- Email: [email protected]
Cyclic nucleotide distribution within rat striatonigral neurons
Neuroscience Vol. 9, No. 1, pp. 23-29, 1983 Printed in Great Britain
0306-4522/83/010023-07$03.00/0
PergamonFWssLtd 0 1983IBRO
CYCLIC NUCLEOTIDE DISTRIBUTION WITHIN RAT STRIATONIGRAL NEURONS MARJORIEA. ARIANO and SUSANK. UFKES Department of Anatomy & Neurobiology, University of Vermont College of Medicine, Burglington, VT 05405, U.S.A. Abstraet-Adenosine cyclic 3’,5’-monophosphate and guanosine cyclic 3’,5’-monophosphate have differential immtmohistochemical distributions within retrogradely-labeled striatonigral neurons of the rat. Adenosine cyclic 3’,5’-monophosphate is localized within more than half of the striatonigral projection neurons. It is also within tbe cytoplasm of other neurons and oligodendroglia. Guanosine cyclic 3’,5’-monophosphate is localized within 80% of the identified striatonigral neurons. These large percentages of cyclic nucleotide immunoreactivity within the striatonigral neurons suggest some of these efferent cells must contain both cyclic nucleotides. The immunofluorescent staining for guanosine cyclic 3’,5’-monophosphate is almost identical to that reported for efferent neurotransmitter-containing neurons of the caudate nucleus. However, the large proportion of striatonigral neurons demonstrating guanosine cyclic 3’,5’-monophosphate immunoreactivity precludes the association of this cyclic nucleotide with a selective neurotransmitter agent. Adenosine cyclic 3’,5’-monophosphate-reactive elements are very different in staining appearance from guanosine cyclic 3’,5’-monophosphate and neurotransmgter-identified somata. The number of striatonigral cells exhibiting reaction for this cyclic nucleotide does not eliminate the possibility that adenosine cyclic 3’,5’-monophosphate might be preferentially co-localized with a specific neurotransmitter, such as y-aminobutyrate, as has been previously suggested through biochemical experimentation.
fluorescent dyes, 32horseradish peroxidase,9*‘5 wheatgerm agglutinin conjugated to horseradish peroxidase,’ Herpes simplex virus6 and following intracellular injection of horseradish peroxidase after electrophysiological identification of individual 1~11s.~~The cytoarchitecture of this efferent population falls into the generic category of the medium spiny neuron of the caudite nucleus, comprising some 90% of the neuronal classification of the rat striatum.” These striatonigral neurons have densely spine-laden dendrites, a centrally placed nucleus and are rounded or oval in somal shape. Ultrastructural examination illustrates a thin rim of cytoplasm, poor in organelles, with sparse axosomatic symmetrical synapses and a smooth nuclear membrane.’ These projection neurons to the substantia nigra contain the neurotransmitters substance p2’ and y-aminobutyrate (GABA).19 In addition, a decrement in dopaminesensitive adenylate cyclase activity in the substantia nigra has been correlated to loss of the striatonigral efferents.‘4*27 The intent of the present study has been to investigate whether one particular cyclic nucleotide is preferentially associated with the striatonigral pathway and further to assess if the individual cyclic nucleotides co-localize with specific neurotransmitter substances. Fluorescent retrograde labeling is compatible with subsequent immunofluorescent characterization of the defined neuronal population,26*30and has demonstrated a large coincident localization of cyclic GMP within the striatonigral projection neurons.
The cyclic nucleotides, adenosine cyclic 3’,5’-monophosphate (cyclic AMP) and guanosine cyclic 3’S’-monophosphate (cyclic GMP), have been implicated as potential modulators of neural transmission. Our previous investigations have shown that these two compounds have a heterogeneous cellular localization and are evenly distributed throughout the rat striatum using immunohistochemical analysis. ‘9’The cyclic nucleotide-contaiuing neurons have been classified as belonging to the medium spiny group” on the basis of their somal diameter, perikaryal shape and ultrastructural characteristics5 However, it is not known if cyclic AMP or cyclic GMP can be linked to a specific group of striatal projection neurons or to a specific class of neurotransmitter synthesizing cells. The immunohistochemical detection of enkephalin-reactive neurons,23 substance P-reactive neurons,“+29 and glutamate decarboxylase-containing neurons25 within the caudate nucleus shows great similarity to one another and are also grouped into the medium spiny neuron class. In an attempt to characterize further the neuronal elements that show cyclic nucleotide immunoreactivity, we have identified specific efferent perikarya through retrograde transport of Evans blue.’ One of the primary efferent target structures of the striatum is the substantia nigra.‘7J**3’ The cellular morphology of the striatonigral neurons has been examined following retrograde transport of Abbreviations: Cyclic AMP, adenosine cyclic 3’,5’-monophosphate; cyclic GMP, guanosine cyclic 3’,5’-monophosphate; GABA, y-aminobutyrate. 23
24
M . A . ARIANO and S. K. UFKES EXPERIMENTAL PROCEDURES
All experiments were performed using male Sprague-Dawley rats (200-250 g). The animals were mainrained under conditions of constant temperature (22°C), relative humidity (50~) and an illumination cycle of 12 h. The rats were anesthetized with sodium pentobarbital (35 mg/kg) and injected with the dye Evans blue (Sigma Chemical Co., St. Louis, MO). Stereotaxic delivery of 0.1/~1 40~ Evans blue in distilled water was made into the substantia nigra using the atlas of Pellegrino, Pellegrino and Cushman. 22 The co-ordinates used for the substantia nigra were: A, -3.1; L, 2.5; and V, -8.6, down from the cortical surface. All injections were made unilaterally using a 1 #1 Hamilton syringe. Seventy-two hours following the Evans blue infusion, animals were re-anesthetized and transcardially perfused with approx 200 ml of cold 4 ~ paraformaldehyde, 0.5~ glutaraldehyde, 0.25 M sucrose in phosphate-buffered saline, pH 7.2. Brains were removed and placed into fresh cold fixative for 1 h, then transferred to cold 30~ sucrose in phosphate-buffered saline overnight. The brains were cut into coronal slabs at the level of the hypophyseal stalk, mounted and frozen at - 2 5 ° C onto brass cryostat chucks until sectioned. Eight #m-thick coronal sections were thaw-mounted onto chrom alum-coated slides and processed for cyclic nucleotide immunohistochemistry according to the method of Ariano, Butcher, and Appleman. 3 Antibody dilutions were 1:100 in phosphate-buffered saline for 1 h at 4°C or 1:500 at 4°C overnight. Antibody characterization and specificity has been previously determined by histochemical methods 5 and using radio-immunoassay these antibodies exhibit low to negligible recognition of their nucleoside triphosphate, diphosphate and monophosphate in competitive binding assay and a 200-fold greater recognition for the appropriate antigen cyclic nucleotide than the other cyclic nucleotide) ° Fluorescein-conjugated secondary immunoglobulins were purchased from Cappel Laboratories (Cochranville, PA) and used at 1:50 dilution in phosphate-buffered saline for I h at 4°C.
Following immunohistochemical processing tissue sections were examined with ultraviolet optics using a Zeiss Photomicroscope 3 equipped with 546 nm primary fiilter/590 nm secondary filter to visualize the Evans blue label and 485 nm primary filter/520 nm secondary filter to visualize the fluorescein-labeled elements. Controls included: (1) examination of contralateral caudate-putamen complex from the nigral injection site; (2) examination of striatal tissue prior to immuno-histochemical processing; and (3) comparison with non-infused caudate-putamen tissue. Photomicrography was performed using HP-5 black and white film (Ilford, Inc., Paramus, N J), pushed one stop and developed using D-19 (Kodak, Inc., Rochester, NY) to increase contrast of the negative. Exposures were made at constant magnification of x 100 and photographic enlargement such that each Evans blue-cyclic nucleotide matched
SN c
SN r
Fig. 1. Coronal section of rat brain illustrating the area of Evans blue injection. The black area indicates the most heavily stained region at the center of the injection site. The hatched area indicates the maximum extent and spread of the Evans blue for the experimental animals. SN r = pars reticulata of the substantia nigra; SNc = pars compacta; CP = cerebral peduncle; ML = medial lemniscus. pair was equivalent in area to 80 mm 2. Regions of densest Evans blue-labeled neurons were chosen for analysis, photographed using the 546/590nm filter combination, then examined with the 485/520 nm fluorescein filter combination. Coincident localization for each cyclic nucleotide in striatonigral neurons was numerically tabulated and averaged. RESULTS T h e injection site a n d extent o f diffusion o f E v a n s blue is illustrated in Fig. 1. M o s t o f the E v a n s blue was c o n t a i n e d within the s u b s t a n t i a nigra, b o t h pars c o m p a c t a a n d pars reticulata, with some spread ventrally into the cerebral peduncle a n d dorsally to e n c r o a c h u p o n the medial lemniscus a n d z o n a incerta. Only those cases in which retrograde labeling of s o m a t a within the ipsilateral c a u d a t e - p u t a m e n complex was f o u n d were used in this analysis. T h e brains o f 9 animals were utilized in this study. R a n d o m coronal sections t h r o u g h the rostral portions o f the rat s t r i a t u m were e x a m i n e d p r i o r to cyclic A M P or cyclic G M P i m m u n o h i s t o c h e m i c a l processing. T h e m o r p h o l o g y o f the brilliant red, E v a n s blue-positive perikarya agreed with previously p u b lished accounts o f striatonigral neurons. 8,9,32 T h e vast majority o f n e u r o n s were medium-sized, 10-20 # m in diameter, ovoid or r o u n d e d in shape a n d the nucleus filled m o s t of the cytoplasm. The cytoplasm was intensely fluorescent with substantial lengths o f neuronal processes evident e m a n a t i n g from the cell soma. T h e nucleus o f E v a n s blue-identified striatonigral n e u r o n s was generally u n s t a i n e d (see Figs 2A a n d 3A). W i t h i n o u r experimental sample o f
Fig. 2. Paired photomicrograph set from a representative region of the caudate-putamen, ipsilateral to the Evans blue injection into the substantia nigra. Calibration bar, 50#m. A. Evans blue-labeled striatonigral neurons within the caudate--putamen visualized with 546--590 nm filter combination. Arrows demonstrate typical reactive somata, fb denotes a fiber bundle. B. Visualization of cyclic AMP-containing elements within the same field as A through use of the 485-520 nm filter combination for fluorescein excitation. The same striatonigral neurons are shown with arrows, fb is the same fiber bundle. Fig. 3. Paired photomicrograph set from a representative region of the caudate-putamen, ipsilateral to the Evans blue injection into the substantia nigra. Calibration bar, 50#m. A. Evans blue-labeled striatonigral neurons within the caudate-putamen, visualized with the 546/590 nm filter combination. Arrows demonstrate three typical reactive somata, fb denotes a fiber bundle. B. Visualization of cyclic GMP-containing elements within the same field as A through use of the 485/520 nm filter combination for fluorescein excitation. The same striatonigral neurons are shown with arrows, fb is the same fiber bundle.
Cyclic AMP and cyclic GMP in striatonigral neurons
27
Table 1. Striatonigral neurons*
Cyclic AMP-containing
Cyclic GMP-containing
54%
81.5%
Striatonigral neurons stained?
*Total issue area examined was 31.4 mm’; each paired photomicrograph set was 80 mm2 x 8 pm tissue section thickness x49 different photomicrograph sets. tTota1 number examined was 3674 neurons from 9 different animals.
31.4 mm3 of caudate-putamen tissue, 3674 striatonigral neurons were unequivocally identified and examined. Cyclic AMP immunohistochemical localization was assessed on random sections throughout the rostra1 striatum. The perikarya and processes of cyclic AMP-stained elements were as described previously. 3,s Uniform immunofluorescence was present throughout the cytoplasm and nucleus of reactive cells (Fig. 2B). Somal dimensions of 8-16pm predominated. More than half of the Evans blue-labeled striatonigral neurons contained cyclic AMP immunoreactivity (Table 1). The cytoarchitecture of striatonigral neurons containing cyclic AMP is illustrated in Fig. 2. The total number of cyclic AMP-containing elements analyzed was 3203 and of that number, less than 30% could be identified as striatonigral neurons (Table 2). Other cyclic AMP-containing elements were identified as oligodendroglia surrounding and within fiber bundles perforating the substance of the caudate-putamen complex. Cyclic AMP-reactive glia were small in diameter (610 pm) and very intensely fluorescent. Cyclic GMP immunohistochemical localization was determined on sections serial to those of cyclic AMP-stained slides within the rostra1 striatum. The cytoarchitecture of cyclic GMP-immunoreactive cells was a previously described.4*5 Predominant staining was within the cytoplasm and proximal processes of cells 15-20 pm in diameter and a lack of nuclear staining was characteristic (Fig. 3B). A minor immunoreactive element was the course fibrous processes of astrocytes.3s4 More than 80% of Evans blue-labeled striatonigral neurons contained cyclic GMPimmunoreactivity (Table 1). The immunofluorescence pattern of staining in striatonigral cyclic GMP-positive neurons is demonstrated in Fig. 3. The total number of cyclic GMP-immunoreactive elements examined and analyzed was 3512 and almost 40% could be identified as components of the striatonigral efferent population (Table 2). Other cyclic Table 2. Cyclic nucleotide-labeled Cyclic AMP-positive? Cyclic GMP-positive1
cells*
EB-labeled neurons
Other cells
28.9% 38.5%
71.1% 61.5%
*Total tissue area examined was 3 1.4 mm3, from 49 different photomicrograph sets. tTota1 cyclic AMP-stained elements was 3203 from 9 different animals. ITotal cyclic GMP-stained elements was 3512 from 9 different animals.
GMP-staining elements were neuronal in nature, but not a part of this efferent system of the caudate-putamen complex to the substantia nigra. DISCUSSION
The cyclic nucleotide distribution within identified rat striatonigral neurons has been examined. The compatibility of fluorescent retrograde labeling with immunohistochemical characterization of the caudate projection neurons has demonstrated that these potential second messengers are not organized solely within this specific efferent cell group. The extremely large percentage of cyclic GMP-positive striatonigral neurons suggests that some of these efferent projection cells must contain both cyclic AMP and cyclic GMP immunohistochemically visualized pools. The greater number of cyclic GMP-stained striatonigral neurons also suggests that this cyclic nucleotide cannot be associated exclusively with only the substance P or solely the GABA efferents to the substantia nigra, but is probably co-localized with both bioactive substances. The report of dissociated topography for these two neurotransmitters within the caudate nucleus’6 adds support to this notion. It is intriguing that the immunochemical staining characteristics of GAD-positive neurons,2s*29 substance P-like reactive cell~,“*~~ and enkephalin-positive somata2) within the striatum have a coincident cellular localization with the cyclic GMP-stained neurons visualized in this study. This suggests that the antigenie determinants for these neurotransmitters and cyclic GMP could be within equivalent cellular organelles. This point awaits confirmation at the subcellular level of resolution. Cyclic AMP immunofluorescent elements exhibit very different staining characteristics when compared to cyclic GMP or the reported neurotransmittersynthesizing cells. Nuclear staining within Evans blue identified striatonigral neuronal somata was characteristic. Slightly more than half of the striatonigral neurons examined in this study showed coincident cyclic AMP-immunoreactivity. Although this does show cyclic AMP is not exclusively associated with the striatonigral projection cells, the cyclic nucleotide may be more closely associated with one of the two identified neurotransmitters utilized by this pathway. Biochemical studies have demonstrated a positive correlation between loss of GAD activity and a decrement in the dopamine-sensitive adenylate cyclase activity in the substantia nigra after interruption of the striatonigral pathway.27 However, the large
28
M. A. ARIANOand S. K.
majority of cyclic AMP-stained elements within the caudate nucleus are not part of the projection system to the substantia nigra. They are oligodendroglia and neurons of the other categories, e.g. interneurons or alternate efferent cells. The predominant energy source for the brain is glucose’* and the role of cyclic AMP in intermediary metabolism has been well studied. However, the immunohistochemically-detected cellular distribution of cyclic AMP does not reflect the ubiquitous function it has in cerebral glycolysis. It may be that the cellular pool of this cyclic nucleotide which can be visualized following immunohistochemical analysis is not directly related to energy metabolism, but subserves another function. The current methodologies will not allow us to address this question appropriately. The ability physiologically to manipulate the visualized cyclic nucleotide pool experimentally is also critical. We know from other studies and our own work, that the concentration of cyclic nucleotides can be selectively increased following synaptic activity. In the experimental system of the rat superior cervical ganglion,’ the elevations in cyclic AMP and cyclic GMP following synaptic activity show cellular selectivity. The cyclic GMP response is confined to postganglionic neurons while the cyclic AMP response is primary associated with the glial-satellite cells. It still needs to be determined if the immunohistochemically visualized neuronal staining of the cyclic nucleotides can be modulated by synaptic activity within the caudate-putamen complex. The selective in vivo acti-
UFKFS
vation of striatal afferents has not yielded consistent cytochemical elevations in cyclic AMP and cyclic GMP in the caudate nucleus which can be localized to defined cell types (unpublished observations). We have shown that selective deafferentation3 does not alter the cellular distribution or pattern of staining of the striatal cyclic AMP and cyclic GMP. The application of antibodies directed against the synthetic enzyme for cyclic GMP, guanylate cyclase (EC 4.6.1.2): demonstrates that this protein exhibits a neuronal localization. It may be that the immunohistochemically-detected cyclic GMP seen in low proportions within fibrous astrocytic endfeet in the striatum has been circumstantially accumulated within that cellular structure in response to production and subsequent diffusion from its active physiological site of action, within neuronal somata and at synaptic terminals. The neuronal distribution of the protein kinase mediating some of the intracellular functions of cyclic GMP (EC 2.7.1.38) would tend to support this contention. ‘**’Cyclic GMP tissue concentrations are usually quite low and two orders of magnitude less than cyclic AMP in the caudate nucleus as determined using radioimmunoassay.’ The selective neuronal elevations in this cyclic nucleotide following synaptic activity’ and the exclusive neuronal distribution of the cyclic GMP system in the striatum are a strong suggestion that it has an important function in neuronal communication. Acknowledgements-This
work was supported
by BNS
81-02648.
REFERENCES 1. Ariano M. A. (1982) Ultrastructural distribution of the cyclic GMP system: a role in neurotransmission in the rat striatum. Sot. Neurosci. 8, 344. 2. Ariano M. A., Briggs C. A. and M&fee D. A. (1982) Cellular localization of cyclic nucleotide changes in rat superior cervical ganglion. Cell Molec. Neurobiol. 2, 143-156. 3. Ariano M. A., Butcher L. L. and Appleman M. M. (1980) Cyclic nucleotides in the rat caudate-putamen complex: histochemical characterization and &ects of deafferentation and kainic acid infusion. Neuroscience 6, 1269-1276. 4. Ariano M. A., Lewicki J. A., Brandwein H. J. and Murad F. (1982) Immuuol&tochemical localizati& of guanytie cyclase within neurons of rat brain. Proc. nutn Acud. Sci. U.S.A. 79, 13161320. Ariano M. A., and Matus A. I. (1981) Ultrastructural localization of cyclic GMP and cyclic AMP in rat striatum. J. Cell Biol. 91. 287-292.
9.
10. 11. 12. 13. 14.
Bak I. J., Markham C. H., Cook M. L. and Stevens J. G. (1978) Ultrastructural and immunoperoxidase study of striatonigral neurons by means of retrograde axonal transport of Herpes simplex virus. Brain Res. i43, 361-369. Bentivoglio M., Kuypers H. G. J. M., Catsman-Berrevoets C. E. and Dann 0. (1979) Fluorescent retrograde neuronal labeling in rat by means of substances binding specifically to adenine-thymine rich DNA. Neurosci. Lerr. 1%.235-245. Bolam J. P., Powell J. F., Totterdell S. and Smith A. D. (1981) The proportion of neurons in the rat neostri#tum that project to the substantia nigra demonstrated using horseradish peroxidase conjugated with wheatgerm agghztinin. B@J Res. 220, 339-343. Bolam J. P., Somogyi P., Totterdell S. and Smith A. D. (1981) A second type of striatonigral neuron: a comparison between retrogradely-labelled and Golgi-stained neurons at the light- and electron-microscopic levels. Neuroscience 6, 2141-2151. Briggs C. A., Whiting G., Ariano M. A. and McAfee D. A. (1982) Cyclic nucleotide metabolism in the sympathetic ganglion. Cell Molec. Neurobibl. 2, 129-142. Cue110A. C. and Kanazawa I. (1978) The distribution of substance P immunoreactive fibers in the rat central nervous system. J. camp. Neural. 178, 129-156. DiFiglia M., Pasik P. and Pasik T. (1976) A Golgi study of neuronal types in the neostriatum of monkeys. Blrain Res. 114, 245-256. Dimova R., Vuillet J. and Seite R. (1980) Study of the rat neostriatum using a combined Golgi-etectron microscope tedmique and &al secticmb. Neuroscisncc 5, 1581-1596. Gale K., Guidotti A. and Costa E. (1977) Dopamine-sensitive adenylate cyclase: location in substantia nigra. Science, N.Y. 195, 503-505.
Cyclic AMP and cyclic GMP in striatonigral neurons
29
15. Grofova I. (1975) The identification of striatal and paltidal neurons projecting to substantia nigra. An experimental study by means of retrograde axonal transport of homemdish peroxidam. Brain Res. 91, 286-291. 16. Jesse11T. M., Emson P. C., Paxinos G. and Cuello A. C. (1978) Topographic projections of substance P and GABA pathways in the striato- and pallido-nigral system: a biochemical and immunohistochemical study. Brain Res. 152, 487498. 17. Kemp J. M. (1970) The termination of strio-pallidal and strio-nigral fibers. Brain Res. 17, 125-128. 18. Kety S. S. (1975) The general metabolism of the brain in I&JO.In Z7reh4etabolism of the Neruous System (ed. Richter D.) Pergamon Press, London. 19. Kim J.-S., Bak I. J., Hassler R. and Okada Y. (1971) Role of y-aminobutyric acid (GABA) in the extra-pyramidal motor system-II. Some evidence for the existence of a type of GABA-rich strio-nigral neurons. Expl Bruin Res. 14, 95-104. 20. Lohmann S. M., Walter U., Miller P. E., Greengard P. and DeCamilli P. (1981) Immunohistochemical localization of cyclic GMPdependent protein kinase in mammalian brain. Proc. natn Acad. Sci. U.S.A. 78, 653-657. 21. Mroz E. A., Brownstein M. J. and Leeman S. E. (1977) Evidence for substance P in the striato-nigral tract. Brain Res. m, 305-311. 22. Pellegrino L. S., Pellegrino A. S. and Cushman A. J. (1979) A Stereotaxic Atlas ofthe Rat Bruin. Plenum Press, New York. 23. Pickel V. M. Sumal K. K., Beckley S. C., Miller R. J. and Reis D. J. (1980) Immunocytochemical localization of enkephahn in the neostriatum of rat brain: a light- and electron-microscopic study. J. camp. Neural. 189, 721-740. 24. Preston R. J., Bishop G. A. and Kitai S. T. (1980) Medium spiny neuron projection from the rat striatum: an intracellular horseradish peroxidase study. Brain Res. 183, 253-263. 25. Ribak C. E., Vaughn J. E. and Roberts E. (1979) The GABA neurons and their axon terminals in rat corpus striatum as demonstrated by GAD immunocytochemistry. J. camp. Neural. 187, 261-284. 26. Sawchenko P. E. and Swanson L. W. (1981) A method for tracing biochemicallydetined pathways in the central nervous system using combined fluorescence retrograde transport and immunohistochemical techniques. Brain Res. 210, 31-51. 27. Spano P. F., Trabucchi M. and DiChiara G. (1977) Localization of nigral dopamine-sensitive adenylate cyclase on neurons originating from the corpus striatum. Science N.Y. I%, 1343-1345. 28. Szabo J. (1962) Topical distribution of the striatal eKerents in the monkey. Expl Neural. 5, 21-36. 29. Utkes S. K. and Ariano M. A. (1982) Cyclic nucleotide distribution in rat strio-nigral neurons: relationship to GADand SP-containing elements. Sot. Neurosci. 8, 961. 30. van der Kooy D. and Steinbusch H. W. M. (1980) Simultaneous fluorescent retrograde axonal tracing and immunofluorescent characterization of neurons. J. Neurosci. Res. 5, 479-485. 31. Voneida T. J. (1960) An experimental study of the course and destination of fibers arising in the head of the caudate nucleus in the cat and monkey. J. camp. Neural. 115, 75-87. 32. Woolf N. J. and Butcher L. L. (1981) Cholinergic neurons in the caudate-putamen complex proper are intrinsically organized: a combined Evans blue and acetylcholinesterase analysis. Brain Res. Bull. 7, 487-507. (Accepted 10 October 1982)
0306-4522/83/010023-07$03.00/0
PergamonFWssLtd 0 1983IBRO
CYCLIC NUCLEOTIDE DISTRIBUTION WITHIN RAT STRIATONIGRAL NEURONS MARJORIEA. ARIANO and SUSANK. UFKES Department of Anatomy & Neurobiology, University of Vermont College of Medicine, Burglington, VT 05405, U.S.A. Abstraet-Adenosine cyclic 3’,5’-monophosphate and guanosine cyclic 3’,5’-monophosphate have differential immtmohistochemical distributions within retrogradely-labeled striatonigral neurons of the rat. Adenosine cyclic 3’,5’-monophosphate is localized within more than half of the striatonigral projection neurons. It is also within tbe cytoplasm of other neurons and oligodendroglia. Guanosine cyclic 3’,5’-monophosphate is localized within 80% of the identified striatonigral neurons. These large percentages of cyclic nucleotide immunoreactivity within the striatonigral neurons suggest some of these efferent cells must contain both cyclic nucleotides. The immunofluorescent staining for guanosine cyclic 3’,5’-monophosphate is almost identical to that reported for efferent neurotransmitter-containing neurons of the caudate nucleus. However, the large proportion of striatonigral neurons demonstrating guanosine cyclic 3’,5’-monophosphate immunoreactivity precludes the association of this cyclic nucleotide with a selective neurotransmitter agent. Adenosine cyclic 3’,5’-monophosphate-reactive elements are very different in staining appearance from guanosine cyclic 3’,5’-monophosphate and neurotransmgter-identified somata. The number of striatonigral cells exhibiting reaction for this cyclic nucleotide does not eliminate the possibility that adenosine cyclic 3’,5’-monophosphate might be preferentially co-localized with a specific neurotransmitter, such as y-aminobutyrate, as has been previously suggested through biochemical experimentation.
fluorescent dyes, 32horseradish peroxidase,9*‘5 wheatgerm agglutinin conjugated to horseradish peroxidase,’ Herpes simplex virus6 and following intracellular injection of horseradish peroxidase after electrophysiological identification of individual 1~11s.~~The cytoarchitecture of this efferent population falls into the generic category of the medium spiny neuron of the caudite nucleus, comprising some 90% of the neuronal classification of the rat striatum.” These striatonigral neurons have densely spine-laden dendrites, a centrally placed nucleus and are rounded or oval in somal shape. Ultrastructural examination illustrates a thin rim of cytoplasm, poor in organelles, with sparse axosomatic symmetrical synapses and a smooth nuclear membrane.’ These projection neurons to the substantia nigra contain the neurotransmitters substance p2’ and y-aminobutyrate (GABA).19 In addition, a decrement in dopaminesensitive adenylate cyclase activity in the substantia nigra has been correlated to loss of the striatonigral efferents.‘4*27 The intent of the present study has been to investigate whether one particular cyclic nucleotide is preferentially associated with the striatonigral pathway and further to assess if the individual cyclic nucleotides co-localize with specific neurotransmitter substances. Fluorescent retrograde labeling is compatible with subsequent immunofluorescent characterization of the defined neuronal population,26*30and has demonstrated a large coincident localization of cyclic GMP within the striatonigral projection neurons.
The cyclic nucleotides, adenosine cyclic 3’,5’-monophosphate (cyclic AMP) and guanosine cyclic 3’S’-monophosphate (cyclic GMP), have been implicated as potential modulators of neural transmission. Our previous investigations have shown that these two compounds have a heterogeneous cellular localization and are evenly distributed throughout the rat striatum using immunohistochemical analysis. ‘9’The cyclic nucleotide-contaiuing neurons have been classified as belonging to the medium spiny group” on the basis of their somal diameter, perikaryal shape and ultrastructural characteristics5 However, it is not known if cyclic AMP or cyclic GMP can be linked to a specific group of striatal projection neurons or to a specific class of neurotransmitter synthesizing cells. The immunohistochemical detection of enkephalin-reactive neurons,23 substance P-reactive neurons,“+29 and glutamate decarboxylase-containing neurons25 within the caudate nucleus shows great similarity to one another and are also grouped into the medium spiny neuron class. In an attempt to characterize further the neuronal elements that show cyclic nucleotide immunoreactivity, we have identified specific efferent perikarya through retrograde transport of Evans blue.’ One of the primary efferent target structures of the striatum is the substantia nigra.‘7J**3’ The cellular morphology of the striatonigral neurons has been examined following retrograde transport of Abbreviations: Cyclic AMP, adenosine cyclic 3’,5’-monophosphate; cyclic GMP, guanosine cyclic 3’,5’-monophosphate; GABA, y-aminobutyrate. 23
24
M . A . ARIANO and S. K. UFKES EXPERIMENTAL PROCEDURES
All experiments were performed using male Sprague-Dawley rats (200-250 g). The animals were mainrained under conditions of constant temperature (22°C), relative humidity (50~) and an illumination cycle of 12 h. The rats were anesthetized with sodium pentobarbital (35 mg/kg) and injected with the dye Evans blue (Sigma Chemical Co., St. Louis, MO). Stereotaxic delivery of 0.1/~1 40~ Evans blue in distilled water was made into the substantia nigra using the atlas of Pellegrino, Pellegrino and Cushman. 22 The co-ordinates used for the substantia nigra were: A, -3.1; L, 2.5; and V, -8.6, down from the cortical surface. All injections were made unilaterally using a 1 #1 Hamilton syringe. Seventy-two hours following the Evans blue infusion, animals were re-anesthetized and transcardially perfused with approx 200 ml of cold 4 ~ paraformaldehyde, 0.5~ glutaraldehyde, 0.25 M sucrose in phosphate-buffered saline, pH 7.2. Brains were removed and placed into fresh cold fixative for 1 h, then transferred to cold 30~ sucrose in phosphate-buffered saline overnight. The brains were cut into coronal slabs at the level of the hypophyseal stalk, mounted and frozen at - 2 5 ° C onto brass cryostat chucks until sectioned. Eight #m-thick coronal sections were thaw-mounted onto chrom alum-coated slides and processed for cyclic nucleotide immunohistochemistry according to the method of Ariano, Butcher, and Appleman. 3 Antibody dilutions were 1:100 in phosphate-buffered saline for 1 h at 4°C or 1:500 at 4°C overnight. Antibody characterization and specificity has been previously determined by histochemical methods 5 and using radio-immunoassay these antibodies exhibit low to negligible recognition of their nucleoside triphosphate, diphosphate and monophosphate in competitive binding assay and a 200-fold greater recognition for the appropriate antigen cyclic nucleotide than the other cyclic nucleotide) ° Fluorescein-conjugated secondary immunoglobulins were purchased from Cappel Laboratories (Cochranville, PA) and used at 1:50 dilution in phosphate-buffered saline for I h at 4°C.
Following immunohistochemical processing tissue sections were examined with ultraviolet optics using a Zeiss Photomicroscope 3 equipped with 546 nm primary fiilter/590 nm secondary filter to visualize the Evans blue label and 485 nm primary filter/520 nm secondary filter to visualize the fluorescein-labeled elements. Controls included: (1) examination of contralateral caudate-putamen complex from the nigral injection site; (2) examination of striatal tissue prior to immuno-histochemical processing; and (3) comparison with non-infused caudate-putamen tissue. Photomicrography was performed using HP-5 black and white film (Ilford, Inc., Paramus, N J), pushed one stop and developed using D-19 (Kodak, Inc., Rochester, NY) to increase contrast of the negative. Exposures were made at constant magnification of x 100 and photographic enlargement such that each Evans blue-cyclic nucleotide matched
SN c
SN r
Fig. 1. Coronal section of rat brain illustrating the area of Evans blue injection. The black area indicates the most heavily stained region at the center of the injection site. The hatched area indicates the maximum extent and spread of the Evans blue for the experimental animals. SN r = pars reticulata of the substantia nigra; SNc = pars compacta; CP = cerebral peduncle; ML = medial lemniscus. pair was equivalent in area to 80 mm 2. Regions of densest Evans blue-labeled neurons were chosen for analysis, photographed using the 546/590nm filter combination, then examined with the 485/520 nm fluorescein filter combination. Coincident localization for each cyclic nucleotide in striatonigral neurons was numerically tabulated and averaged. RESULTS T h e injection site a n d extent o f diffusion o f E v a n s blue is illustrated in Fig. 1. M o s t o f the E v a n s blue was c o n t a i n e d within the s u b s t a n t i a nigra, b o t h pars c o m p a c t a a n d pars reticulata, with some spread ventrally into the cerebral peduncle a n d dorsally to e n c r o a c h u p o n the medial lemniscus a n d z o n a incerta. Only those cases in which retrograde labeling of s o m a t a within the ipsilateral c a u d a t e - p u t a m e n complex was f o u n d were used in this analysis. T h e brains o f 9 animals were utilized in this study. R a n d o m coronal sections t h r o u g h the rostral portions o f the rat s t r i a t u m were e x a m i n e d p r i o r to cyclic A M P or cyclic G M P i m m u n o h i s t o c h e m i c a l processing. T h e m o r p h o l o g y o f the brilliant red, E v a n s blue-positive perikarya agreed with previously p u b lished accounts o f striatonigral neurons. 8,9,32 T h e vast majority o f n e u r o n s were medium-sized, 10-20 # m in diameter, ovoid or r o u n d e d in shape a n d the nucleus filled m o s t of the cytoplasm. The cytoplasm was intensely fluorescent with substantial lengths o f neuronal processes evident e m a n a t i n g from the cell soma. T h e nucleus o f E v a n s blue-identified striatonigral n e u r o n s was generally u n s t a i n e d (see Figs 2A a n d 3A). W i t h i n o u r experimental sample o f
Fig. 2. Paired photomicrograph set from a representative region of the caudate-putamen, ipsilateral to the Evans blue injection into the substantia nigra. Calibration bar, 50#m. A. Evans blue-labeled striatonigral neurons within the caudate--putamen visualized with 546--590 nm filter combination. Arrows demonstrate typical reactive somata, fb denotes a fiber bundle. B. Visualization of cyclic AMP-containing elements within the same field as A through use of the 485-520 nm filter combination for fluorescein excitation. The same striatonigral neurons are shown with arrows, fb is the same fiber bundle. Fig. 3. Paired photomicrograph set from a representative region of the caudate-putamen, ipsilateral to the Evans blue injection into the substantia nigra. Calibration bar, 50#m. A. Evans blue-labeled striatonigral neurons within the caudate-putamen, visualized with the 546/590 nm filter combination. Arrows demonstrate three typical reactive somata, fb denotes a fiber bundle. B. Visualization of cyclic GMP-containing elements within the same field as A through use of the 485/520 nm filter combination for fluorescein excitation. The same striatonigral neurons are shown with arrows, fb is the same fiber bundle.
Cyclic AMP and cyclic GMP in striatonigral neurons
27
Table 1. Striatonigral neurons*
Cyclic AMP-containing
Cyclic GMP-containing
54%
81.5%
Striatonigral neurons stained?
*Total issue area examined was 31.4 mm’; each paired photomicrograph set was 80 mm2 x 8 pm tissue section thickness x49 different photomicrograph sets. tTota1 number examined was 3674 neurons from 9 different animals.
31.4 mm3 of caudate-putamen tissue, 3674 striatonigral neurons were unequivocally identified and examined. Cyclic AMP immunohistochemical localization was assessed on random sections throughout the rostra1 striatum. The perikarya and processes of cyclic AMP-stained elements were as described previously. 3,s Uniform immunofluorescence was present throughout the cytoplasm and nucleus of reactive cells (Fig. 2B). Somal dimensions of 8-16pm predominated. More than half of the Evans blue-labeled striatonigral neurons contained cyclic AMP immunoreactivity (Table 1). The cytoarchitecture of striatonigral neurons containing cyclic AMP is illustrated in Fig. 2. The total number of cyclic AMP-containing elements analyzed was 3203 and of that number, less than 30% could be identified as striatonigral neurons (Table 2). Other cyclic AMP-containing elements were identified as oligodendroglia surrounding and within fiber bundles perforating the substance of the caudate-putamen complex. Cyclic AMP-reactive glia were small in diameter (610 pm) and very intensely fluorescent. Cyclic GMP immunohistochemical localization was determined on sections serial to those of cyclic AMP-stained slides within the rostra1 striatum. The cytoarchitecture of cyclic GMP-immunoreactive cells was a previously described.4*5 Predominant staining was within the cytoplasm and proximal processes of cells 15-20 pm in diameter and a lack of nuclear staining was characteristic (Fig. 3B). A minor immunoreactive element was the course fibrous processes of astrocytes.3s4 More than 80% of Evans blue-labeled striatonigral neurons contained cyclic GMPimmunoreactivity (Table 1). The immunofluorescence pattern of staining in striatonigral cyclic GMP-positive neurons is demonstrated in Fig. 3. The total number of cyclic GMP-immunoreactive elements examined and analyzed was 3512 and almost 40% could be identified as components of the striatonigral efferent population (Table 2). Other cyclic Table 2. Cyclic nucleotide-labeled Cyclic AMP-positive? Cyclic GMP-positive1
cells*
EB-labeled neurons
Other cells
28.9% 38.5%
71.1% 61.5%
*Total tissue area examined was 3 1.4 mm3, from 49 different photomicrograph sets. tTota1 cyclic AMP-stained elements was 3203 from 9 different animals. ITotal cyclic GMP-stained elements was 3512 from 9 different animals.
GMP-staining elements were neuronal in nature, but not a part of this efferent system of the caudate-putamen complex to the substantia nigra. DISCUSSION
The cyclic nucleotide distribution within identified rat striatonigral neurons has been examined. The compatibility of fluorescent retrograde labeling with immunohistochemical characterization of the caudate projection neurons has demonstrated that these potential second messengers are not organized solely within this specific efferent cell group. The extremely large percentage of cyclic GMP-positive striatonigral neurons suggests that some of these efferent projection cells must contain both cyclic AMP and cyclic GMP immunohistochemically visualized pools. The greater number of cyclic GMP-stained striatonigral neurons also suggests that this cyclic nucleotide cannot be associated exclusively with only the substance P or solely the GABA efferents to the substantia nigra, but is probably co-localized with both bioactive substances. The report of dissociated topography for these two neurotransmitters within the caudate nucleus’6 adds support to this notion. It is intriguing that the immunochemical staining characteristics of GAD-positive neurons,2s*29 substance P-like reactive cell~,“*~~ and enkephalin-positive somata2) within the striatum have a coincident cellular localization with the cyclic GMP-stained neurons visualized in this study. This suggests that the antigenie determinants for these neurotransmitters and cyclic GMP could be within equivalent cellular organelles. This point awaits confirmation at the subcellular level of resolution. Cyclic AMP immunofluorescent elements exhibit very different staining characteristics when compared to cyclic GMP or the reported neurotransmittersynthesizing cells. Nuclear staining within Evans blue identified striatonigral neuronal somata was characteristic. Slightly more than half of the striatonigral neurons examined in this study showed coincident cyclic AMP-immunoreactivity. Although this does show cyclic AMP is not exclusively associated with the striatonigral projection cells, the cyclic nucleotide may be more closely associated with one of the two identified neurotransmitters utilized by this pathway. Biochemical studies have demonstrated a positive correlation between loss of GAD activity and a decrement in the dopamine-sensitive adenylate cyclase activity in the substantia nigra after interruption of the striatonigral pathway.27 However, the large
28
M. A. ARIANOand S. K.
majority of cyclic AMP-stained elements within the caudate nucleus are not part of the projection system to the substantia nigra. They are oligodendroglia and neurons of the other categories, e.g. interneurons or alternate efferent cells. The predominant energy source for the brain is glucose’* and the role of cyclic AMP in intermediary metabolism has been well studied. However, the immunohistochemically-detected cellular distribution of cyclic AMP does not reflect the ubiquitous function it has in cerebral glycolysis. It may be that the cellular pool of this cyclic nucleotide which can be visualized following immunohistochemical analysis is not directly related to energy metabolism, but subserves another function. The current methodologies will not allow us to address this question appropriately. The ability physiologically to manipulate the visualized cyclic nucleotide pool experimentally is also critical. We know from other studies and our own work, that the concentration of cyclic nucleotides can be selectively increased following synaptic activity. In the experimental system of the rat superior cervical ganglion,’ the elevations in cyclic AMP and cyclic GMP following synaptic activity show cellular selectivity. The cyclic GMP response is confined to postganglionic neurons while the cyclic AMP response is primary associated with the glial-satellite cells. It still needs to be determined if the immunohistochemically visualized neuronal staining of the cyclic nucleotides can be modulated by synaptic activity within the caudate-putamen complex. The selective in vivo acti-
UFKFS
vation of striatal afferents has not yielded consistent cytochemical elevations in cyclic AMP and cyclic GMP in the caudate nucleus which can be localized to defined cell types (unpublished observations). We have shown that selective deafferentation3 does not alter the cellular distribution or pattern of staining of the striatal cyclic AMP and cyclic GMP. The application of antibodies directed against the synthetic enzyme for cyclic GMP, guanylate cyclase (EC 4.6.1.2): demonstrates that this protein exhibits a neuronal localization. It may be that the immunohistochemically-detected cyclic GMP seen in low proportions within fibrous astrocytic endfeet in the striatum has been circumstantially accumulated within that cellular structure in response to production and subsequent diffusion from its active physiological site of action, within neuronal somata and at synaptic terminals. The neuronal distribution of the protein kinase mediating some of the intracellular functions of cyclic GMP (EC 2.7.1.38) would tend to support this contention. ‘**’Cyclic GMP tissue concentrations are usually quite low and two orders of magnitude less than cyclic AMP in the caudate nucleus as determined using radioimmunoassay.’ The selective neuronal elevations in this cyclic nucleotide following synaptic activity’ and the exclusive neuronal distribution of the cyclic GMP system in the striatum are a strong suggestion that it has an important function in neuronal communication. Acknowledgements-This
work was supported
by BNS
81-02648.
REFERENCES 1. Ariano M. A. (1982) Ultrastructural distribution of the cyclic GMP system: a role in neurotransmission in the rat striatum. Sot. Neurosci. 8, 344. 2. Ariano M. A., Briggs C. A. and M&fee D. A. (1982) Cellular localization of cyclic nucleotide changes in rat superior cervical ganglion. Cell Molec. Neurobiol. 2, 143-156. 3. Ariano M. A., Butcher L. L. and Appleman M. M. (1980) Cyclic nucleotides in the rat caudate-putamen complex: histochemical characterization and &ects of deafferentation and kainic acid infusion. Neuroscience 6, 1269-1276. 4. Ariano M. A., Lewicki J. A., Brandwein H. J. and Murad F. (1982) Immuuol&tochemical localizati& of guanytie cyclase within neurons of rat brain. Proc. nutn Acud. Sci. U.S.A. 79, 13161320. Ariano M. A., and Matus A. I. (1981) Ultrastructural localization of cyclic GMP and cyclic AMP in rat striatum. J. Cell Biol. 91. 287-292.
9.
10. 11. 12. 13. 14.
Bak I. J., Markham C. H., Cook M. L. and Stevens J. G. (1978) Ultrastructural and immunoperoxidase study of striatonigral neurons by means of retrograde axonal transport of Herpes simplex virus. Brain Res. i43, 361-369. Bentivoglio M., Kuypers H. G. J. M., Catsman-Berrevoets C. E. and Dann 0. (1979) Fluorescent retrograde neuronal labeling in rat by means of substances binding specifically to adenine-thymine rich DNA. Neurosci. Lerr. 1%.235-245. Bolam J. P., Powell J. F., Totterdell S. and Smith A. D. (1981) The proportion of neurons in the rat neostri#tum that project to the substantia nigra demonstrated using horseradish peroxidase conjugated with wheatgerm agghztinin. B@J Res. 220, 339-343. Bolam J. P., Somogyi P., Totterdell S. and Smith A. D. (1981) A second type of striatonigral neuron: a comparison between retrogradely-labelled and Golgi-stained neurons at the light- and electron-microscopic levels. Neuroscience 6, 2141-2151. Briggs C. A., Whiting G., Ariano M. A. and McAfee D. A. (1982) Cyclic nucleotide metabolism in the sympathetic ganglion. Cell Molec. Neurobibl. 2, 129-142. Cue110A. C. and Kanazawa I. (1978) The distribution of substance P immunoreactive fibers in the rat central nervous system. J. camp. Neural. 178, 129-156. DiFiglia M., Pasik P. and Pasik T. (1976) A Golgi study of neuronal types in the neostriatum of monkeys. Blrain Res. 114, 245-256. Dimova R., Vuillet J. and Seite R. (1980) Study of the rat neostriatum using a combined Golgi-etectron microscope tedmique and &al secticmb. Neuroscisncc 5, 1581-1596. Gale K., Guidotti A. and Costa E. (1977) Dopamine-sensitive adenylate cyclase: location in substantia nigra. Science, N.Y. 195, 503-505.
Cyclic AMP and cyclic GMP in striatonigral neurons
29
15. Grofova I. (1975) The identification of striatal and paltidal neurons projecting to substantia nigra. An experimental study by means of retrograde axonal transport of homemdish peroxidam. Brain Res. 91, 286-291. 16. Jesse11T. M., Emson P. C., Paxinos G. and Cuello A. C. (1978) Topographic projections of substance P and GABA pathways in the striato- and pallido-nigral system: a biochemical and immunohistochemical study. Brain Res. 152, 487498. 17. Kemp J. M. (1970) The termination of strio-pallidal and strio-nigral fibers. Brain Res. 17, 125-128. 18. Kety S. S. (1975) The general metabolism of the brain in I&JO.In Z7reh4etabolism of the Neruous System (ed. Richter D.) Pergamon Press, London. 19. Kim J.-S., Bak I. J., Hassler R. and Okada Y. (1971) Role of y-aminobutyric acid (GABA) in the extra-pyramidal motor system-II. Some evidence for the existence of a type of GABA-rich strio-nigral neurons. Expl Bruin Res. 14, 95-104. 20. Lohmann S. M., Walter U., Miller P. E., Greengard P. and DeCamilli P. (1981) Immunohistochemical localization of cyclic GMPdependent protein kinase in mammalian brain. Proc. natn Acad. Sci. U.S.A. 78, 653-657. 21. Mroz E. A., Brownstein M. J. and Leeman S. E. (1977) Evidence for substance P in the striato-nigral tract. Brain Res. m, 305-311. 22. Pellegrino L. S., Pellegrino A. S. and Cushman A. J. (1979) A Stereotaxic Atlas ofthe Rat Bruin. Plenum Press, New York. 23. Pickel V. M. Sumal K. K., Beckley S. C., Miller R. J. and Reis D. J. (1980) Immunocytochemical localization of enkephahn in the neostriatum of rat brain: a light- and electron-microscopic study. J. camp. Neural. 189, 721-740. 24. Preston R. J., Bishop G. A. and Kitai S. T. (1980) Medium spiny neuron projection from the rat striatum: an intracellular horseradish peroxidase study. Brain Res. 183, 253-263. 25. Ribak C. E., Vaughn J. E. and Roberts E. (1979) The GABA neurons and their axon terminals in rat corpus striatum as demonstrated by GAD immunocytochemistry. J. camp. Neural. 187, 261-284. 26. Sawchenko P. E. and Swanson L. W. (1981) A method for tracing biochemicallydetined pathways in the central nervous system using combined fluorescence retrograde transport and immunohistochemical techniques. Brain Res. 210, 31-51. 27. Spano P. F., Trabucchi M. and DiChiara G. (1977) Localization of nigral dopamine-sensitive adenylate cyclase on neurons originating from the corpus striatum. Science N.Y. I%, 1343-1345. 28. Szabo J. (1962) Topical distribution of the striatal eKerents in the monkey. Expl Neural. 5, 21-36. 29. Utkes S. K. and Ariano M. A. (1982) Cyclic nucleotide distribution in rat strio-nigral neurons: relationship to GADand SP-containing elements. Sot. Neurosci. 8, 961. 30. van der Kooy D. and Steinbusch H. W. M. (1980) Simultaneous fluorescent retrograde axonal tracing and immunofluorescent characterization of neurons. J. Neurosci. Res. 5, 479-485. 31. Voneida T. J. (1960) An experimental study of the course and destination of fibers arising in the head of the caudate nucleus in the cat and monkey. J. camp. Neural. 115, 75-87. 32. Woolf N. J. and Butcher L. L. (1981) Cholinergic neurons in the caudate-putamen complex proper are intrinsically organized: a combined Evans blue and acetylcholinesterase analysis. Brain Res. Bull. 7, 487-507. (Accepted 10 October 1982)