- Email: [email protected]
Collateral Sprouting of Central Noradrenergic Neurons during Aging: Histochemical and Neurochemical Studies in Intraocular Triple Transplants
EXPERIMENTAL NEUROLOGY ARTICLE NO.
145, 524–535 (1997)
EN976485
Collateral Sprouting of Central Noradrenergic Neurons during Aging: Histochemical and Neurochemical Studies in Intraocular Triple Transplants Nisha Srivastava,*,† Ann-Charlotte Granholm,*,†,‡,§ and Greg A. Gerhardt†,‡,§,¶ *Department of Basic Science, ¶Department of Psychiatry, and ‡Department of Pharmacology, §the Neuroscience Training Program, and †the Rocky Mountain Center for Sensor Technology, University of Colorado Health Sciences Center, Denver, Colorado 80262
The sprouting capacity of aged noradrenergic neurons of the brain-stem nucleus locus coeruleus (LC) was examined using intraocular transplants of fetal tissues. Fetal hippocampal tissue (E18) and LC tissue (E15) were transplanted together as a double transplant into the anterior chamber of the eye of young adult Fischer 344 rats. The double transplants were allowed to mature for 14–18 months, after which an additional fetal hippocampal transplant was placed next to the LC graft. The triple transplants were monitored for overall growth and vascularization for an additional 2–6 months. Immunohistochemical examinations showed that both young (2–6 months old) and aged (16–24 months old) hippocampal cografts contained a plexus of thin varicose tyrosine hydroxylase (TH)-immunoreactive fibers extending throughout the grafted hippocampal tissues. However, the aged hippocampal grafts contained a denser uniform plexus of TH-positive fibers compared to the young transplants. Immunohistochemistry with synapsin antibodies demonstrated that both the young and the aged hippocampal transplants contained much higher densities of synaptic elements than the LC grafts. In vivo electrochemical measurements of potassium-evoked overflow of norepinephrine (NE) in the grafts showed that similar amounts of NE overflow were detected in both the young and the aged hippocampal grafts. HPLC–EC measurements of NE levels in the grafts revealed that there were similar amounts of NE in the young and the aged grafts, and the grafts did not contain serotonin or dopamine. In summary, the findings of the present study show that aged LC neurons are capable of undergoing collateral sprouting producing a functional NE neuronal system when introduced to an appropriate young target. r 1997 Academic Press
INTRODUCTION
Aging is associated with localized, region-specific degenerative changes in the central nervous system (21). The process of neurodegeneration in aging or 0014-4886/97 $25.00 Copyright r 1997 by Academic Press All rights of reproduction in any form reserved.
disease has been associated with intrinsic changes in neurons, leading to loss of plasticity and/or alterations in the target tissues supplied by the neurons, such as reduced production of neurotrophic factors. In order to develop appropriate therapeutic treatments for neurodegenerative diseases during aging, the cause and effect of this neurodegeneration needs to be closely studied. The hippocampal formation has been widely used to study neuronal plasticity and target dependence of such plasticity (12, 40, 41, 55, 61). This region is particularly well suited for such studies due to the abundant knowledge of its anatomic and physiologic properties, as well as its function in memory and learning processes (11, 60). Several different transmitter-specific pathways innervating the hippocampal formation have been implicated in these learning and memory paradigms such as the cholinergic and the noradrenergic (NE) pathways (1, 34, 39, 44, 65). NE fibers in the hippocampus arise from distinct clusters of NE cell bodies in the pontine and medullary reticular formation. The largest group of NE neurons is located in a single nucleus, the locus coeruleus (LC), located beneath the floor of the fourth ventricle anterior to the facial colliculus, containing an estimated 1600 cells per hemisphere in the rat brain (29, 48, 63). This NE transmitter system represents a highly collateralized neuronal system, with axonal collaterals extending both rostrally into the forebrain, hippocampal formation, and cerebral cortex and caudally into the medulla and spinal cord (15, 46). The NE neurites from the LC reach the hippocampal formation by means of both a ventral and a dorsal pathway (40, 50). Although the phenotypic expression of these NE neurons appears early in embryonic development, the cell bodies undergo a significant histological reorganization until they have attained their definite position within the brain stem in the late postnatal period, as late as 3 weeks postnatally in the rat (3, 59). It has also been shown that plasticity of these neurons remains throughout adulthood. Pharmacological treatments (7, 16), physiological perturbations, or lesions (42, 54, 57) can
524
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS
give rise to enhanced TH expression in the LC, or to sprouting of NE neurites. However, it is not known whether NE neurons in the LC maintain this plasticity during aging. Numerous studies of the aged brain in man and animals have demonstrated a significant cell loss in the NE-containing neurons of the LC during aging (8–10, 36, 62, 64). In a recent study, Manaye and collaborators (43) demonstrated that this cell loss was significantly greater in the rostral than in the caudal NE cells in the LC of aged human brain. Thus, this represents a systematic rather than a random loss of cells in the nucleus (43). In addition, morphological abnormalities of the NE innervation of the hippocampal formation have been observed, both in normal aged humans and in patients with senile dementia of the Alzheimer’s type (53). In contrast to morphometric studies, neurochemical studies in the aged human brain have indicated a general sparing of noradrenergic indices in the target areas, with possible exception of the hippocampal formation (50). This suggests a regional difference in agerelated changes of NE mechanisms and also a potential high compensatory ability of these neurons. The sparing of NE levels and its metabolites could arise from compensation for NE denervation by increased NE turnover (49). In experimental animals, on the other hand, significant reductions in NE activity and innervation have been reported throughout the aged brain (45, 56), even though some studies have not shown any differences in LC cell numbers in aged animals (30). Furthermore, a number of other indices of NE function have been found to be decreased in the aged rat, such as postsynaptic a- and b-receptor responsivity (4, 37), behavioral response to NE agonists in the aged rat (6), and basal firing rates of LC neurons (47). In order to determine whether these changes are due to an age-related decrease in the function of the NE neurons of the LC or to changes in the hippocampal target, a model system in which the age of the target versus the age of the neuronal population could be varied independently is needed. Sequential double or triple grafting of tissues into the anterior chamber of the eye of the rat offers such a unique model system (17, 18, 26, 31). In several series of experiments we have previously demonstrated that many characteristics that appear during aging of the NE transmitter system of the LC in situ are duplicated in intraocular nervous tissue transplants. For example, we found that aged intraocular cerebellar (31) and hippocampal transplants (17, 18) develop significant postsynaptic subsensitivities to applied a-receptor agonists, suggesting a postsynaptic decrease in receptor affinity or density. A subsensitivity to a-adrenergic agonists was not observed in young transplants placed in aged hosts, indicating that the changes occurring to the adrenergic
525
receptors are due to factors within the transplanted CNS tissue, rather than system-related extrinsic factors in the animal such as hormone levels. This work has been further substantiated by a recent paper by Eriksdotter-Jo¨nhagen et al. (19) who found that postsynaptic binding of a-adrenergic ligands was reduced in aged compared to young hippocampal transplants in oculo. In addition, we have also shown that aged LC grafts in oculo exhibit a reduction in release amplitude as well as prolonged rise times for the potassiumevoked release of NE, compared to young transplants (26). Finally, the LC–NE pathway between hippocampal and brain-stem cografts in oculo was found to be significantly perturbed in aged double grafts, in terms of both pre- and postsynaptic properties (17). These experiments have demonstrated that there are both pre- and postsynaptic changes in NE neurotransmission during aging and that the intraocular transplant model represents a valid model system for studies of aging. In the present study we have continued the previous work by constructing triple transplants containing both young and aged hippocampal transplants in contact with an aged NE-containing brain-stem transplant. By varying the age of the transplanted tissues versus the host, it is possible to create so called ‘‘old–young chimeras,’’ in which intrinsic versus extrinsic determinants of aging can be investigated in detail. Thus, the same NE neurons can potentially innervate a young and an aged target simultaneously. The specific questions addressed in this study were: (1) Can aged LC neurons innervate a fetal target with NE fibers? (2) If innervated, are the densities and distributions of NE fibers and other neuronal markers the same in the young versus the aged hippocampal grafts? (3) Are NE release and synthesis the same in the young versus aged hippocampal tissues? MATERIALS AND METHODS
Animals. Young adult female Fischer 344 rats (5–6 weeks of age; Harlan Laboratories) were used as recipients for intraocular fetal grafts. Tissue donors were timed pregnant Fischer 344 dams of Embryonic Day 15–17 (E15–E17) for brain-stem tissue and E18 for hippocampal tissue. The recipient rats were sympathectomized prior to transplantation by means of surgical removal of the superior cervical ganglia (32). This procedure eliminates all sympathetic NE fibers in the host iris, which could potentially interfere with the central LC–NE innervation of target grafts in the eye. All animals were maintained on standard food pellets and water that were available ad libitum, and animals were kept on a standard 12:12-h light:dark cycle. Following surgery, the host rats were given paracetamol
526
SRIVASTAVA, GRANHOLM, AND GERHARDT
(6 mg/L) in their drinking water to alleviate any pain associated with the surgery. Dissections and transplantations. Pregnant rats were euthanized by an overdose of halothane; fetuses were removed and kept on ice until they were dissected. The fetal brain was removed, and brain-stem tissues containing the LC neurons and hippocampal CA1 tissue were dissected as previously described (32). Recipient host rats were anesthetized with Xylazine (Rompun, 12 mg/kg, ip) and Ketamine (Ketaset, 80 mg/kg, ip), and 1 drop of 1% atropine solution was placed on the cornea to dilate the pupil. Tissue pieces of fetal brain-stem (‘‘LC graft’’) and hippocampal (‘‘Hi1 graft’’) tissue were placed adjacent to each other into the anterior chamber of the eye through a razor blade slit in the cornea using a modified glass pipette as previously described (see 32). The LC grafts were first placed, followed by the Hi1 grafts 2–3 weeks later. The double transplants were allowed to mature for 14–18 months after which a second hippocampal tissue graft (‘‘Hi2 graft’’) was placed next to the brain-stem transplant. The triple transplants were allowed to grow for an additional 2–6 months in oculo. Thus, the LC and the Hi1 grafts were 16–24 months old, and the young Hi2 grafts were 2–6 months old at the time of study. See Fig. 1A for an outline of the experimental design. Immunohistochemical evaluation of tyrosine hydroxylase, neurofilament, glial fibrillary acidic protein, and synapsin I and II in the hippocampal and LC grafts. A number of immunohistochemical markers were used to investigate the intraocular triple transplants. Transplants (seven per group) were rapidly removed from the anterior eye chamber of urethane-anesthetized hosts and immersion fixed in 4% paraformaldehyde prepared in 0.1 M phosphate-buffered saline (PBS). After overnight fixation, the transplants were transferred to 10% sucrose in PBS for at least 16 h prior to sectioning on a cryostat. Twelve-micrometer-thick sections were collected on slides, washed in Sorensen’s buffer (0.1 M phosphate-buffered saline with pH 7.4), and incubated with antibodies directed against tyrosine hydroxylase (TH; 1:100; Eugene Tech Inc.), neurofilament (NF; 1:100; Dakopatts, Denmark), glial fibrillary acidic protein (GFAP; 1:100; see 14), or synapsin I and II (1:200; see 2) for 48 h. After washing in Sorensen’s buffer (3 3 10 min), the sections were incubated with antirabbit or anti-mouse IgG conjugated with fluorescein isothiocyanate (1:30) for 1 h at room temperature. The sections were washed again for 3 3 10 min and coverslipped with glycerol:PBS (9:1). Light-microscopical evaluations were performed on a Nikon Optiphot fluorescence microscope. The sections were devoid of immunoreactivity when the first antibody was omitted. In vivo electrochemical recordings. Host rats were anesthetized with urethane (1.25 g/kg, ip), tracheoto-
FIG. 1. (A) Schematic drawing outlining the experimental design of the triple brain grafts. Brain-stem (LC-NE) and hippocampal (Hi1) tissues from fetuses were transplanted to the anterior chamber of the eye of young adult Fischer 344 rats. The double transplants were allowed to mature for 14–18 months, after which a second hippocampal graft (Hi2) was placed next to the brain-stem graft. The triple transplants were evaluated through the translucent cornea for an additional 2–6 months. Thus, the aged hippocampal and brain-stem grafts were 16–24 months old, and the young hippocampal grafts were 2–6 months old at the time of sacrifice. (B) Photograph of a triple transplant in oculo. Blood vessels can be seen on the surface of all three transplants. No clear borders between the three grafts can be seen. (LC, brain stem; Hi1, aged hippocampal graft; Hi2, young hippocampal graft.) Scale bar represents 1 mm.
mized, and placed in a stereotaxic frame. The cornea overlying the transplants was removed and a plexiglas chamber was placed over the eye (24–26). The eye was continually superfused with Earle’s balanced salt solution maintained at 37°C. In vivo electrochemical recordings were performed in the hippocampal grafts using a high-speed chronoamperometric recording system (IVEC-10; Medical System Corp.). The working electrodes employed were the Nafion-coated single-carbon fiber type (30-µm o.d.), which were attached to single-barrel micropipettes with a tip separation of 250–300 µm (26). The carbon fiber electrodes were coated with Nafion and characterized for sensitivity and selectivity as previously described (24–28). The sensitivities, selectivities, and linearities of all recording electrodes were determined using 0.1–6.0 µM levels of NE prepared in pH 7.4 PBS with 250 µM ascorbic acid (26, 28). The high-speed chronoamperometric electrochemical measurements consisted of applying square-wave pulses of 0.0 to 10.55 V with respect to an Ag/AgCl reference electrode (in contact with the recording chamber) for 0.1 s and were repeated at 5 Hz. The resulting oxidation and
527
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS
reduction currents were digitally integrated during the final 80% of the recording periods. Data were averaged over 1-s epochs to improve the signal-to-noise ratios of the recordings. Only electrodes that exhibited selectivities of greater than 250:1 for NE versus ascorbic acid were used for the recording in the grafts. The average detection limits of all sensors ranged from 25 to 50 nanomolar levels of NE (signal-to-noise ratio of 3). The overflow of NE was elicited by local application of excess potassium via pressure ejection using micropipettes with an outer tip diameter of 10–15 µm (26). A typical experiment involved placement of the electrode recording assembly into regions of the young and aged hippocampal grafts. Electrochemical recordings were initiated and carried out for 5–10 min to establish a stable baseline. Once the signals were stable, the potassium-evoked overflow of NE was elicited by pressure ejection of 100–250 nl of a potassium solution (70 mM KCl, 79 mM NaCl, and 2.5 mM CaCl2, prepared in distilled water and adjusted to pH 7.3–7.4). Recordings were continued until the signals returned to baseline. The assembly was then moved to a new region of the grafts. The potassium-evoked overflow of NE was expressed quantitatively in terms of NE calibration curves determined in vitro for each electrode prior to experiments; the initial preexperiment calibration factors were used for data analysis. Only electrode placements that yielded signals greater than or equal to 50 nM NE following maximal potassium activation (26) were included in the final data analysis. In addition, the rise times of the signals (TR) and the 80% decay times of the signals (T80) were used as data analysis parameters (26, 27). All statistical comparisons of the NE overflow data from the young and aged transplant groups were made using a two-tailed Student’s t test. High-performance liquid chromatography with electrochemical detection (HPLC–EC). Hippocampal transplants from all groups (n 5 9 per group) were rapidly removed from the anterior eye chamber of anesthetized hosts, and the animals were euthanized by an overdose of urethane (2 g/kg, ip). The grafts were divided into Hi1 (16–24 months old—‘‘old’’) and Hi2 (2–6 months old—‘‘young’’) portions, weighed, and frozen on dry ice for HPLC–EC measurements. The whole-tissue weights of the hippocampal grafts averaged 2.1 6 0.4 mg for the young grafts and 2.3 6 0.5 mg for the aged grafts. The tissue levels of NE, dopamine (DA), and serotonin (5-HT) were measured in the transplants using previously reported HPLC–EC methods (35). Monoamine values were calculated as total nanograms per gram wet weight of tissue (ng/g), and statistical comparisons were made using a two-tailed Student’s t test.
RESULTS
Intraocular growth. The intraocular transplants survived well and became vascularized within a week from grafting (Fig. 1B). The hippocampal grafts (Hi1 and Hi2) grew extensively in oculo, while the brainstem grafts remained approximately the same size throughout the experiment. The surface area of LC grafts ranged from 1 to 2 mm2 after cessation of growth, and the hippocampal grafts were 4–8 mm2, regardless of whether the grafts were young (Hi2) or aged (Hi1). Thus, the age of the recipient rat did not determine growth rate of the fetal hippocampal tissue. Blood vessels could be seen on the surface of all three transplant groups. Both the Hi1 and the Hi2 grafts attached well to the brain-stem grafts and no distinct border could be seen between the three grafts (Fig. 1B). Morphology. Routine histological evaluation with cresyl violet staining revealed a distinct cell layer resembling the pyramidal cells in both hippocampal grafts, as well as the seamless integration of the three different tissues with each other (Fig. 2). The pyramidallike cell layer consisted of a two- to three-cell-thick, dense layer of large cells (approximate axial diameter perpendicular to the apical dendrites was 20–30 µm) that extended along the longitudinal axis of the grafts, a few hundred µm from the surface. There was no apparent difference in the morphological organization of the young and aged hippocampal grafts. Thus, the age of the recipient at the time of grafting does not appear to influence this parameter. The brain-stem grafts contained a single group of large neurons organized in a fashion similar to LC neurons in situ (Fig. 2). There were no signs of rejection in any of the grafted CNS tissues (n 5 7). Blood vessels could be found throughout the transplanted tissue (see Fig. 2). Immunohistochemistry. Clusters of large TH-positive cells were found in the LC portion of the transplants (Fig. 3). These TH-positive cells resembled LC neurons in situ and exhibited extensive dendritic arborizations in the vicinity of the cell bodies. Axons were thin and varicose and innervated the brain-stem tissue with a dense plexus of fibers (Fig. 4). A plexus of TH-like immunoreactive fibers could be seen innervating both the Hi2 (2–6 months old) and the Hi1 (16–24 months old) hippocampal transplants. The density of the THpositive fibers appeared greater in the Hi1 grafts, compared to the Hi2 grafts, as seen in Figs. 3, 4, and 5. This differential density was seen in all triple transplants examined and is summarized in Fig. 5. However, neurites could be found throughout both transplant groups. Club-like, larger end processes could sometimes be seen in the Hi2 (young) hippocampal grafts, indicative of active neuritic growth in the young target. The Hi2 grafts exhibited a patchy and gradient TH-
528
SRIVASTAVA, GRANHOLM, AND GERHARDT
FIG. 2. Cresyl violet-stained section of a triple transplant; Hi1, old (16-month) hippocampal graft; Hi2, young (2-month) hippocampal graft; LC, brain-stem graft. Note the pyramidal-like cell layer in both hippocampal grafts as well as the close attachment of the three different tissues with each other. The host iris can be seen attached to the ventral surface of the tissues. Scale bar represents 0.5 mm.
positive fiber distribution, with highest densities of fibers found close to the LC–hippocampal borders (Fig. 3). This could not be seen in the aged hippocampal grafts, which were found to contain a dense innervation pattern throughout the grafted tissue (Fig. 3). In Fig. 4, higher magnification micrographs from all three grafts seen in Fig. 3 are shown. As seen in Fig. 4, the individual fiber morphology appeared to be similar in the Hi1 and Hi2 grafts, despite the difference in the fiber density. The overall neuronal fiber network, shown with NF immunohistochemistry, was distributed through the transplants with no observable differences in terms of either density or distribution between the three graft groups (Fig. 6). The abundance of NF-like neurites in the grafts strongly suggests the presence of other innervation modalities, apart from the LC-NE neurons.
Also the brain-stem grafts contained a large number of NF-immunoreactive neurites (Fig. 6). These neurites differed from the TH-like neurites in their morphology, with both smooth, thick process and thin, varicose processes (Fig. 6). This further suggests that there are other neuronal populations in the grafts. A particularly dense NF-immunoreactive network of fibers could be found surrounding the neurons of the pyramidal-like cell layer in both hippocampal grafts (Fig. 6). Morphological studies using immunohistochemical methods with antibodies directed against synapsin (I and II) showed that both hippocampal transplants contained much higher densities of synaptic elements than the LC grafts (Fig. 7). The density of synapsinimmunoreactive profiles appeared to be somewhat greater in the old versus young hippocampal grafts. However, quantitative measurements were not per-
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS
529
FIG. 3. Photomontage of a section from a triple transplant, incubated with TH antibodies. Hippo1, old (16-month) hippocampal graft; Hippo2, young (2-month) hippocampal graft; LC, old brain-stem graft]. Note the difference in innervation density between the two hippocampal grafts. There are, however, TH-immunoreactive fibers extending throughout the length of both grafts. Scale bar represents 0.5 mm.
formed in the present studies. The distribution and density of glial elements was also examined in the triple transplants, using antibodies directed against GFAP (Fig. 8B). The Hi1 and brain-stem transplants exhibited a marked gliosis with a denser pattern of astrocytes than the young hippocampal grafts (Fig. 8B). No significant glial scarring could be seen in the borders between the brain-stem grafts and either Hi1 or Hi2 grafts. Finally, autofluorescent lipofuchsin granules were present in all of the aged transplants, but were not observed in any of the young hippocampal tissues (Fig. 8A). In vivo electrochemical recordings. In vivo electrochemical recordings of potassium-evoked overflow of NE were carried out in five young (2–6 months) and five aged (16–24 months) hippocampal grafts in order to investigate the functional properties of the NE fibers in the grafts (26). Potassium was seen to produce detectable electrochemical signals in both the young and the aged hippocampal grafts. The recorded oxidation and reduction currents of the signals supported the detection of NE signals and not the measurement of ascorbic acid, serotonin, or dopamine (26, 28). Interestingly, the
average NE-like signals were slightly but not significantly larger in the aged tissues (see Fig. 9A). The potassium-evoked electrochemical signals averaged 1.05 6 0.25 µM (NE equivalents, n 5 20 signals from five grafts) versus 0.60 6 0.10 µM in the young hippocampal tissues (n 5 20 releases from five grafts). The temporal properties of the potassium-evoked signals were also evaluated to investigate potential differences in NE release dynamics in both the young and the aged hippocampal grafts. The TR and the T80 of the potassium-evoked NE-like signals were also similar in the young and aged grafts. However, as seen in Fig. 9B, the temporal properties of the signals recorded in the young grafts tended to be longer than those recorded in the aged grafts. The TR and T80 times were 25 6 3 and 53 6 5 s (n 5 20), respectively, in the young hippocampal tissues versus 17 6 2 and 43 6 2 s (n 5 20), respectively, in the aged grafts. This trend was statistically significant for only the rise times of the signals (P , 0.01). Thus, the time courses and amplitudes of the potassium-evoked NE-like signals were similar in both the young and the aged tissues, indicating that the NE fibers in both sets of grafts were
FIG. 4. Micrographs at higher magnification showing details from the TH-incubated triple transplant section shown in Fig. 3, illustrating the density of innervation in the Hi1 graft (A), the LC graft (B), and the Hi2 graft (C). Note the sparse network of fibers in the Hi2 versus the Hi1 graft. Scale bar represents 35 µm.
530
SRIVASTAVA, GRANHOLM, AND GERHARDT
FIG. 5. Schematic drawing illustrating the overall TH-positive fiber distribution found in the Hi1 versus the Hi2 hippocampal transplants. This schematic drawing represents objective estimations from seven different grafts.
capable of supporting depolarization-induced release of NE. HPLC–EC measures of NE, DA, and 5-HT in the young and aged grafts. HPLC–EC measurements of the whole tissue levels of NE, DA, and 5-HT that were contained in the grafts were carried out in nine sets of young and aged tissues to investigate the neurochemical characteristics of the hippocampal grafts. All the hippocampal tissues contained large levels of NE, and there was no statistically significant difference between the NE levels in the young and aged hippocampal grafts. The average NE content of the young hippocam-
pal grafts was 1530 6 648 ng/g wet weight (n 5 9) and the NE content of the aged grafts was 1401 6 544 ng/g wet weight (n 5 9). These levels represent an approximately fourfold greater amount of NE in the grafts compared to the normal NE innervation of the hippocampus (5). In addition, there were no detectable levels of 5-HT in both sets of tissues (detection limit ,1 ng/gram wet weight of tissue). Moreover, only one set of grafts contained significant levels of DA (,100 ng/g wet weight). Thus, the young and aged hippocampal grafts were seen to contain high levels of NE without contamination from other monoamine neurotransmitters such as 5-HT and DA. These data also support the hypothesis that NE was the predominant monoamine that contributed to the potassium-evoked electrochemical signals. DISCUSSION
The present study is, to the best of our knowledge, the first study showing that LC neurons are capable of maintaining their regenerative capacity during aging in the rat. We found both morphological and neurochemical evidence for functioning noradrenergic terminals throughout the young hippocampal tissue that
FIG. 6. Sections of triple graft incubated with neurofilament antibodies. The overall pattern of nerve fibers was similar in the Hi1 (A) and the Hi2 (B) grafts. A dense plexus of nerve fibers could also be seen in the brain-stem graft (C) and around the pyramidal-like cell layer in the hippocampal grafts (D). The graft–graft border can be seen in (D). Scale bar in D represents 100 µm for A, B, and C, and 75 µm for D.
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS
FIG. 7. Synapsin immunoreactivity in intraocular triple transplants. The brain-stem LC tissue contained a sparse plexus of synapsin-positive profiles (the tissue seen on the right side in (C)), while the hippocampal grafts both exhibited dense aggregations of synapsin immunoreactivity. (A) Hi1 and (B) Hi2. Scale bar represents 50 µm.
531
was innervated by an old LC. This suggests that collateral sprouts had been generated from the aged LC neurons in the brain-stem grafts to innervate the new hippocampal target. The in vivo electrochemical recordings of potassium-evoked NE-like releases in both sets of grafts and the HPLC–EC measures of NE content in the grafts also support the hypothesis that aged NEcontaining neurons can innervate a young target with functional NE fibers. Thus, our histological and neurochemical data support the hypothesis that aged NEcontaining LC neurons retain the ability to innervate an appropriate target with functional NE fibers. There are contradicting results in the literature concerning neuronal plasticity during aging. A reduction in the collateral sprouting (38, 55), neuronal regeneration (20), and formation of new synapses (56) was reported in aged animals. In contrast, when aged sympathetic neurons were grafted in contact with young targets, they were found to retain a high degree of plasticity (13, 22). However, this maintenance of sprouting capacity could be a function of peripheral neurons only. In addition, most studies to date have been performed on aged animals in situ, a model system that does not allow for separation of target and neuron age independently. With the use of our intraocular grafting paradigm, we are able to vary the age of the neuron versus its target and its host independently, and thus determine the dependence of intrinsic versus extrinsic factors during aging. Our findings in the present study, in which we found that aged LC neurons retain the ability to reinnervate a fetal target, correlate well with a recent study by Gavazzi (23) showing identical results for aged peripheral sympathetic neurons. Thus, it is possible that both peripheral and central neurons retain a certain amount of plasticity during aging, when presented with appropriate targets, but that there is a limit to this pruning or collateral sprouting ability. The NE innervation in both the young and the old hippocampal grafts resembled that seen in young LC– hippocampal grafts (17) with a plexus of thin, varicose fibers. In a recent study, we found morphological alterations, such as axonal swellings and degenerative debris, in aged brain-stem grafts compared to young grafts (26). Since we did not detect such changes in the LC–hippocampal grafts in the present study, nor in a previous study of aged LC–hippocampal double grafts (17), it is possible that changes occur in single brainstem grafts during aging due to the lack of an appropriate target to innervate. We conclude from the morphological findings presented here that aged LC neurons have retained their ability to innervate a fetal target, and that the pattern of innervation resembles that seen in young grafts and in the hippocampal formation, even though the density is somewhat lower. In the present study, the potassium-evoked NE over-
532
SRIVASTAVA, GRANHOLM, AND GERHARDT
FIG. 8. (A) Autofluorescent lipofuchsin granulae in an aged hippocampal graft. (B) GFAP immunoreactivity at the border between the aged (Hi1) and the young (Hi2) hippocampal portion of a triple transplant. Note that the aged hippocampal tissue contains a significantly higher density of GFAP immunoreactivity than the young hippocampal tissue. Scale bar represents 50 µm.
flow data and the whole-tissue measures of NE levels also support the hypothesis that the old LC grafts innervate the young and aged hippocampal tissues with functional NE fibers. The signals recorded from the young grafts support the hypothesis that aged LC neurons can innervate a young host with functional NE fibers that are comparable to the NE innervation of the aged hippocampal grafts. The time courses and amplitudes of the NE responses recorded from the young and aged grafts were very similar. The differences in the
rise times of the signals recorded in the young hippocampal grafts, although significant, were small and do not reflect a major difference in the functional properties of the NE neurons in the young and aged tissues. However, in a previous study, we showed that potassiumevoked overflow of NE-like signals in aged single LC grafts in oculo were decreased compared to young grafts. In that study, we found major differences in the potassium-evoked NE signals recorded from young and aged brain-stem grafts (26). By contrast, the NE-
FIG. 9. In vivo electrochemical recordings of potassium-evoked NE overflow in the young and aged hippocampal grafts. (A) Average amplitudes of potassium-evoked NE overflow in young (2–6 months old) and aged (16–24 months old) hippocampal grafts. (B) Average rise times (TR) and 80% decay times (T80) of the potassium-evoked NE signals recorded from young and aged tissues. **P , 0.01 using a two-tailed Student’s t test.
533
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS
containing fibers in the present study were derived from the aged LC for both the young and the aged hippocampal grafts. Interestingly, the amplitudes of NE overflow recorded in the present study were greatly diminished compared to those recorded in young LC grafts (26). However, the temporal properties of the responses recorded in the present study are comparable to those recorded from the aged single LC grafts. Thus, although the old LC appears to have the capability to innervate a young graft with functional NE fibers, the functional properties of these NE fibers appear to retain aged NE release properties and they do not release NE analogous to young NE-containing fibers. There are several possible explanations why the aged LC neurons are capable of innervating the fetal hippocampal tissue after such a long time of maturation in the grafts. The most likely hypothesis would be that a centrally active adrenergic neurotrophic factor is released from the fetal hippocampal tissue, and thus attracts the ingrown collaterals from the LC graft. An alternative hypothesis is also possible. Previous work by others has shown that partial damage to the noradrenergic afferent fibers to cerebellum resulted in a proliferation of the remaining LC axons in this structure as well as in a nondenervated structure, the hippocampus (51, 52). This phenomenon has been called neuronal ‘‘pruning’’ (58) and was proposed to be a property of highly collateralized neuronal systems, such as the LC. If there are properties of the aged target tissue that are detrimental to the NE innervation, for example, lack of trophic or tropic support, perhaps the fetal tissue provides a more beneficial environment in which the LC neurons can retreat. Future studies, using for example electron microscopy in combination with intracellular labeling techniques, can be used to determine if there is an abandonment of synaptic sites in the aged hippocampal transplants, in favor of the young target tissue. However, the immunohistochemical labeling of synaptic proteins performed here does not indicate that this is the case, since a dense plexus of synaptic profiles was still present in the aged hippocampal target. On the other hand, previous work from our group (33) has shown that the levels of synapsin proteins in the hippocampal portion of aged LC–hippocampal double grafts are significantly decreased compared to the levels seen in young double grafts. This could indicate a functional reorganization of synaptic sites in the aged hippocampal grafts. In conclusion, we have shown that NE-containing LC neurons maintain plasticity during aging, since these neurons were able to reinnervate fetal hippocampal transplants that were introduced at the age of 14–18 months. These data have significant implications for development of therapeutic strategies in age-related neurodegenerative disorders. The next step will be to
try to identify the factor or factors that stimulate the regenerative capacity of LC neurons in the brain. ACKNOWLEDGMENTS We thank Justin Mott for expert technical assistance with the microphotography. We thank Dr. Doris Dahl, Department of Neuropathology, VA Medical Center, West Roxbury, Massachusetts, for supplying the GFAP antibody and Dr. Michael Browning, Department of Pharmacology, UCHSC, Denver, Colorado, for supplying the synapsin I and II antibodies. This work was supported by USPHS Grants AG12122 and MH49661 to A.C.G. and AG-06434 and NS09199 to G.A.G. Dr. Gerhardt received support from a Level II National Research Service Scientific Development Award (MH01245). Dr. Srivastava was supported by a fellowship from a Drug Abuse training grant (5T32AAO7464-20).
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Bartus, R. T., R. L., Dean, B. Beer, and A. S. Lippa. 1982. The cholinergic hypothesis of geriatric memory dysfunction. Science 217: 408–417. Bergman, H., M. Browning, and A-Ch. Granholm. 1992. Development of synapsin I and synapsin II in intraocular hippocampal transplants. Hippocampus 2: 339–348. Bezin, L., D. Marcel, L. I. Debure, N. Ginovart, C. Rousset, J-F. Pujol, and D. Weissman. 1994. Postnatal development of the tyrosine hydoxylase-containing cell population within the rat locus coeruleus: Topological organization and phenotypic plasticity. J. Neurosci. 14: 7486–7501. Bickford-Wimer, P. C., J. A. Miller, R. Freedman, and G. M. Rose. 1988. Age related reduction in responses of rat hippocampal neurons to locally applied monoamines. Neurobiol. Aging 9: 173–179. Bickford, P. C., C. Heron, D. A. Young, G. A. Gerhardt, and R. De La Garza. 1992. Impaired acquisition of novel locomotor tasks in aged and norepinephrine-depleted F344 rats. Neurobiol. Aging 13: 475–481. Bickford, P. C. 1993. Motor learning deficits in aged rats are correlated with loss of cerebellar noradrenergic functions. Brain Res. 620: 133–138. Blanchard, V., R. Raisman-Vorazi, M. Savasta, E. Hirsch, F. Jovoy-Agid, C. Feuestein, and Y. Agid. 1993. Cellular quantification of tyrosine hydroxylase in the rat brain by immunoautoradiography. J. Neurochem. 61: 617–626. Bondareff, W., C. Q. Mountjoy, and M. Roth. 1982. Loss of neurons of origin of the adrenergic projection to cerebral cortex (nucleus locus coeruleus) in senile dementia. Neurology 32: 164–168. Chan-Palay, V., and E. Asan. 1989. Alterations in catecholamine neurons of the locus coeruleus in senile dementia of the Alzheimer type and in Parkinson’s disease with and without dementia and depression. J. Comp. Neurol. 287: 373–392. Collier, T. J., D. M. Gash, and J. R. Sladek. 1987. Norepinephrine deficiency and behavioral senescence in aged rats: Transplanted locus coeruleus neurons as an experimental replacement therapy. Ann. N. Y. Acad. Sci. 495: 396–403. Cotman, C. W., and G. S. Lynch. 1976. Reactive synaptogenesis in the adult nervous system. Pages 69–108. in S. H. Barondes, Ed., Neuronal Recognition. Chapman & Hall, London. Crutcher, K. A., and C. F. Marfurt. 1988. Non-regenerative axonal growth within the mature mammalian brain: Ultrastructural identification of sympathohippocampal sprouts. J. Neuroscience 8: 2289–2302.
534 13.
14. 15.
16.
17.
18.
19.
20. 21. 22.
23.
24.
25.
26.
27.
28.
29.
30.
SRIVASTAVA, GRANHOLM, AND GERHARDT Crutcher, K. A. 1990. Age-related decrease in sympathetic sprouting is primarily due to decreased target receptivity: Implications for understanding brain aging. Neurobiol. Aging 11: 175–183. Dahl, D., and A. Bignami. 1976. Immunogenic properties of the glial fibrillary acidic protein. Brain Res. 116: 150–157. Dahlstro¨m, A., and K. Fuxe. 1964. Evidence for the existence of monoamine-containing neurons in the central nervous system. I. Demonstration of monoamines in the cell bodies of brain stem neurons. Acta Physiol. Scand. 62(Suppl. 232): 1–55. Debure, L. I., E. Moyse, M. Fevre-Montagne, H. Hardin, M. F. Belin, C. Rousset, J. F. Pujol, and D. Weissmann. 1992. Somatotopic organization of tyrosine hydroxylase expression in the rat locus coeruleus: Long term effect of ZRU24722. Brain Res. 581: 19–32. Eriksdotter-Nilsson, M., G. A. Gerhardt, A. Seiger, B. J. Hoffer, and A-Ch. Granholm. 1989. Multiple changes in noradrenergic mechanisms in the coeruleo–hippocampal pathway during aging: Structural and functional correlates from intraocular double transplants. Neurobiol. Aging 10: 117–124. Eriksdotter-Nilsson, M., G. A. Gerhardt, A. Seiger, L. Olson, B. J. Hoffer, and A-Ch. Granholm. 1989. Age-related alterations in noradrenergic input to the hippocampal formation: Structural and functional studies on intraocular transplants. Brain Res. 478: 269–280. Eriksdotter-Jo¨nhagen, M., B. J. Hoffer, and J. Luthman. 1995. Alterations in alpha-adrenoceptors in aging intraocular hippocampal grafts. Neurobiol. Aging 16: 633–638. Fawcett, J. W. 1992. Intrinsic neuronal determinants of regeneration. Trends Neurosci. 15: 5–8. Finch, C. E. 1993. Neuron atrophy during aging: Programmed or sporadic. Trends Neurosci. 16: 104–110. Gavazzi, I., and T. Cowen. 1993. Axonal regeneration from transplanted sympathetic ganglia is not impaired by age. Exp. Neurol. 122: 57–64. Gavazzi, I. 1995. Collateral sprouting and responsiveness to nerve growth factor of aging neurons. Neurosci. Lett. 189: 47–50. Gerhardt, G. A., A. Oke, B. Moghaddam, and R. N. Adams. 1984. Nafion-coated electrodes with high selectivity for CNS electrochemistry. Brain Res. 290: 390–395. Gerhardt, G. A., M. R. Palmer, A. Seiger, R. N. Adams, L. Olson, and B. J. Hoffer. 1984. Adrenergic transmission in hippocampus– locus coeruleus double grafts in oculo: Demonstration by in vivo electrochemical detection. Brain Res. 306: 319–325. Gerhardt, G. A., M. R. Palmer, and A-Ch. Granholm. 1991. Age-induced changes in single locus coeruleus brain transplants grown in oculo: An in vivo electrochemical study. Neurobiol. Aging 12: 487–494. Gerhardt, G. A., W. A. Cass, J. Hudson, M. Henson, Z. Zhang, A. Ovadia, B. J. Hoffer, and D. M. Gash. 1996. In vivo electrochemical studies of dopamine overflow and clearance in the striatum of normal and MPTP-treated rhesus monkeys. J. Neurochem. 66: 579–588. Gerhardt, G. A. 1995. Rapid chronocoulometric measurements of norepinephrine overflow and clearance in CNS tissues. Pages 117–151 in A. Boulton, G. Baker, and R. N. Adams, Eds., Neuromethods, Vol. 27, Voltammetric Methods in Brain Systems. Humana Press, Clifton, NJ. German, D. C., B. S. Walker, K. Manaye, W. K. Smith, D. J. Woodward, and A. J. North. 1988. The human locus coeruleus: Computer reconstruction of cellular distribution. J. Neurosci. 8: 1776–1788. Goldman G., and P. D. Coleman. 1981. Neuron numbers in locus
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
coeruleus do not change with age in Fischer 344 rat. Neurobiol. Aging 2: 33–36. Granholm, A-Ch., G. A. Gerhardt, M. Eriksdotter-Nilsson, P. C. Bickford Wimer, M. R. Palmer, A. Seiger, L. Olson, and B. J. Hoffer. 1987. Age-related changes in cerebellar noradrenergic pre- and postsynaptic mechanisms: Intrinsic vs. extrinsic determinants evaluated with brain grafts in oculo. Brain Res. 423: 71–78. Granholm, A-Ch. 1991. Hippocampal transplants in oculo: A model for establishment of isolated circuits. Pages 327–345 in P. M. Conn, Ed., Methods in Neuroscience, Vol. 7. Academic Press, San Diego. Granholm, A-Ch., H. Bergman, E. Dudek, and M. Browing. 1995. Synapsin I in intraocular hippocampal transplants during maturation and aging: Effects of brainstem cografts. Cell Transplant. 4: 3–12. Greenberg, L. H., and B. Weiss. 1978. Beta-adrenergic receptors in aged rat brain: Reduced number and capacity of pineal gland to develop supersensitivity. Science 201: 61–63. Hall, M. E., B. J. Hoffer, and G. A. Gerhardt. 1989. Rapid and sensitive determination of catecholamines in small tissue samples by high pressure liquid chromatography coupled with electrochemical detection. LC/GC 7: 258–265. Iversen, I. L., M. N. Rosser, G. P. Reynolds, R. Hills, M. Roth, C. Q. Mountjoy, S. L. Foote, J. H. Morrison, and F. F. Bloom. 1983. Loss of pigmented dopamine-b hydroxylase positive cells from locus coeruleus in senile dementia of the Alzheimer’s type. Neurosci. Lett. 39: 95–100. Jones, R. S. G., and H. R. Olpe. 1983. Altered sensitivity of forebrain neurons to iontophoretically applied noradrenaline in aging rats. Neurobiol. Aging 4: 97–99. Kuchel, G. A. 1993. Alterations in target innervation and collateral sprouting in the aging sympathetic nervous system. Exp. Neurol. 124: 381–386. Leslie, F. M., S. E. Loughlin, D. B. Sternberg, J. L. McGaugh, L. Young, and S. F. Zornetzer. 1985. Noradrenergic changes and memory loss in aged mice. Brain Res. 359: 292–299. Loy, R., D. A. Kozielli, J. D. Lindsey, and R. Y. Moore. 1980. Noradrenergic innervation of the adult rat hippocampal formation. J. Comp. Neurol. 189: 699–710. Loy, R., T. A. Milner, and R. Y. Moore. 1980. Sprouting of sympathetic axons in the hippocampal formation: Conditions necessary to elicit ingrowth. Exp. Neurol. 67: 399–411. Madison, R., and J. N. Davis. 1983. Sprouting of noradrenergic fibers in hippocampus after medial septal lesions: Contributions of the central and peripheral nervous systems. Exp. Neurol. 80: 167–177. Manaye, K. F., D. D. MacIntire, D. M. A. Mann, and D. C. German. 1995. Locus coeruleus cell loss in the aging human brain: A non-random process. J. Comp. Neurol. 358: 79–87. McGaugh, J. L., K. C. Liang, C. Bennett, and D. B. Sternberg. 1984. Adrenergic influences on memory storage: Interaction of peripheral and central systems. Pages 313–332 in G. Lynch, J. L. McGaugh, and N. N. Weinberger, Eds., Neurobiology of Learning and Memory. Gilford Press, New York. McGeer, E. G., and P. L. McGeer. 1981. Neurotransmitters in the aging brain. Pages 631–648 in R. D. Terry and S. Gershon, Eds., The Molecular Basis of Neuropathology. Igaku-shoin, Tokyo. Nobin, A., and A. Bjo¨rklund. 1973. Monoamine neuron systems in the human brain as revealed in fetuses. Acta Physiol. Scand. 88(Suppl. 388):1–40. Olpe, H. R., and M. W. Steinmann. 1982. Age-related decline in the activity of noradrenergic neurons of the rat locus coeruleus. Brain Res. 251: 174–176.
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS 48.
Olszewski, J., and D. Baxter. 1954. Pages 170–173 in Cytoarchitecture of the Human Brain Stem. Lippincott, Philadelphia.
49.
Palmer, A. M., G. C. Stratmann, A. W. Procter, and D. M. Bowen. 1988. Possible neurotransmitter basis of behavioral changes in Alzheimer’s disease. Ann. Neurol. 23: 616–620.
50.
Palmer, A. M., and S. T. Dekosky. 1993. Monoamine neurons in aging and Alzheimer’s disease. J. Neural. Transm. 91: 135–159.
51.
Pickel, V. M., H. Krebs, and F. E. Bloom. 1973. Proliferation of norepinephrine-containing axons in rat cerebellar cortex after peduncular lesions. Brain Res. 59: 169–179. Pickel, V. M., M. Segal, and F. E. Bloom. 1974. Axonal proliferation following lesions of cerebellar peduncles: A combined fluorescence microscopic and radioautographic study. J. Comp. Neurol. 155: 43–60. Powers, R. E., R. G. Struble, M. F. Casanova, D. T. O’Connor, C. A. Kitt, and D. L. Price. 1988. Innervation of human hippocampus by noradrenergic systems: Normal anatomy and structural abnormalities in aging and Alzheimer’s disease. Neuroscience 25: 401–417. Richarde, F., N. Faucon-Biguet, R. Labatut, D. Rollet, J. Mallet, and M. Buda. 1988. Modulation of tyrosine hydroxylase gene expression in rat brain and adrenal by exposure to cold. J. Neurosci. Res. 260: 32–37. Scheff, S. W., L. S. Bernardo, and C. W. Cotman. 1978. Decrease in adrenergic axon sprouting in the senescent rat. Science 202: 775–778. Scheff, S. W., L. S. Bernado, and C. W. Cotman. 1980. Decline in reactive fiber growth in the dentate gyrus of aged rats compared
52.
53.
54.
55.
56.
57.
58. 59.
60. 61.
62.
63.
64.
65.
535
to young adult rats following entorhinal cortex removal. Brain Res. 199: 21–38. Schmitt, P., V. Reny-Palsse, O. Bourde, C. Garcia, and J. F. Pujol. 1993. Further characterization of the long-term effect of RU24722 on tyrosine hydroxylase in the rat locus coeruleus. J. Neurochem. 61: 1423–1429. Schneider, G. E. 1981. Early lesions and abnormal neuronal connections in mammals. Trends Neurosci. 4: 187–192. Sievers, J., I. Lolova, S. Jenner, H. P. Klemm, and H. Sievers. 1981. Morphological and biochemical studies on the ontogenesis of the nucleus locus coeruleus. Bibl. Anat. 19: 52–130. Squire, L. R. 1986. Mechanisms of memory. Science 232: 1612– 1619. Stenevi, U., and A. Bjo¨rklund. 1978. Growth of vascular sympathetic axons into the hippocampus after lesions of the septohippocampal pathway: Pitfall in brain lesion studies. Neurosci. Lett. 7: 219–224. Sturrock, R. R., and K. A. A. Rao. 1985. A quantitative histological study of neuronal loss from the locus coeruleus of ageing mice. Neuropathol. Appl. Neurobiol. 11: 55–60. Swanson, L. W. 1976. The locus coeruleus: A cytoarchitectonic, Golgi and immunohistochemical study in the albino rat. Brain Res. 110: 39–56. Tomlinson, B. E., D. Irving, and G. Blessed. 1981. Cell loss in the locus coeruleus in senile dementia of the Alzheimer type. J. Neurol. Sci. 49: 419–428. Zornetzer, S. F. 1985. Catecholamine system involvement in age-related memory dysfunction. Ann. N. Y. Acad. Sci. 444: 242–254.
145, 524–535 (1997)
EN976485
Collateral Sprouting of Central Noradrenergic Neurons during Aging: Histochemical and Neurochemical Studies in Intraocular Triple Transplants Nisha Srivastava,*,† Ann-Charlotte Granholm,*,†,‡,§ and Greg A. Gerhardt†,‡,§,¶ *Department of Basic Science, ¶Department of Psychiatry, and ‡Department of Pharmacology, §the Neuroscience Training Program, and †the Rocky Mountain Center for Sensor Technology, University of Colorado Health Sciences Center, Denver, Colorado 80262
The sprouting capacity of aged noradrenergic neurons of the brain-stem nucleus locus coeruleus (LC) was examined using intraocular transplants of fetal tissues. Fetal hippocampal tissue (E18) and LC tissue (E15) were transplanted together as a double transplant into the anterior chamber of the eye of young adult Fischer 344 rats. The double transplants were allowed to mature for 14–18 months, after which an additional fetal hippocampal transplant was placed next to the LC graft. The triple transplants were monitored for overall growth and vascularization for an additional 2–6 months. Immunohistochemical examinations showed that both young (2–6 months old) and aged (16–24 months old) hippocampal cografts contained a plexus of thin varicose tyrosine hydroxylase (TH)-immunoreactive fibers extending throughout the grafted hippocampal tissues. However, the aged hippocampal grafts contained a denser uniform plexus of TH-positive fibers compared to the young transplants. Immunohistochemistry with synapsin antibodies demonstrated that both the young and the aged hippocampal transplants contained much higher densities of synaptic elements than the LC grafts. In vivo electrochemical measurements of potassium-evoked overflow of norepinephrine (NE) in the grafts showed that similar amounts of NE overflow were detected in both the young and the aged hippocampal grafts. HPLC–EC measurements of NE levels in the grafts revealed that there were similar amounts of NE in the young and the aged grafts, and the grafts did not contain serotonin or dopamine. In summary, the findings of the present study show that aged LC neurons are capable of undergoing collateral sprouting producing a functional NE neuronal system when introduced to an appropriate young target. r 1997 Academic Press
INTRODUCTION
Aging is associated with localized, region-specific degenerative changes in the central nervous system (21). The process of neurodegeneration in aging or 0014-4886/97 $25.00 Copyright r 1997 by Academic Press All rights of reproduction in any form reserved.
disease has been associated with intrinsic changes in neurons, leading to loss of plasticity and/or alterations in the target tissues supplied by the neurons, such as reduced production of neurotrophic factors. In order to develop appropriate therapeutic treatments for neurodegenerative diseases during aging, the cause and effect of this neurodegeneration needs to be closely studied. The hippocampal formation has been widely used to study neuronal plasticity and target dependence of such plasticity (12, 40, 41, 55, 61). This region is particularly well suited for such studies due to the abundant knowledge of its anatomic and physiologic properties, as well as its function in memory and learning processes (11, 60). Several different transmitter-specific pathways innervating the hippocampal formation have been implicated in these learning and memory paradigms such as the cholinergic and the noradrenergic (NE) pathways (1, 34, 39, 44, 65). NE fibers in the hippocampus arise from distinct clusters of NE cell bodies in the pontine and medullary reticular formation. The largest group of NE neurons is located in a single nucleus, the locus coeruleus (LC), located beneath the floor of the fourth ventricle anterior to the facial colliculus, containing an estimated 1600 cells per hemisphere in the rat brain (29, 48, 63). This NE transmitter system represents a highly collateralized neuronal system, with axonal collaterals extending both rostrally into the forebrain, hippocampal formation, and cerebral cortex and caudally into the medulla and spinal cord (15, 46). The NE neurites from the LC reach the hippocampal formation by means of both a ventral and a dorsal pathway (40, 50). Although the phenotypic expression of these NE neurons appears early in embryonic development, the cell bodies undergo a significant histological reorganization until they have attained their definite position within the brain stem in the late postnatal period, as late as 3 weeks postnatally in the rat (3, 59). It has also been shown that plasticity of these neurons remains throughout adulthood. Pharmacological treatments (7, 16), physiological perturbations, or lesions (42, 54, 57) can
524
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS
give rise to enhanced TH expression in the LC, or to sprouting of NE neurites. However, it is not known whether NE neurons in the LC maintain this plasticity during aging. Numerous studies of the aged brain in man and animals have demonstrated a significant cell loss in the NE-containing neurons of the LC during aging (8–10, 36, 62, 64). In a recent study, Manaye and collaborators (43) demonstrated that this cell loss was significantly greater in the rostral than in the caudal NE cells in the LC of aged human brain. Thus, this represents a systematic rather than a random loss of cells in the nucleus (43). In addition, morphological abnormalities of the NE innervation of the hippocampal formation have been observed, both in normal aged humans and in patients with senile dementia of the Alzheimer’s type (53). In contrast to morphometric studies, neurochemical studies in the aged human brain have indicated a general sparing of noradrenergic indices in the target areas, with possible exception of the hippocampal formation (50). This suggests a regional difference in agerelated changes of NE mechanisms and also a potential high compensatory ability of these neurons. The sparing of NE levels and its metabolites could arise from compensation for NE denervation by increased NE turnover (49). In experimental animals, on the other hand, significant reductions in NE activity and innervation have been reported throughout the aged brain (45, 56), even though some studies have not shown any differences in LC cell numbers in aged animals (30). Furthermore, a number of other indices of NE function have been found to be decreased in the aged rat, such as postsynaptic a- and b-receptor responsivity (4, 37), behavioral response to NE agonists in the aged rat (6), and basal firing rates of LC neurons (47). In order to determine whether these changes are due to an age-related decrease in the function of the NE neurons of the LC or to changes in the hippocampal target, a model system in which the age of the target versus the age of the neuronal population could be varied independently is needed. Sequential double or triple grafting of tissues into the anterior chamber of the eye of the rat offers such a unique model system (17, 18, 26, 31). In several series of experiments we have previously demonstrated that many characteristics that appear during aging of the NE transmitter system of the LC in situ are duplicated in intraocular nervous tissue transplants. For example, we found that aged intraocular cerebellar (31) and hippocampal transplants (17, 18) develop significant postsynaptic subsensitivities to applied a-receptor agonists, suggesting a postsynaptic decrease in receptor affinity or density. A subsensitivity to a-adrenergic agonists was not observed in young transplants placed in aged hosts, indicating that the changes occurring to the adrenergic
525
receptors are due to factors within the transplanted CNS tissue, rather than system-related extrinsic factors in the animal such as hormone levels. This work has been further substantiated by a recent paper by Eriksdotter-Jo¨nhagen et al. (19) who found that postsynaptic binding of a-adrenergic ligands was reduced in aged compared to young hippocampal transplants in oculo. In addition, we have also shown that aged LC grafts in oculo exhibit a reduction in release amplitude as well as prolonged rise times for the potassiumevoked release of NE, compared to young transplants (26). Finally, the LC–NE pathway between hippocampal and brain-stem cografts in oculo was found to be significantly perturbed in aged double grafts, in terms of both pre- and postsynaptic properties (17). These experiments have demonstrated that there are both pre- and postsynaptic changes in NE neurotransmission during aging and that the intraocular transplant model represents a valid model system for studies of aging. In the present study we have continued the previous work by constructing triple transplants containing both young and aged hippocampal transplants in contact with an aged NE-containing brain-stem transplant. By varying the age of the transplanted tissues versus the host, it is possible to create so called ‘‘old–young chimeras,’’ in which intrinsic versus extrinsic determinants of aging can be investigated in detail. Thus, the same NE neurons can potentially innervate a young and an aged target simultaneously. The specific questions addressed in this study were: (1) Can aged LC neurons innervate a fetal target with NE fibers? (2) If innervated, are the densities and distributions of NE fibers and other neuronal markers the same in the young versus the aged hippocampal grafts? (3) Are NE release and synthesis the same in the young versus aged hippocampal tissues? MATERIALS AND METHODS
Animals. Young adult female Fischer 344 rats (5–6 weeks of age; Harlan Laboratories) were used as recipients for intraocular fetal grafts. Tissue donors were timed pregnant Fischer 344 dams of Embryonic Day 15–17 (E15–E17) for brain-stem tissue and E18 for hippocampal tissue. The recipient rats were sympathectomized prior to transplantation by means of surgical removal of the superior cervical ganglia (32). This procedure eliminates all sympathetic NE fibers in the host iris, which could potentially interfere with the central LC–NE innervation of target grafts in the eye. All animals were maintained on standard food pellets and water that were available ad libitum, and animals were kept on a standard 12:12-h light:dark cycle. Following surgery, the host rats were given paracetamol
526
SRIVASTAVA, GRANHOLM, AND GERHARDT
(6 mg/L) in their drinking water to alleviate any pain associated with the surgery. Dissections and transplantations. Pregnant rats were euthanized by an overdose of halothane; fetuses were removed and kept on ice until they were dissected. The fetal brain was removed, and brain-stem tissues containing the LC neurons and hippocampal CA1 tissue were dissected as previously described (32). Recipient host rats were anesthetized with Xylazine (Rompun, 12 mg/kg, ip) and Ketamine (Ketaset, 80 mg/kg, ip), and 1 drop of 1% atropine solution was placed on the cornea to dilate the pupil. Tissue pieces of fetal brain-stem (‘‘LC graft’’) and hippocampal (‘‘Hi1 graft’’) tissue were placed adjacent to each other into the anterior chamber of the eye through a razor blade slit in the cornea using a modified glass pipette as previously described (see 32). The LC grafts were first placed, followed by the Hi1 grafts 2–3 weeks later. The double transplants were allowed to mature for 14–18 months after which a second hippocampal tissue graft (‘‘Hi2 graft’’) was placed next to the brain-stem transplant. The triple transplants were allowed to grow for an additional 2–6 months in oculo. Thus, the LC and the Hi1 grafts were 16–24 months old, and the young Hi2 grafts were 2–6 months old at the time of study. See Fig. 1A for an outline of the experimental design. Immunohistochemical evaluation of tyrosine hydroxylase, neurofilament, glial fibrillary acidic protein, and synapsin I and II in the hippocampal and LC grafts. A number of immunohistochemical markers were used to investigate the intraocular triple transplants. Transplants (seven per group) were rapidly removed from the anterior eye chamber of urethane-anesthetized hosts and immersion fixed in 4% paraformaldehyde prepared in 0.1 M phosphate-buffered saline (PBS). After overnight fixation, the transplants were transferred to 10% sucrose in PBS for at least 16 h prior to sectioning on a cryostat. Twelve-micrometer-thick sections were collected on slides, washed in Sorensen’s buffer (0.1 M phosphate-buffered saline with pH 7.4), and incubated with antibodies directed against tyrosine hydroxylase (TH; 1:100; Eugene Tech Inc.), neurofilament (NF; 1:100; Dakopatts, Denmark), glial fibrillary acidic protein (GFAP; 1:100; see 14), or synapsin I and II (1:200; see 2) for 48 h. After washing in Sorensen’s buffer (3 3 10 min), the sections were incubated with antirabbit or anti-mouse IgG conjugated with fluorescein isothiocyanate (1:30) for 1 h at room temperature. The sections were washed again for 3 3 10 min and coverslipped with glycerol:PBS (9:1). Light-microscopical evaluations were performed on a Nikon Optiphot fluorescence microscope. The sections were devoid of immunoreactivity when the first antibody was omitted. In vivo electrochemical recordings. Host rats were anesthetized with urethane (1.25 g/kg, ip), tracheoto-
FIG. 1. (A) Schematic drawing outlining the experimental design of the triple brain grafts. Brain-stem (LC-NE) and hippocampal (Hi1) tissues from fetuses were transplanted to the anterior chamber of the eye of young adult Fischer 344 rats. The double transplants were allowed to mature for 14–18 months, after which a second hippocampal graft (Hi2) was placed next to the brain-stem graft. The triple transplants were evaluated through the translucent cornea for an additional 2–6 months. Thus, the aged hippocampal and brain-stem grafts were 16–24 months old, and the young hippocampal grafts were 2–6 months old at the time of sacrifice. (B) Photograph of a triple transplant in oculo. Blood vessels can be seen on the surface of all three transplants. No clear borders between the three grafts can be seen. (LC, brain stem; Hi1, aged hippocampal graft; Hi2, young hippocampal graft.) Scale bar represents 1 mm.
mized, and placed in a stereotaxic frame. The cornea overlying the transplants was removed and a plexiglas chamber was placed over the eye (24–26). The eye was continually superfused with Earle’s balanced salt solution maintained at 37°C. In vivo electrochemical recordings were performed in the hippocampal grafts using a high-speed chronoamperometric recording system (IVEC-10; Medical System Corp.). The working electrodes employed were the Nafion-coated single-carbon fiber type (30-µm o.d.), which were attached to single-barrel micropipettes with a tip separation of 250–300 µm (26). The carbon fiber electrodes were coated with Nafion and characterized for sensitivity and selectivity as previously described (24–28). The sensitivities, selectivities, and linearities of all recording electrodes were determined using 0.1–6.0 µM levels of NE prepared in pH 7.4 PBS with 250 µM ascorbic acid (26, 28). The high-speed chronoamperometric electrochemical measurements consisted of applying square-wave pulses of 0.0 to 10.55 V with respect to an Ag/AgCl reference electrode (in contact with the recording chamber) for 0.1 s and were repeated at 5 Hz. The resulting oxidation and
527
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS
reduction currents were digitally integrated during the final 80% of the recording periods. Data were averaged over 1-s epochs to improve the signal-to-noise ratios of the recordings. Only electrodes that exhibited selectivities of greater than 250:1 for NE versus ascorbic acid were used for the recording in the grafts. The average detection limits of all sensors ranged from 25 to 50 nanomolar levels of NE (signal-to-noise ratio of 3). The overflow of NE was elicited by local application of excess potassium via pressure ejection using micropipettes with an outer tip diameter of 10–15 µm (26). A typical experiment involved placement of the electrode recording assembly into regions of the young and aged hippocampal grafts. Electrochemical recordings were initiated and carried out for 5–10 min to establish a stable baseline. Once the signals were stable, the potassium-evoked overflow of NE was elicited by pressure ejection of 100–250 nl of a potassium solution (70 mM KCl, 79 mM NaCl, and 2.5 mM CaCl2, prepared in distilled water and adjusted to pH 7.3–7.4). Recordings were continued until the signals returned to baseline. The assembly was then moved to a new region of the grafts. The potassium-evoked overflow of NE was expressed quantitatively in terms of NE calibration curves determined in vitro for each electrode prior to experiments; the initial preexperiment calibration factors were used for data analysis. Only electrode placements that yielded signals greater than or equal to 50 nM NE following maximal potassium activation (26) were included in the final data analysis. In addition, the rise times of the signals (TR) and the 80% decay times of the signals (T80) were used as data analysis parameters (26, 27). All statistical comparisons of the NE overflow data from the young and aged transplant groups were made using a two-tailed Student’s t test. High-performance liquid chromatography with electrochemical detection (HPLC–EC). Hippocampal transplants from all groups (n 5 9 per group) were rapidly removed from the anterior eye chamber of anesthetized hosts, and the animals were euthanized by an overdose of urethane (2 g/kg, ip). The grafts were divided into Hi1 (16–24 months old—‘‘old’’) and Hi2 (2–6 months old—‘‘young’’) portions, weighed, and frozen on dry ice for HPLC–EC measurements. The whole-tissue weights of the hippocampal grafts averaged 2.1 6 0.4 mg for the young grafts and 2.3 6 0.5 mg for the aged grafts. The tissue levels of NE, dopamine (DA), and serotonin (5-HT) were measured in the transplants using previously reported HPLC–EC methods (35). Monoamine values were calculated as total nanograms per gram wet weight of tissue (ng/g), and statistical comparisons were made using a two-tailed Student’s t test.
RESULTS
Intraocular growth. The intraocular transplants survived well and became vascularized within a week from grafting (Fig. 1B). The hippocampal grafts (Hi1 and Hi2) grew extensively in oculo, while the brainstem grafts remained approximately the same size throughout the experiment. The surface area of LC grafts ranged from 1 to 2 mm2 after cessation of growth, and the hippocampal grafts were 4–8 mm2, regardless of whether the grafts were young (Hi2) or aged (Hi1). Thus, the age of the recipient rat did not determine growth rate of the fetal hippocampal tissue. Blood vessels could be seen on the surface of all three transplant groups. Both the Hi1 and the Hi2 grafts attached well to the brain-stem grafts and no distinct border could be seen between the three grafts (Fig. 1B). Morphology. Routine histological evaluation with cresyl violet staining revealed a distinct cell layer resembling the pyramidal cells in both hippocampal grafts, as well as the seamless integration of the three different tissues with each other (Fig. 2). The pyramidallike cell layer consisted of a two- to three-cell-thick, dense layer of large cells (approximate axial diameter perpendicular to the apical dendrites was 20–30 µm) that extended along the longitudinal axis of the grafts, a few hundred µm from the surface. There was no apparent difference in the morphological organization of the young and aged hippocampal grafts. Thus, the age of the recipient at the time of grafting does not appear to influence this parameter. The brain-stem grafts contained a single group of large neurons organized in a fashion similar to LC neurons in situ (Fig. 2). There were no signs of rejection in any of the grafted CNS tissues (n 5 7). Blood vessels could be found throughout the transplanted tissue (see Fig. 2). Immunohistochemistry. Clusters of large TH-positive cells were found in the LC portion of the transplants (Fig. 3). These TH-positive cells resembled LC neurons in situ and exhibited extensive dendritic arborizations in the vicinity of the cell bodies. Axons were thin and varicose and innervated the brain-stem tissue with a dense plexus of fibers (Fig. 4). A plexus of TH-like immunoreactive fibers could be seen innervating both the Hi2 (2–6 months old) and the Hi1 (16–24 months old) hippocampal transplants. The density of the THpositive fibers appeared greater in the Hi1 grafts, compared to the Hi2 grafts, as seen in Figs. 3, 4, and 5. This differential density was seen in all triple transplants examined and is summarized in Fig. 5. However, neurites could be found throughout both transplant groups. Club-like, larger end processes could sometimes be seen in the Hi2 (young) hippocampal grafts, indicative of active neuritic growth in the young target. The Hi2 grafts exhibited a patchy and gradient TH-
528
SRIVASTAVA, GRANHOLM, AND GERHARDT
FIG. 2. Cresyl violet-stained section of a triple transplant; Hi1, old (16-month) hippocampal graft; Hi2, young (2-month) hippocampal graft; LC, brain-stem graft. Note the pyramidal-like cell layer in both hippocampal grafts as well as the close attachment of the three different tissues with each other. The host iris can be seen attached to the ventral surface of the tissues. Scale bar represents 0.5 mm.
positive fiber distribution, with highest densities of fibers found close to the LC–hippocampal borders (Fig. 3). This could not be seen in the aged hippocampal grafts, which were found to contain a dense innervation pattern throughout the grafted tissue (Fig. 3). In Fig. 4, higher magnification micrographs from all three grafts seen in Fig. 3 are shown. As seen in Fig. 4, the individual fiber morphology appeared to be similar in the Hi1 and Hi2 grafts, despite the difference in the fiber density. The overall neuronal fiber network, shown with NF immunohistochemistry, was distributed through the transplants with no observable differences in terms of either density or distribution between the three graft groups (Fig. 6). The abundance of NF-like neurites in the grafts strongly suggests the presence of other innervation modalities, apart from the LC-NE neurons.
Also the brain-stem grafts contained a large number of NF-immunoreactive neurites (Fig. 6). These neurites differed from the TH-like neurites in their morphology, with both smooth, thick process and thin, varicose processes (Fig. 6). This further suggests that there are other neuronal populations in the grafts. A particularly dense NF-immunoreactive network of fibers could be found surrounding the neurons of the pyramidal-like cell layer in both hippocampal grafts (Fig. 6). Morphological studies using immunohistochemical methods with antibodies directed against synapsin (I and II) showed that both hippocampal transplants contained much higher densities of synaptic elements than the LC grafts (Fig. 7). The density of synapsinimmunoreactive profiles appeared to be somewhat greater in the old versus young hippocampal grafts. However, quantitative measurements were not per-
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS
529
FIG. 3. Photomontage of a section from a triple transplant, incubated with TH antibodies. Hippo1, old (16-month) hippocampal graft; Hippo2, young (2-month) hippocampal graft; LC, old brain-stem graft]. Note the difference in innervation density between the two hippocampal grafts. There are, however, TH-immunoreactive fibers extending throughout the length of both grafts. Scale bar represents 0.5 mm.
formed in the present studies. The distribution and density of glial elements was also examined in the triple transplants, using antibodies directed against GFAP (Fig. 8B). The Hi1 and brain-stem transplants exhibited a marked gliosis with a denser pattern of astrocytes than the young hippocampal grafts (Fig. 8B). No significant glial scarring could be seen in the borders between the brain-stem grafts and either Hi1 or Hi2 grafts. Finally, autofluorescent lipofuchsin granules were present in all of the aged transplants, but were not observed in any of the young hippocampal tissues (Fig. 8A). In vivo electrochemical recordings. In vivo electrochemical recordings of potassium-evoked overflow of NE were carried out in five young (2–6 months) and five aged (16–24 months) hippocampal grafts in order to investigate the functional properties of the NE fibers in the grafts (26). Potassium was seen to produce detectable electrochemical signals in both the young and the aged hippocampal grafts. The recorded oxidation and reduction currents of the signals supported the detection of NE signals and not the measurement of ascorbic acid, serotonin, or dopamine (26, 28). Interestingly, the
average NE-like signals were slightly but not significantly larger in the aged tissues (see Fig. 9A). The potassium-evoked electrochemical signals averaged 1.05 6 0.25 µM (NE equivalents, n 5 20 signals from five grafts) versus 0.60 6 0.10 µM in the young hippocampal tissues (n 5 20 releases from five grafts). The temporal properties of the potassium-evoked signals were also evaluated to investigate potential differences in NE release dynamics in both the young and the aged hippocampal grafts. The TR and the T80 of the potassium-evoked NE-like signals were also similar in the young and aged grafts. However, as seen in Fig. 9B, the temporal properties of the signals recorded in the young grafts tended to be longer than those recorded in the aged grafts. The TR and T80 times were 25 6 3 and 53 6 5 s (n 5 20), respectively, in the young hippocampal tissues versus 17 6 2 and 43 6 2 s (n 5 20), respectively, in the aged grafts. This trend was statistically significant for only the rise times of the signals (P , 0.01). Thus, the time courses and amplitudes of the potassium-evoked NE-like signals were similar in both the young and the aged tissues, indicating that the NE fibers in both sets of grafts were
FIG. 4. Micrographs at higher magnification showing details from the TH-incubated triple transplant section shown in Fig. 3, illustrating the density of innervation in the Hi1 graft (A), the LC graft (B), and the Hi2 graft (C). Note the sparse network of fibers in the Hi2 versus the Hi1 graft. Scale bar represents 35 µm.
530
SRIVASTAVA, GRANHOLM, AND GERHARDT
FIG. 5. Schematic drawing illustrating the overall TH-positive fiber distribution found in the Hi1 versus the Hi2 hippocampal transplants. This schematic drawing represents objective estimations from seven different grafts.
capable of supporting depolarization-induced release of NE. HPLC–EC measures of NE, DA, and 5-HT in the young and aged grafts. HPLC–EC measurements of the whole tissue levels of NE, DA, and 5-HT that were contained in the grafts were carried out in nine sets of young and aged tissues to investigate the neurochemical characteristics of the hippocampal grafts. All the hippocampal tissues contained large levels of NE, and there was no statistically significant difference between the NE levels in the young and aged hippocampal grafts. The average NE content of the young hippocam-
pal grafts was 1530 6 648 ng/g wet weight (n 5 9) and the NE content of the aged grafts was 1401 6 544 ng/g wet weight (n 5 9). These levels represent an approximately fourfold greater amount of NE in the grafts compared to the normal NE innervation of the hippocampus (5). In addition, there were no detectable levels of 5-HT in both sets of tissues (detection limit ,1 ng/gram wet weight of tissue). Moreover, only one set of grafts contained significant levels of DA (,100 ng/g wet weight). Thus, the young and aged hippocampal grafts were seen to contain high levels of NE without contamination from other monoamine neurotransmitters such as 5-HT and DA. These data also support the hypothesis that NE was the predominant monoamine that contributed to the potassium-evoked electrochemical signals. DISCUSSION
The present study is, to the best of our knowledge, the first study showing that LC neurons are capable of maintaining their regenerative capacity during aging in the rat. We found both morphological and neurochemical evidence for functioning noradrenergic terminals throughout the young hippocampal tissue that
FIG. 6. Sections of triple graft incubated with neurofilament antibodies. The overall pattern of nerve fibers was similar in the Hi1 (A) and the Hi2 (B) grafts. A dense plexus of nerve fibers could also be seen in the brain-stem graft (C) and around the pyramidal-like cell layer in the hippocampal grafts (D). The graft–graft border can be seen in (D). Scale bar in D represents 100 µm for A, B, and C, and 75 µm for D.
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS
FIG. 7. Synapsin immunoreactivity in intraocular triple transplants. The brain-stem LC tissue contained a sparse plexus of synapsin-positive profiles (the tissue seen on the right side in (C)), while the hippocampal grafts both exhibited dense aggregations of synapsin immunoreactivity. (A) Hi1 and (B) Hi2. Scale bar represents 50 µm.
531
was innervated by an old LC. This suggests that collateral sprouts had been generated from the aged LC neurons in the brain-stem grafts to innervate the new hippocampal target. The in vivo electrochemical recordings of potassium-evoked NE-like releases in both sets of grafts and the HPLC–EC measures of NE content in the grafts also support the hypothesis that aged NEcontaining neurons can innervate a young target with functional NE fibers. Thus, our histological and neurochemical data support the hypothesis that aged NEcontaining LC neurons retain the ability to innervate an appropriate target with functional NE fibers. There are contradicting results in the literature concerning neuronal plasticity during aging. A reduction in the collateral sprouting (38, 55), neuronal regeneration (20), and formation of new synapses (56) was reported in aged animals. In contrast, when aged sympathetic neurons were grafted in contact with young targets, they were found to retain a high degree of plasticity (13, 22). However, this maintenance of sprouting capacity could be a function of peripheral neurons only. In addition, most studies to date have been performed on aged animals in situ, a model system that does not allow for separation of target and neuron age independently. With the use of our intraocular grafting paradigm, we are able to vary the age of the neuron versus its target and its host independently, and thus determine the dependence of intrinsic versus extrinsic factors during aging. Our findings in the present study, in which we found that aged LC neurons retain the ability to reinnervate a fetal target, correlate well with a recent study by Gavazzi (23) showing identical results for aged peripheral sympathetic neurons. Thus, it is possible that both peripheral and central neurons retain a certain amount of plasticity during aging, when presented with appropriate targets, but that there is a limit to this pruning or collateral sprouting ability. The NE innervation in both the young and the old hippocampal grafts resembled that seen in young LC– hippocampal grafts (17) with a plexus of thin, varicose fibers. In a recent study, we found morphological alterations, such as axonal swellings and degenerative debris, in aged brain-stem grafts compared to young grafts (26). Since we did not detect such changes in the LC–hippocampal grafts in the present study, nor in a previous study of aged LC–hippocampal double grafts (17), it is possible that changes occur in single brainstem grafts during aging due to the lack of an appropriate target to innervate. We conclude from the morphological findings presented here that aged LC neurons have retained their ability to innervate a fetal target, and that the pattern of innervation resembles that seen in young grafts and in the hippocampal formation, even though the density is somewhat lower. In the present study, the potassium-evoked NE over-
532
SRIVASTAVA, GRANHOLM, AND GERHARDT
FIG. 8. (A) Autofluorescent lipofuchsin granulae in an aged hippocampal graft. (B) GFAP immunoreactivity at the border between the aged (Hi1) and the young (Hi2) hippocampal portion of a triple transplant. Note that the aged hippocampal tissue contains a significantly higher density of GFAP immunoreactivity than the young hippocampal tissue. Scale bar represents 50 µm.
flow data and the whole-tissue measures of NE levels also support the hypothesis that the old LC grafts innervate the young and aged hippocampal tissues with functional NE fibers. The signals recorded from the young grafts support the hypothesis that aged LC neurons can innervate a young host with functional NE fibers that are comparable to the NE innervation of the aged hippocampal grafts. The time courses and amplitudes of the NE responses recorded from the young and aged grafts were very similar. The differences in the
rise times of the signals recorded in the young hippocampal grafts, although significant, were small and do not reflect a major difference in the functional properties of the NE neurons in the young and aged tissues. However, in a previous study, we showed that potassiumevoked overflow of NE-like signals in aged single LC grafts in oculo were decreased compared to young grafts. In that study, we found major differences in the potassium-evoked NE signals recorded from young and aged brain-stem grafts (26). By contrast, the NE-
FIG. 9. In vivo electrochemical recordings of potassium-evoked NE overflow in the young and aged hippocampal grafts. (A) Average amplitudes of potassium-evoked NE overflow in young (2–6 months old) and aged (16–24 months old) hippocampal grafts. (B) Average rise times (TR) and 80% decay times (T80) of the potassium-evoked NE signals recorded from young and aged tissues. **P , 0.01 using a two-tailed Student’s t test.
533
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS
containing fibers in the present study were derived from the aged LC for both the young and the aged hippocampal grafts. Interestingly, the amplitudes of NE overflow recorded in the present study were greatly diminished compared to those recorded in young LC grafts (26). However, the temporal properties of the responses recorded in the present study are comparable to those recorded from the aged single LC grafts. Thus, although the old LC appears to have the capability to innervate a young graft with functional NE fibers, the functional properties of these NE fibers appear to retain aged NE release properties and they do not release NE analogous to young NE-containing fibers. There are several possible explanations why the aged LC neurons are capable of innervating the fetal hippocampal tissue after such a long time of maturation in the grafts. The most likely hypothesis would be that a centrally active adrenergic neurotrophic factor is released from the fetal hippocampal tissue, and thus attracts the ingrown collaterals from the LC graft. An alternative hypothesis is also possible. Previous work by others has shown that partial damage to the noradrenergic afferent fibers to cerebellum resulted in a proliferation of the remaining LC axons in this structure as well as in a nondenervated structure, the hippocampus (51, 52). This phenomenon has been called neuronal ‘‘pruning’’ (58) and was proposed to be a property of highly collateralized neuronal systems, such as the LC. If there are properties of the aged target tissue that are detrimental to the NE innervation, for example, lack of trophic or tropic support, perhaps the fetal tissue provides a more beneficial environment in which the LC neurons can retreat. Future studies, using for example electron microscopy in combination with intracellular labeling techniques, can be used to determine if there is an abandonment of synaptic sites in the aged hippocampal transplants, in favor of the young target tissue. However, the immunohistochemical labeling of synaptic proteins performed here does not indicate that this is the case, since a dense plexus of synaptic profiles was still present in the aged hippocampal target. On the other hand, previous work from our group (33) has shown that the levels of synapsin proteins in the hippocampal portion of aged LC–hippocampal double grafts are significantly decreased compared to the levels seen in young double grafts. This could indicate a functional reorganization of synaptic sites in the aged hippocampal grafts. In conclusion, we have shown that NE-containing LC neurons maintain plasticity during aging, since these neurons were able to reinnervate fetal hippocampal transplants that were introduced at the age of 14–18 months. These data have significant implications for development of therapeutic strategies in age-related neurodegenerative disorders. The next step will be to
try to identify the factor or factors that stimulate the regenerative capacity of LC neurons in the brain. ACKNOWLEDGMENTS We thank Justin Mott for expert technical assistance with the microphotography. We thank Dr. Doris Dahl, Department of Neuropathology, VA Medical Center, West Roxbury, Massachusetts, for supplying the GFAP antibody and Dr. Michael Browning, Department of Pharmacology, UCHSC, Denver, Colorado, for supplying the synapsin I and II antibodies. This work was supported by USPHS Grants AG12122 and MH49661 to A.C.G. and AG-06434 and NS09199 to G.A.G. Dr. Gerhardt received support from a Level II National Research Service Scientific Development Award (MH01245). Dr. Srivastava was supported by a fellowship from a Drug Abuse training grant (5T32AAO7464-20).
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Bartus, R. T., R. L., Dean, B. Beer, and A. S. Lippa. 1982. The cholinergic hypothesis of geriatric memory dysfunction. Science 217: 408–417. Bergman, H., M. Browning, and A-Ch. Granholm. 1992. Development of synapsin I and synapsin II in intraocular hippocampal transplants. Hippocampus 2: 339–348. Bezin, L., D. Marcel, L. I. Debure, N. Ginovart, C. Rousset, J-F. Pujol, and D. Weissman. 1994. Postnatal development of the tyrosine hydoxylase-containing cell population within the rat locus coeruleus: Topological organization and phenotypic plasticity. J. Neurosci. 14: 7486–7501. Bickford-Wimer, P. C., J. A. Miller, R. Freedman, and G. M. Rose. 1988. Age related reduction in responses of rat hippocampal neurons to locally applied monoamines. Neurobiol. Aging 9: 173–179. Bickford, P. C., C. Heron, D. A. Young, G. A. Gerhardt, and R. De La Garza. 1992. Impaired acquisition of novel locomotor tasks in aged and norepinephrine-depleted F344 rats. Neurobiol. Aging 13: 475–481. Bickford, P. C. 1993. Motor learning deficits in aged rats are correlated with loss of cerebellar noradrenergic functions. Brain Res. 620: 133–138. Blanchard, V., R. Raisman-Vorazi, M. Savasta, E. Hirsch, F. Jovoy-Agid, C. Feuestein, and Y. Agid. 1993. Cellular quantification of tyrosine hydroxylase in the rat brain by immunoautoradiography. J. Neurochem. 61: 617–626. Bondareff, W., C. Q. Mountjoy, and M. Roth. 1982. Loss of neurons of origin of the adrenergic projection to cerebral cortex (nucleus locus coeruleus) in senile dementia. Neurology 32: 164–168. Chan-Palay, V., and E. Asan. 1989. Alterations in catecholamine neurons of the locus coeruleus in senile dementia of the Alzheimer type and in Parkinson’s disease with and without dementia and depression. J. Comp. Neurol. 287: 373–392. Collier, T. J., D. M. Gash, and J. R. Sladek. 1987. Norepinephrine deficiency and behavioral senescence in aged rats: Transplanted locus coeruleus neurons as an experimental replacement therapy. Ann. N. Y. Acad. Sci. 495: 396–403. Cotman, C. W., and G. S. Lynch. 1976. Reactive synaptogenesis in the adult nervous system. Pages 69–108. in S. H. Barondes, Ed., Neuronal Recognition. Chapman & Hall, London. Crutcher, K. A., and C. F. Marfurt. 1988. Non-regenerative axonal growth within the mature mammalian brain: Ultrastructural identification of sympathohippocampal sprouts. J. Neuroscience 8: 2289–2302.
534 13.
14. 15.
16.
17.
18.
19.
20. 21. 22.
23.
24.
25.
26.
27.
28.
29.
30.
SRIVASTAVA, GRANHOLM, AND GERHARDT Crutcher, K. A. 1990. Age-related decrease in sympathetic sprouting is primarily due to decreased target receptivity: Implications for understanding brain aging. Neurobiol. Aging 11: 175–183. Dahl, D., and A. Bignami. 1976. Immunogenic properties of the glial fibrillary acidic protein. Brain Res. 116: 150–157. Dahlstro¨m, A., and K. Fuxe. 1964. Evidence for the existence of monoamine-containing neurons in the central nervous system. I. Demonstration of monoamines in the cell bodies of brain stem neurons. Acta Physiol. Scand. 62(Suppl. 232): 1–55. Debure, L. I., E. Moyse, M. Fevre-Montagne, H. Hardin, M. F. Belin, C. Rousset, J. F. Pujol, and D. Weissmann. 1992. Somatotopic organization of tyrosine hydroxylase expression in the rat locus coeruleus: Long term effect of ZRU24722. Brain Res. 581: 19–32. Eriksdotter-Nilsson, M., G. A. Gerhardt, A. Seiger, B. J. Hoffer, and A-Ch. Granholm. 1989. Multiple changes in noradrenergic mechanisms in the coeruleo–hippocampal pathway during aging: Structural and functional correlates from intraocular double transplants. Neurobiol. Aging 10: 117–124. Eriksdotter-Nilsson, M., G. A. Gerhardt, A. Seiger, L. Olson, B. J. Hoffer, and A-Ch. Granholm. 1989. Age-related alterations in noradrenergic input to the hippocampal formation: Structural and functional studies on intraocular transplants. Brain Res. 478: 269–280. Eriksdotter-Jo¨nhagen, M., B. J. Hoffer, and J. Luthman. 1995. Alterations in alpha-adrenoceptors in aging intraocular hippocampal grafts. Neurobiol. Aging 16: 633–638. Fawcett, J. W. 1992. Intrinsic neuronal determinants of regeneration. Trends Neurosci. 15: 5–8. Finch, C. E. 1993. Neuron atrophy during aging: Programmed or sporadic. Trends Neurosci. 16: 104–110. Gavazzi, I., and T. Cowen. 1993. Axonal regeneration from transplanted sympathetic ganglia is not impaired by age. Exp. Neurol. 122: 57–64. Gavazzi, I. 1995. Collateral sprouting and responsiveness to nerve growth factor of aging neurons. Neurosci. Lett. 189: 47–50. Gerhardt, G. A., A. Oke, B. Moghaddam, and R. N. Adams. 1984. Nafion-coated electrodes with high selectivity for CNS electrochemistry. Brain Res. 290: 390–395. Gerhardt, G. A., M. R. Palmer, A. Seiger, R. N. Adams, L. Olson, and B. J. Hoffer. 1984. Adrenergic transmission in hippocampus– locus coeruleus double grafts in oculo: Demonstration by in vivo electrochemical detection. Brain Res. 306: 319–325. Gerhardt, G. A., M. R. Palmer, and A-Ch. Granholm. 1991. Age-induced changes in single locus coeruleus brain transplants grown in oculo: An in vivo electrochemical study. Neurobiol. Aging 12: 487–494. Gerhardt, G. A., W. A. Cass, J. Hudson, M. Henson, Z. Zhang, A. Ovadia, B. J. Hoffer, and D. M. Gash. 1996. In vivo electrochemical studies of dopamine overflow and clearance in the striatum of normal and MPTP-treated rhesus monkeys. J. Neurochem. 66: 579–588. Gerhardt, G. A. 1995. Rapid chronocoulometric measurements of norepinephrine overflow and clearance in CNS tissues. Pages 117–151 in A. Boulton, G. Baker, and R. N. Adams, Eds., Neuromethods, Vol. 27, Voltammetric Methods in Brain Systems. Humana Press, Clifton, NJ. German, D. C., B. S. Walker, K. Manaye, W. K. Smith, D. J. Woodward, and A. J. North. 1988. The human locus coeruleus: Computer reconstruction of cellular distribution. J. Neurosci. 8: 1776–1788. Goldman G., and P. D. Coleman. 1981. Neuron numbers in locus
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
coeruleus do not change with age in Fischer 344 rat. Neurobiol. Aging 2: 33–36. Granholm, A-Ch., G. A. Gerhardt, M. Eriksdotter-Nilsson, P. C. Bickford Wimer, M. R. Palmer, A. Seiger, L. Olson, and B. J. Hoffer. 1987. Age-related changes in cerebellar noradrenergic pre- and postsynaptic mechanisms: Intrinsic vs. extrinsic determinants evaluated with brain grafts in oculo. Brain Res. 423: 71–78. Granholm, A-Ch. 1991. Hippocampal transplants in oculo: A model for establishment of isolated circuits. Pages 327–345 in P. M. Conn, Ed., Methods in Neuroscience, Vol. 7. Academic Press, San Diego. Granholm, A-Ch., H. Bergman, E. Dudek, and M. Browing. 1995. Synapsin I in intraocular hippocampal transplants during maturation and aging: Effects of brainstem cografts. Cell Transplant. 4: 3–12. Greenberg, L. H., and B. Weiss. 1978. Beta-adrenergic receptors in aged rat brain: Reduced number and capacity of pineal gland to develop supersensitivity. Science 201: 61–63. Hall, M. E., B. J. Hoffer, and G. A. Gerhardt. 1989. Rapid and sensitive determination of catecholamines in small tissue samples by high pressure liquid chromatography coupled with electrochemical detection. LC/GC 7: 258–265. Iversen, I. L., M. N. Rosser, G. P. Reynolds, R. Hills, M. Roth, C. Q. Mountjoy, S. L. Foote, J. H. Morrison, and F. F. Bloom. 1983. Loss of pigmented dopamine-b hydroxylase positive cells from locus coeruleus in senile dementia of the Alzheimer’s type. Neurosci. Lett. 39: 95–100. Jones, R. S. G., and H. R. Olpe. 1983. Altered sensitivity of forebrain neurons to iontophoretically applied noradrenaline in aging rats. Neurobiol. Aging 4: 97–99. Kuchel, G. A. 1993. Alterations in target innervation and collateral sprouting in the aging sympathetic nervous system. Exp. Neurol. 124: 381–386. Leslie, F. M., S. E. Loughlin, D. B. Sternberg, J. L. McGaugh, L. Young, and S. F. Zornetzer. 1985. Noradrenergic changes and memory loss in aged mice. Brain Res. 359: 292–299. Loy, R., D. A. Kozielli, J. D. Lindsey, and R. Y. Moore. 1980. Noradrenergic innervation of the adult rat hippocampal formation. J. Comp. Neurol. 189: 699–710. Loy, R., T. A. Milner, and R. Y. Moore. 1980. Sprouting of sympathetic axons in the hippocampal formation: Conditions necessary to elicit ingrowth. Exp. Neurol. 67: 399–411. Madison, R., and J. N. Davis. 1983. Sprouting of noradrenergic fibers in hippocampus after medial septal lesions: Contributions of the central and peripheral nervous systems. Exp. Neurol. 80: 167–177. Manaye, K. F., D. D. MacIntire, D. M. A. Mann, and D. C. German. 1995. Locus coeruleus cell loss in the aging human brain: A non-random process. J. Comp. Neurol. 358: 79–87. McGaugh, J. L., K. C. Liang, C. Bennett, and D. B. Sternberg. 1984. Adrenergic influences on memory storage: Interaction of peripheral and central systems. Pages 313–332 in G. Lynch, J. L. McGaugh, and N. N. Weinberger, Eds., Neurobiology of Learning and Memory. Gilford Press, New York. McGeer, E. G., and P. L. McGeer. 1981. Neurotransmitters in the aging brain. Pages 631–648 in R. D. Terry and S. Gershon, Eds., The Molecular Basis of Neuropathology. Igaku-shoin, Tokyo. Nobin, A., and A. Bjo¨rklund. 1973. Monoamine neuron systems in the human brain as revealed in fetuses. Acta Physiol. Scand. 88(Suppl. 388):1–40. Olpe, H. R., and M. W. Steinmann. 1982. Age-related decline in the activity of noradrenergic neurons of the rat locus coeruleus. Brain Res. 251: 174–176.
AGING IN INTRAOCULAR TRIPLE TRANSPLANTS 48.
Olszewski, J., and D. Baxter. 1954. Pages 170–173 in Cytoarchitecture of the Human Brain Stem. Lippincott, Philadelphia.
49.
Palmer, A. M., G. C. Stratmann, A. W. Procter, and D. M. Bowen. 1988. Possible neurotransmitter basis of behavioral changes in Alzheimer’s disease. Ann. Neurol. 23: 616–620.
50.
Palmer, A. M., and S. T. Dekosky. 1993. Monoamine neurons in aging and Alzheimer’s disease. J. Neural. Transm. 91: 135–159.
51.
Pickel, V. M., H. Krebs, and F. E. Bloom. 1973. Proliferation of norepinephrine-containing axons in rat cerebellar cortex after peduncular lesions. Brain Res. 59: 169–179. Pickel, V. M., M. Segal, and F. E. Bloom. 1974. Axonal proliferation following lesions of cerebellar peduncles: A combined fluorescence microscopic and radioautographic study. J. Comp. Neurol. 155: 43–60. Powers, R. E., R. G. Struble, M. F. Casanova, D. T. O’Connor, C. A. Kitt, and D. L. Price. 1988. Innervation of human hippocampus by noradrenergic systems: Normal anatomy and structural abnormalities in aging and Alzheimer’s disease. Neuroscience 25: 401–417. Richarde, F., N. Faucon-Biguet, R. Labatut, D. Rollet, J. Mallet, and M. Buda. 1988. Modulation of tyrosine hydroxylase gene expression in rat brain and adrenal by exposure to cold. J. Neurosci. Res. 260: 32–37. Scheff, S. W., L. S. Bernardo, and C. W. Cotman. 1978. Decrease in adrenergic axon sprouting in the senescent rat. Science 202: 775–778. Scheff, S. W., L. S. Bernado, and C. W. Cotman. 1980. Decline in reactive fiber growth in the dentate gyrus of aged rats compared
52.
53.
54.
55.
56.
57.
58. 59.
60. 61.
62.
63.
64.
65.
535
to young adult rats following entorhinal cortex removal. Brain Res. 199: 21–38. Schmitt, P., V. Reny-Palsse, O. Bourde, C. Garcia, and J. F. Pujol. 1993. Further characterization of the long-term effect of RU24722 on tyrosine hydroxylase in the rat locus coeruleus. J. Neurochem. 61: 1423–1429. Schneider, G. E. 1981. Early lesions and abnormal neuronal connections in mammals. Trends Neurosci. 4: 187–192. Sievers, J., I. Lolova, S. Jenner, H. P. Klemm, and H. Sievers. 1981. Morphological and biochemical studies on the ontogenesis of the nucleus locus coeruleus. Bibl. Anat. 19: 52–130. Squire, L. R. 1986. Mechanisms of memory. Science 232: 1612– 1619. Stenevi, U., and A. Bjo¨rklund. 1978. Growth of vascular sympathetic axons into the hippocampus after lesions of the septohippocampal pathway: Pitfall in brain lesion studies. Neurosci. Lett. 7: 219–224. Sturrock, R. R., and K. A. A. Rao. 1985. A quantitative histological study of neuronal loss from the locus coeruleus of ageing mice. Neuropathol. Appl. Neurobiol. 11: 55–60. Swanson, L. W. 1976. The locus coeruleus: A cytoarchitectonic, Golgi and immunohistochemical study in the albino rat. Brain Res. 110: 39–56. Tomlinson, B. E., D. Irving, and G. Blessed. 1981. Cell loss in the locus coeruleus in senile dementia of the Alzheimer type. J. Neurol. Sci. 49: 419–428. Zornetzer, S. F. 1985. Catecholamine system involvement in age-related memory dysfunction. Ann. N. Y. Acad. Sci. 444: 242–254.