SYNAPSE 8:162-168 (1991)
Ultrastructure of Aberrant SerotoninImmunoreactive Fibers in the Caudate Putamen Complex of the Aged Rat MARIELLE G.P.A. VAN LUIJTELAAR, FLORIS G. WOUTERLOOD, JEROEN A.D.M. TONNAER, AND HARRY W.M. STEINBUSCH Departments of Pharmacology (M.G.P.A.V.L., H. W.M.S.) and of Anatomy (F.G. W.), Faculty of Medicine, Free University, Amsterdam, and Department of CNS Pharmacology, Organon Int BV, Oss (J.A.D.M.T.), The Netherlands
KEY WORDS
Agng, Degeneration, Electron microscopy, Immunocytochemistry
ABSTRACT Degeneration of neurons in the central nervous system is associated with morphological changes. Previous observations made at the light microscopical level indicated degeneration of serotonin-immunoreactive (IR) fibers in the aged rat brain. In this study, a comparison at the ultrastructural level was made between serotonin-IR normal thin and aberrant swollen varicose fibers in the caudate-putamen complex of the aged rat. Ultrastructural features such as the size and content of the thin varicose fibers resembled those in the caudate-putamen complex of the young rat as reported by others. The aberrant profiles were swollen, reaching a size of 6 pm. Their vesicles varied in size and were no longer uniformly round. Moreover, distorted mitochondria and membranefilled vacuolelike structures were a common feature of the aberrant profiles. These changes are indicative of a degenerative process and give further evidence that, whereas many serotonergic fibers are preserved at high age, other serotonergic fibers are degenerating in the caudate-putamen complex of the aged rat. INTRODUCTION During the process of agmg there is a change of morphological characteristics of neurons in the central nervous system (CNS). At the light microscopical level, a loss of cell density, shrinkage of neurons, and a deposition of pigment bodies such as lipofuscin can be observed. Changes in axon terminals have been described in the catecholaminergic system in the brain of aged mice (Masuoka et al., 1979) and in the cholinergic system of the senescent rat (Armstrong et al., 1988).At the ultrastructural level, aging neurons show a decrease of cellular and nuclear diameter, a decrease of ribosomes and Nissl material, an alteration of the internal membrane system (Golgi complex and endoplasmic reticulum), abnormal mitochondria that sometimes contain inclusions, areas of degeneration consisting of granules surrounded by vacuoles or tangles of neurofibrils (for reviews see Johnson, 1985; Rogers and Styron, 1987). Recently, we reported on the occurrence of morphologically aberrant serotonin-immunoreactive (IR) fibers in the forebrain of aged rats (Van Luijtelaar et al., 1988). These fibers show swollen varicosities and swollen intervaricose connections when compared with nonaffected fibers in the aged or normal fibers in the @3 1991 WILEY-LISS, INC.
young adult brain. The aberrant fibers are often located in clusterlike structures. In constrast to the cholinergic and the catecholaminergic fibers, the serotonin-IR fibers in the aged brain are frequently affected. An overview of the occurrence of aberrant serotonin-IR fibers appeared in a previous report (Van Luijtelaar et al., 19911. The similarity at the light microscopical level between the morphology of aberrant serotonergic fibers and neurotoxin-induced degenerating serotonerac fibers indicates the degenerative character of the aged aberrant fibers (Van Luijtelaar et al., 1989). At the ultrastructural level, lesion-induced degenerating serotonergic fibers have been shown to contain disrupted mitochondria and enlarged vesicles (Aghajanian et al., 1969; Mallggrd et al., 1978; Rapisardi et al., 1990) and similar features of degeneration have been described in other neurotransmitter systems (Mugnaini and Friedrich Jr., 1981; Wisniewski et al., 1972). The aim of the present study was to further characterize the aber-
Received November 9,1990; accepted in revised form December 31,1990. Address reprint requests to H.W.M. Steinbusch, Dept. of Pharmacolo Faculty of Medicine, Free University, van der Boechorststraat 7, 1081 BT, E s t e r dam, The Netherlands.
ULTRASTRUCTURE OF ABERRANT SEROTONIN-IR FIBERS
rant fibers in the brains of aged rats. As an example, swollen varicose fibers in the caudate-putamen complex were compared with normal thin serotonin-IR varicose fibers. MATERIALS AND METHODS Five aged male rats (28 months and older, Wistar, Harlan CPB, Zeist, The Netherlands) were deeply anaesthetized with sodium pentobarbital (Nembutal, 60 mgkg body weight). They were perfused transcardially with Tyrode’s buffer (approximately 1 minute), and subsequently with 450 ml fixative consisting of 4% paraformaldehyde, 1%glutaraldehyde, and 0.2% picric acid in 0.1 M phosphate buffer (pH 7.4). The brains were dissected and postfixed for 2 hours in the same solution with omission of the glutaraldehyde. Transverse 50-pm sections containing the caudate-putamen complex were cut on a Vibratome, collected in Tris-buffered saline (TBS, pH 7.6), and treated for 30 minutes with 1% NaBH, in TBS. The sections obtained from each brain were divided in two sets. One set was stained according to standard procedure (Steinbusch and Tilders, 1987) and used for light microscopical investigation. For this set, the incubation and rinsing solutions consisted of 0.5%Triton-X 100 containing TBS (TBS-T).The second set of sections was incubated and rinsed without Triton-X 100. Immunostaining with an antibody directed against serotonin was performed according to an unlabeled peroxidaseantiperoxidase method described elsewhere (Steinbusch and Tilders, 1987). Sections from the second set were subsequently processed for electron microscopy, i.e. rinsed two times with 0,l M cacodylate buffer (pH 7.4) and postfixed in 1%osmiumtetroxide in 0.1 M cacodylate buffer, pH 7.4 (60 min, 4°C). They were then stained in 2% aqueous uranyl acetate (60 min, PC), dehydrated in ethanol, and via propylene oxide embedded in Epon-Araldite (Ciba Geigy, Basel, Switzerland) on glass slides between polyethylene foils. After the curing of the epoxy resin, small areas of the caudateputamen complex were selected for the presence or absence of aberrant (swollen, varicose) serotonin-IR fibers. These areas were dissected and remounted with a drop of fresh resin on the flat tops of pre-cured resin blocks. After one night of extra curing, serial ultrathin sections were cut, mounted on formvar-coated slot grids, contrasted with lead-citrate, and examined in a Philips EM 301 electron microscope. RESULTS In pieces of the caudate-putamen complex that did not contain aberrant fibers, small serotonin-IR profiles were observed with a diameter up to 1pm (Fig. 1).These profiles were surrounded by elements that are common in the caudate-putamen complex: perikarya, dendritic profiles, and spines, myelinated and unmyelinated axons, axon terminals, glial processes, glial perivascular
163
endfeet (when the serotonin-IR profiles occurred next to capillaries), and thin unidentified processes. We observed no appositions of membranes or membrane specializations. Infrequently we saw small serotonin-IR, vesicle-loaded profiles forming an asymmetrical synaps with a non-IR profile. The serotonin-IR profiles contained, next to the electron dense immunoreaction product, mitochondria, and vesicles; the intervaricose fibers also contained microtubuli. The electron-dense, fuzzy immunoreaction product had a cytoplasmic location. It appeared preferentially bound to the outer membranes of vesicles and mitochondria. The synaptic vesicles in the serotonin-IR profiles were mostly spherical, although we sometimes observed a pleomorphic vesicle. Their diameter ranged from 20 to 70 nm, comparable t o the diameter of vesicles in axon terminals in the surrounding neuropil. The mitochondria in serotonin-IR fibers similarly resembled those in the adjacent neuropil. The mitochondria1 matrix was of medium electron density and the cristae were well preserved in both IR and non-IR profiles (Fig. 1A-D), though occasionally mitochondria with a more electron lucent content and lacking the normal arrangement of cristae were observed in the serotonin-IR profiles (Fig. 1D). In the caudate-putamen complex from the aged rat, other areas were selected for the presence of aberrant, swollen serotonin-IR fibers. In these areas, the serotonin-IR profiles showed a much wider range of diameters. Occasional small profiles, comparable to those in the nonaffected tissue blocks, were observed (Fig. 2). However, there were also large serotonin-IR profiles ranging in size from 2-6 pm. The swollen varicosities had smooth (Fig. 2) or undulating (see Fig. 4) outlines. They were interconnected by relatively thick and sometimes swollen fibers. In one case, an apposition was seen of an aberrant fiber with a dendritic profile including a puncturn udhuerens (desmosome).Otherwise, no membrane specializations (such as synapselike appositions) between swollen varicosities and adjacent, non-IR structures were observed. Occasionally, the serotoninIR aberrant fibers apposed perikarya of striatal neurons and glial cells. Specific appositions of glial cells with aberrant fibers were not apparent. The large serotonin-IR profiles contained vesicles, mitochondria, and vacuolelike structures. The swollen varicosities were packed with electron-lucent vesicles of varying size. The shape of these vesicles was spherical in most cases, the majority being approximately 50% larger than synaptic vesicles in axon terminals surrounding the IR profile or in small serotonin-IR profiles. Infrequently larger vesicles were observed. The largest vesicle that we observed, was 180 nm in diameter. The morphology of the mitochondria in the swollen varicosities often differed from those in the adjacent neuropil or in small serotonin-IR profiles. The number of cristae was reduced as compared with the mitochondria in the surrounding IR and non-IR varicosities. Some mito-
164
M.G.P.A. VAN LUIJTELAAR ET AL.
errant serotonin-IR terminals in the caudate-putamen complex of aged rats. These aberrant fibers were examined only in the caudate-putamen complex, since at the light microscopical level, their morphology did not differ from aberrant fibers occurring in other brain areas. Small serotonin-IR profiles appeared to be vesicleloaded varicosities or intervaricose fibers containing mitochondria and only infrequently forming a synapse type of apposition with an adjacent profile. Thus the size of the profiles, the shape of the vesicles, and the presence of a minority of large granular vesicles reported in this study fit within the generally accepted description of serotonergic axon terminals based on immunocyDISCUSSION tochemical or autoradiographic studies in the young rat Previous light microscopical observations on the mor- (Arluison and De la Manche, 1980; Beaudet and Descarphology of aberrant serotonergic fibers in the brains of ries, 1987; Chazal and Ma, 1989; Soghomonian et al., aged rats indicated the degenerative character of these 1989; Maley et al., 19901, cat (Calas et al., 1976), or fibers (Van Luijtelaar et al., 1988, 1989, 1991). In the monkey (Roberts Jr. et al., 1990)brain. The immunoreaction product was localized on the present work, this suggestion was further evaluated by an ultrastructural examination of aberrant and nonab- outer membranes of vesicles and mitochondria. This is
chondria showed distorted cristae in an electron lucent, swollen matrix. However, mitochondria with a normal appearance were also encountered in the large IR profiles. In many aberrant fibers membrane-bound, small vacuoles of the size of mitochondria, containing membrane whorls or fragments of membranes were observed (Fig. 3). In addition to these small, membrane-filled vacuoles, several clusters contained a large to very large membrane-bound, electron-lucent compartment limited by a single membrane. In most cases these large vacuoles contained irregularly shaped membranewhirls, or packs of fragments of membranes.
Fig. 1. Examples of nonaffected serotonin-IR profiles in the caudate-putamen complex of an aged rat. Serotonin-IR fiber in close contact with a cell body (A), varicosities apposing a dendritic shaft (B), myelinated axons ( C ) ,or unidentified profiles (D). Note the occurrence of a distorted mitochondrium in a nonimmunoreactive profile (arrow in D). Bar = 1 pm. Magnification in B-D is the same as in A.
ULTRASTRUCTURE OF ABERRANT SEROTONIN-IR FIBERS
Fig. 2. Photomontage (A) showing swollen serotonin-IR profiles in the caudate-putamen complex of an aged rat. Note that small serotonin-IR profiles are also present (arrows).Bar = 1pm. The inset (B) shows aberrant serotonin-IR fibers among thin varicose fibers at the light microscopic level. Bar = 50 pm.
165
166
M.G.P.A. VAN LUIJTELAAR ET AL.
Fig. 3. Detail of a swollen serotonin-IR varicosity. Both healthy (arrowhead) and distorted (arrow) mitochondria and a small membrane-filled vacuole (double arrow) are present. The shape and the size of the vesicles is a t variance with those in the adjacent neuropil. Bar = 1 pm.
ULTRASTRUCTURE OF ABERRANT SEROTOIL’IN-IRFIBERS
167
Fig. 4. Photomontage of a swollen serotonin-IRprofile in the caudate-putamen complex of an aged rat, showing abnormal mitochondria: The mitochondriumindicated by an arrow has an electron lucent matrix and has lost its internal membrane organization. The open asterisk indicates a mitochondrium that is swollenas well. Small arrows indicate swollen vesicles.Theundulating outline of the profile is indicated by arrowheads or, when unclear, by filled asterisks. Bar = 1 wm.
comparable to the localization of the immunoreaction product in studies on other neurotransmitters (Voorn and Buys, 1987). The parallellism between young and non-swollen aged serotonergic fibers indicates that during aging many fibers are preserved. The swollen serotonin-IR profiles in the aged rat caudate-putamen complex differed from the small profiles in that they showed distorted and enlarged mitochondria, membrane-filled vacuolelike structures, and many enlarged vesicles. A tenfold increase of the size of the swollen profiles was no exception. In the swollen profiles the size of the vesicles could be as large as 180 nm and could vary more than in the small serotonin-IR profiles. Moreover, it was often impossible to make a discrimination between swollen vesicles and vacuolelike structures. Similar alterations in the structure of mitochondria and size and number of vesicles have been used as the major indicators for axonal degeneration, either age-induced (Miquel et al., 1983) or lesion-induced (Aghajanian et al., 1969; Wisniewski et al., 1972; Mugnaini and Friedrich Jr., 1981; Rapisardi et al., 1990). The morphology of the swollen serotonin-IR profiles thus resembles that of degenerating axons.
It has been suggested that mitochondria1injury is one of the first events in the process of degeneration (Miquel et al., 1983) and can be regarded as a marker for early stages of degeneration. Indeed, in the present study, distorted mitochondria have been observed in small immunoreactive profiles, and it is tempting to speculate that these profiles represent a stage of degeneration occurring before the swelling of the varicosities. However, mitochondria are very vulnerable toward osmotic shocks and care should be taken that artifacts of tissue preparation are not interpreted as degeneration of the tissue. The presence of both healthy and distorted mitochondria in immunoreactive profiles argues against a deleterious effect of the immunoreaction or the immunoreaction product on the structure of the mitochondria. Similarly, the presence of healthy and distorted mitochondria in nonimmunoreactive profiles indicates that processing for electron microscopy did not interfere with the ultrastructure of mitochondria. Conversely, the presence of distorted mitochondria in nonimmunoreactive fibers points toward the degeneration of other systems in the caudate-putamen complex of the aged rat, as can be expected in an aged animal.
168
M.G.P.A. VAN LUIJTELAAR ET A t .
To our knowledge, only one previous report was published on the ultrastructure of serotonergic fibers in the brain of aged rats (Calas and Van Den Bosch de Aguilar, 1980).In that radioautographic study, no differences in the pattern and intensity of the label were observed in the periventricular part of the caudate-putamen complex, though granulo-vacuolar degeneration was reported to occur in the raphe region. In the caudateputamen complex of the aged rat, many normal, unaffected serotonergic fibers occur, apart from the aberrant fibers (Van Luijtelaar et al., 1988, 1991), and a randomly selected area of the caudate-putamen complex does not necessarily include aberrant fibers. This might explain the failure of these authors to find any agerelated degeneration in the serotonergic system. Also, radioautographic labelling depends on the uptake mechanism in the serotonergic terminals. A decreased uptake capacity of the aberrant fibers would hamper the labelling of these fibers and this possibility cannot yet be excluded. Pilot studies in our own lab (unpublished data) gave no indication of a reduced uptake capacity in the caudate-putamen complex of the aged rat. However, the aberrant fibers constitute a subpopulation of the total serotonergic innervation of the caudate-putamen complex and the lack of an overall change cannot exclude a decrease uptake capacity of the aberrant fibers. As the present EM observations indicate that small profiles exist with degenerating mitochondria, the absence of aberrant fibers a t the LM level does not implicate a healthy, nondegenerating neurotransmitter system. In contrast, a t the EM level, all swollen varicosities were shown to contain distorted mitochondria, swollen vesicles, and vacuolelike structures. Thus degeneration of a neurotransmitter system is strongly suggested at the LM level by the presence of crumpled and folded fibers with swollen varicosities. ACKNOWLEDGMENTS The authors thank Anaatje Pattisalanno, Ineke Aarts, and Alice Rinkens for their technical assistance and Shimon Paniry for preparing the microphotographs. This work was supported by grant no. 900552-072 from the Netherlands Foundation for Medical Research (MEDIGON, NWO). REFERENCES Aghajanian, G.K., Bloom, F.E., and Sheard, M.H. (1969) Electron microscopy of degeneration within the serotoninergic pathway of rat brain. Brain Res., 13:26f%273. Arluison, M., and De la Manche, IS. (1980)High resolution autoradiographic study of the serotoninergic innervation of the rat cor us striatum after intraventricular administration of [3H]-5%ydroxytryptamine. Neuroscience, 5:229-240. Armstrong, D.M., Hersch, L.B., and Gage, F.H. (1988) Morphologic alterations of cholinergic processes in the neocortex of aged rats. Neurobiol Aging, 9:199-205.
Beaudet, A,, and Descarries, L. (1987)Ultrastructural identification of serotonin neurons. In: Monoaminergc Neurons: Light Microscopy and Ultrastructure. IBRO Handbook Series: Methods in Neurosciences, vol. 10. H.W.M. Steinbusch, ed. John Wiley & Sons, ChiChester, pp. 265-313. Calas, A., and Van Den Bosch de Aguilar, P. (1980) Comparative radioautogra hic study of serotoner 'c neurons in young and senescent rats. The psychobiology opagng: problems and perspectives. D.G. Stein, ed. Elsevier Amsterdam, pp. 59-80. Calas, A,, Besson, M.J., Gauchy, C., Alonso, G., Glowinsky, J., and Cheramy, A. (1976) Radioautography study of in vivo incorporation of [3H]-monoamines in the cat caudate nucleus: identification of serotoninergic fibers. Brain Res., 118:l-13. Chazal, G., and Ma, W. (1989)An ultrastructural analysis of serotoninergic neurons in the nucleus raphe magnus of the rat. Neuroscience, 33:301-310. Johnson! R.J. (1985)Anatom of the a g n g nerve cell. In: Handbook of Cell Biology ofAging. V.J. Jristafalo, ed. CRC Press, Boca Raton, pp. 149-178. Maley, B.E., Gail Engle, M., Humpreys, S., et al. (1990) Monoamine synaptic structure and localization in the central nervous system. J. Electron Microsc. Techn., 15:20-33. Masuoka, D.T., Jonsson, G., and Finch, C.E. (1979)Agingandunusual catecholamine-containing structures in the mouse brain. Brain Res., 16913355341. M.& i uel, J ' ,Johnson, . J.E., and Cervos-Navarro, J . (1983)Comparisonof S a g n g in humans and experimental animals. Aging, 21:231258. M@llg5rd, K., Lundberg, J.J., Wiklund, L., Lachenmayer, L., and Baumgarten, H.G. (1978) Morphologic consequences of serotonin neurotoxin administration: neuron-tar et cell interaction in the rat subcommissural o y a n . &n. N.Y. A c a f Sci., 305:262-288. M u y n i , E:, and riedrich Jr., V.L. (1981) Electron microscopy: i entification and study of normal and degeneratin neural ele ments by electron microsco y In Neuroanatomical &act-Tracing Methods. L. Heimer and hJ.! Robards, eds. Plenum Press, New York, p 377406. Rapisar&,'S.C., Warrington, V.O.P., and Wilson, J.S. (1990) Effect of MPTP on the fine structure of neurons in substantia nigra of dogs. Brain Res., 512:147-154. Roberts J r . , E.L., Rosenthal, M., and Sick, T.J. (1990) Age-related modifications of potassium homeostasis and synaptic transmission during and after anoxia in rat hippocampal slices. Brain Res., 514:111-118. Rogers, J.,and Styren, S.D. (1987) Neuroanatomy of agingand dementia. Rev. Biol. Res. Aging, 3:223-253. Soghomonian,J-J., Descarries, L., and Watkins, K.C. (1989)Serotonin innervation in adult rat neostriatum. 11. Ultrastructual features: A radioautographic and immunocytochemical study. Brain Res., 481337-87. Steinbusch, H.W.M., and Tilders, F.J.H. (1987) Immunohistochemical techniques for light-microscopical localization of dopamine, noradrenaline, adrenaline, serotonin and histamine in the central nervous system. In: Monoaminergic Neurons: Li ht Microscopy and Ultrastructure. IBRO Handbook Series: Methojs in Neurosciences, vol. 10. H.W.M. Steinbusch, ed. John Wiley & Sons, Chichester, pp. 125-166. Van Luijtelaar, M.G.P.A., Tonnaer, J.A.D.M., and Steinbusch, H.W.M. (1988)Aberrant morphology of serotonergic fibers in the forebrain of the aged rat. Neurosci. Lett., 95:93-96. Van LuijteIaar, M.G.P.A., Tonnaer, J.A.D.M., and Steinbusch, H.W.M. (1989) Similarities between aberrant serotonerGc fibers in the aged and 5,7-DHT denervated young adult rat brain. Exp. Brain Res., 78:81-89. VanLuijtelaar, M.G.P.A., Tonnaer, J.A.D.M., andsteinbusch, H.W.M. (1991) Aging of the serotonerpic svstem in the rat forebrain. Neurobiol Aging (Cn ress) Voorn, P., and [uys, 'R.M. (1987) Ultrastructural demonstration of dopamine in the central nervous system. In: Monoaminer 'c Neu rons: Light Microscopy and Ultrastructure. IBRO Handboo Series. ' 1 Methods in Neuroscience,vol10. H.W.M. Steinbusch, ed. John Wiley & Sons, Chichester, pp. 241-264. Wisniewski, H.M., Ghetti, B., and Horoupian, D.S. (1972) The fate of s naptic membranes of degenerating optic nerve terminals, and tgeir role in the mechanism of trans-synaptic changes. J. Neurocyto].. 1~297-310.
A:
-
I
f
Ultrastructure of Aberrant SerotoninImmunoreactive Fibers in the Caudate Putamen Complex of the Aged Rat MARIELLE G.P.A. VAN LUIJTELAAR, FLORIS G. WOUTERLOOD, JEROEN A.D.M. TONNAER, AND HARRY W.M. STEINBUSCH Departments of Pharmacology (M.G.P.A.V.L., H. W.M.S.) and of Anatomy (F.G. W.), Faculty of Medicine, Free University, Amsterdam, and Department of CNS Pharmacology, Organon Int BV, Oss (J.A.D.M.T.), The Netherlands
KEY WORDS
Agng, Degeneration, Electron microscopy, Immunocytochemistry
ABSTRACT Degeneration of neurons in the central nervous system is associated with morphological changes. Previous observations made at the light microscopical level indicated degeneration of serotonin-immunoreactive (IR) fibers in the aged rat brain. In this study, a comparison at the ultrastructural level was made between serotonin-IR normal thin and aberrant swollen varicose fibers in the caudate-putamen complex of the aged rat. Ultrastructural features such as the size and content of the thin varicose fibers resembled those in the caudate-putamen complex of the young rat as reported by others. The aberrant profiles were swollen, reaching a size of 6 pm. Their vesicles varied in size and were no longer uniformly round. Moreover, distorted mitochondria and membranefilled vacuolelike structures were a common feature of the aberrant profiles. These changes are indicative of a degenerative process and give further evidence that, whereas many serotonergic fibers are preserved at high age, other serotonergic fibers are degenerating in the caudate-putamen complex of the aged rat. INTRODUCTION During the process of agmg there is a change of morphological characteristics of neurons in the central nervous system (CNS). At the light microscopical level, a loss of cell density, shrinkage of neurons, and a deposition of pigment bodies such as lipofuscin can be observed. Changes in axon terminals have been described in the catecholaminergic system in the brain of aged mice (Masuoka et al., 1979) and in the cholinergic system of the senescent rat (Armstrong et al., 1988).At the ultrastructural level, aging neurons show a decrease of cellular and nuclear diameter, a decrease of ribosomes and Nissl material, an alteration of the internal membrane system (Golgi complex and endoplasmic reticulum), abnormal mitochondria that sometimes contain inclusions, areas of degeneration consisting of granules surrounded by vacuoles or tangles of neurofibrils (for reviews see Johnson, 1985; Rogers and Styron, 1987). Recently, we reported on the occurrence of morphologically aberrant serotonin-immunoreactive (IR) fibers in the forebrain of aged rats (Van Luijtelaar et al., 1988). These fibers show swollen varicosities and swollen intervaricose connections when compared with nonaffected fibers in the aged or normal fibers in the @3 1991 WILEY-LISS, INC.
young adult brain. The aberrant fibers are often located in clusterlike structures. In constrast to the cholinergic and the catecholaminergic fibers, the serotonin-IR fibers in the aged brain are frequently affected. An overview of the occurrence of aberrant serotonin-IR fibers appeared in a previous report (Van Luijtelaar et al., 19911. The similarity at the light microscopical level between the morphology of aberrant serotonergic fibers and neurotoxin-induced degenerating serotonerac fibers indicates the degenerative character of the aged aberrant fibers (Van Luijtelaar et al., 1989). At the ultrastructural level, lesion-induced degenerating serotonergic fibers have been shown to contain disrupted mitochondria and enlarged vesicles (Aghajanian et al., 1969; Mallggrd et al., 1978; Rapisardi et al., 1990) and similar features of degeneration have been described in other neurotransmitter systems (Mugnaini and Friedrich Jr., 1981; Wisniewski et al., 1972). The aim of the present study was to further characterize the aber-
Received November 9,1990; accepted in revised form December 31,1990. Address reprint requests to H.W.M. Steinbusch, Dept. of Pharmacolo Faculty of Medicine, Free University, van der Boechorststraat 7, 1081 BT, E s t e r dam, The Netherlands.
ULTRASTRUCTURE OF ABERRANT SEROTONIN-IR FIBERS
rant fibers in the brains of aged rats. As an example, swollen varicose fibers in the caudate-putamen complex were compared with normal thin serotonin-IR varicose fibers. MATERIALS AND METHODS Five aged male rats (28 months and older, Wistar, Harlan CPB, Zeist, The Netherlands) were deeply anaesthetized with sodium pentobarbital (Nembutal, 60 mgkg body weight). They were perfused transcardially with Tyrode’s buffer (approximately 1 minute), and subsequently with 450 ml fixative consisting of 4% paraformaldehyde, 1%glutaraldehyde, and 0.2% picric acid in 0.1 M phosphate buffer (pH 7.4). The brains were dissected and postfixed for 2 hours in the same solution with omission of the glutaraldehyde. Transverse 50-pm sections containing the caudate-putamen complex were cut on a Vibratome, collected in Tris-buffered saline (TBS, pH 7.6), and treated for 30 minutes with 1% NaBH, in TBS. The sections obtained from each brain were divided in two sets. One set was stained according to standard procedure (Steinbusch and Tilders, 1987) and used for light microscopical investigation. For this set, the incubation and rinsing solutions consisted of 0.5%Triton-X 100 containing TBS (TBS-T).The second set of sections was incubated and rinsed without Triton-X 100. Immunostaining with an antibody directed against serotonin was performed according to an unlabeled peroxidaseantiperoxidase method described elsewhere (Steinbusch and Tilders, 1987). Sections from the second set were subsequently processed for electron microscopy, i.e. rinsed two times with 0,l M cacodylate buffer (pH 7.4) and postfixed in 1%osmiumtetroxide in 0.1 M cacodylate buffer, pH 7.4 (60 min, 4°C). They were then stained in 2% aqueous uranyl acetate (60 min, PC), dehydrated in ethanol, and via propylene oxide embedded in Epon-Araldite (Ciba Geigy, Basel, Switzerland) on glass slides between polyethylene foils. After the curing of the epoxy resin, small areas of the caudateputamen complex were selected for the presence or absence of aberrant (swollen, varicose) serotonin-IR fibers. These areas were dissected and remounted with a drop of fresh resin on the flat tops of pre-cured resin blocks. After one night of extra curing, serial ultrathin sections were cut, mounted on formvar-coated slot grids, contrasted with lead-citrate, and examined in a Philips EM 301 electron microscope. RESULTS In pieces of the caudate-putamen complex that did not contain aberrant fibers, small serotonin-IR profiles were observed with a diameter up to 1pm (Fig. 1).These profiles were surrounded by elements that are common in the caudate-putamen complex: perikarya, dendritic profiles, and spines, myelinated and unmyelinated axons, axon terminals, glial processes, glial perivascular
163
endfeet (when the serotonin-IR profiles occurred next to capillaries), and thin unidentified processes. We observed no appositions of membranes or membrane specializations. Infrequently we saw small serotonin-IR, vesicle-loaded profiles forming an asymmetrical synaps with a non-IR profile. The serotonin-IR profiles contained, next to the electron dense immunoreaction product, mitochondria, and vesicles; the intervaricose fibers also contained microtubuli. The electron-dense, fuzzy immunoreaction product had a cytoplasmic location. It appeared preferentially bound to the outer membranes of vesicles and mitochondria. The synaptic vesicles in the serotonin-IR profiles were mostly spherical, although we sometimes observed a pleomorphic vesicle. Their diameter ranged from 20 to 70 nm, comparable t o the diameter of vesicles in axon terminals in the surrounding neuropil. The mitochondria in serotonin-IR fibers similarly resembled those in the adjacent neuropil. The mitochondria1 matrix was of medium electron density and the cristae were well preserved in both IR and non-IR profiles (Fig. 1A-D), though occasionally mitochondria with a more electron lucent content and lacking the normal arrangement of cristae were observed in the serotonin-IR profiles (Fig. 1D). In the caudate-putamen complex from the aged rat, other areas were selected for the presence of aberrant, swollen serotonin-IR fibers. In these areas, the serotonin-IR profiles showed a much wider range of diameters. Occasional small profiles, comparable to those in the nonaffected tissue blocks, were observed (Fig. 2). However, there were also large serotonin-IR profiles ranging in size from 2-6 pm. The swollen varicosities had smooth (Fig. 2) or undulating (see Fig. 4) outlines. They were interconnected by relatively thick and sometimes swollen fibers. In one case, an apposition was seen of an aberrant fiber with a dendritic profile including a puncturn udhuerens (desmosome).Otherwise, no membrane specializations (such as synapselike appositions) between swollen varicosities and adjacent, non-IR structures were observed. Occasionally, the serotoninIR aberrant fibers apposed perikarya of striatal neurons and glial cells. Specific appositions of glial cells with aberrant fibers were not apparent. The large serotonin-IR profiles contained vesicles, mitochondria, and vacuolelike structures. The swollen varicosities were packed with electron-lucent vesicles of varying size. The shape of these vesicles was spherical in most cases, the majority being approximately 50% larger than synaptic vesicles in axon terminals surrounding the IR profile or in small serotonin-IR profiles. Infrequently larger vesicles were observed. The largest vesicle that we observed, was 180 nm in diameter. The morphology of the mitochondria in the swollen varicosities often differed from those in the adjacent neuropil or in small serotonin-IR profiles. The number of cristae was reduced as compared with the mitochondria in the surrounding IR and non-IR varicosities. Some mito-
164
M.G.P.A. VAN LUIJTELAAR ET AL.
errant serotonin-IR terminals in the caudate-putamen complex of aged rats. These aberrant fibers were examined only in the caudate-putamen complex, since at the light microscopical level, their morphology did not differ from aberrant fibers occurring in other brain areas. Small serotonin-IR profiles appeared to be vesicleloaded varicosities or intervaricose fibers containing mitochondria and only infrequently forming a synapse type of apposition with an adjacent profile. Thus the size of the profiles, the shape of the vesicles, and the presence of a minority of large granular vesicles reported in this study fit within the generally accepted description of serotonergic axon terminals based on immunocyDISCUSSION tochemical or autoradiographic studies in the young rat Previous light microscopical observations on the mor- (Arluison and De la Manche, 1980; Beaudet and Descarphology of aberrant serotonergic fibers in the brains of ries, 1987; Chazal and Ma, 1989; Soghomonian et al., aged rats indicated the degenerative character of these 1989; Maley et al., 19901, cat (Calas et al., 1976), or fibers (Van Luijtelaar et al., 1988, 1989, 1991). In the monkey (Roberts Jr. et al., 1990)brain. The immunoreaction product was localized on the present work, this suggestion was further evaluated by an ultrastructural examination of aberrant and nonab- outer membranes of vesicles and mitochondria. This is
chondria showed distorted cristae in an electron lucent, swollen matrix. However, mitochondria with a normal appearance were also encountered in the large IR profiles. In many aberrant fibers membrane-bound, small vacuoles of the size of mitochondria, containing membrane whorls or fragments of membranes were observed (Fig. 3). In addition to these small, membrane-filled vacuoles, several clusters contained a large to very large membrane-bound, electron-lucent compartment limited by a single membrane. In most cases these large vacuoles contained irregularly shaped membranewhirls, or packs of fragments of membranes.
Fig. 1. Examples of nonaffected serotonin-IR profiles in the caudate-putamen complex of an aged rat. Serotonin-IR fiber in close contact with a cell body (A), varicosities apposing a dendritic shaft (B), myelinated axons ( C ) ,or unidentified profiles (D). Note the occurrence of a distorted mitochondrium in a nonimmunoreactive profile (arrow in D). Bar = 1 pm. Magnification in B-D is the same as in A.
ULTRASTRUCTURE OF ABERRANT SEROTONIN-IR FIBERS
Fig. 2. Photomontage (A) showing swollen serotonin-IR profiles in the caudate-putamen complex of an aged rat. Note that small serotonin-IR profiles are also present (arrows).Bar = 1pm. The inset (B) shows aberrant serotonin-IR fibers among thin varicose fibers at the light microscopic level. Bar = 50 pm.
165
166
M.G.P.A. VAN LUIJTELAAR ET AL.
Fig. 3. Detail of a swollen serotonin-IR varicosity. Both healthy (arrowhead) and distorted (arrow) mitochondria and a small membrane-filled vacuole (double arrow) are present. The shape and the size of the vesicles is a t variance with those in the adjacent neuropil. Bar = 1 pm.
ULTRASTRUCTURE OF ABERRANT SEROTOIL’IN-IRFIBERS
167
Fig. 4. Photomontage of a swollen serotonin-IRprofile in the caudate-putamen complex of an aged rat, showing abnormal mitochondria: The mitochondriumindicated by an arrow has an electron lucent matrix and has lost its internal membrane organization. The open asterisk indicates a mitochondrium that is swollenas well. Small arrows indicate swollen vesicles.Theundulating outline of the profile is indicated by arrowheads or, when unclear, by filled asterisks. Bar = 1 wm.
comparable to the localization of the immunoreaction product in studies on other neurotransmitters (Voorn and Buys, 1987). The parallellism between young and non-swollen aged serotonergic fibers indicates that during aging many fibers are preserved. The swollen serotonin-IR profiles in the aged rat caudate-putamen complex differed from the small profiles in that they showed distorted and enlarged mitochondria, membrane-filled vacuolelike structures, and many enlarged vesicles. A tenfold increase of the size of the swollen profiles was no exception. In the swollen profiles the size of the vesicles could be as large as 180 nm and could vary more than in the small serotonin-IR profiles. Moreover, it was often impossible to make a discrimination between swollen vesicles and vacuolelike structures. Similar alterations in the structure of mitochondria and size and number of vesicles have been used as the major indicators for axonal degeneration, either age-induced (Miquel et al., 1983) or lesion-induced (Aghajanian et al., 1969; Wisniewski et al., 1972; Mugnaini and Friedrich Jr., 1981; Rapisardi et al., 1990). The morphology of the swollen serotonin-IR profiles thus resembles that of degenerating axons.
It has been suggested that mitochondria1injury is one of the first events in the process of degeneration (Miquel et al., 1983) and can be regarded as a marker for early stages of degeneration. Indeed, in the present study, distorted mitochondria have been observed in small immunoreactive profiles, and it is tempting to speculate that these profiles represent a stage of degeneration occurring before the swelling of the varicosities. However, mitochondria are very vulnerable toward osmotic shocks and care should be taken that artifacts of tissue preparation are not interpreted as degeneration of the tissue. The presence of both healthy and distorted mitochondria in immunoreactive profiles argues against a deleterious effect of the immunoreaction or the immunoreaction product on the structure of the mitochondria. Similarly, the presence of healthy and distorted mitochondria in nonimmunoreactive profiles indicates that processing for electron microscopy did not interfere with the ultrastructure of mitochondria. Conversely, the presence of distorted mitochondria in nonimmunoreactive fibers points toward the degeneration of other systems in the caudate-putamen complex of the aged rat, as can be expected in an aged animal.
168
M.G.P.A. VAN LUIJTELAAR ET A t .
To our knowledge, only one previous report was published on the ultrastructure of serotonergic fibers in the brain of aged rats (Calas and Van Den Bosch de Aguilar, 1980).In that radioautographic study, no differences in the pattern and intensity of the label were observed in the periventricular part of the caudate-putamen complex, though granulo-vacuolar degeneration was reported to occur in the raphe region. In the caudateputamen complex of the aged rat, many normal, unaffected serotonergic fibers occur, apart from the aberrant fibers (Van Luijtelaar et al., 1988, 1991), and a randomly selected area of the caudate-putamen complex does not necessarily include aberrant fibers. This might explain the failure of these authors to find any agerelated degeneration in the serotonergic system. Also, radioautographic labelling depends on the uptake mechanism in the serotonergic terminals. A decreased uptake capacity of the aberrant fibers would hamper the labelling of these fibers and this possibility cannot yet be excluded. Pilot studies in our own lab (unpublished data) gave no indication of a reduced uptake capacity in the caudate-putamen complex of the aged rat. However, the aberrant fibers constitute a subpopulation of the total serotonergic innervation of the caudate-putamen complex and the lack of an overall change cannot exclude a decrease uptake capacity of the aberrant fibers. As the present EM observations indicate that small profiles exist with degenerating mitochondria, the absence of aberrant fibers a t the LM level does not implicate a healthy, nondegenerating neurotransmitter system. In contrast, a t the EM level, all swollen varicosities were shown to contain distorted mitochondria, swollen vesicles, and vacuolelike structures. Thus degeneration of a neurotransmitter system is strongly suggested at the LM level by the presence of crumpled and folded fibers with swollen varicosities. ACKNOWLEDGMENTS The authors thank Anaatje Pattisalanno, Ineke Aarts, and Alice Rinkens for their technical assistance and Shimon Paniry for preparing the microphotographs. This work was supported by grant no. 900552-072 from the Netherlands Foundation for Medical Research (MEDIGON, NWO). REFERENCES Aghajanian, G.K., Bloom, F.E., and Sheard, M.H. (1969) Electron microscopy of degeneration within the serotoninergic pathway of rat brain. Brain Res., 13:26f%273. Arluison, M., and De la Manche, IS. (1980)High resolution autoradiographic study of the serotoninergic innervation of the rat cor us striatum after intraventricular administration of [3H]-5%ydroxytryptamine. Neuroscience, 5:229-240. Armstrong, D.M., Hersch, L.B., and Gage, F.H. (1988) Morphologic alterations of cholinergic processes in the neocortex of aged rats. Neurobiol Aging, 9:199-205.
Beaudet, A,, and Descarries, L. (1987)Ultrastructural identification of serotonin neurons. In: Monoaminergc Neurons: Light Microscopy and Ultrastructure. IBRO Handbook Series: Methods in Neurosciences, vol. 10. H.W.M. Steinbusch, ed. John Wiley & Sons, ChiChester, pp. 265-313. Calas, A., and Van Den Bosch de Aguilar, P. (1980) Comparative radioautogra hic study of serotoner 'c neurons in young and senescent rats. The psychobiology opagng: problems and perspectives. D.G. Stein, ed. Elsevier Amsterdam, pp. 59-80. Calas, A,, Besson, M.J., Gauchy, C., Alonso, G., Glowinsky, J., and Cheramy, A. (1976) Radioautography study of in vivo incorporation of [3H]-monoamines in the cat caudate nucleus: identification of serotoninergic fibers. Brain Res., 118:l-13. Chazal, G., and Ma, W. (1989)An ultrastructural analysis of serotoninergic neurons in the nucleus raphe magnus of the rat. Neuroscience, 33:301-310. Johnson! R.J. (1985)Anatom of the a g n g nerve cell. In: Handbook of Cell Biology ofAging. V.J. Jristafalo, ed. CRC Press, Boca Raton, pp. 149-178. Maley, B.E., Gail Engle, M., Humpreys, S., et al. (1990) Monoamine synaptic structure and localization in the central nervous system. J. Electron Microsc. Techn., 15:20-33. Masuoka, D.T., Jonsson, G., and Finch, C.E. (1979)Agingandunusual catecholamine-containing structures in the mouse brain. Brain Res., 16913355341. M.& i uel, J ' ,Johnson, . J.E., and Cervos-Navarro, J . (1983)Comparisonof S a g n g in humans and experimental animals. Aging, 21:231258. M@llg5rd, K., Lundberg, J.J., Wiklund, L., Lachenmayer, L., and Baumgarten, H.G. (1978) Morphologic consequences of serotonin neurotoxin administration: neuron-tar et cell interaction in the rat subcommissural o y a n . &n. N.Y. A c a f Sci., 305:262-288. M u y n i , E:, and riedrich Jr., V.L. (1981) Electron microscopy: i entification and study of normal and degeneratin neural ele ments by electron microsco y In Neuroanatomical &act-Tracing Methods. L. Heimer and hJ.! Robards, eds. Plenum Press, New York, p 377406. Rapisar&,'S.C., Warrington, V.O.P., and Wilson, J.S. (1990) Effect of MPTP on the fine structure of neurons in substantia nigra of dogs. Brain Res., 512:147-154. Roberts J r . , E.L., Rosenthal, M., and Sick, T.J. (1990) Age-related modifications of potassium homeostasis and synaptic transmission during and after anoxia in rat hippocampal slices. Brain Res., 514:111-118. Rogers, J.,and Styren, S.D. (1987) Neuroanatomy of agingand dementia. Rev. Biol. Res. Aging, 3:223-253. Soghomonian,J-J., Descarries, L., and Watkins, K.C. (1989)Serotonin innervation in adult rat neostriatum. 11. Ultrastructual features: A radioautographic and immunocytochemical study. Brain Res., 481337-87. Steinbusch, H.W.M., and Tilders, F.J.H. (1987) Immunohistochemical techniques for light-microscopical localization of dopamine, noradrenaline, adrenaline, serotonin and histamine in the central nervous system. In: Monoaminergic Neurons: Li ht Microscopy and Ultrastructure. IBRO Handbook Series: Methojs in Neurosciences, vol. 10. H.W.M. Steinbusch, ed. John Wiley & Sons, Chichester, pp. 125-166. Van Luijtelaar, M.G.P.A., Tonnaer, J.A.D.M., and Steinbusch, H.W.M. (1988)Aberrant morphology of serotonergic fibers in the forebrain of the aged rat. Neurosci. Lett., 95:93-96. Van LuijteIaar, M.G.P.A., Tonnaer, J.A.D.M., and Steinbusch, H.W.M. (1989) Similarities between aberrant serotonerGc fibers in the aged and 5,7-DHT denervated young adult rat brain. Exp. Brain Res., 78:81-89. VanLuijtelaar, M.G.P.A., Tonnaer, J.A.D.M., andsteinbusch, H.W.M. (1991) Aging of the serotonerpic svstem in the rat forebrain. Neurobiol Aging (Cn ress) Voorn, P., and [uys, 'R.M. (1987) Ultrastructural demonstration of dopamine in the central nervous system. In: Monoaminer 'c Neu rons: Light Microscopy and Ultrastructure. IBRO Handboo Series. ' 1 Methods in Neuroscience,vol10. H.W.M. Steinbusch, ed. John Wiley & Sons, Chichester, pp. 241-264. Wisniewski, H.M., Ghetti, B., and Horoupian, D.S. (1972) The fate of s naptic membranes of degenerating optic nerve terminals, and tgeir role in the mechanism of trans-synaptic changes. J. Neurocyto].. 1~297-310.
A:
-
I
f
