Mechanisms of Ageing and Development, 61 (1991) 23-32 Elsevier ScientificPublishers Ireland Ltd.
23
LIGHT AND E L E C T R O N M I C R O S C O P I C FEATURES OF P E R I P H E R A L GANGLION C E L L S C U L T U R E D FROM YOUNG AND AGED WISTAR RATS
JARI KOISTINAHO, K I M M O HATANP.~,,~ and ANTTI H E R V O N E N LaboratorY of Gerontology, Department t~ Public Health, University of ?'ampere. P. O. Box 607. SF-33101 Tampere (Finland)
(Received January 15th, 1991) (Revision received March 28th, 1991) SUMMARY Neurons of the dorsal root ganglion (DRG) and sympathetic superior cervical ganglion (SCG) were cultured as explants from young adult (3 months old) and aged (28 months old) Wistar rats. Both aged D R G and SCG neurons showed delayed neurite outgrowth compared to young adult neurons. Young and some aged cultured neurons had an ultrastructure similar to that found in normal uncultured cells, but most of the aged cultured neurons displayed a heavy accumulation of homogenous lipid-like inclusions in addition to classic age pigments. Occasionally, large neurofilament aggregates were seen in aged D R G neurons. They were sometimes localized perinuclearly, resembling neurofibrillary tangles. The results show that even very old peripheral neurons survive in culture. As aged cultured neurons show degenerative changes not observed in young adult neurons, it is suggested that cultured peripheral neurons of different ages could provide useful means for neuronal aging studies.
Key words: Aging; Lipofuscin; Peripheral neuron; In culture; Ultrastructure
INTRODUCTION Neurons from the peripheral and central nervous system of embryonic and perinatal mammals have been used extensively in culture studies because of the success in dissociating and maintaining undifferentiated and immature neurons alive in vitro [1-3]. Even though these cells can be maintained in culture for several months, Correspondence to: Jari Koistinaho, Laboratoryof Gerontology,Departmentof Public Health, University of Tampere, P.O. Box 607, SF-33101 Tampere, Finland.
0047-6374/91/$03.50 Printed and Published in Ireland
© 1991 ElsevierScientific Publishers Ireland Ltd.
24 it is unusual to find any structure that could be clearly associated with aging changes in neurons, such as lipofuscin granules, neuroaxonal dystrophy or neurofibrillary degeneration [2,4,5]. Over the last 15 years there has been an increasing number of studies on cultured adult peripheral ganglion cells [2,4-9]. The advantage of using adult neurons in culture is that they better represent mature cells normally present in vivo, enabling more relevant experiments with a reflection of functions in the adult nervous system. In addition, the survival of differentiated neurons may be better than that of undifferentiated neuroblasts [5-7]. Recently, there have been some reports on successful short-term cultivation of aged peripheral neurons [9,10] but no histological studies on cultured senescent peripheral neurons have been reported. In the present study, we describe some age-related morphological features of dorsal root ganglion (DRG) cells and sympathetic neurons cultured from 28-month-old rats. MATERIALS AND METHODS Animals and tissue culture A total of eight male Wistar rats aged three (N = 4) and 28 months (N = 4), were used. They were anesthetized by ether inhalation and decapitated. Superior cervical ganglia (SCG) and dorsal root ganglia (DRG) were dissected out, desheathed and placed in plastic petri dishes containing Ca-Mg-free Dulbecco's phosphate buffer medium (GIBCO, Grand Island, NY). The ganglia were sliced using a sterilized razor blade and the ganglion tissues were placed in separate plastic culture dishes (NUNC, Denmark) coated with poly-L-lysine (Sigma Chemical Co., St. Louis, MO) and then covered by glass cover slides. The ganglion explants were plated in Eagle's minimum essential medium (MEM) (GIBCO) with 6 g/I glucose, 10% (v/v) horse serum (GIBCO), 5% fetal calf serum (GIBCO), 1% glutamine (GIBCO) 30 nM Nerve growth factor (Sigma) and 0.1% gentamycin. The medium was changed twice a week using the same medium with 5'¼, horse serum but without fetal calf serum. The cultures were maintained at +37°C in humidified 10% CO~-90% air (v/v) and viewed daily using an Olympus CK2 phase-contrast microscope. Fluorescence microscopy At 21 days, four SCG cultures were treated with 10 -'~ M noradrenaline bitartrate (Sigma) for 20 min. Then all cultures were rinsed in phosphate buffered saline. Half of the cultures were rapidly dried with a hair dryer and exposed to paraformaldehyde vapor at + 60°C for 60 min and embedded in xylene. The fluorescence properties of lipopigments and noradrenaline were then examined from these cultures using a Vanox T fluorescence microscope. Some cultures were also stained with toluidine blue and studied using the same microscope with visible light. Electron microscopy The other half of the cultures were fixed with 3% glutaraldehyde in buffer at room
25 temperature for 30 min, postfixed in osmium tetroxide, dehydrated and embedded in Epon. Thin sections were stained with lead citrate-uranyl acetate and examined using Jeol J E M 2000 electron microscope. RESULTS The neurons grew neurites from 3-month-old ganglia already at day 3, whereas outgrowths from 28-month-old rats were seen at day 6 - 7 (Fig. l a - d ) . All explants also grew out fibroblasts and Schwann cells, and occasional migrated neurons were seen outside the explants (Fig. l e - f ) . After 14 days, a n u m b e r o f enlarged neurons containing granular material were seen in phase-contrast microscopy in S C G and D R G explants, especially in those
Fig. I. Phase contrast (a-d,f) and fluorescence (e) micrographs of living young adult and aged peripheral neurons in culture. Nerve fiber outgrowths from a 28-month-old DRG explant maintained for 6 (a), 8 (b) and 10 (c) days in culture. For comparison nerve fiber outgrowths from 3-month-old DRG explant maintained for 4 days is shown in (d). (e): Heavy accumulations of autofluorescent lipopigments (arrows) are seen in living 21-day-old cultures of aged sympathetic ganglion expLant. (f): Migrated healthy-looking young sympathetic neurons and nonneuronal cells at 21-days in culture, x 500.
26
from old animals. These cells remained attached and had neurite-like extensions for several days, although they gradually vacuolized and disappeared later on. At 21 days, cultures from younger animals also had some vacuolized neurons. Vacuolated fibroblasts and Schwann cells were seen in cultures from aged animals, but not in those from young adult rats. In living cultures, fluorescence microscopy revealed a number of neurons containing autofluorescent lipopigments throughout the cytoplasm (Fig. 2a-b). This feature was more frequently observed in cultures from aged animals. Heavily pigmented large cells, presumably neurons, were observed to have migrated out of the explants also in young adult tissues. With further aging of the cultures, autofluorescent material accumulated even in nonneuronai cells, e.g. in fibroblasts and Schwann cells (Fig. 2c). In these cells, the autofluorescent material was usually located perinuclearly. The sympathetic neurons treated with noradrenaline showed strong catecholamine-specific fluorescence both in the somata and dendrites as well as in axon varicosities (Fig. 3a,b). Non-neuronal cells displayed catecholamine fluorescence which was much less than in neurons (Fig. 3b). Toluidine blue staining revealed several small groups of neurons without evident growth of processes (Fig. 4a-c). However, these cells were present throughout the time that the cultures were maintained. Sympathetic neurons, in particular, formed small pseudoganglia even in cultures of aged neurons (Fig. 4b), whereas D R G neurons were usually separately attached to non-neuronal cells (4a). In electron microscopy, both neurons of normal in vivo fine structure and neurons showing several pathological alterations were seen. The neurons classified as normal had a round and large nucleus, intact mitocondria and Goigi complex, regularly arranged neurofilaments and endoplasmic reticulum, and a varying number of lysosomes or lipofuscin granules (not shown). In neurons of large regions of several explants as well as in many dissociated neurons, nuclei showing a considerable invagination of cytoplasma were seen (Fig. 5a). These cells frequently contained a cytoplasmic accumulation of large lipid-like vacuoles in addition to clumps of classic lipofuscin (Fig. 6a,b). The surrounding nonneuronal cells also contained very large homogenous vacuoles (Fig 613. In aged D R G neurons, very heterogeneous inclusion bodies were sometimes observed (6c-e). They were composed of a round lipid-like component, a more electron-dense matrix resembling that of the classic pigment, and an array of thin filament-like strings (Fig. 6c-e). The size of these inclusions (1.5-2.5 #m) was greater than that of the classic lipofuscin (0.4-1.0 ~m). Finally, about 25% Figs. 2-4. Figs. 2a-c show autofluorescence micrographs of lipopigment accumulations in sympathetic neurons and fibroblasts (asterisks) which have migrated from aged sympathetic ganglion explants at 20 days in culture. Figs. 3a,b show catecholamine fluorescent sympathetic neurons (brightly fluorescent cells1 after noradrenaline incubation. Also non-neuronal cells are slightly fluorescent. Figs. 4 a - c show toluidine blue stained aged DRG (4a) and sympathetic (4b,c) ganglion cells cultured for I 0 days. Most of them have not grown neurites. (2a;3b): x 130; (2b,c: 3a;4a-c): x 500.
28
Fig. 5. ( a - d ) Electron micrographs of an aged D R G neuron cultured for 21 days. An accumulation of homogenous inclusion bodies (arrow heads), large vacuoles (open arrows), a cytoplasmic invagination (star) into the nucleus and a perinuclear aggregation of neurofilaments (black arrows) are the pathological changes observed in this cell. (5a): x 8000; (5b,c): x 20 000; (5c): × 160 000.
29
Fig. 6. (a-f) Electron micrographs from an aged sympathetic ganglion cxplant (a.b) and D R G explant (c-f) maintained in culture for 21 days. Pigment inclusions in sympathetic neurons are large and homogenous, whereas in D R G neurons heterogenenous pigment bodies are also seen. In (61") a Schwann cell containing very large inclusion bodies, between of which an intact mitochondrion can still be seen. (6a): x8000; (6b): x 4 0 000; (6c): ×30 000: (6d): ×60 000: (6e): x 5 0 000: (61"): x 16 000.
30 of the aged D R G neurons showed neurofilament aggregates, which were sometimes localized around the nucleus (Fig. 5c,d). This alteration was not seen in sympathetic neurons. DISCUSSION In the present study, no specific neuronal or gliai cell stainings were used to distinguish between neurons and non-neuronal cells in culture. Unlike flat nonneuronal cells, D R G neurons and sympathetic ganglion cells are characterized by large and round somata. Proof of neuronal origin of most of the neuron-like cells was provided by the neuronal processes. Furthermore, the catecholamine-specific fluorescence that was observed after incubating sympathetic neurons with noradrenaline as well as the fine structure typical of peripheral ganglion cells ascertain that our cultures contained detectable and living neurons even in preparations of aged rat tissues. Ganglion explants from aged animals grew out neurites later than the explants from young adult rats, which is in agreement with earlier studies [2,5,7]. This indicates an impaired ability of aged peripheral neurons to regenerate after axotomy. Enhanced degeneration processes after axotomy may contribute to the late response of regeneration of aged neurons, since pathological structural changes were more pronounced and more rapid in aged ganglion explants. In spite of this, the survival of aged peripheral neurons was comparable to that of young adult neurons. The accumulation of autofluorescent pigment bodies like lipofuscin and ceroid pigments in postmitotic cells including neurons is thought to be an indication of agerelated degeneration in which free radicals and lipid peroxidation are involved [11-14]. Even though there is a body of evidence suggesting that lipid peroxidation and free radical production increase in ratio to free radical scavengers during aging [14], there is as yet no direct evidence that the attack of these radicals on cellular lipids and proteins are the cause of the pigment formation. Instead, recent studies on the experimental production of lipofuscin point to the role of impaired protease function in a rapid pigment accumulation [t5,16], whereas a longer period of time for the lack of essential antioxidants is needed to lead to pigment formation [13,17]. In the present study, autofluorescent pigment bodies occupied the cytoplasm of ganglion cells in a few days, particularly in aged neurons. These granules were either homogenous lipid-like large inclusions or composed of heterogenous material, differing thus from the classic lipofuscin in structure. Similar type of pigment granules are found in neurons after treatment with leupetin, a protease inhibitor [15,16], suggesting that the altered function of proteases rather than increased lipid peroxidation may be responsible for the pigment formation. Some of the cultured aged D R G neurons contained neurofilament aggregates, which sometimes were localized perinuclearly, thus resembling neurofibrillary tangles [6,18,19]. In humans, similar morphological changes are present in various
31
neurological disorders like Alzheimer's disease, and have been found to a lesser extent in the brains of healthy individuals [6,18]. In the rat, neuro-fibrillary changes in neurons are not normally found, but after treatment with aluminium lactate or inhibitors of mitotic spindle (e.g. colchicine, vincristine, vinblastine) striking cytoplasmic masses of neurofilaments can be seen [18,20-22]. The present results show that aged DRG neurons are more prone to the stress caused by axotomy and the culture procedure than young adult DRG neurons, and that under these circumstances they react by forming neurofibrillary aggregates typical of neurotoxintreated rodent neurons. In conclusion, both aged DRG and sympathetic neurons survive in culture, although regeneration after axotomy is delayed compared with neurons from young adult rats. Aged neurons in culture displayed pathological changes such as heavy accumulation of autofluorescent inclusions and neurofilament aggregates, which are absent from normal peripheral neurons in vivo, probably due to age-related impairment of regeneration ability after axotomy and to the stress induced by the culture procedure. The culture of neurons of different ages could provide useful means for studying neuronal aging. ACKNOWLEDGEM ENTS
The study was financially supported by the Finnish Cultural Foundation of Pirkanmaa. REFERENCES 1 2 3 4 5
S.M. Crain, Neurophysiologic studies in Tissue Culture, Raven Press, 1976. S.U. Kim, Neuronal ageing in tissue and cell cultures: a review. In Vitro. 19 (1983) 73-82. M. Murray, Nervous tissue in vitro. In E.N. Willner (ed.), Cells and Tissues in Culture, Vol. II, Academic Press, New York, 1965, pp. 373-455. S.U. Kim, K.G. Warren and M. Kalia, Tissue culture of adult human neurons. Neurosci. Lett., II (1979) 137-141.
B.S. Scott, Adult neurons in cell culture: electrophysiological characterization and use in neurobiological research. Prog. NeurobioZ. 19 (1982) 187-211. 6 J. Dodd, D. Solter and T,M. Jessel, Monoclonal antibodies against carbohydrate diffrentation antigens identify subsets of primary sensory neurones. Nature. 279 (1984) 469-472. 7 B.S. Scott, Adult mouse dorsal root ganglia neurons in cell culture. J. NeurohioL. 8 ( 1977~ 417-427. 8 R.A. Smith amd I.B. Mclnnes, Phase contrast and electron microscopical observations of adult dorsal root ganglion cells maintained in primary culture. J. Anat.. 145 (1986t 1-12, 9 Y. Uchita and M. Tomonaga, Effect of nerve growth factor and heart cell conditioned medium on neurite regeneration of aged sympathetic neurons in culture. Brain Res.. 348 (1985) 100-106. 10 D.H. Silberberg and S.U. Kim, Studies of aging in cultured nervous system. Intern. Rev. C:vtol., Suppl. 10 (1979) 117-130. 11 K.R. Brizzee, J.M. Ordy and B. Kaack, Early appearance and regional differences in intraneuronal and extraneuronal lipofuscin accumulation with age in the brain of nonhuman primate (Macaca mulatta). J. Gerontol., 29 (1974) 366-381. 12 J. Koistinaho. Difference in the age-related accumulation of lipopigments in the adrenergic and nonadrenergic peripheral neurons in the male rat. Gerontology, 32 (1986) 300-307.
32 13 14 15 16 17
18 19
20 21 22
J. Koistinaho, H. A l h o a n d A. Hervonen, Effect of vitamin E and selenium supplement on the aging peripheral neurons of the male Sprague-Dawley rats. Mech. Ageing Dev.. 51 (1990) 63-72. A.L. Tappel, Lipid peroxidation and fluorescent molecular damage to membranes. In B. Trumps and A. Arstila (eds.), Pathology ~[ Cell Membranes, Academic Press, New York, 1975, pp. 145-170. G.O. Ivy, J. Wenzel. M. Baydry and G. Lynch, Inhibitors of lysosomal enzymes: accumulation of lipofuscin-like dense bodies in the brain. Science. 226 (1984) 985-988. M. Katz and M.J. Shankler, Development of lipofuscin-like fluorescence in the retinal epithelium in response to protease inhibitor treatment. Mech. Ageing Dev.. 49 (1989) 23-40. M,L. Katz. W.G. Robinson, R.L. Herrman, A.B. Groome and J.G. Bieri, Lipofuscin accumulation resulting from senescence and vitamin E deficiency: spectral properites and tissue distribution. Mech. Ageing Dev., 25 (1984) 149-159. R.D. Terry, The fine structure of neurofibrillary tangles in Alzheimer's disease. J. Neuroputhol. EYp. Neurol., 22 (1963) 629. H. Matsuyama and S. Nakamura, Senile changes in the brains in the Japanesc: Incidence of Alzheimer's neurofibrillary change and senile plaques. In R. Katzman, R. Terry and K. Bick (eds.). Alzheimer's Disease, Senile Dementia and Related Disorders. New York, Raven Press, 1976, pp. 183-204, M.P. Daniels, Fine structural changes in neurons and nerve fibers associated with colchicine inhibition of nerve fiber formation in vitro. J. Cell Biol., 58 (1973) 463-470. B. Ghetti, Induction of neurofibrillary degeneration following treatment with maytansine in vivo. Brain Res., 163 (1979) 9-19. I. Klatzo, H. Wisnieski and E. Streicker, Experimental production of neurofibrillary degeneration I. Light microscopic observations. J. Neuropathol. Exp. Neurol.. 24 (1965) 187-199.
23
LIGHT AND E L E C T R O N M I C R O S C O P I C FEATURES OF P E R I P H E R A L GANGLION C E L L S C U L T U R E D FROM YOUNG AND AGED WISTAR RATS
JARI KOISTINAHO, K I M M O HATANP.~,,~ and ANTTI H E R V O N E N LaboratorY of Gerontology, Department t~ Public Health, University of ?'ampere. P. O. Box 607. SF-33101 Tampere (Finland)
(Received January 15th, 1991) (Revision received March 28th, 1991) SUMMARY Neurons of the dorsal root ganglion (DRG) and sympathetic superior cervical ganglion (SCG) were cultured as explants from young adult (3 months old) and aged (28 months old) Wistar rats. Both aged D R G and SCG neurons showed delayed neurite outgrowth compared to young adult neurons. Young and some aged cultured neurons had an ultrastructure similar to that found in normal uncultured cells, but most of the aged cultured neurons displayed a heavy accumulation of homogenous lipid-like inclusions in addition to classic age pigments. Occasionally, large neurofilament aggregates were seen in aged D R G neurons. They were sometimes localized perinuclearly, resembling neurofibrillary tangles. The results show that even very old peripheral neurons survive in culture. As aged cultured neurons show degenerative changes not observed in young adult neurons, it is suggested that cultured peripheral neurons of different ages could provide useful means for neuronal aging studies.
Key words: Aging; Lipofuscin; Peripheral neuron; In culture; Ultrastructure
INTRODUCTION Neurons from the peripheral and central nervous system of embryonic and perinatal mammals have been used extensively in culture studies because of the success in dissociating and maintaining undifferentiated and immature neurons alive in vitro [1-3]. Even though these cells can be maintained in culture for several months, Correspondence to: Jari Koistinaho, Laboratoryof Gerontology,Departmentof Public Health, University of Tampere, P.O. Box 607, SF-33101 Tampere, Finland.
0047-6374/91/$03.50 Printed and Published in Ireland
© 1991 ElsevierScientific Publishers Ireland Ltd.
24 it is unusual to find any structure that could be clearly associated with aging changes in neurons, such as lipofuscin granules, neuroaxonal dystrophy or neurofibrillary degeneration [2,4,5]. Over the last 15 years there has been an increasing number of studies on cultured adult peripheral ganglion cells [2,4-9]. The advantage of using adult neurons in culture is that they better represent mature cells normally present in vivo, enabling more relevant experiments with a reflection of functions in the adult nervous system. In addition, the survival of differentiated neurons may be better than that of undifferentiated neuroblasts [5-7]. Recently, there have been some reports on successful short-term cultivation of aged peripheral neurons [9,10] but no histological studies on cultured senescent peripheral neurons have been reported. In the present study, we describe some age-related morphological features of dorsal root ganglion (DRG) cells and sympathetic neurons cultured from 28-month-old rats. MATERIALS AND METHODS Animals and tissue culture A total of eight male Wistar rats aged three (N = 4) and 28 months (N = 4), were used. They were anesthetized by ether inhalation and decapitated. Superior cervical ganglia (SCG) and dorsal root ganglia (DRG) were dissected out, desheathed and placed in plastic petri dishes containing Ca-Mg-free Dulbecco's phosphate buffer medium (GIBCO, Grand Island, NY). The ganglia were sliced using a sterilized razor blade and the ganglion tissues were placed in separate plastic culture dishes (NUNC, Denmark) coated with poly-L-lysine (Sigma Chemical Co., St. Louis, MO) and then covered by glass cover slides. The ganglion explants were plated in Eagle's minimum essential medium (MEM) (GIBCO) with 6 g/I glucose, 10% (v/v) horse serum (GIBCO), 5% fetal calf serum (GIBCO), 1% glutamine (GIBCO) 30 nM Nerve growth factor (Sigma) and 0.1% gentamycin. The medium was changed twice a week using the same medium with 5'¼, horse serum but without fetal calf serum. The cultures were maintained at +37°C in humidified 10% CO~-90% air (v/v) and viewed daily using an Olympus CK2 phase-contrast microscope. Fluorescence microscopy At 21 days, four SCG cultures were treated with 10 -'~ M noradrenaline bitartrate (Sigma) for 20 min. Then all cultures were rinsed in phosphate buffered saline. Half of the cultures were rapidly dried with a hair dryer and exposed to paraformaldehyde vapor at + 60°C for 60 min and embedded in xylene. The fluorescence properties of lipopigments and noradrenaline were then examined from these cultures using a Vanox T fluorescence microscope. Some cultures were also stained with toluidine blue and studied using the same microscope with visible light. Electron microscopy The other half of the cultures were fixed with 3% glutaraldehyde in buffer at room
25 temperature for 30 min, postfixed in osmium tetroxide, dehydrated and embedded in Epon. Thin sections were stained with lead citrate-uranyl acetate and examined using Jeol J E M 2000 electron microscope. RESULTS The neurons grew neurites from 3-month-old ganglia already at day 3, whereas outgrowths from 28-month-old rats were seen at day 6 - 7 (Fig. l a - d ) . All explants also grew out fibroblasts and Schwann cells, and occasional migrated neurons were seen outside the explants (Fig. l e - f ) . After 14 days, a n u m b e r o f enlarged neurons containing granular material were seen in phase-contrast microscopy in S C G and D R G explants, especially in those
Fig. I. Phase contrast (a-d,f) and fluorescence (e) micrographs of living young adult and aged peripheral neurons in culture. Nerve fiber outgrowths from a 28-month-old DRG explant maintained for 6 (a), 8 (b) and 10 (c) days in culture. For comparison nerve fiber outgrowths from 3-month-old DRG explant maintained for 4 days is shown in (d). (e): Heavy accumulations of autofluorescent lipopigments (arrows) are seen in living 21-day-old cultures of aged sympathetic ganglion expLant. (f): Migrated healthy-looking young sympathetic neurons and nonneuronal cells at 21-days in culture, x 500.
26
from old animals. These cells remained attached and had neurite-like extensions for several days, although they gradually vacuolized and disappeared later on. At 21 days, cultures from younger animals also had some vacuolized neurons. Vacuolated fibroblasts and Schwann cells were seen in cultures from aged animals, but not in those from young adult rats. In living cultures, fluorescence microscopy revealed a number of neurons containing autofluorescent lipopigments throughout the cytoplasm (Fig. 2a-b). This feature was more frequently observed in cultures from aged animals. Heavily pigmented large cells, presumably neurons, were observed to have migrated out of the explants also in young adult tissues. With further aging of the cultures, autofluorescent material accumulated even in nonneuronai cells, e.g. in fibroblasts and Schwann cells (Fig. 2c). In these cells, the autofluorescent material was usually located perinuclearly. The sympathetic neurons treated with noradrenaline showed strong catecholamine-specific fluorescence both in the somata and dendrites as well as in axon varicosities (Fig. 3a,b). Non-neuronal cells displayed catecholamine fluorescence which was much less than in neurons (Fig. 3b). Toluidine blue staining revealed several small groups of neurons without evident growth of processes (Fig. 4a-c). However, these cells were present throughout the time that the cultures were maintained. Sympathetic neurons, in particular, formed small pseudoganglia even in cultures of aged neurons (Fig. 4b), whereas D R G neurons were usually separately attached to non-neuronal cells (4a). In electron microscopy, both neurons of normal in vivo fine structure and neurons showing several pathological alterations were seen. The neurons classified as normal had a round and large nucleus, intact mitocondria and Goigi complex, regularly arranged neurofilaments and endoplasmic reticulum, and a varying number of lysosomes or lipofuscin granules (not shown). In neurons of large regions of several explants as well as in many dissociated neurons, nuclei showing a considerable invagination of cytoplasma were seen (Fig. 5a). These cells frequently contained a cytoplasmic accumulation of large lipid-like vacuoles in addition to clumps of classic lipofuscin (Fig. 6a,b). The surrounding nonneuronal cells also contained very large homogenous vacuoles (Fig 613. In aged D R G neurons, very heterogeneous inclusion bodies were sometimes observed (6c-e). They were composed of a round lipid-like component, a more electron-dense matrix resembling that of the classic pigment, and an array of thin filament-like strings (Fig. 6c-e). The size of these inclusions (1.5-2.5 #m) was greater than that of the classic lipofuscin (0.4-1.0 ~m). Finally, about 25% Figs. 2-4. Figs. 2a-c show autofluorescence micrographs of lipopigment accumulations in sympathetic neurons and fibroblasts (asterisks) which have migrated from aged sympathetic ganglion explants at 20 days in culture. Figs. 3a,b show catecholamine fluorescent sympathetic neurons (brightly fluorescent cells1 after noradrenaline incubation. Also non-neuronal cells are slightly fluorescent. Figs. 4 a - c show toluidine blue stained aged DRG (4a) and sympathetic (4b,c) ganglion cells cultured for I 0 days. Most of them have not grown neurites. (2a;3b): x 130; (2b,c: 3a;4a-c): x 500.
28
Fig. 5. ( a - d ) Electron micrographs of an aged D R G neuron cultured for 21 days. An accumulation of homogenous inclusion bodies (arrow heads), large vacuoles (open arrows), a cytoplasmic invagination (star) into the nucleus and a perinuclear aggregation of neurofilaments (black arrows) are the pathological changes observed in this cell. (5a): x 8000; (5b,c): x 20 000; (5c): × 160 000.
29
Fig. 6. (a-f) Electron micrographs from an aged sympathetic ganglion cxplant (a.b) and D R G explant (c-f) maintained in culture for 21 days. Pigment inclusions in sympathetic neurons are large and homogenous, whereas in D R G neurons heterogenenous pigment bodies are also seen. In (61") a Schwann cell containing very large inclusion bodies, between of which an intact mitochondrion can still be seen. (6a): x8000; (6b): x 4 0 000; (6c): ×30 000: (6d): ×60 000: (6e): x 5 0 000: (61"): x 16 000.
30 of the aged D R G neurons showed neurofilament aggregates, which were sometimes localized around the nucleus (Fig. 5c,d). This alteration was not seen in sympathetic neurons. DISCUSSION In the present study, no specific neuronal or gliai cell stainings were used to distinguish between neurons and non-neuronal cells in culture. Unlike flat nonneuronal cells, D R G neurons and sympathetic ganglion cells are characterized by large and round somata. Proof of neuronal origin of most of the neuron-like cells was provided by the neuronal processes. Furthermore, the catecholamine-specific fluorescence that was observed after incubating sympathetic neurons with noradrenaline as well as the fine structure typical of peripheral ganglion cells ascertain that our cultures contained detectable and living neurons even in preparations of aged rat tissues. Ganglion explants from aged animals grew out neurites later than the explants from young adult rats, which is in agreement with earlier studies [2,5,7]. This indicates an impaired ability of aged peripheral neurons to regenerate after axotomy. Enhanced degeneration processes after axotomy may contribute to the late response of regeneration of aged neurons, since pathological structural changes were more pronounced and more rapid in aged ganglion explants. In spite of this, the survival of aged peripheral neurons was comparable to that of young adult neurons. The accumulation of autofluorescent pigment bodies like lipofuscin and ceroid pigments in postmitotic cells including neurons is thought to be an indication of agerelated degeneration in which free radicals and lipid peroxidation are involved [11-14]. Even though there is a body of evidence suggesting that lipid peroxidation and free radical production increase in ratio to free radical scavengers during aging [14], there is as yet no direct evidence that the attack of these radicals on cellular lipids and proteins are the cause of the pigment formation. Instead, recent studies on the experimental production of lipofuscin point to the role of impaired protease function in a rapid pigment accumulation [t5,16], whereas a longer period of time for the lack of essential antioxidants is needed to lead to pigment formation [13,17]. In the present study, autofluorescent pigment bodies occupied the cytoplasm of ganglion cells in a few days, particularly in aged neurons. These granules were either homogenous lipid-like large inclusions or composed of heterogenous material, differing thus from the classic lipofuscin in structure. Similar type of pigment granules are found in neurons after treatment with leupetin, a protease inhibitor [15,16], suggesting that the altered function of proteases rather than increased lipid peroxidation may be responsible for the pigment formation. Some of the cultured aged D R G neurons contained neurofilament aggregates, which sometimes were localized perinuclearly, thus resembling neurofibrillary tangles [6,18,19]. In humans, similar morphological changes are present in various
31
neurological disorders like Alzheimer's disease, and have been found to a lesser extent in the brains of healthy individuals [6,18]. In the rat, neuro-fibrillary changes in neurons are not normally found, but after treatment with aluminium lactate or inhibitors of mitotic spindle (e.g. colchicine, vincristine, vinblastine) striking cytoplasmic masses of neurofilaments can be seen [18,20-22]. The present results show that aged DRG neurons are more prone to the stress caused by axotomy and the culture procedure than young adult DRG neurons, and that under these circumstances they react by forming neurofibrillary aggregates typical of neurotoxintreated rodent neurons. In conclusion, both aged DRG and sympathetic neurons survive in culture, although regeneration after axotomy is delayed compared with neurons from young adult rats. Aged neurons in culture displayed pathological changes such as heavy accumulation of autofluorescent inclusions and neurofilament aggregates, which are absent from normal peripheral neurons in vivo, probably due to age-related impairment of regeneration ability after axotomy and to the stress induced by the culture procedure. The culture of neurons of different ages could provide useful means for studying neuronal aging. ACKNOWLEDGEM ENTS
The study was financially supported by the Finnish Cultural Foundation of Pirkanmaa. REFERENCES 1 2 3 4 5
S.M. Crain, Neurophysiologic studies in Tissue Culture, Raven Press, 1976. S.U. Kim, Neuronal ageing in tissue and cell cultures: a review. In Vitro. 19 (1983) 73-82. M. Murray, Nervous tissue in vitro. In E.N. Willner (ed.), Cells and Tissues in Culture, Vol. II, Academic Press, New York, 1965, pp. 373-455. S.U. Kim, K.G. Warren and M. Kalia, Tissue culture of adult human neurons. Neurosci. Lett., II (1979) 137-141.
B.S. Scott, Adult neurons in cell culture: electrophysiological characterization and use in neurobiological research. Prog. NeurobioZ. 19 (1982) 187-211. 6 J. Dodd, D. Solter and T,M. Jessel, Monoclonal antibodies against carbohydrate diffrentation antigens identify subsets of primary sensory neurones. Nature. 279 (1984) 469-472. 7 B.S. Scott, Adult mouse dorsal root ganglia neurons in cell culture. J. NeurohioL. 8 ( 1977~ 417-427. 8 R.A. Smith amd I.B. Mclnnes, Phase contrast and electron microscopical observations of adult dorsal root ganglion cells maintained in primary culture. J. Anat.. 145 (1986t 1-12, 9 Y. Uchita and M. Tomonaga, Effect of nerve growth factor and heart cell conditioned medium on neurite regeneration of aged sympathetic neurons in culture. Brain Res.. 348 (1985) 100-106. 10 D.H. Silberberg and S.U. Kim, Studies of aging in cultured nervous system. Intern. Rev. C:vtol., Suppl. 10 (1979) 117-130. 11 K.R. Brizzee, J.M. Ordy and B. Kaack, Early appearance and regional differences in intraneuronal and extraneuronal lipofuscin accumulation with age in the brain of nonhuman primate (Macaca mulatta). J. Gerontol., 29 (1974) 366-381. 12 J. Koistinaho. Difference in the age-related accumulation of lipopigments in the adrenergic and nonadrenergic peripheral neurons in the male rat. Gerontology, 32 (1986) 300-307.
32 13 14 15 16 17
18 19
20 21 22
J. Koistinaho, H. A l h o a n d A. Hervonen, Effect of vitamin E and selenium supplement on the aging peripheral neurons of the male Sprague-Dawley rats. Mech. Ageing Dev.. 51 (1990) 63-72. A.L. Tappel, Lipid peroxidation and fluorescent molecular damage to membranes. In B. Trumps and A. Arstila (eds.), Pathology ~[ Cell Membranes, Academic Press, New York, 1975, pp. 145-170. G.O. Ivy, J. Wenzel. M. Baydry and G. Lynch, Inhibitors of lysosomal enzymes: accumulation of lipofuscin-like dense bodies in the brain. Science. 226 (1984) 985-988. M. Katz and M.J. Shankler, Development of lipofuscin-like fluorescence in the retinal epithelium in response to protease inhibitor treatment. Mech. Ageing Dev.. 49 (1989) 23-40. M,L. Katz. W.G. Robinson, R.L. Herrman, A.B. Groome and J.G. Bieri, Lipofuscin accumulation resulting from senescence and vitamin E deficiency: spectral properites and tissue distribution. Mech. Ageing Dev., 25 (1984) 149-159. R.D. Terry, The fine structure of neurofibrillary tangles in Alzheimer's disease. J. Neuroputhol. EYp. Neurol., 22 (1963) 629. H. Matsuyama and S. Nakamura, Senile changes in the brains in the Japanesc: Incidence of Alzheimer's neurofibrillary change and senile plaques. In R. Katzman, R. Terry and K. Bick (eds.). Alzheimer's Disease, Senile Dementia and Related Disorders. New York, Raven Press, 1976, pp. 183-204, M.P. Daniels, Fine structural changes in neurons and nerve fibers associated with colchicine inhibition of nerve fiber formation in vitro. J. Cell Biol., 58 (1973) 463-470. B. Ghetti, Induction of neurofibrillary degeneration following treatment with maytansine in vivo. Brain Res., 163 (1979) 9-19. I. Klatzo, H. Wisnieski and E. Streicker, Experimental production of neurofibrillary degeneration I. Light microscopic observations. J. Neuropathol. Exp. Neurol.. 24 (1965) 187-199.
