- Email: [email protected]
Localization of ARPP-90, a major 90 kiloDalton basal ganglion-enriched substrate for cyclic AMP-dependent protein kinase, in striatonigral neurons in the rat brain
Molecular Brain Research, 5 (1989) 149-157 Elsevier
149
BRM 70124
Localization of ARPP-90, a major 90 kiloDalton basal ganglionenriched substrate for cyclic AMP-dependent protein kinase, in striatonigral neurons in the rat brain S. Ivar Walaas, Steve Cala and Paul Greengard Laboratory of Molecular and Cellular Neuroscience, The Rochefeller University, New York, NY IO021 (U.S.A.)
(Accepted 25 October 1988) Key words: Neostriatum; Substantia nigra; Quinolinic acid; Striatonigral fiber; Phosphoprotein; Protein phosphorylation
Cyclic AMP-regulated phosphoproteins with specific cellular localizations in brain represent important targets through which this second messenger system can mediate or modulate distinct neurotransmitter signals. This study reports that two cyclic AMP-regulated phosphoproteins (Mr 90,000 and 93,000) found in brain share several properties, including similar isoelectric points and similar phosphopeptide maps. This protein doublet is particularly enriched in the forebrain basal ganglia, but it can also be found in the substantia nigra, a brainstem region which is a major target for fibers from the forebrain basal ganglia. Quinolinic acid lesions of neurons in the neostriatum decrease the levels of the 90/93 kDa phosphoprotein doublet to about the same extent as they reduce the levels of DARPP-32, a phosphoprotein specifically enriched in striatonigral medium-sized spiny neurons. These reductions are seen in both the neostriatum and the substantia nigra. Therefore, within the basal ganglia, the 90/93 kDa phosphoprotein doublet, termed adenosine 3' :5'-monophosphate-regulated phosphoprotein, Mr = 90,000 (ARPP-90), is largely, if not solely, present in striatonigral cells and fibers. The specific localization in these neurons suggests that ARPP-90 could be important in receptor-regulated, cyclic AMP-mediated functions in the striatonigral neurons. INTRODUCTION Extensive evidence indicates that receptor-regulated, cyclic AMP-mediated protein phosphorylation is of considerable importance in the functions of the basal ganglia of mammalian brain t6,36. In regions such as the neostriatum and substantia nigra, neurotransmitter release, regulation of neuronal excitability and cellular metabolism appear to be sensitive to changes in intracellular cyclic A M P levels 6,11'12,31,38. It is of great interest, therefore, that the basal ganglia of mammalian brain contain several protein substrates for cyclic AMP-dependent protein kinase, including two phosphoprotein species which migrate closely together on sodium dodecylsulfate-polyacrylamide gel electrophoresis ( S D S - P A G E ) with approximate molecular masses of 90 and 93 kilodalton (kDa) (previously reported as 96/98 kDa, the differences in apparent molecular weights being attrib-
utable to different electrophoretic conditions) 5,32. In the rat brain, the 90/93 k D a doublet is particularly enriched in forebrain basal-ganglion regions such as the caudatoputamen, nucleus accumbens and olfactory tubercle, but is not easily detected by onedimensional S D S - P A G E in the output regions of the basal ganglia, such as the globus pallidus and substantia nigra 3z. This localization contrasts with that of another basal ganglion-enriched phosphoprotein, D A R P P - 3 2 , which is specifically concentrated in the medium-sized spiny neurons of the neostriatum as well as in their efferent striatopallidal and striatonigral axons 26'35. Thus, D A R P P - 3 2 is easily detected by in vitro phosphorylation and S D S - P A G E in both the forebrain neostriatum and the substantia nigra 33'35. Therefore, it appeared possible that the 90/93 kDa phosphoproteins might have a cellular distribution in the rat basal ganglia which was distinct from that of DARPP-3232. The present study has ex-
Correspondence: S.I. Walaas, Institute of Medical Biochemistry, University of Oslo, P.O. Box 1112, 0317 Oslo 3, Norway.
0169-328X/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
150 amined this question, using various gel electrophoretic and peptide mapping techniques combined with chemical lesioning of the medium-sized spiny striatonigral neurons and their efferent axons 35. MATERIALS AND METHODS
nylfluoride (PMSF, 0.1 mM), and homogenized with 15 strokes in glass-Teflon tissue grinders rotating at 1200 rpm. Soluble fractions were prepared by centrifugation at 150,000 g for 30 min, washing the pellet once, and pooling the supernatants as described 32,34. The soluble fractions were desalted by gel filtration on minicolumns TM.
Materials [},-3zp]ATP was from New England Nuclear, Boston. Staphylococcus aureus V8 protease was from Miles Biochemicals, aprotinin (Trasylol) was from Mobay Chemical Corporation, and TPCK-trypsin was from Cooper Biomedical. Ampholines (Servalyte 3-10) were from Serva, Heidelberg, F.R.G. The catalytic subunit of cyclic AMP-dependent protein kinase was purified as described TM, and was a gift from Dr. A.C. Nairn, The Rockefeller University. Other chemicals of analytical grade were from standard commercial suppliers.
Preparation of lesioned animals Male Sprague-Dawley rats (150-200 g b. wt.) were anesthetized with chloral hydrate (0.35 g/kg, i.p.) and mounted on a stereotactic frame. 1 #1 of 0.2 M quinolinic acid (adjusted to pH 7.4 with NaOH), an excitotoxin which produces 'axon-sparing' lesions of local cells in the neostriatum and the striatofugal fibers emanating from these neurons 3°'39, was infused into the rostral neostriatum at coordinates A:8900; L:2500 and V:500, following the atlas of K6nig and Klippel2°. All animals survived for 5-7 days before analysis.
Preparation of tissue Rabbits were killed under barbiturate anesthesia, while lesioned and unlesioned rats were stunned and decapitated. The brains were rapidly removed and cooled on ice. The neocortex, neostriatum (caudatoputamen), substantia nigra and other brain regions were dissected as described 34, and the brain samples were weighed and put on ice. Similar samples were dissected from frozen rhesus monkey brain (kindly provided by Dr. T. Bartfai, University of Stockholm, Sweden). Brain samples were added (approx. 1:10, w/v) to ice-cold standard buffer containing 10 mM HEPES (pH 7.4), 2 mM EDTA, 1 mM dithiothreitol (DTT), and the protease inhibitors aprotinin (50 kallikrein inhibitor units/ml) and phenylmethyl sulfo-
Phosphorylation of proteins Proteins in the tissue samples were phosphorylated with [T-32p]ATP as described 32'3~. Briefly, aliquots of homogenates (50/~g protein from rat substantia nigra, 100-200 ktg protein for other regions) or soluble fractions (20-50 #g protein) from the different brain regions were incubated in a medium (final vol. 100 Izl) containing (final conc.) 25 mM HEPES (pH 7.4), 10 mM MgC12, 1 mM EDTA, 1 mM EGTA, 1 mM DTT and 10/~M [7-32P]ATP (5-15 Ci/mmol). Protein phosphorylation systems were activated by the addition of either 10/~M 8-bromo cyclic AMP, 2/~M cyclic AMP plus 1 mM isobutylmethylxanthine, or 1.25 mM CaCI 2. In analyses of samples from the lesioned animals and their controls, catalytic subunit of cyclic AMP-dependent protein kinase (50 nM final conc.) was added to the reaction mixture. The samples were preincubated at 30 °C for 60 s, the reactions were initiated by the addition of the [y-32p]ATP, and terminated after a 30 s incubation period by the addition of a stop-solution containing (final conc.) 3% SDS, 60 mM Tris-HC1 (pH 6.8), 5% glycerol, 2% 2-mercaptoethanol, and a trace of Bromophenol blue. The samples were then heated for 2 min in a boiling water bath, and cooled to room temperature before analysis of phosphorylated proteins.
Analytical separation of proteins Phosphoproteins were separated by one- or twodimensional polyacrylamide gel electrophoresis (PAGE). For one-dimensional SDS-PAGE, aliquots were directly applied to gels prepared with different amounts of acrylamide as described 19'21'34, and proteins were separated by constant current electrophoresis 1°. For two-dimensional separations, samples were first precipitated twice with 90% cold acetone (-20 °C) to remove ATP and salts. Proteins were then solubilized with 0.5% SDS, 9 M urea, 0.1% ampholines and 1% 2-mercaptoethanol, followed by an equal volume of 4% NP-40, 9 M urea,
151 0.1% ampholines and 1% 2-mercaptoethanol, as described 25. Separation was achieved in 12 cm long tube gels (1.5 mm i.d.) containing 2% ampholines (pH 3-10) 24. The gels were subjected to isoelectrofocusing (IEF) at 500 V for 16 h 24 or non-equilibrium pHgradient gel electrophoresis (NEPHGE) at 400 V for 4 h 25, carefully extruded and directly loaded onto the second dimension gels. The proteins were then separated by S D S - P A G E as described above. All slab gels were stained, fixed and destained as described 34. Phosphoproteins were visualized by autoradiography of the dried gels. Apparent molecular masses of different proteins were determined by comparison with standard proteins (Sigma) included in separate lanes, while apparent isoelectric points were determined by comparison with a pH gradient standard curve determined on separate tube gels run in parallel 24. Quantitation of phosphoproteins was achieved by cutting the gel pieces containing the protein of interest out of the gel, using the autoradiogram as a guide, and analysing 32p-content by Cerenkov counting. Preliminary experiments, using an incubation time of 30 s, showed that the incorporation of 32p into phosphoproteins of apparent molecular masses between 90 and 98 kDa was linear with amount of protein up to at least 50/~g protein in cytosolic fractions, while homogenate samples showed greater variability (not shown). Therefore, analysis of the level of ARPP-90 in the neostriatum was performed on cytosol preparations. Due to the small amount of material available, homogenate samples were used in assays of single substantiae nigrae. The changes observed in these phosphoproteins were compared to the changes observed in synapsin I, a phosphoprotein enriched in all nerve terminals 8,9, DARPP-32, a phosphoprotein marker for the striatonigral neurons 26'35, and pyruvate dehydrogenase, a widely distributed mitochondrial phosphoprotein 34.
Tryptic phosphopeptide mapping Dried and washed gel pieces containing phosphoproteins 17 were incubated in 1 ml of 50 mM NH4HCO 3 (pH 8.0) containing TPCK-trypsin (0.075 mg/ml) at 37 °C for 24 h. The gel slices were washed with 0.5 ml of 50 mM NH4CO 3 at 37 °C for 4 h, and the collected supernatants were lyophilized and resuspended in electrophoresis buffer (10% acetic
acid, 1% pyridine, pH 3.5). Phosphopeptides were separated on thin layer cellulose sheets (Kodak Eastman) by electrophoresis at 400 V for 90 min, followed by chromatography in the second dimension in a buffer containing pyridine, 1-butanol, water, acetic acid (15:10:12:3) 17. 32p-labeled peptides were identified by autoradiography of the dried cellulose sheets.
Miscellaneous methods Protein was analysed by the method of Bradford 4 or the method of Peterson 27, using bovine serum albumin as standard. Peptide mapping of phosphoproteins by means of incomplete proteolysis was performed using 5 ktg of Staphylococcus aureus V8 protease, essentially as described 7'17. RESULTS Incubation of homogenates (Fig. 1) or cytosol fractions (not shown) from rat caudatoputamen and cerebral cortex with [y-32p]ATP and cyclic AMP, followed by one-dimensional S D S - P A G E and autoradiographic detection of phosphoproteins, showed an apparent doublet of phosphorylated protein bands of apparent molecular masses of 90 and 93 kDa (using Mr = 94,000 for phosphorylase as a standard). Addition of CaC12 (250/~M flee) to the reaction mixture did not effect the phosphorylation of this same protein (Fig. 1A). Phosphopeptide mapping of these two phosphoprotein bands, using V8 protease from S. aureus to achieve incomplete digestion in SDS, showed that the two protein bands generated identical phosphopeptides (Fig. 1B). Two-dimensional gel electrophoresis of phosphorylated samples from the rat caudate nucleus, using IEF in the first dimension, showed that the two phosphoproteins comigrated in the first dimension, displaying an apparent isoelectric point of approximately 6.0 (not shown). Because these two protein bands appear to be so closely related, we have designated them collectively as ARPP-90 (adenosine 3':5'-monophosphate-regulated phosphoprotein, M r = 90,000). Analysis of various rat brain regions indicated ARPP-90 to be highly enriched in the neostriatum, including the caudatoputamen, nucleus accumbens and olfactory tubercle 32. A much lower amount of this phosphoprotein doublet could be detected in
152
B.
A. STRIKnJM CC~]rEX I
it
DARPF
cAM, ~,_,,
•
---
I
÷--
--
+
Fig. 1. Autoradiogram showing the presence of ARPP-90, a 90/93 kDa protein substrate for cyclic AMP-dependent protein kinase, in rat neostriatum. A: homogenates from neostriatum or cerebral cortex were incubated with 10/aM [7-32p]ATP in the presence of either 2/aM cyclic AMP plus 1 mM isobutylmethylxanthine (IBMX) or 0.25 mM free Ca 2÷ as described in the text. Proteins were separated by SDS-PAGE, and phosphoproteins were visualized by autoradiography. Identified phosphoproteins are indicated. ARPP-90, basal ganglion-enriched phosphoprotein (M r 90,000); synapsin I, a synaptic vesicle-associated phosphoprotein; PDH, a-subunit of pyruvate dehydrogenase; DARPP-32, dopamine- and cyclic AMP-regulated phosphoprotein (M r 32,000); MBP, myelin basic protein. B: peptide mapping of ARPP-90, showing homology between the two phosphoprotein bands. Gel pieces containing the 90 and 93 kDa bands were separately cut out of the gel shown in A, and subjected to peptide mapping with S. aureus V8 protease as described in the text. Molecular masses of phosphopeptide fragments are given in kiloDalton.
homogenates of rat cerebral cortex (Fig. 1A), diencephalic and mesencephalic regions, and the protein was not detected in brainstem or cerebellum by onedimensional S D S - P A G E 32. Similar analysis of rabbit brain, monkey brain (Fig. 2) and human brain 37 showed that phosphoprotein bands having similar apparent molecular masses and similar phosphopeptide maps were present in the neostriatum of these species as well. In the rabbit brain, ARPP-90 displayed a regional distribution similar to that seen in the rat brain, with considerably more of the phosphoprotein being seen in the neostriatum than in the cortical sample (not shown). Monkey (Fig. 2) and human brains 37, in contrast, contained detectable amounts of ARPP-90 in both cortical and basal ganglion regions. These distribution patterns are somewhat similar to that of DARPP-32 (Fig. 2), which in the rat brain is known to be highly enriched in the mediumsized spiny neurons of the neostriatum and the striatofugal fibers emanating from these neurons 26, but
A i
B ......~
t
C ,J
D i
i
i
Fig. 2. Autoradiogram showing the presence of ARPP-90 in rat, rabbit and monkey brain. Homogenates from neostriatum of rat (A), rabbit (B) and monkey (C) and cerebral cortex from monkey (D) were phosphorylated in the presence of absence of 10/aM 8-bromo cyclic AMP, and phosphoproteins were separated by S D S - P A G E and visualized by autoradiography. Abbreviations as in the legend to Fig. 1.
153
Cellular localization of ARPP-90 in the neostriatum
-
+
Fig. 3. Autoradiogram showing the effect of quinolinic acid lesion on phosphoproteins in rat neostriatum. Homogenates from neostriatum from control or lesioned animals were phosphorylated in the absence or presence of 50 nM catalytic subunit of cyclic AMP-dependent protein kinase as described in the text. Phosphoproteins were separated by SDS-PAGE and visualized by autoradiography. Some of the identified proteins are indicated. Abbreviations as in the legend to Fig. 1. which appears to be present in m o n k e y cerebral cortex (Fig. 2) in somewhat higher amounts than in rat cortex.
A.
The regional and species distribution of ARPP-90 indicated that it might be enriched in specific cell types in rat brain. The cellular localization of the protein was further examined in the rat basal ganglia, which are well suited to various lesioning techniques due to the organization of the afferent and efferent connections and intrinsic neurons in these brain regions 22'35. Infusion of quinolinic acid into the caudatoputamen was used to lesion the medium-sized spiny neurons of the neostriatum, the cells of origin of the striatopallidal and striatonigral fiber tracts 3,13. Following 5 - 7 days survival, ARPP-90 was analysed by phosphorylation of homogenate or supernatant samples in the absence or presence of exogenous catalytic subunit of cyclic AMP-dependent protein kinase. This lesion led to a pronounced decrease in phosphoproteins of 90-93 kDa in homogenates (Fig. 3) and in supernatant fractions (not shown). Scintillation counting of excised gel pieces showed that the levels of the 90/93 kDa phosphoproteins amounted to 34 + 10% (mean + S.D., n = 6) of that in control samples. Tryptic peptide mapping of the phosphoproteins contained in these gel pieces showed that this decrease was due to disappearance of ARPP-90, since the distinct tryptic phosphopeptides which decreased in the lesioned animals (Fig. 4) are the same
B,
Fig. 4. Autoradiogram showing two-dimensional peptide maps of ARPP-90 in phosphorylated cytosol fractions of neostriatum from control (A) and quinolinic acid-lesioned (B) rats. Neostriatal cytosol was phosphorylated in the presence of the catalytic subunit of cyclic AMP-dependent protein kinase, and proteins were separated by SDS-PAGE. Gel pieces containing 90/93 kDa phosphoproteins were cut out of the gels and digested with trypsin, and the resulting phosphopeptides (labeled 1, 2, 3) were separated by electrophoresis and chromatography as described in the text. Small circles indicate application points.
154 as those obtained by mapping of purified ARPP-90 from both rat and bovine brain 5. The decrease in ARPP-90 following local quinolinic acid infusion was also compared with that of other phosphoproteins in the neostriatum, including DARPP-32 and synapsin I. Lesioned animals displayed 27 + 9% of 32p-incorporation into DARPP-32 and 59 + 6% into synapsin I when compared to control samples (mean + S.D., n = 5). Thus, the decreases of ARPP-90 and DARPP32 were comparable, and much greater than that of synapsin I.
nigra (Fig. 5), which was identified as ARPP-90 by peptide mapping (not shown). However, the amount of ARPP-90 appeared considerably lower in the substantia nigra than in the neostriatum, the phosphorylated ARPP-90 in the substantia nigra representing only 40-50% of phosphorylated DARPP-32 (values from 5 experiments), whereas in the caudatoputamen ARPP-90 consistently incorporated 1.5-2 times the amount of 32p that DARPP-32 did (Figs. I and 3; data not shown). In the substantia nigra from animals which had received injections of quinolinic acid into the caudate nucleus, the amount of ARPP-90 was greatly decreased (Fig. 5), indicating that destruction of the striatonigral fibers led to a disappearance of the protein. We compared the phosphorylation of ARPP-90 in each gel to the phosphorylation of several wellcharacterized proteins present in the striatonigral fibers, including synapsin I and DARPP-32 (ref. 35), in the same gels. Because the total amount of protein which entered the two-dimensional gels varied considerably between different experiments (not shown), the ratios of lesioned to control levels of these various phosphoproteins were normalized against the level of the phosphorylated 42 kDa a-sub-
Cellular localization of ARPP-90 in substantia nigra The localization of ARPP-90 in basal ganglion regions outside of the neostriatum was studied in the substantia nigra, the target region of the striatonigrai fiber tract and the origin of major output tracts from the mammalian basal ganglia 22. Analysis of phosphoproteins in homogenates from the substantia nigra was performed with two-dimensional N E P H G E , since one-dimensional S D S - P A G E was unable to identify ARPP-90 in this region 32. Comparison of autoradiograms of such two-dimensional gels from rat substantia nigra and neostriatum showed the presence of a 90-93 kDa phosphoprotein in the substantia
A-
~
, i
B.
~
, I
SYNAI
BASIC '1 ACIDIC Fig. 5. Autoradiogram showing effect of quinolinic acid infusion into rat neostriatum on phosphoproteins in substantia nlgra. Homogenates of substantia nigra from control (A) or lesioned (B) rats were phosphorylated in the presence of 50 nM of the catalytic subunit of cyclic AMP-dependent protein kinase, and separated by non-equilibrium pH-gradient electrophoresis (NEPHGE, arrows) in the first dimension and by SDS-PAGE in the second dimension. Phosphorylated proteins were visualized by autoradiography. Some of the identified proteins are indicated. Stippled circle indicates location of ARPP-90.21K, phosphoprotein enriched in the basal ganglia32. Other abbreviations as in the legend to Fig. 1.
155 unit of the mitochondrial enzyme pyruvate dehydrogenase (PDH), since the latter protein, a widely distributed phosphoprotein present in all cell types, was found not to change in the substantia nigra following the lesions (Fig. 5). This analysis showed that the ratio ARPP-90/PDH in the lesioned animals represented 22 + 5% of that found in control animals (mean + S.D., n = 5), while the ratios of DARPP32/PDH and synapsin I/PDH also were considerably decreased (18 + 4% and 31 + 3% of control values, respectively, mean + S.D., n = 3). Since the decreases in the values for DARPP-32 and synapsin I were similar to the lesion-induced decreases in the absolute levels of these proteins measured by other analytical methods 35, the ratio ARPP-90/PDH appears to give a valid estimate of the changes in ARPP-90 induced by the lesions. These results, therefore, indicate that ARPP-90 in the rat caudatoputamen is highly enriched in the medium-sized spiny neurons and, albeit in lower concentrations, in the axons and terminals of the striatonigral fibers. DISCUSSION ARPP-90 is one of a group of phosphoproteins found in the rat brain which are substrates for cyclic AMP-dependent protein kinase and which are enriched in the basal ganglia 32,34. This group of proteins was originally observed to display two distinct distribution patterns. Some proteins in this group, including DARPP-32, were enriched to the same extent in both forebrain ('input') basal ganglia regions (e.g. the caudatoputamen) and in 'output' basal ganglia regions (e.g. the globus pallidus and substantia nigra). Other basal ganglion-enriched proteins, including ARPP-90, were seen only in the forebrain basal ganglion regions. These observations implied that the two groups of proteins either were present in different groups of basal ganglion cells, or that they were localized to different subcellular compartments within the same basal ganglion neurons. Recent studies have shown that DARPP-32, a soluble protein belonging to the first group, is present throughout the cytosol of medium-sized spiny neurons of the neostriatum and the striatonigral axons 15,26,35. By implication, the distinct distribution of ARPP-90 raised the possibility that this protein might be present in a different cell population.
The present results suggest a different conclusion. Using higher resolution techniques, we have now found that ARPP-90 is present not only in forebrain, but also in brainstem basal ganglia, albeit in lower apparent concentration in the brainstem region. Furthermore, the effects of local lesions in the neostriatum induced by quinolinic acid, a chemical excitotoxic agent which preferentially destroys the medium-sized spiny neurons and their efferent axons 30'39, strongly indicate that this cell type contains most of the ARPP-90 found in the basal ganglia. The percent decreases of ARPP-90 seen both in the caudatoputamen and in the substantia nigra were comparable to those of DARPP-32, which is specifically enriched in the medium-sized spiny neurons 26, and more pronounced than those of synapsin I, a protein enriched in virtually all nerve terminals 8,9. However, in spite of the same cellular localization, the apparent enrichment of ARPP-90 in the neostriatum where these neurons originate, but not in the substantia nigra, indicates that ARPP-90 has an intracellular distribution different from that of DARPP-32. Our data also suggest that these are species differences in the distribution of ARPP-90, since primate brains appear to contain considerable amounts of the protein in cortical regions lying outside the basal ganglia. Given that DARPP-32 is also present outside the basal ganglia in primate brains 37, there may exist species differences in the regional distribution of signal transduction systems involving this family of cyclic AMP-regulated phosphoproteins. ARPP-90 is distinct from several other brain proteins of M r = 90,000-100,000, presumed or shown to be phosphoproteins, which have recently been reported to be present in these brain regions. We have labeled the dopamine-binding subunit of the D 2 dopamine receptor, a protein of 94 kDa molecular mass 2, in rat striatal membranes with the photoaffinity label 8-N3-NAPS 2, and found that the receptor subunit migrates with a distinctly slower mobility upon S D S - P A G E than does ARPP-90 (data not shown). Similarly, a 96 kDa synaptosomal phosphoprotein which undergoes dephosphorylation following calcium influx 29 also displays a distinctly slower mobility upon S D S - P A G E than does ARPP-90 (not shown), in addition to being equally enriched in rat brain regions outside the basal ganglia (not shown). Finally, O'Carroll et al. 23 have reported a 94 kDa
156 phosphoprotein substrate for cyclic A M P - d e p e n d e n t protein kinase which was partially purified from the rat neostriatum. Since this protein was enriched in astroglial cells in neostriatum, cortex and cerebellum, it appears to be distinct from ARPP-90, which is enriched in neurons (this study) and is not detectable in the cerebellum with the methods used in this study (data not shown). In conclusion, our results present strong evidence for a localization of ARPP-90 in the medium-sized
activation)-containing neurons in the basal ganglia l, and that they exhibit D 1 receptor regulation of both neurotransmitter release and neuronal excitability 6'11A2'28'31.The possible role of ARPP-90 in the regulation of such functions awaits further study.
ACKNOWLEDGEMENTS
spiny neurons of the neostriatum and in their striatonigral axons. Recent reports indicate that these cells
This study was supported by USPHS G r a n t M H 40899, E P A CR-813826-01-0, and the Norwegian Council for Science and the Humanities (SIW). We
represent the predominant dopamine D 1 receptor (dopamine receptors coupled to adenylate cyclase
thank Ms. Irene Slizys and Mr. Barry Swimar for help in the preparation of the figures.
REFERENCES
10 Freifelder, D., Physical Biochemistry, Applications to Biochemistry and Molecular Biology, Freeman, New York, 1982. 11 Girault, J.-A., Spampinato, U., Savaki, H.E., Glowinski, J. and Besson, M.J., In vivo release of [3H]~-aminobutyric acid in the rat neostriatum - - I. Characterization and topographical heterogeneity of the effects of dopaminergic and cholinergic agents, Neuroscience, 19 (1986) 1101-1108. 12 Girault, J.-A., Spampinato, U., Glowinski, J. and Besson, M.J., In vivo relese of [3H]y-aminobutyric acid in the rat neostratum - - II. Opposing effects of D1 and D2 dopamine receptor stimulation in the dorsal caudate putamen, Neuroscience, 19 (1986) 1109-1117. 13 Grofova, I., Extrinsic connections of the neostriatum. In I. Divac and G.E. Oberg (Eds.), The Neostriatum, Pergamon, New York, 1979, pp. 37-51. 14 Helmerhorst, E. and Stokes, G.B., Microcentrifuge desalting: a rapid, quantitative method for desalting small amounts of protein, Anal. Biochem., 104 (1980) 130-135. 15 Hemmings, H.C., Jr. and Greengard, P., DARPP-32, a dopamine- and cyclic AMP-regulated phosphoprotein. Regional, tissue, and phylogenetic distribution, J. Neurosci., 6 (1986) 1469-1481. 16 Hemmings, H.C., Jr., Walaas, S.I., Ouimet, C.C. and Greengard, P., Dopamine receptors: regulation of protein phosphorylation. In I. Creese and C.M. Fraser (Eds.), Receptor Biochemistry and Methodology, Vol. 9: Structure and Function of Dopamine Receptors, Liss, New York, 1987, pp. 115-151. 17 Huttner, W.B. and Greengard, P., Multiple phosphorylation sites in Protein I, and their differential regulation by cyclic AMP and calcium, Proc. Natl. Acad. Sci. U.S.A., 76 (1979) 5402-5406. 18 Kaczmarek, L.K., Jennings, K.R., Strumwasser, F., Nairn, A.C., Walter, U., Wilson, F.D. and Greengard, P., Microinjection of catalytic subunit of cyclic AMP-dependent protein kinase enhances calcium action potentials of bag cell neurons in cell culture, Proc. Natl. Acad. Sci. U.S.A., 77 (1980) 7487-7491. 19 Kelly, P.T. and Lunges, M.W., Electrophoretic separation of nervous system proteins on exponential gradient polyacrylamide gels, J. Neurochem., 24 (1975) 1077-1079.
1 Altar, C.A. and Hauser, K., Topography of substantia nigra innervation by D1 receptor-containing striatal neurons, Brain Res., 410 (1987) 1-11. 2 Amlaiky, N. and Caron, M.G., Identification of the D2dopamine receptor binding subunit in several mammalian tissues and species by photoaffinity labeling, J. Neurochem., 47 (1986) 196-204. 3 Bolam, J.P., Powell, J.F., Totterdell, S. and Smith, A.D., The proportion of neurons in the rat neostriatum that project to the substantia nigra demonstrated using horseradish peroxidase conjugated with wheatgerm agglutinin, Brain Res., 220 (1981) 339-343. 4 Bradford, M.M., A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem., 72 (1976) 248-254. 5 Cala, S. and Greengard, P., Purification and biochemical studies of a basal ganglia-enriched phosphoprotein, Soc. Neurosci. Abstr., 13 (1987) 901. 6 Calabresi, P., Mercuri, N., Stanzione, P., Stefani, A. and Bernardi, G., Intracellular studies on the dopamine-induced firing inhibition of neostriatal neurons in vitro: evidence for D1 receptor involvement, Neuroscience, 20 (1987) 757-771. 7 Cleveland, D.W., Fischer, S.G., Kirschner, M.W. and Laemmli, U.K., Peptide mapping by limited proteolysis in sodium dodecyl sulfate, and analysis by gel electrophoresis, J. Biol. Chem., 252 (1977) 1102-1106. 8 DeCamilli, P., Harris, S.M., Huttner, W.B. and Greengard, P., Synapsin I (Protein I), a nerve terminal-specific phosphoprotein: II. Its specific association with synaptic vesicles demonstrated by immunocytochemistry in agaroseembedded synaptosomes, J. Cell Biol., 96 (1983) 1355-1373. 9 DeCamilli, P., Cameron, R. and Greengard, P., Synapsin I (Protein I), a nerve terminal-specific phosphoprotein: I. Its general distribution in synapses of the central and peripheral nervous system demonstrated by immunofluorescence in frozen and plastic sections, J. Cell Biol., 96 (1983) 1337-1354.
157 20 Krnig, J.F.R. and Klippel, R.A., The Rat Brain. A Stereotaxic Atlas of the Forebrain and Lower Parts of the Brain Stem, Williams and Wilkins, Baltimore, 1963. 21 Liu, A.Y.-C. and Greengard, P., Regulation by steroid hormones of phosphorylation of specific proteins common to several target organs, Proc. Natl. Acad. Sci. U.S.A., 73 (1976) 568-572. 22 Nauta, W.J.H. and Domesick, V.B., Afferent and efferent relationships of the basal ganglia. In Functions of the Basal Ganglia, Ciba Foundation Symposium 107, Pitman, London, 1984, pp. 3-23. 23 O'Carroll, A.-M., Patel, J. and Jacobowitz, D.M., Identification, partial purification and immunohistochemical localization of a 94 kD phosphoprotein present in astrocyte-like cells in the brain, Soc. Neurosci. Abstr., 13 (1987) 1134. 24 O'Farrell, P.H., High resolution two-dimensional electrophoresis of proteins, J. Biol. Chem., 250 (1975) 4007-4021. 25 O'Farrell, P.Z., Goodman, H.W. and O'Farrell, P.H., High resolution two-dimensional electrophoresis of basic as well as acidic proteins, Cell, 12 (1977) 1133-1142. 26 Ouimet, C.C., Miller, P.E., Hemmings, H.C., Jr., Walaas, S.I. and Greengard, P., DARPP-32, a dopamine- and adenosine 3':5'-monophosphate-regulated phosphoprotein enriched in dopamine-innervated brain regions. III. Immunocytochemical localization, J. Neurosci., 4 (1984) 111-124. 27 Peterson, G., A simplification of the protein assay of Lowry et al. which is more generally applicable, Anal. Biochem., 83 (1977) 346-356. 28 Reubi, J.-C., Iversen, L.L. and Jessel, T.M., Dopamine selectively increases 3H-GABA release from slices of rat substantia nigra in vitro, Nature (Lond.), 268 (1977) 652-654. 29 Robinson, P.J., Hauptschein, R., Lovenberg, W. and Dunkley, P.R., Dephosphorylation of synaptosomal proteins p96 and p139 is regulated by both depolarization and calcium, but not by a rise in cytosolic calcium alone, J. Neurochem., 48 (1987) 187-195. 30 Schwarcz, R., Whetsell, W.D. and Mangano, R.M., Quinolinic acid: an endogenous metabolite that produces axon-
sparing lesions in rat brain, Science, 219 (1983) 316-318. 31 Starr, M., Opposing roles of dopamine D1 and D2 receptors in nigral y-[3H]aminobutyric acid release?, J. Neurochem., 49 (1987) 1042-1049. 32 Walaas, S.I., Nairn, A.C. and Greengard, P., Regional distribution of calcium- and adenosine 3':5'-monophosphateregulated protein phosphorylation systems in mammalian brain. II. Soluble systems, J. Neurosci., 3 (1983) 302-311. 33 Walaas, S.I., Aswad, D.W. and Greengard, P., A dopamine- and cyclic AMP-regulated phosphoprotein enriched in dopamine-innervated brain regions, Nature (Lond.), 301 (1983) 69-71. 34 Walaas, S.I., Nairn, A.C. and Greengard, P., Regional distribution of calcium- and adenosine 3':5'-monophosphateregulated protein phosphorylation systems in mammalian brain. I. Particulate systems, J. Neurosci., 3 (1983) 291-301. 35 Walaas, S.I. and Greengard, P., DARPP-32, a dopamineand adenosine 3':5'-monophosphate-regulated phosphoprotein enriched in dopamine-innervated brain regions. I. Regional and cellular distribution in rat brain, J. Neurosci., 4 (1984) 84-98. 36 Walaas, S.I., Ouimet, C.C., Hemmings, H.C., Jr. and Greengard, P., DARPP-32: a dopamine-regulated phosphoprotein in the basal ganglia. In G.N. Woodruff, J.A. Poat and P.J. Roberts (Eds.), Dopaminergic Systems and their Regulation, MacMillan, London, 1986, pp. 165-179. 37 Walaas, S.I., Perdahl-Wallace, E., Winblad, B. and Greengard, P., Protein phosphorylation systems in postmortem human brain, J. Mol. Neurosci., in press. 38 Wilkening, D. and Makman, M.H., Activation of glycogen phosphorylase in rat caudate nucleus slices by 1-isopropylnorepinephrine and dibutyryl cyclic AMP, J. Neurochem., 28 (1977) 1001-1007. 39 Worley, P.F., Baraban, J.M. and Snyder, S.H., Heterogenous localization of protein kinase C in rat brain: autoradiographic analysis of phorbol ester binding, J. Neurosci., 6 (1986) 199-207.
149
BRM 70124
Localization of ARPP-90, a major 90 kiloDalton basal ganglionenriched substrate for cyclic AMP-dependent protein kinase, in striatonigral neurons in the rat brain S. Ivar Walaas, Steve Cala and Paul Greengard Laboratory of Molecular and Cellular Neuroscience, The Rochefeller University, New York, NY IO021 (U.S.A.)
(Accepted 25 October 1988) Key words: Neostriatum; Substantia nigra; Quinolinic acid; Striatonigral fiber; Phosphoprotein; Protein phosphorylation
Cyclic AMP-regulated phosphoproteins with specific cellular localizations in brain represent important targets through which this second messenger system can mediate or modulate distinct neurotransmitter signals. This study reports that two cyclic AMP-regulated phosphoproteins (Mr 90,000 and 93,000) found in brain share several properties, including similar isoelectric points and similar phosphopeptide maps. This protein doublet is particularly enriched in the forebrain basal ganglia, but it can also be found in the substantia nigra, a brainstem region which is a major target for fibers from the forebrain basal ganglia. Quinolinic acid lesions of neurons in the neostriatum decrease the levels of the 90/93 kDa phosphoprotein doublet to about the same extent as they reduce the levels of DARPP-32, a phosphoprotein specifically enriched in striatonigral medium-sized spiny neurons. These reductions are seen in both the neostriatum and the substantia nigra. Therefore, within the basal ganglia, the 90/93 kDa phosphoprotein doublet, termed adenosine 3' :5'-monophosphate-regulated phosphoprotein, Mr = 90,000 (ARPP-90), is largely, if not solely, present in striatonigral cells and fibers. The specific localization in these neurons suggests that ARPP-90 could be important in receptor-regulated, cyclic AMP-mediated functions in the striatonigral neurons. INTRODUCTION Extensive evidence indicates that receptor-regulated, cyclic AMP-mediated protein phosphorylation is of considerable importance in the functions of the basal ganglia of mammalian brain t6,36. In regions such as the neostriatum and substantia nigra, neurotransmitter release, regulation of neuronal excitability and cellular metabolism appear to be sensitive to changes in intracellular cyclic A M P levels 6,11'12,31,38. It is of great interest, therefore, that the basal ganglia of mammalian brain contain several protein substrates for cyclic AMP-dependent protein kinase, including two phosphoprotein species which migrate closely together on sodium dodecylsulfate-polyacrylamide gel electrophoresis ( S D S - P A G E ) with approximate molecular masses of 90 and 93 kilodalton (kDa) (previously reported as 96/98 kDa, the differences in apparent molecular weights being attrib-
utable to different electrophoretic conditions) 5,32. In the rat brain, the 90/93 k D a doublet is particularly enriched in forebrain basal-ganglion regions such as the caudatoputamen, nucleus accumbens and olfactory tubercle, but is not easily detected by onedimensional S D S - P A G E in the output regions of the basal ganglia, such as the globus pallidus and substantia nigra 3z. This localization contrasts with that of another basal ganglion-enriched phosphoprotein, D A R P P - 3 2 , which is specifically concentrated in the medium-sized spiny neurons of the neostriatum as well as in their efferent striatopallidal and striatonigral axons 26'35. Thus, D A R P P - 3 2 is easily detected by in vitro phosphorylation and S D S - P A G E in both the forebrain neostriatum and the substantia nigra 33'35. Therefore, it appeared possible that the 90/93 kDa phosphoproteins might have a cellular distribution in the rat basal ganglia which was distinct from that of DARPP-3232. The present study has ex-
Correspondence: S.I. Walaas, Institute of Medical Biochemistry, University of Oslo, P.O. Box 1112, 0317 Oslo 3, Norway.
0169-328X/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
150 amined this question, using various gel electrophoretic and peptide mapping techniques combined with chemical lesioning of the medium-sized spiny striatonigral neurons and their efferent axons 35. MATERIALS AND METHODS
nylfluoride (PMSF, 0.1 mM), and homogenized with 15 strokes in glass-Teflon tissue grinders rotating at 1200 rpm. Soluble fractions were prepared by centrifugation at 150,000 g for 30 min, washing the pellet once, and pooling the supernatants as described 32,34. The soluble fractions were desalted by gel filtration on minicolumns TM.
Materials [},-3zp]ATP was from New England Nuclear, Boston. Staphylococcus aureus V8 protease was from Miles Biochemicals, aprotinin (Trasylol) was from Mobay Chemical Corporation, and TPCK-trypsin was from Cooper Biomedical. Ampholines (Servalyte 3-10) were from Serva, Heidelberg, F.R.G. The catalytic subunit of cyclic AMP-dependent protein kinase was purified as described TM, and was a gift from Dr. A.C. Nairn, The Rockefeller University. Other chemicals of analytical grade were from standard commercial suppliers.
Preparation of lesioned animals Male Sprague-Dawley rats (150-200 g b. wt.) were anesthetized with chloral hydrate (0.35 g/kg, i.p.) and mounted on a stereotactic frame. 1 #1 of 0.2 M quinolinic acid (adjusted to pH 7.4 with NaOH), an excitotoxin which produces 'axon-sparing' lesions of local cells in the neostriatum and the striatofugal fibers emanating from these neurons 3°'39, was infused into the rostral neostriatum at coordinates A:8900; L:2500 and V:500, following the atlas of K6nig and Klippel2°. All animals survived for 5-7 days before analysis.
Preparation of tissue Rabbits were killed under barbiturate anesthesia, while lesioned and unlesioned rats were stunned and decapitated. The brains were rapidly removed and cooled on ice. The neocortex, neostriatum (caudatoputamen), substantia nigra and other brain regions were dissected as described 34, and the brain samples were weighed and put on ice. Similar samples were dissected from frozen rhesus monkey brain (kindly provided by Dr. T. Bartfai, University of Stockholm, Sweden). Brain samples were added (approx. 1:10, w/v) to ice-cold standard buffer containing 10 mM HEPES (pH 7.4), 2 mM EDTA, 1 mM dithiothreitol (DTT), and the protease inhibitors aprotinin (50 kallikrein inhibitor units/ml) and phenylmethyl sulfo-
Phosphorylation of proteins Proteins in the tissue samples were phosphorylated with [T-32p]ATP as described 32'3~. Briefly, aliquots of homogenates (50/~g protein from rat substantia nigra, 100-200 ktg protein for other regions) or soluble fractions (20-50 #g protein) from the different brain regions were incubated in a medium (final vol. 100 Izl) containing (final conc.) 25 mM HEPES (pH 7.4), 10 mM MgC12, 1 mM EDTA, 1 mM EGTA, 1 mM DTT and 10/~M [7-32P]ATP (5-15 Ci/mmol). Protein phosphorylation systems were activated by the addition of either 10/~M 8-bromo cyclic AMP, 2/~M cyclic AMP plus 1 mM isobutylmethylxanthine, or 1.25 mM CaCI 2. In analyses of samples from the lesioned animals and their controls, catalytic subunit of cyclic AMP-dependent protein kinase (50 nM final conc.) was added to the reaction mixture. The samples were preincubated at 30 °C for 60 s, the reactions were initiated by the addition of the [y-32p]ATP, and terminated after a 30 s incubation period by the addition of a stop-solution containing (final conc.) 3% SDS, 60 mM Tris-HC1 (pH 6.8), 5% glycerol, 2% 2-mercaptoethanol, and a trace of Bromophenol blue. The samples were then heated for 2 min in a boiling water bath, and cooled to room temperature before analysis of phosphorylated proteins.
Analytical separation of proteins Phosphoproteins were separated by one- or twodimensional polyacrylamide gel electrophoresis (PAGE). For one-dimensional SDS-PAGE, aliquots were directly applied to gels prepared with different amounts of acrylamide as described 19'21'34, and proteins were separated by constant current electrophoresis 1°. For two-dimensional separations, samples were first precipitated twice with 90% cold acetone (-20 °C) to remove ATP and salts. Proteins were then solubilized with 0.5% SDS, 9 M urea, 0.1% ampholines and 1% 2-mercaptoethanol, followed by an equal volume of 4% NP-40, 9 M urea,
151 0.1% ampholines and 1% 2-mercaptoethanol, as described 25. Separation was achieved in 12 cm long tube gels (1.5 mm i.d.) containing 2% ampholines (pH 3-10) 24. The gels were subjected to isoelectrofocusing (IEF) at 500 V for 16 h 24 or non-equilibrium pHgradient gel electrophoresis (NEPHGE) at 400 V for 4 h 25, carefully extruded and directly loaded onto the second dimension gels. The proteins were then separated by S D S - P A G E as described above. All slab gels were stained, fixed and destained as described 34. Phosphoproteins were visualized by autoradiography of the dried gels. Apparent molecular masses of different proteins were determined by comparison with standard proteins (Sigma) included in separate lanes, while apparent isoelectric points were determined by comparison with a pH gradient standard curve determined on separate tube gels run in parallel 24. Quantitation of phosphoproteins was achieved by cutting the gel pieces containing the protein of interest out of the gel, using the autoradiogram as a guide, and analysing 32p-content by Cerenkov counting. Preliminary experiments, using an incubation time of 30 s, showed that the incorporation of 32p into phosphoproteins of apparent molecular masses between 90 and 98 kDa was linear with amount of protein up to at least 50/~g protein in cytosolic fractions, while homogenate samples showed greater variability (not shown). Therefore, analysis of the level of ARPP-90 in the neostriatum was performed on cytosol preparations. Due to the small amount of material available, homogenate samples were used in assays of single substantiae nigrae. The changes observed in these phosphoproteins were compared to the changes observed in synapsin I, a phosphoprotein enriched in all nerve terminals 8,9, DARPP-32, a phosphoprotein marker for the striatonigral neurons 26'35, and pyruvate dehydrogenase, a widely distributed mitochondrial phosphoprotein 34.
Tryptic phosphopeptide mapping Dried and washed gel pieces containing phosphoproteins 17 were incubated in 1 ml of 50 mM NH4HCO 3 (pH 8.0) containing TPCK-trypsin (0.075 mg/ml) at 37 °C for 24 h. The gel slices were washed with 0.5 ml of 50 mM NH4CO 3 at 37 °C for 4 h, and the collected supernatants were lyophilized and resuspended in electrophoresis buffer (10% acetic
acid, 1% pyridine, pH 3.5). Phosphopeptides were separated on thin layer cellulose sheets (Kodak Eastman) by electrophoresis at 400 V for 90 min, followed by chromatography in the second dimension in a buffer containing pyridine, 1-butanol, water, acetic acid (15:10:12:3) 17. 32p-labeled peptides were identified by autoradiography of the dried cellulose sheets.
Miscellaneous methods Protein was analysed by the method of Bradford 4 or the method of Peterson 27, using bovine serum albumin as standard. Peptide mapping of phosphoproteins by means of incomplete proteolysis was performed using 5 ktg of Staphylococcus aureus V8 protease, essentially as described 7'17. RESULTS Incubation of homogenates (Fig. 1) or cytosol fractions (not shown) from rat caudatoputamen and cerebral cortex with [y-32p]ATP and cyclic AMP, followed by one-dimensional S D S - P A G E and autoradiographic detection of phosphoproteins, showed an apparent doublet of phosphorylated protein bands of apparent molecular masses of 90 and 93 kDa (using Mr = 94,000 for phosphorylase as a standard). Addition of CaC12 (250/~M flee) to the reaction mixture did not effect the phosphorylation of this same protein (Fig. 1A). Phosphopeptide mapping of these two phosphoprotein bands, using V8 protease from S. aureus to achieve incomplete digestion in SDS, showed that the two protein bands generated identical phosphopeptides (Fig. 1B). Two-dimensional gel electrophoresis of phosphorylated samples from the rat caudate nucleus, using IEF in the first dimension, showed that the two phosphoproteins comigrated in the first dimension, displaying an apparent isoelectric point of approximately 6.0 (not shown). Because these two protein bands appear to be so closely related, we have designated them collectively as ARPP-90 (adenosine 3':5'-monophosphate-regulated phosphoprotein, M r = 90,000). Analysis of various rat brain regions indicated ARPP-90 to be highly enriched in the neostriatum, including the caudatoputamen, nucleus accumbens and olfactory tubercle 32. A much lower amount of this phosphoprotein doublet could be detected in
152
B.
A. STRIKnJM CC~]rEX I
it
DARPF
cAM, ~,_,,
•
---
I
÷--
--
+
Fig. 1. Autoradiogram showing the presence of ARPP-90, a 90/93 kDa protein substrate for cyclic AMP-dependent protein kinase, in rat neostriatum. A: homogenates from neostriatum or cerebral cortex were incubated with 10/aM [7-32p]ATP in the presence of either 2/aM cyclic AMP plus 1 mM isobutylmethylxanthine (IBMX) or 0.25 mM free Ca 2÷ as described in the text. Proteins were separated by SDS-PAGE, and phosphoproteins were visualized by autoradiography. Identified phosphoproteins are indicated. ARPP-90, basal ganglion-enriched phosphoprotein (M r 90,000); synapsin I, a synaptic vesicle-associated phosphoprotein; PDH, a-subunit of pyruvate dehydrogenase; DARPP-32, dopamine- and cyclic AMP-regulated phosphoprotein (M r 32,000); MBP, myelin basic protein. B: peptide mapping of ARPP-90, showing homology between the two phosphoprotein bands. Gel pieces containing the 90 and 93 kDa bands were separately cut out of the gel shown in A, and subjected to peptide mapping with S. aureus V8 protease as described in the text. Molecular masses of phosphopeptide fragments are given in kiloDalton.
homogenates of rat cerebral cortex (Fig. 1A), diencephalic and mesencephalic regions, and the protein was not detected in brainstem or cerebellum by onedimensional S D S - P A G E 32. Similar analysis of rabbit brain, monkey brain (Fig. 2) and human brain 37 showed that phosphoprotein bands having similar apparent molecular masses and similar phosphopeptide maps were present in the neostriatum of these species as well. In the rabbit brain, ARPP-90 displayed a regional distribution similar to that seen in the rat brain, with considerably more of the phosphoprotein being seen in the neostriatum than in the cortical sample (not shown). Monkey (Fig. 2) and human brains 37, in contrast, contained detectable amounts of ARPP-90 in both cortical and basal ganglion regions. These distribution patterns are somewhat similar to that of DARPP-32 (Fig. 2), which in the rat brain is known to be highly enriched in the mediumsized spiny neurons of the neostriatum and the striatofugal fibers emanating from these neurons 26, but
A i
B ......~
t
C ,J
D i
i
i
Fig. 2. Autoradiogram showing the presence of ARPP-90 in rat, rabbit and monkey brain. Homogenates from neostriatum of rat (A), rabbit (B) and monkey (C) and cerebral cortex from monkey (D) were phosphorylated in the presence of absence of 10/aM 8-bromo cyclic AMP, and phosphoproteins were separated by S D S - P A G E and visualized by autoradiography. Abbreviations as in the legend to Fig. 1.
153
Cellular localization of ARPP-90 in the neostriatum
-
+
Fig. 3. Autoradiogram showing the effect of quinolinic acid lesion on phosphoproteins in rat neostriatum. Homogenates from neostriatum from control or lesioned animals were phosphorylated in the absence or presence of 50 nM catalytic subunit of cyclic AMP-dependent protein kinase as described in the text. Phosphoproteins were separated by SDS-PAGE and visualized by autoradiography. Some of the identified proteins are indicated. Abbreviations as in the legend to Fig. 1. which appears to be present in m o n k e y cerebral cortex (Fig. 2) in somewhat higher amounts than in rat cortex.
A.
The regional and species distribution of ARPP-90 indicated that it might be enriched in specific cell types in rat brain. The cellular localization of the protein was further examined in the rat basal ganglia, which are well suited to various lesioning techniques due to the organization of the afferent and efferent connections and intrinsic neurons in these brain regions 22'35. Infusion of quinolinic acid into the caudatoputamen was used to lesion the medium-sized spiny neurons of the neostriatum, the cells of origin of the striatopallidal and striatonigral fiber tracts 3,13. Following 5 - 7 days survival, ARPP-90 was analysed by phosphorylation of homogenate or supernatant samples in the absence or presence of exogenous catalytic subunit of cyclic AMP-dependent protein kinase. This lesion led to a pronounced decrease in phosphoproteins of 90-93 kDa in homogenates (Fig. 3) and in supernatant fractions (not shown). Scintillation counting of excised gel pieces showed that the levels of the 90/93 kDa phosphoproteins amounted to 34 + 10% (mean + S.D., n = 6) of that in control samples. Tryptic peptide mapping of the phosphoproteins contained in these gel pieces showed that this decrease was due to disappearance of ARPP-90, since the distinct tryptic phosphopeptides which decreased in the lesioned animals (Fig. 4) are the same
B,
Fig. 4. Autoradiogram showing two-dimensional peptide maps of ARPP-90 in phosphorylated cytosol fractions of neostriatum from control (A) and quinolinic acid-lesioned (B) rats. Neostriatal cytosol was phosphorylated in the presence of the catalytic subunit of cyclic AMP-dependent protein kinase, and proteins were separated by SDS-PAGE. Gel pieces containing 90/93 kDa phosphoproteins were cut out of the gels and digested with trypsin, and the resulting phosphopeptides (labeled 1, 2, 3) were separated by electrophoresis and chromatography as described in the text. Small circles indicate application points.
154 as those obtained by mapping of purified ARPP-90 from both rat and bovine brain 5. The decrease in ARPP-90 following local quinolinic acid infusion was also compared with that of other phosphoproteins in the neostriatum, including DARPP-32 and synapsin I. Lesioned animals displayed 27 + 9% of 32p-incorporation into DARPP-32 and 59 + 6% into synapsin I when compared to control samples (mean + S.D., n = 5). Thus, the decreases of ARPP-90 and DARPP32 were comparable, and much greater than that of synapsin I.
nigra (Fig. 5), which was identified as ARPP-90 by peptide mapping (not shown). However, the amount of ARPP-90 appeared considerably lower in the substantia nigra than in the neostriatum, the phosphorylated ARPP-90 in the substantia nigra representing only 40-50% of phosphorylated DARPP-32 (values from 5 experiments), whereas in the caudatoputamen ARPP-90 consistently incorporated 1.5-2 times the amount of 32p that DARPP-32 did (Figs. I and 3; data not shown). In the substantia nigra from animals which had received injections of quinolinic acid into the caudate nucleus, the amount of ARPP-90 was greatly decreased (Fig. 5), indicating that destruction of the striatonigral fibers led to a disappearance of the protein. We compared the phosphorylation of ARPP-90 in each gel to the phosphorylation of several wellcharacterized proteins present in the striatonigral fibers, including synapsin I and DARPP-32 (ref. 35), in the same gels. Because the total amount of protein which entered the two-dimensional gels varied considerably between different experiments (not shown), the ratios of lesioned to control levels of these various phosphoproteins were normalized against the level of the phosphorylated 42 kDa a-sub-
Cellular localization of ARPP-90 in substantia nigra The localization of ARPP-90 in basal ganglion regions outside of the neostriatum was studied in the substantia nigra, the target region of the striatonigrai fiber tract and the origin of major output tracts from the mammalian basal ganglia 22. Analysis of phosphoproteins in homogenates from the substantia nigra was performed with two-dimensional N E P H G E , since one-dimensional S D S - P A G E was unable to identify ARPP-90 in this region 32. Comparison of autoradiograms of such two-dimensional gels from rat substantia nigra and neostriatum showed the presence of a 90-93 kDa phosphoprotein in the substantia
A-
~
, i
B.
~
, I
SYNAI
BASIC '1 ACIDIC Fig. 5. Autoradiogram showing effect of quinolinic acid infusion into rat neostriatum on phosphoproteins in substantia nlgra. Homogenates of substantia nigra from control (A) or lesioned (B) rats were phosphorylated in the presence of 50 nM of the catalytic subunit of cyclic AMP-dependent protein kinase, and separated by non-equilibrium pH-gradient electrophoresis (NEPHGE, arrows) in the first dimension and by SDS-PAGE in the second dimension. Phosphorylated proteins were visualized by autoradiography. Some of the identified proteins are indicated. Stippled circle indicates location of ARPP-90.21K, phosphoprotein enriched in the basal ganglia32. Other abbreviations as in the legend to Fig. 1.
155 unit of the mitochondrial enzyme pyruvate dehydrogenase (PDH), since the latter protein, a widely distributed phosphoprotein present in all cell types, was found not to change in the substantia nigra following the lesions (Fig. 5). This analysis showed that the ratio ARPP-90/PDH in the lesioned animals represented 22 + 5% of that found in control animals (mean + S.D., n = 5), while the ratios of DARPP32/PDH and synapsin I/PDH also were considerably decreased (18 + 4% and 31 + 3% of control values, respectively, mean + S.D., n = 3). Since the decreases in the values for DARPP-32 and synapsin I were similar to the lesion-induced decreases in the absolute levels of these proteins measured by other analytical methods 35, the ratio ARPP-90/PDH appears to give a valid estimate of the changes in ARPP-90 induced by the lesions. These results, therefore, indicate that ARPP-90 in the rat caudatoputamen is highly enriched in the medium-sized spiny neurons and, albeit in lower concentrations, in the axons and terminals of the striatonigral fibers. DISCUSSION ARPP-90 is one of a group of phosphoproteins found in the rat brain which are substrates for cyclic AMP-dependent protein kinase and which are enriched in the basal ganglia 32,34. This group of proteins was originally observed to display two distinct distribution patterns. Some proteins in this group, including DARPP-32, were enriched to the same extent in both forebrain ('input') basal ganglia regions (e.g. the caudatoputamen) and in 'output' basal ganglia regions (e.g. the globus pallidus and substantia nigra). Other basal ganglion-enriched proteins, including ARPP-90, were seen only in the forebrain basal ganglion regions. These observations implied that the two groups of proteins either were present in different groups of basal ganglion cells, or that they were localized to different subcellular compartments within the same basal ganglion neurons. Recent studies have shown that DARPP-32, a soluble protein belonging to the first group, is present throughout the cytosol of medium-sized spiny neurons of the neostriatum and the striatonigral axons 15,26,35. By implication, the distinct distribution of ARPP-90 raised the possibility that this protein might be present in a different cell population.
The present results suggest a different conclusion. Using higher resolution techniques, we have now found that ARPP-90 is present not only in forebrain, but also in brainstem basal ganglia, albeit in lower apparent concentration in the brainstem region. Furthermore, the effects of local lesions in the neostriatum induced by quinolinic acid, a chemical excitotoxic agent which preferentially destroys the medium-sized spiny neurons and their efferent axons 30'39, strongly indicate that this cell type contains most of the ARPP-90 found in the basal ganglia. The percent decreases of ARPP-90 seen both in the caudatoputamen and in the substantia nigra were comparable to those of DARPP-32, which is specifically enriched in the medium-sized spiny neurons 26, and more pronounced than those of synapsin I, a protein enriched in virtually all nerve terminals 8,9. However, in spite of the same cellular localization, the apparent enrichment of ARPP-90 in the neostriatum where these neurons originate, but not in the substantia nigra, indicates that ARPP-90 has an intracellular distribution different from that of DARPP-32. Our data also suggest that these are species differences in the distribution of ARPP-90, since primate brains appear to contain considerable amounts of the protein in cortical regions lying outside the basal ganglia. Given that DARPP-32 is also present outside the basal ganglia in primate brains 37, there may exist species differences in the regional distribution of signal transduction systems involving this family of cyclic AMP-regulated phosphoproteins. ARPP-90 is distinct from several other brain proteins of M r = 90,000-100,000, presumed or shown to be phosphoproteins, which have recently been reported to be present in these brain regions. We have labeled the dopamine-binding subunit of the D 2 dopamine receptor, a protein of 94 kDa molecular mass 2, in rat striatal membranes with the photoaffinity label 8-N3-NAPS 2, and found that the receptor subunit migrates with a distinctly slower mobility upon S D S - P A G E than does ARPP-90 (data not shown). Similarly, a 96 kDa synaptosomal phosphoprotein which undergoes dephosphorylation following calcium influx 29 also displays a distinctly slower mobility upon S D S - P A G E than does ARPP-90 (not shown), in addition to being equally enriched in rat brain regions outside the basal ganglia (not shown). Finally, O'Carroll et al. 23 have reported a 94 kDa
156 phosphoprotein substrate for cyclic A M P - d e p e n d e n t protein kinase which was partially purified from the rat neostriatum. Since this protein was enriched in astroglial cells in neostriatum, cortex and cerebellum, it appears to be distinct from ARPP-90, which is enriched in neurons (this study) and is not detectable in the cerebellum with the methods used in this study (data not shown). In conclusion, our results present strong evidence for a localization of ARPP-90 in the medium-sized
activation)-containing neurons in the basal ganglia l, and that they exhibit D 1 receptor regulation of both neurotransmitter release and neuronal excitability 6'11A2'28'31.The possible role of ARPP-90 in the regulation of such functions awaits further study.
ACKNOWLEDGEMENTS
spiny neurons of the neostriatum and in their striatonigral axons. Recent reports indicate that these cells
This study was supported by USPHS G r a n t M H 40899, E P A CR-813826-01-0, and the Norwegian Council for Science and the Humanities (SIW). We
represent the predominant dopamine D 1 receptor (dopamine receptors coupled to adenylate cyclase
thank Ms. Irene Slizys and Mr. Barry Swimar for help in the preparation of the figures.
REFERENCES
10 Freifelder, D., Physical Biochemistry, Applications to Biochemistry and Molecular Biology, Freeman, New York, 1982. 11 Girault, J.-A., Spampinato, U., Savaki, H.E., Glowinski, J. and Besson, M.J., In vivo release of [3H]~-aminobutyric acid in the rat neostriatum - - I. Characterization and topographical heterogeneity of the effects of dopaminergic and cholinergic agents, Neuroscience, 19 (1986) 1101-1108. 12 Girault, J.-A., Spampinato, U., Glowinski, J. and Besson, M.J., In vivo relese of [3H]y-aminobutyric acid in the rat neostratum - - II. Opposing effects of D1 and D2 dopamine receptor stimulation in the dorsal caudate putamen, Neuroscience, 19 (1986) 1109-1117. 13 Grofova, I., Extrinsic connections of the neostriatum. In I. Divac and G.E. Oberg (Eds.), The Neostriatum, Pergamon, New York, 1979, pp. 37-51. 14 Helmerhorst, E. and Stokes, G.B., Microcentrifuge desalting: a rapid, quantitative method for desalting small amounts of protein, Anal. Biochem., 104 (1980) 130-135. 15 Hemmings, H.C., Jr. and Greengard, P., DARPP-32, a dopamine- and cyclic AMP-regulated phosphoprotein. Regional, tissue, and phylogenetic distribution, J. Neurosci., 6 (1986) 1469-1481. 16 Hemmings, H.C., Jr., Walaas, S.I., Ouimet, C.C. and Greengard, P., Dopamine receptors: regulation of protein phosphorylation. In I. Creese and C.M. Fraser (Eds.), Receptor Biochemistry and Methodology, Vol. 9: Structure and Function of Dopamine Receptors, Liss, New York, 1987, pp. 115-151. 17 Huttner, W.B. and Greengard, P., Multiple phosphorylation sites in Protein I, and their differential regulation by cyclic AMP and calcium, Proc. Natl. Acad. Sci. U.S.A., 76 (1979) 5402-5406. 18 Kaczmarek, L.K., Jennings, K.R., Strumwasser, F., Nairn, A.C., Walter, U., Wilson, F.D. and Greengard, P., Microinjection of catalytic subunit of cyclic AMP-dependent protein kinase enhances calcium action potentials of bag cell neurons in cell culture, Proc. Natl. Acad. Sci. U.S.A., 77 (1980) 7487-7491. 19 Kelly, P.T. and Lunges, M.W., Electrophoretic separation of nervous system proteins on exponential gradient polyacrylamide gels, J. Neurochem., 24 (1975) 1077-1079.
1 Altar, C.A. and Hauser, K., Topography of substantia nigra innervation by D1 receptor-containing striatal neurons, Brain Res., 410 (1987) 1-11. 2 Amlaiky, N. and Caron, M.G., Identification of the D2dopamine receptor binding subunit in several mammalian tissues and species by photoaffinity labeling, J. Neurochem., 47 (1986) 196-204. 3 Bolam, J.P., Powell, J.F., Totterdell, S. and Smith, A.D., The proportion of neurons in the rat neostriatum that project to the substantia nigra demonstrated using horseradish peroxidase conjugated with wheatgerm agglutinin, Brain Res., 220 (1981) 339-343. 4 Bradford, M.M., A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem., 72 (1976) 248-254. 5 Cala, S. and Greengard, P., Purification and biochemical studies of a basal ganglia-enriched phosphoprotein, Soc. Neurosci. Abstr., 13 (1987) 901. 6 Calabresi, P., Mercuri, N., Stanzione, P., Stefani, A. and Bernardi, G., Intracellular studies on the dopamine-induced firing inhibition of neostriatal neurons in vitro: evidence for D1 receptor involvement, Neuroscience, 20 (1987) 757-771. 7 Cleveland, D.W., Fischer, S.G., Kirschner, M.W. and Laemmli, U.K., Peptide mapping by limited proteolysis in sodium dodecyl sulfate, and analysis by gel electrophoresis, J. Biol. Chem., 252 (1977) 1102-1106. 8 DeCamilli, P., Harris, S.M., Huttner, W.B. and Greengard, P., Synapsin I (Protein I), a nerve terminal-specific phosphoprotein: II. Its specific association with synaptic vesicles demonstrated by immunocytochemistry in agaroseembedded synaptosomes, J. Cell Biol., 96 (1983) 1355-1373. 9 DeCamilli, P., Cameron, R. and Greengard, P., Synapsin I (Protein I), a nerve terminal-specific phosphoprotein: I. Its general distribution in synapses of the central and peripheral nervous system demonstrated by immunofluorescence in frozen and plastic sections, J. Cell Biol., 96 (1983) 1337-1354.
157 20 Krnig, J.F.R. and Klippel, R.A., The Rat Brain. A Stereotaxic Atlas of the Forebrain and Lower Parts of the Brain Stem, Williams and Wilkins, Baltimore, 1963. 21 Liu, A.Y.-C. and Greengard, P., Regulation by steroid hormones of phosphorylation of specific proteins common to several target organs, Proc. Natl. Acad. Sci. U.S.A., 73 (1976) 568-572. 22 Nauta, W.J.H. and Domesick, V.B., Afferent and efferent relationships of the basal ganglia. In Functions of the Basal Ganglia, Ciba Foundation Symposium 107, Pitman, London, 1984, pp. 3-23. 23 O'Carroll, A.-M., Patel, J. and Jacobowitz, D.M., Identification, partial purification and immunohistochemical localization of a 94 kD phosphoprotein present in astrocyte-like cells in the brain, Soc. Neurosci. Abstr., 13 (1987) 1134. 24 O'Farrell, P.H., High resolution two-dimensional electrophoresis of proteins, J. Biol. Chem., 250 (1975) 4007-4021. 25 O'Farrell, P.Z., Goodman, H.W. and O'Farrell, P.H., High resolution two-dimensional electrophoresis of basic as well as acidic proteins, Cell, 12 (1977) 1133-1142. 26 Ouimet, C.C., Miller, P.E., Hemmings, H.C., Jr., Walaas, S.I. and Greengard, P., DARPP-32, a dopamine- and adenosine 3':5'-monophosphate-regulated phosphoprotein enriched in dopamine-innervated brain regions. III. Immunocytochemical localization, J. Neurosci., 4 (1984) 111-124. 27 Peterson, G., A simplification of the protein assay of Lowry et al. which is more generally applicable, Anal. Biochem., 83 (1977) 346-356. 28 Reubi, J.-C., Iversen, L.L. and Jessel, T.M., Dopamine selectively increases 3H-GABA release from slices of rat substantia nigra in vitro, Nature (Lond.), 268 (1977) 652-654. 29 Robinson, P.J., Hauptschein, R., Lovenberg, W. and Dunkley, P.R., Dephosphorylation of synaptosomal proteins p96 and p139 is regulated by both depolarization and calcium, but not by a rise in cytosolic calcium alone, J. Neurochem., 48 (1987) 187-195. 30 Schwarcz, R., Whetsell, W.D. and Mangano, R.M., Quinolinic acid: an endogenous metabolite that produces axon-
sparing lesions in rat brain, Science, 219 (1983) 316-318. 31 Starr, M., Opposing roles of dopamine D1 and D2 receptors in nigral y-[3H]aminobutyric acid release?, J. Neurochem., 49 (1987) 1042-1049. 32 Walaas, S.I., Nairn, A.C. and Greengard, P., Regional distribution of calcium- and adenosine 3':5'-monophosphateregulated protein phosphorylation systems in mammalian brain. II. Soluble systems, J. Neurosci., 3 (1983) 302-311. 33 Walaas, S.I., Aswad, D.W. and Greengard, P., A dopamine- and cyclic AMP-regulated phosphoprotein enriched in dopamine-innervated brain regions, Nature (Lond.), 301 (1983) 69-71. 34 Walaas, S.I., Nairn, A.C. and Greengard, P., Regional distribution of calcium- and adenosine 3':5'-monophosphateregulated protein phosphorylation systems in mammalian brain. I. Particulate systems, J. Neurosci., 3 (1983) 291-301. 35 Walaas, S.I. and Greengard, P., DARPP-32, a dopamineand adenosine 3':5'-monophosphate-regulated phosphoprotein enriched in dopamine-innervated brain regions. I. Regional and cellular distribution in rat brain, J. Neurosci., 4 (1984) 84-98. 36 Walaas, S.I., Ouimet, C.C., Hemmings, H.C., Jr. and Greengard, P., DARPP-32: a dopamine-regulated phosphoprotein in the basal ganglia. In G.N. Woodruff, J.A. Poat and P.J. Roberts (Eds.), Dopaminergic Systems and their Regulation, MacMillan, London, 1986, pp. 165-179. 37 Walaas, S.I., Perdahl-Wallace, E., Winblad, B. and Greengard, P., Protein phosphorylation systems in postmortem human brain, J. Mol. Neurosci., in press. 38 Wilkening, D. and Makman, M.H., Activation of glycogen phosphorylase in rat caudate nucleus slices by 1-isopropylnorepinephrine and dibutyryl cyclic AMP, J. Neurochem., 28 (1977) 1001-1007. 39 Worley, P.F., Baraban, J.M. and Snyder, S.H., Heterogenous localization of protein kinase C in rat brain: autoradiographic analysis of phorbol ester binding, J. Neurosci., 6 (1986) 199-207.