABLE
AGE-RELATED
CHANGES
I
IN THE RELATIVE VOLUME FRACTION INFERIOR OLIVE NUCLEUS
24
30
3
6
Mean _CS.D. of Terminal per Rat
RI R2 R3 R4
28rt 30t 29-c 28%
9 9 9 8
282 282 292 282
9 7 9 8
28% 272 25+ 292
8 8 7 9
28 27 28 28
292
9
28-+
8
272
8
28 rt 9
27 + 7*
24 r 7*
28 25 26 28
28 28 25 29
25 26 24 29
Mean rt SD. of Dendrite per Rat
RI R2 R3 R4
Mean rt S.D. per Group Mean rt SD. of Glial Process per Rat Mean t S.D. per Group Unclassified and other axon profiles except for axon
RI R2 R3 R4
32rt 8 34Icr: 11 34 * 10 32-c 8
33 -c 31+ 31-c 312
10 9 6 8
32~
8
33%
6rt 6r: 6-t 6r
3 3 4 3
6rt
3
34
18
IN THE
Age (month)
Mean r S.D. per Group
12
(410)OF NEUROPIL
29k 32 f 34 -c 282
8 10 10 7
9
31+-
9
7F 7* 6+ 7k
3 4 4 4
7k
4
32
r 9 -t- 8 r. 8 f 9
r * it rt
8 7 8 9
23 26 26 22
r f k f
i + + +
6 7 8 7
8 9 7 7
24 23 22 25
+ 7 _’ 6 -e 6 2 8
+ 2 2 +
8 9 7 8
27 +- 8*
28 + 8*
26 r 8*
10-i- 5 9+ 4 11-t 5 9t 4
13 rt 5 954 13 + 5 12 t 5
11 14 10 13
15 + 4 17 z!z5 17 t 4 16 -t 4
lo*
12 c 5*
12 r 5*
16 t 4*
34
36
34
5 32
25 1?16 k 4 k 4
terminals n=160. *p
nals tends to decrease over the lifespan under study. Compared with 3 month-old rats, the volume fraction of axon terminals at the ages of 24 and 30 months significantly decreased (p
amount of lipofnscin granules are observed. The degree of accum~ation of the lipofuscin granules seemed to increase with aging. Figure 6 shows a neuronal cell body from the inferior olive nucleus at the age of 24 months in which large mass of lipofuscin granules are observed. As far as we observed, there was no tendency to increase dark cells or other degenerative nerve cells in the inferior olive nucleus of 30 month old rats. DISCUSSION
In the present study, we observed relative decrease in the areas of axon terminals and dendrites, and relative increase in the area of astroglial processes in the inferior olive nucleus of the senescent rat. These findings may represent the common changes in the aging process of neuronal tissue. It seems reasonable to suggest that the decrease in the neuronal cell processes and the increase in the astroglial processes in the inferior olive nucleus with advancing age are the glio-neuronal or neuro-glial age-related changes which indicate the gradual reduction in the cellular efficiency. Concerning the axon terminals, a few age-related changes have been reported. They were the decrease in synaptic density [8, 9, 111, decrease in the terminal area [8,16] and synaptic vesicles depletion [I 1,161. Fujisawa and Shiraki [8] reported the slow progressive decrease in the synaptic density and the relative volume fraction of axon terminals in the gracile nucleus of aged rats. Geinisman and Bondareff [9]
IKARI, HAYASHI
AND OCiATA
FIG. 3. An example of neuropil of the inferior olive nudeus, 3 month old rat, with presynaptic terminals fT), dendrites (D) and as&o&al processes (G). Small amount of astroglial processes are observed. M: myelin sheath. 30,ooox
NEUROPIL
IN THE
RAT INFERIOR
OLIVE
FIG. 4. An example ofneuropil of the inferior olive nucieus, 30 month ofd rat, with presynaptic terminals (T), dendrites (D) and astroglial processes (G). Note the increase of astroglial profiles which shows clear cytoplasmic matrix containing glycogen particles (gly) and glial filaments (fil). Arrows indicate gap junctions between adjacent astrocytic processes. 30$00x
386
IKARI,
HAYASHI
AND OGATA
FIG. 5. An electron micrograph of a neuronal cell body of a rat inferior olive nucleus, 5 weeks old. Small amount of lipofuscin granules are observed. N: nucleus, L: lipofuscin granules.
reported a loss of synapses in the molecular layer of the dentate gyrus in aged rats. Hassan and Glees [ll] in their study of age changes in the hippocampus noted that while axo-somatic contacts were a prominent feature of younger rats, these decreased in aged rats along with characteristic membrane thickening and accumulution of synaptic vesicles. Landfield et al. [16] also reported reduced numbers of synaptic vesicles in the terminals and a trend toward reduced terminal size of aged rat hippocampus. Neurochemical studies reported that changes in the acetylcholinesterase activity, norepinephline, serotonine and GABA contents in the brain have been associated with increasing age [22,28]. These findings may suggest reduced functioning of synapse with aging. One possible mechanism for the axonal atrophy and reduction of synaptic activity which accompanies advancing age is a “dying-back” degeneration (or retrograde axonal degeneration) of neurons. This could be explained by the perikaryal insufficiency or axonal transport system damage. Namely, there is a symmetrical axonal degeneration, beginning distally in vulnerable regions of the nervous system, spreading proximally, and sometimes involving the neuron cell body [IS, 25, 26, 321. Bondareff et al. reported age-related changes in axonal transport of glycoprotein in the rat septodentate pathway [I]. Preceeding the neuronal cell loss, the atrophic changes
may occur in the axon terminals by just such a very slow “dying-back” process of aging. some other studies have shown constant synaptic density throughout the adult life (ages 16-72 years) [ 131and at the age of 89 years [5] in human cortex. Any speculation, therefore, must be made carefully. Concerning the dendrites, age-related dendritic atrophy has been reported from the investigations of Golgi preparations [6, 7, 29, 30, 33, 341 and electron microscopic studies [7, 8, lo]. By modifications of Golgi methods, age-related changes in the dendritic domen of neocortical [29] and archicortical [30] neurons in human materials were characterized by patchy loss of dendritic spines, progressive decrement of dendritic systems and subsequent degeneration of apical shaft with final loss of the cell body. Some experimental studies in Golgi preparations [6, 7, 33, 341 and quantitative electron microscopic analyses [7,8, IO] also reported that atrophic changes affect more peripheral portions of neuronal dendritic tree, especially spines. The age-related dendritic atrophy could be understood as the result of partial detierentation which occurs in axo-spinous synapses [ 10,291. Concerning the astroglial processes, several studies reported an increase in the number of their profiles or of their volume in the dentate gyrus [I 1,161, glacile nucleus of the senescent rats [8], and cerebral cortex of the senescent
NEUROPLL IN THE RAT INFERIOR
OLIVE
FIG. 6. An electron micrograpb showing a neuronal cell body of a rat inferior olive nucleus, 30 months otd. Large mass of lipofuscin granules are observed. N: nucleus, L: lipofuscin granules.
human brain 143. However there were no sign~~ant aherations in astrc ~ytic cell soma [4,11]. Terminals showing early processes of degenerative alterations are often seen in proximity to reac :tive astrocytic processes [Hi]. Glial cells have characteristic J membrane processes for amino acid transport which can in aeract with the microenvironment. During glial proliferation or gliosis which can cause disturbance of the
micr~nvironment and ne~on~ function, selective increase in amino acid uptake may occur [ 171. The possible f‘unctions of astrocytes are as follows: (I) insulation of unm! yrelinated neuronal processes fkom each other, (2) isolat :ion and removal of degenerative debris, (3) a channel for * moving metabolites from blood vessels to neurons, (4) regu Iation of synaptic activity. Thus, there is a cross relation between
388
IKARI. HAYASHI
AND OGATA
neuron and astroglia [14]. Ravens 1271 insisted on the So far as the physiological aging process is concerned, it angio-Rio-ne~on~ unit theory and that the pathological seems reasonable to suggest that the mo~hologica1 manifeschanges of any one of these three elements resulted in subtations are the accumulation of Iipofusein granules in ceU sequent pathological changes of the rest of the two elements. soma, atrophy of neuronal celI processes and proliferation of The relationships of the intracellular lipofuscin concenastroglial processes in neuropil. It is therefore necessary to tration with the reduction of cell number of cellular efficlarify the relationships between the lipofuscin concentraciency in aged rat have not been clear [23, 24, 311. In the tion in nerve cells and the changes in the neuropil. present study, we observed the progressive accumulation of the lipofuscin granules with advancing age in the neuronal cell bodies. We did not count the nerve cell numbers, so that ACKNOWLEDGEMENTS we can not comment on the change in nerve cell population in the inferior olive. There was however, no tendency to We should like to thank Prof. Hiroyuki Nakao and Prof. Yutaka increase dark cells or other degenerative nerve cells in aged Oomura for important suggestions, Ms. Junko Ono for technical assistance and Ms. Yoko Kondo for expert secretarial help. rats. REFERENCES I. Bondareff,
2. 3.
4.
5.
W., Y. Geinisman and A. Telser. Age-related changes in axonal transport of glycoprotein in the rat septodentate pathway. J. Cell Biol. 70: 332, I976. Bradbury, M. The Concept of a Blood-Brain Barrier. Chichester: John Wiley & Sons, 1979, pp. 8-12. Brody, H. Accumulation and distribution of lipofuscin, amyloid and senile plaques in the aging nervous system. In: Aging, vol. I, edited by H. Brady, D. Harman and J. M. Ordy. New York: Raven Press, 1974, pp. 39-78. Brody, H. and N. Vijayashankar. Anatomical changes in the nervous system. In: Handbook of the Biology of Aging, edited bv C. E. Finch and L. Havflick. New York: Von Nostrand Reinhold, 1975, pp. 241-261: Cragg, B. G. The density of synapses and neurons in normal, mentally defective and aging human brains. Brain 98: 81-90, 1975.
6. Feldman, L. M. Loss of dendritic spines in aging cerebral cortex. Anat. Embtyof. 148: 279-301, 1975. 7. Feldman, L. M. Dendritic changes in aging rat brain. In: The Aging Brain and Senile Dementia, edited by K. Nandy and I. Sherwin. New York: Plenum Press, 1977. 8. Fujisawa, K. and H. Shiraki. Study of axonal dystrophy I. Pathology of the graciie and cuneate nuclei in aging and old rat: A stereological study. Neuropath. appl. Neurobioi. 4: L-20, 1978. 9 Geinisman, Y. and W. Bondareff. Decrease in the number of synapses in the senescent brain: A quantitative electron microscopic analysis of the dentate gytus molecular layer of the rat. Mech Age DPV. 5: 11-23, 1976. 10 Geinisman, Y., W. Bondareff and J. T. Dodge. Dendritic atrophy in the dentate gyrus of the senescent rat. Am. J. Anat. 152: 321-330, 1978. t 1. Hassan, M. and P. Glees. Ultrastructural age changes in hippocampal neurons, synapses and neuroglia. Expt. Geront. S: 75-83, 1973. 12. Hirano, T. The structure of astrocytes in the adult staggerer. 1. Neuropath. exp. Neural. 35: 63-74, 1976. 13. Huttenlocher, P. R. Synaptic density in human frontal cortex: Developmental changes and effects of aging. Brain Res. 163:
19s.2O$ 1979. 14. Hyden, H. Dynamic aspects on the neuron-&a relationship: In: The Neuron, edited by H. Hyden. Amsterdam: Elsevier, 1967, pp. 179-219. 15. Ikari, K. and M. Takeichi. Electron microscopic study of the nervous svstem in rats with acrylamide intoxication. Kyushu Neuropsy~hjat. 22: 17-30, 1976 (in Japanese). 16. Landfield. P. W.. C. Wurtzand J. D. Lindsey. Qu~tification of synaptic vesicles’in hippocampus of aging rat and initial studies of possible relations to neurophysiology. Brain Res. Bull. 4: 75.3-763, 1980. 17. Logan, W. J. Amino acid transport by two glial cell lines and by pr&ferating gha. Expl Neural.. 53: 431-443, 1976. 18. Lund. R. D. Development and Plasticity of the brain: An Introduction. New Yard: Oxford University Press, 1978, pp. 179-
219. 19. Machado-Salas, J. P. and A, B. Scheiber. Limbic system of the aged mouse. Expl Neurol. 63: 347-355, 1979.
20. Mann, D. N. A. and P. 0. Yates. Lipoprotein
pigment-their relationship to aging in the human nervous system. I. The lipofuscin content of nerve cells. Brain 97: 481-488. 1974. 21. Mann, D. M. A., P. 0. Yates and J. E. Stamp. The relationship between lipofuscin pigment and aging in human nervous system. J. neural. Sci. 37: 83-93, 1978. 22. Mcgeer, E. G. and P. L. Mcgeer. Age changes in the human for some enzymes associated with metabolism of catecholamines, GABA and acetylcholine. In: Neurobiology ofAging (Advances in Behavioral Biology, vol. 16), edited by J. M. Ordy and K. R. Brizzee. New York: Plenum Press, 1975, pp. 287-305. 23. Moatamed, F. Cell frequencies in human inferior &vary complex. J. camp. Neurol. 128: 109-116, 1966. 24. Monagle, R. D. and H. Brody. The effects of age upon the main nucleus of the inferior olive in the human. J. cont~. Neuroi. 155: 61-66, 1973. 25. Prineas, J. The pathogenesis of dying-back polyneuropathies. i. An ultrastructural study of experimental tri-ortho-cresyl phosphate intoxication in the cat. J. Neuropath. exp. Neural. 28: 571-597, 1969. 26. Prineas, J. The pathogenesis
of dying-back polyneuropathies. II. An ultrastructural study of the experimental acrylamide intoxication in the cat. J. Neur~~path. exp. Neurof. 28: 598621. 1469. 27. Ravens, J. R. The angio-glio-neuronal unit: Morphology, physiology and pathology. In: 6th International Cungress of Neuropathology. Parts: Masson, 1970, pp. 417-418. 28. Samoraiski. , T.., C. Roisten and J. M. Ordv. Changes in behavior, b&t and neuroendocrine chemistry with age and stress in C57 BLiiO male mice. J. Geront. 26: 168-175, 1971. 29. Scheibel, M. E., R. D. Lindsay, U. Tomiyasu and A. B. Scheibel. Progressive dendritic changes in aging human cortex. Expl Neural. 47: 392403, 1975. 30. Scheibel, M. E., R. D. Lindsay,
U. Tomiyasu and A. B. Scheibel. Progressive dendritic changes in the aging human limbic system. Expl Neural. 53: 420-430, 1976. 31. Siakotos, A. N., D. Armstrong, N. Koppang and J. Mueller. Biochemical significance of age pigment in neurons. In: The Aging Brain and Senile Dementia, edited by K. Nandy and I. Sherwin. New York: Plenum Press, 1977, pp. 99-l 18. 32. Spencer, P. S. and H. H. Schaumburg. Central-peripheral distal axonopathy. The pathology of dying-back polyneuropathies. In: Progress in Neuropathology, vol. 3, edited by H. M. Zimmerman. New York: Grune & Stratton, 1976, pp. 253-295. 33. Sturrk, R. R. Quantitative and morphological changes in neurons and neuroglia iu the indusium griseum of aging mice. J. Geront. 32: 647-652, 1977. 34. Vaughan, D. W. Age-related deterioration of pyramidal celi basal dendrites in rat auditory cortex. J. cump. Nemo!. 171: 501-516, 1977. 35. Zeman, W. The ceroid lipofuscinosis-Batton-Vogst Syndrome. A model for human aging? In: Advance in Gerontological Research. vol. 3. edited bv B. L. Strehler. New York: Academic Press,‘I971, pp. 147-169.
Age-Related Changes in the Neuropil in the Rat Inferior Olive Nucleus: A Quantitative Electron Microscopic Study KOICHI
IKARI,
MICHIHIKO
HAYASHI
AND RYO OGATA
Department of Neuropsychiatry, Faculty of Medicine, Kyushu University, I-I, Maidashi I-chome, Higashi-ku, Fukuoka 812, Japan Received
I July
198 I
IKARI, K., hf. HAYASHI AND R. OGATA. Age-related changes in the neurapil in the rat inferior olive nucleus: A quantitative electron microscopic study. BRAIN RES. BULL 8(4) 381-388, 1982.-Age-related ultrastructural changes in the neuropil in the rat inferior olive nucleus were examined at 3, 6. 12, 18, 24 and 30 months old. The profiles of axon terminals, dendrites and nstroglial processes from random samplings within the neuropif were traced. Subsequently, the percentages of these pro&d areas in relation to the area of neuropil (relative volume fraction) were examined using the image analyzer system. The relative volume fractions of both axon terminals and dendrites in relation to the neuropil were found to have decreased in the aged rats, while the relative volume fraction of astroglial processes had progressively increased with aging. Inferior olive nucleus
Aging
Quantitative electron microscopic study
THE inferior olive in the human brain, one of the li~p~~ic nuclei, contains lipofuscin granules (aged pigment) in the cell soma beginning in the fetal stage [3]. The intracellular accumulation of lipofuscin granules were reported to be correlated directly with aging in this nucleus [21]. Mann and Yates [20) proposed that cell death may be a result of lipofuscin pigment accum~ation. The inferior olive nucleus is also reported capable of mantling its rno~holo~~~ integrity into advanced age, without significant change in nerve cell number [23,24]. The physiological aging process in inferior olive must be, therefore, examined from another point of view. Zeman [35] reported that a stunting of the dendritic apparatus occurs when the rate of pigment production is higher than the rate of regenerative molecular synthesis. Several quantitative morphometrical studies have been carried out on the changes in cell processes in the neuropil with aging [6, 7, 9, 10, 19, 29, 30, 331. These studies reported atrophic or dystrophic changes in cell processes, or proliferation of astrogiiai cell processes. As to the inferior olive, quantitative uItr~t~ctur~ studies on the neuropil of aged mammals are very few. The present study was undertaken to assess the age-related ultrastructural changes in the neuropil in the rat inferior olive nucleus. METHOD
The animals used were male Wistar King rats. They were reared in closed colony with water and a standard laboratory rat chow. Four animals from each of the following ages were prepared: 3,6, 12,18,24 and 30 months old. Animals exhibiting any grossly detectable pathology at the time of perfusion
Copyright
Q 1982 ANKHO
Intemation~
Neuropil
were not used. The animals were anesthetized with pentobarbital sodium (35 mg/kg IP) and perfused through the heart with buffered saline followed by 4% paraformaldehyde in Millonig buffer at pH 7.3. Blocks of the brain stem were cut from the level of C, to about 0.3 cm rostral to the pontine-medulhuy junction, sliced and postfixed in 1% osmium tetraoxide, dehyd~ted and embedded in epon 812. Ultrathin sections were cut and stained with many1 acetate and lead citrate. The inferior olive nucleus was identified in 1 Frn thick toluidine blue stained sections (Fig. 1) from which areas were selected for ultrastructural examinations. An electron microscope (JEOL-100-B) was used to examine and photograph areas within the neuropil of the inferior olive at a direct rn~~~tion of 10,000~. Random samphngs within the neuropil were performed on two to three different tissue blocks from each rat producing 40 electron micrographs per subject animal, totalhng 960 electron micrographs. Photographic enlargement resulted in a final magnification of 30,000x on the prints. The profiles of axon terminals, dendrites and astrogliai processes in neuropil of each rni~r~~phs were traced. The percentages of the profiled areas in relation to the area of neuropil (named relative volume fraction) were then calculated using the image analyzer system. The “neuropil” in this context means the remainder of the inferior olive nucleus after exclusion of cell bodies, myelinated fibers and blood vessels. The criteria for identification of axon terminals were the presence of synaptic vesicles and/or of synaptic membrane differentiation. The criteria for identification of dendrites were as follows: any round or cylindrical profiles of cell processes which showed postsynaptic membrane differentiation on the surface and/or
Inc .-0361-9230/82/~381-~$03.~/0
382
IKARl,
HAYASNl
AND OGATA
any round or cylindrical profiles of greater than 0.3 pm in diameter which contained evenly spaced microtubl~s~ or any round or cylindrical profiles of smaller than 0.3 pm in diameter which contained no microtubles or glio~l~ent (the smallest dendrite and spine) 1181 and showed rather clear cytoplasmic matrix. The criteria for identification of astroglial cell processes were as follows: round, cylindri~ai or irregumrly shaped profibs which filled up the space between neuronal cell processes. These showed clear cytoplasmic matrix and contained contents, glycogen particles, and were poor in other organelIes [2, 12, IS]. The profIles which could not be ident~ed by the above mentioned criteria remained as unclassified. All of the jdent~able profiles except for unmyel~ted axons were traced and the percentage of the area of each profile for the area of neuropil (named relative volume fraction) was measured (Figs. 2,3,4). Figure 2 is a line drawing of the profiles of neuronal and astroglial processes in the ~e~ropil ~hoto~aphed in Fig. 4. Additional electron microscopic observations were made on the lipofuscin granules in the cell soma for one or two animals at 5 weeks, 3, 6, 18 and 30 months of age. FIG. I. inferior olive nucleus of a 3 month old rat. Epon 1 t.lm, toluidine blue stained. P: pyramis meduliae oblongatae. medz n. olivaris accesorius medialis. inf: n. olivaris inferior. dors: n. olivaris accessorius dorsalis. 50x
r
RESULTS
TabIe 1 shows the relative volume fractions of cell processes within the neuropil at various ages. Mean values (percentage) for each animal were derived from analyses of 40 electron micrographs. Mean values for groups of each age were calcuiated from ~dividual mean values and treated statistica~y with the ~de~ndent f-test for multiple comparison test (Dunnet’s test). The volume fraction of axon termi-
(B)
FIG.2.An example of profiles tracing the (A) presynaptic terminals and dendrites, (B) astroglial processes ofthe same rn~c~~~h
as in Fig. 4.
NEUROPIL
IN THE RAT INFERIOR
383
OLIVE TABLE
AGE-RELATED
CHANGES
I
IN THE RELATIVE VOLUME FRACTION INFERIOR OLIVE NUCLEUS
24
30
3
6
Mean _CS.D. of Terminal per Rat
RI R2 R3 R4
28rt 30t 29-c 28%
9 9 9 8
282 282 292 282
9 7 9 8
28% 272 25+ 292
8 8 7 9
28 27 28 28
292
9
28-+
8
272
8
28 rt 9
27 + 7*
24 r 7*
28 25 26 28
28 28 25 29
25 26 24 29
Mean rt SD. of Dendrite per Rat
RI R2 R3 R4
Mean rt S.D. per Group Mean rt SD. of Glial Process per Rat Mean t S.D. per Group Unclassified and other axon profiles except for axon
RI R2 R3 R4
32rt 8 34Icr: 11 34 * 10 32-c 8
33 -c 31+ 31-c 312
10 9 6 8
32~
8
33%
6rt 6r: 6-t 6r
3 3 4 3
6rt
3
34
18
IN THE
Age (month)
Mean r S.D. per Group
12
(410)OF NEUROPIL
29k 32 f 34 -c 282
8 10 10 7
9
31+-
9
7F 7* 6+ 7k
3 4 4 4
7k
4
32
r 9 -t- 8 r. 8 f 9
r * it rt
8 7 8 9
23 26 26 22
r f k f
i + + +
6 7 8 7
8 9 7 7
24 23 22 25
+ 7 _’ 6 -e 6 2 8
+ 2 2 +
8 9 7 8
27 +- 8*
28 + 8*
26 r 8*
10-i- 5 9+ 4 11-t 5 9t 4
13 rt 5 954 13 + 5 12 t 5
11 14 10 13
15 + 4 17 z!z5 17 t 4 16 -t 4
lo*
12 c 5*
12 r 5*
16 t 4*
34
36
34
5 32
25 1?16 k 4 k 4
terminals n=160. *p
nals tends to decrease over the lifespan under study. Compared with 3 month-old rats, the volume fraction of axon terminals at the ages of 24 and 30 months significantly decreased (p
amount of lipofnscin granules are observed. The degree of accum~ation of the lipofuscin granules seemed to increase with aging. Figure 6 shows a neuronal cell body from the inferior olive nucleus at the age of 24 months in which large mass of lipofuscin granules are observed. As far as we observed, there was no tendency to increase dark cells or other degenerative nerve cells in the inferior olive nucleus of 30 month old rats. DISCUSSION
In the present study, we observed relative decrease in the areas of axon terminals and dendrites, and relative increase in the area of astroglial processes in the inferior olive nucleus of the senescent rat. These findings may represent the common changes in the aging process of neuronal tissue. It seems reasonable to suggest that the decrease in the neuronal cell processes and the increase in the astroglial processes in the inferior olive nucleus with advancing age are the glio-neuronal or neuro-glial age-related changes which indicate the gradual reduction in the cellular efficiency. Concerning the axon terminals, a few age-related changes have been reported. They were the decrease in synaptic density [8, 9, 111, decrease in the terminal area [8,16] and synaptic vesicles depletion [I 1,161. Fujisawa and Shiraki [8] reported the slow progressive decrease in the synaptic density and the relative volume fraction of axon terminals in the gracile nucleus of aged rats. Geinisman and Bondareff [9]
IKARI, HAYASHI
AND OCiATA
FIG. 3. An example of neuropil of the inferior olive nudeus, 3 month old rat, with presynaptic terminals fT), dendrites (D) and as&o&al processes (G). Small amount of astroglial processes are observed. M: myelin sheath. 30,ooox
NEUROPIL
IN THE
RAT INFERIOR
OLIVE
FIG. 4. An example ofneuropil of the inferior olive nucieus, 30 month ofd rat, with presynaptic terminals (T), dendrites (D) and astroglial processes (G). Note the increase of astroglial profiles which shows clear cytoplasmic matrix containing glycogen particles (gly) and glial filaments (fil). Arrows indicate gap junctions between adjacent astrocytic processes. 30$00x
386
IKARI,
HAYASHI
AND OGATA
FIG. 5. An electron micrograph of a neuronal cell body of a rat inferior olive nucleus, 5 weeks old. Small amount of lipofuscin granules are observed. N: nucleus, L: lipofuscin granules.
reported a loss of synapses in the molecular layer of the dentate gyrus in aged rats. Hassan and Glees [ll] in their study of age changes in the hippocampus noted that while axo-somatic contacts were a prominent feature of younger rats, these decreased in aged rats along with characteristic membrane thickening and accumulution of synaptic vesicles. Landfield et al. [16] also reported reduced numbers of synaptic vesicles in the terminals and a trend toward reduced terminal size of aged rat hippocampus. Neurochemical studies reported that changes in the acetylcholinesterase activity, norepinephline, serotonine and GABA contents in the brain have been associated with increasing age [22,28]. These findings may suggest reduced functioning of synapse with aging. One possible mechanism for the axonal atrophy and reduction of synaptic activity which accompanies advancing age is a “dying-back” degeneration (or retrograde axonal degeneration) of neurons. This could be explained by the perikaryal insufficiency or axonal transport system damage. Namely, there is a symmetrical axonal degeneration, beginning distally in vulnerable regions of the nervous system, spreading proximally, and sometimes involving the neuron cell body [IS, 25, 26, 321. Bondareff et al. reported age-related changes in axonal transport of glycoprotein in the rat septodentate pathway [I]. Preceeding the neuronal cell loss, the atrophic changes
may occur in the axon terminals by just such a very slow “dying-back” process of aging. some other studies have shown constant synaptic density throughout the adult life (ages 16-72 years) [ 131and at the age of 89 years [5] in human cortex. Any speculation, therefore, must be made carefully. Concerning the dendrites, age-related dendritic atrophy has been reported from the investigations of Golgi preparations [6, 7, 29, 30, 33, 341 and electron microscopic studies [7, 8, lo]. By modifications of Golgi methods, age-related changes in the dendritic domen of neocortical [29] and archicortical [30] neurons in human materials were characterized by patchy loss of dendritic spines, progressive decrement of dendritic systems and subsequent degeneration of apical shaft with final loss of the cell body. Some experimental studies in Golgi preparations [6, 7, 33, 341 and quantitative electron microscopic analyses [7,8, IO] also reported that atrophic changes affect more peripheral portions of neuronal dendritic tree, especially spines. The age-related dendritic atrophy could be understood as the result of partial detierentation which occurs in axo-spinous synapses [ 10,291. Concerning the astroglial processes, several studies reported an increase in the number of their profiles or of their volume in the dentate gyrus [I 1,161, glacile nucleus of the senescent rats [8], and cerebral cortex of the senescent
NEUROPLL IN THE RAT INFERIOR
OLIVE
FIG. 6. An electron micrograpb showing a neuronal cell body of a rat inferior olive nucleus, 30 months otd. Large mass of lipofuscin granules are observed. N: nucleus, L: lipofuscin granules.
human brain 143. However there were no sign~~ant aherations in astrc ~ytic cell soma [4,11]. Terminals showing early processes of degenerative alterations are often seen in proximity to reac :tive astrocytic processes [Hi]. Glial cells have characteristic J membrane processes for amino acid transport which can in aeract with the microenvironment. During glial proliferation or gliosis which can cause disturbance of the
micr~nvironment and ne~on~ function, selective increase in amino acid uptake may occur [ 171. The possible f‘unctions of astrocytes are as follows: (I) insulation of unm! yrelinated neuronal processes fkom each other, (2) isolat :ion and removal of degenerative debris, (3) a channel for * moving metabolites from blood vessels to neurons, (4) regu Iation of synaptic activity. Thus, there is a cross relation between
388
IKARI. HAYASHI
AND OGATA
neuron and astroglia [14]. Ravens 1271 insisted on the So far as the physiological aging process is concerned, it angio-Rio-ne~on~ unit theory and that the pathological seems reasonable to suggest that the mo~hologica1 manifeschanges of any one of these three elements resulted in subtations are the accumulation of Iipofusein granules in ceU sequent pathological changes of the rest of the two elements. soma, atrophy of neuronal celI processes and proliferation of The relationships of the intracellular lipofuscin concenastroglial processes in neuropil. It is therefore necessary to tration with the reduction of cell number of cellular efficlarify the relationships between the lipofuscin concentraciency in aged rat have not been clear [23, 24, 311. In the tion in nerve cells and the changes in the neuropil. present study, we observed the progressive accumulation of the lipofuscin granules with advancing age in the neuronal cell bodies. We did not count the nerve cell numbers, so that ACKNOWLEDGEMENTS we can not comment on the change in nerve cell population in the inferior olive. There was however, no tendency to We should like to thank Prof. Hiroyuki Nakao and Prof. Yutaka increase dark cells or other degenerative nerve cells in aged Oomura for important suggestions, Ms. Junko Ono for technical assistance and Ms. Yoko Kondo for expert secretarial help. rats. REFERENCES I. Bondareff,
2. 3.
4.
5.
W., Y. Geinisman and A. Telser. Age-related changes in axonal transport of glycoprotein in the rat septodentate pathway. J. Cell Biol. 70: 332, I976. Bradbury, M. The Concept of a Blood-Brain Barrier. Chichester: John Wiley & Sons, 1979, pp. 8-12. Brody, H. Accumulation and distribution of lipofuscin, amyloid and senile plaques in the aging nervous system. In: Aging, vol. I, edited by H. Brady, D. Harman and J. M. Ordy. New York: Raven Press, 1974, pp. 39-78. Brody, H. and N. Vijayashankar. Anatomical changes in the nervous system. In: Handbook of the Biology of Aging, edited bv C. E. Finch and L. Havflick. New York: Von Nostrand Reinhold, 1975, pp. 241-261: Cragg, B. G. The density of synapses and neurons in normal, mentally defective and aging human brains. Brain 98: 81-90, 1975.
6. Feldman, L. M. Loss of dendritic spines in aging cerebral cortex. Anat. Embtyof. 148: 279-301, 1975. 7. Feldman, L. M. Dendritic changes in aging rat brain. In: The Aging Brain and Senile Dementia, edited by K. Nandy and I. Sherwin. New York: Plenum Press, 1977. 8. Fujisawa, K. and H. Shiraki. Study of axonal dystrophy I. Pathology of the graciie and cuneate nuclei in aging and old rat: A stereological study. Neuropath. appl. Neurobioi. 4: L-20, 1978. 9 Geinisman, Y. and W. Bondareff. Decrease in the number of synapses in the senescent brain: A quantitative electron microscopic analysis of the dentate gytus molecular layer of the rat. Mech Age DPV. 5: 11-23, 1976. 10 Geinisman, Y., W. Bondareff and J. T. Dodge. Dendritic atrophy in the dentate gyrus of the senescent rat. Am. J. Anat. 152: 321-330, 1978. t 1. Hassan, M. and P. Glees. Ultrastructural age changes in hippocampal neurons, synapses and neuroglia. Expt. Geront. S: 75-83, 1973. 12. Hirano, T. The structure of astrocytes in the adult staggerer. 1. Neuropath. exp. Neural. 35: 63-74, 1976. 13. Huttenlocher, P. R. Synaptic density in human frontal cortex: Developmental changes and effects of aging. Brain Res. 163:
19s.2O$ 1979. 14. Hyden, H. Dynamic aspects on the neuron-&a relationship: In: The Neuron, edited by H. Hyden. Amsterdam: Elsevier, 1967, pp. 179-219. 15. Ikari, K. and M. Takeichi. Electron microscopic study of the nervous svstem in rats with acrylamide intoxication. Kyushu Neuropsy~hjat. 22: 17-30, 1976 (in Japanese). 16. Landfield. P. W.. C. Wurtzand J. D. Lindsey. Qu~tification of synaptic vesicles’in hippocampus of aging rat and initial studies of possible relations to neurophysiology. Brain Res. Bull. 4: 75.3-763, 1980. 17. Logan, W. J. Amino acid transport by two glial cell lines and by pr&ferating gha. Expl Neural.. 53: 431-443, 1976. 18. Lund. R. D. Development and Plasticity of the brain: An Introduction. New Yard: Oxford University Press, 1978, pp. 179-
219. 19. Machado-Salas, J. P. and A, B. Scheiber. Limbic system of the aged mouse. Expl Neurol. 63: 347-355, 1979.
20. Mann, D. N. A. and P. 0. Yates. Lipoprotein
pigment-their relationship to aging in the human nervous system. I. The lipofuscin content of nerve cells. Brain 97: 481-488. 1974. 21. Mann, D. M. A., P. 0. Yates and J. E. Stamp. The relationship between lipofuscin pigment and aging in human nervous system. J. neural. Sci. 37: 83-93, 1978. 22. Mcgeer, E. G. and P. L. Mcgeer. Age changes in the human for some enzymes associated with metabolism of catecholamines, GABA and acetylcholine. In: Neurobiology ofAging (Advances in Behavioral Biology, vol. 16), edited by J. M. Ordy and K. R. Brizzee. New York: Plenum Press, 1975, pp. 287-305. 23. Moatamed, F. Cell frequencies in human inferior &vary complex. J. camp. Neurol. 128: 109-116, 1966. 24. Monagle, R. D. and H. Brody. The effects of age upon the main nucleus of the inferior olive in the human. J. cont~. Neuroi. 155: 61-66, 1973. 25. Prineas, J. The pathogenesis of dying-back polyneuropathies. i. An ultrastructural study of experimental tri-ortho-cresyl phosphate intoxication in the cat. J. Neuropath. exp. Neural. 28: 571-597, 1969. 26. Prineas, J. The pathogenesis
of dying-back polyneuropathies. II. An ultrastructural study of the experimental acrylamide intoxication in the cat. J. Neur~~path. exp. Neurof. 28: 598621. 1469. 27. Ravens, J. R. The angio-glio-neuronal unit: Morphology, physiology and pathology. In: 6th International Cungress of Neuropathology. Parts: Masson, 1970, pp. 417-418. 28. Samoraiski. , T.., C. Roisten and J. M. Ordv. Changes in behavior, b&t and neuroendocrine chemistry with age and stress in C57 BLiiO male mice. J. Geront. 26: 168-175, 1971. 29. Scheibel, M. E., R. D. Lindsay, U. Tomiyasu and A. B. Scheibel. Progressive dendritic changes in aging human cortex. Expl Neural. 47: 392403, 1975. 30. Scheibel, M. E., R. D. Lindsay,
U. Tomiyasu and A. B. Scheibel. Progressive dendritic changes in the aging human limbic system. Expl Neural. 53: 420-430, 1976. 31. Siakotos, A. N., D. Armstrong, N. Koppang and J. Mueller. Biochemical significance of age pigment in neurons. In: The Aging Brain and Senile Dementia, edited by K. Nandy and I. Sherwin. New York: Plenum Press, 1977, pp. 99-l 18. 32. Spencer, P. S. and H. H. Schaumburg. Central-peripheral distal axonopathy. The pathology of dying-back polyneuropathies. In: Progress in Neuropathology, vol. 3, edited by H. M. Zimmerman. New York: Grune & Stratton, 1976, pp. 253-295. 33. Sturrk, R. R. Quantitative and morphological changes in neurons and neuroglia iu the indusium griseum of aging mice. J. Geront. 32: 647-652, 1977. 34. Vaughan, D. W. Age-related deterioration of pyramidal celi basal dendrites in rat auditory cortex. J. cump. Nemo!. 171: 501-516, 1977. 35. Zeman, W. The ceroid lipofuscinosis-Batton-Vogst Syndrome. A model for human aging? In: Advance in Gerontological Research. vol. 3. edited bv B. L. Strehler. New York: Academic Press,‘I971, pp. 147-169.