Modulation of protein tyrosine phosphorylation in rat insular cortex ...

6 downloads 0 Views 1MB Size Report
Jan 10, 1994 - cortex after conditioned taste aversion training ..... to accelerated blood supply and enhanced metabolic activity in the activated insular cortex.
Proc. Natl. Acad. Sci. USA Vol. 92, pp. 1157-1161, February 1995 Neurobiology

Modulation of protein tyrosine phosphorylation in rat insular cortex after conditioned taste aversion training (taste learning/taste cortex/cortical plasticity/tyrosine kinase)

KOBI ROSENBLUM*, RINA SCHUL*, NOAM MEIRI*, YARON R. HADARIt, YEHIEL ZICKt, AND YADIN DUDAI* Departments of *Neurobiology and tChemical Immunology, The Weizmann Institute of Science, Rehovot 76100, Israel

Communicated by James L. McGaugh, University of California, Irvine, CA, June 14, 1994 (received for review January 10, 1994)

Analysis of the modulation of protein tyrosine phosphorylation in brain areas that subserve distinct experiencedependent behaviors can contribute to our understanding of the function of this signal transduction process in neuronal plasticity. In our laboratory we investigate the role of the insular cortex in taste memory in the rat (14). A behavioral paradigm offering numerous experimental advantages for the analysis of taste learning and memory is conditioned taste aversion (CTA; refs. 15 and 16). CTA taps an ecologically critical behavior, in which an animal learns to avoid a taste if the first encounter with that taste is followed by poisoning. In CTA, learning is very fast and memory robust, and, in contrast with classical and instrumental conditioning, the association between the taste and the negative reinforcer tolerates a delay of many hours. CTA thus permits separate analysis in time of the acquisition of information about the taste and about the associability of that taste with a reinforcer. We have demonstrated that the formation of the memory for the novel taste can be disrupted by local microinjection of a protein synthesis inhibitor into the gustatory area in the insular cortex (14). The effect of the protein synthesis inhibitor is specific to the gustatory cortex and to the time window of training (14). The finding that critical molecular processes involving protein synthesis take place in the insular cortex during taste learning, combined with earlier reports that gustatory cortex is involved in the formation of memory of CTA (17, 18), led us to search for modulation by CTA training of signal transduction systems that might trigger neuronal remodeling in that cortex. Toward that end, we have analyzed protein tyrosine phosphorylation in the insular cortex after CTA training and after separate exposure to the conditioned stimulus (CS; the novel taste) and the unconditioned stimulus (US; transient poisoning) used in CTA training. The methodology involved behavioral manipulation followed by excision of the appropriate brain area and immunoblot analysis of the homogenate with specific anti-phosphotyrosine (aPY) antibodies. We report here that CTA training leads to a marked increase in tyrosine phosphorylation of several proteins in the insular cortex but not in other brain areas. Exposure only to a novel taste also increases the phosphorylation of these proteins but to a smaller degree, whereas administration of the malaiseinducing agent only has no such effect. A major protein so modulated, of 180 kDa, is a membrane component that remains modulated for at least an hour after training. To the best of our knowledge this is the first demonstration of modulation of protein tyrosine phosphorylation in the brain of an animal after a sensory and behavioral experience.

ABSTRACT Protein tyrosine phosphorylation is a major signal transduction pathway involved in cellular metabolism, growth, and differentiation. Recent data indicate that tyrosine phosphorylation also plays a role in neuronal plasticity. We are using conditioned taste aversion, a fast and robust associative learning paradigm subserved among other brain areas by the insular cortex, to investigate molecular correlates of learning and memory in the rat cortex. In conditioned taste aversion, rats learn to associate a novel taste (e.g., saccharin) with delayed poisoning (e.g., by LiCl injection). Here we report that after conditioned taste aversion training, there is a rapid and marked increase in tyrosine phosphorylation of a set of proteins in the insular cortex but not in other brain areas. A major protein so modulated, of 180 kDa, is abundant in a membrane fraction and remains modulated for more than an hour after training. Exposure of the rats to the novel taste alone results in only a small modulation of the aforementioned proteins whereas administration of the malaise-inducing agent per se has no effect. To the best of our knowledge, this is the first demonstration of modulation of protein tyrosine phosphorylation in the brain after a behavioral experience.

Protein tyrosine phosphorylation is a major signal transduction pathway involved in cellular metabolism, growth, and differentiation (1). Recently, it became apparent that tyrosine phosphorylation also plays a key function in neuronal activity and plasticity. Activation of receptor tyrosine kinases by growth and neurotrophic factors is required for development, remodeling, and possibly also maintenance of neural tissue. Prominent among this class of molecules are members of the trk family that respond to a variety of neurotrophic factors including nerve growth factor, brain-derived neurotrophic factor, and neurotrophin 3 (2-4). Protein tyrosine kinases appear also to be involved in mediating the effect of neural cell adhesion molecules on neuronal membrane-associated cytoskeleton (5) and in modulating neurotransmitter receptors (6, 7). In addition, activation of nonreceptor protein tyrosine kinases may play a role in encoding intracellular neuronal responses to specific ligands, probably in the context of cross-talk between signal transduction cascades (8-10). The function of protein tyrosine phosphorylation in longterm structural modification of neurons is especially pertinent to the study of behavioral plasticity, since consolidation of long-term memory is postulated to involve modulation of gene expression, culminating in altered synaptic morphology and physiology (11, 12). Indeed, inhibitors of protein tyrosine kinase were reported to block long-term potentiation in hippocampal CAl region in the guinea pig (9) and knockout of the fyn gene, encoding a nonreceptor tyrosine kinase, was found to impair long-term potentiation, spatial learning, and hippocampal morphology in transgenic mice (13).

MATERIALS AND METHODS Animals. Male Wistar rats ('60 days old, 200-250 g) were used. They were individually caged at 22 ± 2°C in a 12-hr light/12-hr dark cycle.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Abbreviations: aPY, anti-phosphotyrosine antibody; CS, conditioned stimulus; CTA, conditioned taste aversion; US, unconditioned stimulus.

1157

1158

Neurobiology: Rosenblum et al.

Chemicals. Polyclonal aPY (rabbit) was prepared as detailed in Heffetz and Zick (19) or purchased from Zymed. Horseradish peroxidase-linked protein A and the enhanced chemiluminiscence (ECL) kit were from Amersham. All other chemicals were of analytical grade. Behavioral Procedures. All the rats used for behavioral experiments were first trained to get their daily 20-ml water rations for 10 min in their home cage. In the CTA protocol used in our laboratory, the rats receive on the day of conditioning 0.1% saccharin or 0.1 M NaCl as the novel taste and 50 min later are injected with LiCl (i.p. 150 mM, 2% body weight). In a single experiment, CuS04 (i.p. 1.6 mM, 2% body weight) was injected instead of LiCl, to test whether the effects observed were restricted to CTA training with LiCl. Three days after training, the conditioned rats prefer water to saccharin at a ratio of 9:1 in a multiple choice test situation (three pipettes, each with 5 ml of saccharin and three pipettes, each with 5 ml of water; ref. 14), whereas nonconditioned rats or controls injected with saline prefer saccharin to water. In the biochemical experiments described in this work, the rats subjected to CTA training were sacrificed 10 min after completion of training, the rats experiencing the novel taste in the absence of subsequent LiCl injection were sacrificed 60 min after the exposure to the novel taste, and the rats subjected to the LiCl injection in the absence of previous exposure to a novel taste were sacrificed 10 min after the injection. In the experiments intended to test the effect of odor, a gustoolfactometer was used (20), and a saturated odorized air stream carrying 0.5% of isopropyl acetate flushed the rat's nose while it was drinking from a spout. Preparation of Homogenates. Rats were sacrificed by decapitation and the insular cortex and other brain areas, as indicated in Results, were dissected out. The crossing of the rhinal fissure and the medial cerebral artery was used as a reference point, and cortical tissue 1.0 mm rostral, 0.5 mm caudal, and 1.5 mm dorsal to it was excised. This region corresponds to the area containing taste-responsive and visceral and somatosensory neurons (21, 22). The tissue dissected from both hemispheres of each rat was homogenized separately in SDS buffer [2.3% (wt/vol) SDS/5% (vol/vol) 2-mercaptoethanol/10% (vol/vol) glycerol/0.025 M Tris-HCl, pH 6.8]. The homogenate was immediately processed further (see below) or placed at -80°C until used. Subcellular Fractionation. The insular cortex was dissected out and homogenized on ice in 2 mM Na3VO4/50 mM NaF/2 mM EDTA/2 mM EGTA/10 mM Na4P207/1 mM phenylmethylsulfonyl fluoride/50 mM Hepes, pH 7.6. The resulting homogenate was centrifuged for 10 min at 3000 x g and the supernatant thus obtained was further centrifuged at 100,000 x g for 1 hr. The pellet was resuspended to yield the crude membrane fraction. Synaptosomes were prepared from rat forebrain as described by Rao and Steward (23). Western Blot Analysis. Aliquots (equal amounts of protein) of boiled homogenization solution were subjected to PAGE (24) (7.5% polyacrylamide in the presence of SDS and 2-mercaptoethanol) and Western blot analysis (25). After blocking with 1% bovine serum albumin, the blot was reacted either ovemight at 4°C or for 1 hr at room temperature with affinity-purified ciPY as described (19). The presence of aPY was determined with horseradish peroxidase-linked protein A and the ECL kit. Quantification was carried out using a computerized densitometer and image analyzer (Molecular Dynamics). Statistics. A two-way t test was used for comparison of two mean scores and Schaffe's test for multiple comparisons.

RESULTS CTA Training Increases Tyrosine Phosphorylation in the Insular Cortex. A single session of saccharin sampling (8-10 ml), followed 50 min later by LiCl i.p., resulted in a marked

Proc. Natl. Acad Sci. USA 92

(1995)

increase in the amount of tyrosine phosphorylation in insular cortex tissue, as detected by immunoblot analysis with specific aPY antibodies (Fig. 1 A). The major tyrosine-phosphorylated polypeptides altered were of 100 kDa, 115 kDa, and 180 kDa (hereafter termed plOOgc, pll5gc, and pl80gc, respectively). This effect was specific, as several other polypeptides detected on the immunoblot (e.g., of 35 kDa and 40 kDa) were unaffected by the experience (Fig. 1A). Compared to rats given no oral input, mere drinking of water had only a minor effect or no effect at all on tyrosine phosphorylation in the insular cortex. Sampling of the novel taste had a larger effect, but still smaller than that of CTA training (Table 1). Since the most intensive tyrosine phosphorylation signal on the immunoblot was often detected on pl80gc, unless otherwise indicated, the experiments below refer to this molecular species. The Modulation of Protein Tyrosine Phosphorylation Is Not Due to Injection of the US. Since CTA training routinely involves LiCl administration, and the latter may affect metabolism and behavior (26), it is pertinent to enquire whether LiCl alone is capable of inducing the alteration in protein tyrosine phosphorylation in our system. Our data show that LiCl alone does not bring about the enhancing effect observed after CTA training (Fig. 1B and Table 1). Still, the possibility exists that the effects seen after taste-LiCl pairing are due to some unique properties of LiCl. We have therefore replaced LiCl with another malaise-inducing agent, CuS04 (27). CTA obtained with this US was similar to that obtained in the routine CTA training procedure using LiCl (the conditioned rats preferred water to saccharin at a ratio of 9:1 in a multiple choice test). We found that the effect on protein tyrosine phosphorylation of CTA training using CuS04 as the US was similar to that obtained with LiCl as the US (Table 1). CuS04 i.p. injection per se had no effect on protein tyrosine phosphorylation in the insular cortex (data not shown). The Modulation of Protein Tyrosine Phosphorylation Is Observed with Different Novel Tastes but Not with a Novel Odor or Environment. Since it can be reasoned that saccharin, either alone or in combination with the malaise-inducing agent, may have a unique effect on tyrosine phosphorylation in cortex, we have repeated the CTA experiments with another novel taste, 0.1 M NaCl. The sensation of NaCl is transmitted to the cortex by peripheral pathways and central chemotopic maps which are at least partially different from those that transmit the sensation of saccharin (21). We found that irrespective of the novel taste used, CTA increased tyrosine phosphorylation in the insular cortex (Fig. 1B). In contrast to the taste and the CTA training effect, exposure to a distinct odor (isopropyl acetate) or exposure to a new environment by transfer to a new cage did not affect the level of tyrosine phosphorylation in the insular cortex (data not shown). The Effect Is Specific to the Insular Cortex. Protein tyrosine phosphorylation in brain areas other than the insular cortex was not affected by CTA training. Areas tested were the olfactory bulb, the piriform cortex, the occipital cortex, and the frontal cortex (Fig. 1C Inset). The Effect Lasts for More Than an Hour. We then followed the time course of the CTA effect. As can be seen in Fig. 1C, the modulation of tyrosine phosphorylation of pl80gc was maximal in aliquots taken 10 min after training and persisted for more than 1 hr afterward. The Experience-Dependent Modulation of the Tyrosine Phosphorylation of pl80gc Is Observed in a Membrane Fraction. By using subcellular fractionation of insular cortex tissue, we have found that the proteins whose tyrosine phosphorylation is modulated by experience are localized in the particulate fraction; furthermore, the effect of CTA was clearly observed in such an isolated fraction (Fig. 2). Further subcellular fractionation combined with affinity chromatography on either lentil lectin coupled to Sepharose or wheat germ agglutinin coupled to agarose indicated that tyrosine-phosphory-

sp

Neurobiology: Rosenblum et al.

Proc. NatL Acad Sci USA 92 (1995)

1159

Q c

o

z

:.:

0

3:

U,

cc

kDa

:.

.. ;....

- 200

|_: .

. F_ - 97

ms

3Z

c

.0 L03

- 69

ssp

_L _L1:

-46

t .

. ;.....

180

115

100 40 Molecular mass (kDa)

_ .

B 400

:

300

a

.-

200

>

100

2.0

El

*

V

A_

CTA (NaCI) CTA (Sacch) LiCI Water

35

co zaa

co

un

OkDa

1 I 11I IEEEll-200

~~II

0

a.

-97

(L

Time (hr)

FIG. 1. Protein tyrosine phosphorylation in the insular cortex after a behavioral experience. (A) Rats were subjected to the following experiences: CTA, consumption of saccharin only (Sacch), consumption of water only, and no oral input (No Input). The rats were sacrificed and the insular cortex was isolated, homogenized, and subjected to PAGE and immunoblot analysis with aPY antibodies. Here and in B a representative blot is depicted on the right and quantification of the level of phosphotyrosine is on the left. Values in the graph are mean + SEM. *, P < 0.05 for the difference from both no-input and water samples (n = 3). (B) The effect of different tastes as CSs in CTA training (saccharin and NaCl) and of the UC (LiCl) only, on protein tyrosine phosphorylation in the insular cortex. *, P < 0.05 for the difference from LiCl and water (n = 3). (C) The modulation of pl8Ogc as a function of the time after the completion of CTA training, i.e., after the LiCl injection. The point at t = 0 is for groups of rats receiving water at t = 50 min but no LiCl injection afterwards (see the time table of the behavioral maniuplations in Materials and Methods). (Inset) The level of PY in pl8Ogc in various brain regions 10 min after the completion of CTA training. *, P < 0.05 for the difference from 1.0 (n = 3). OB, olfactory bulb; PC, piriform cortex; OL, occipital lobe; FC, frontal cortex; IC, insular cortex.

lated glycoproteins of 100 and 180 kDa are abundant in the postsynaptic density fraction (data not shown). However, we do not yet know whether these proteins are identical to the ones modulated by experience in the homogenate or in the crude membrane fraction. Table 1. Effect of gustatory and visceral experience on protein tyrosine phosphorylation in rat insular cortex Substrate Sacch/Water CTA/Water CTA/LiCl CTA/CuSO4 2.05* 2.54* 2.16* 1.58* pl8Ogc 1.19 2.10* 2.17* 1.80* pll5gc 2.16* 1.21 1.88* 1.70* plOOgc Values are ratios of the magnitude of protein tyrosine phosphorylation in the indicated substrates as a function of gustatory and visceral experience. As detailed in the text, LiCl was routinely used as the malaise-inducing agent in CTA training; CuS04 was employed only to demonstrate that the effect of the association of the malaiseinducing agent and taste on protein tyrosine phosphorylation is not restricted to LiCl. Sacch, saccharin. *, Significantly different from 1.0 (P < 0.05).

DISCUSSION The combination of CS (taste) and US (injection of a malaiseinducing agent) in a CTA paradigm enhances net tyrosine phosphorylation in several proteins in the rat insular cortex. This effect is specific for that cortex and lasts for at least an hour. The CS alone has a weaker though still significant effect, whereas the US alone has no effect. The effect of the US may involve convergence of signal transduction cascade(s) activated by the sensation of malaise on the signal transduction cascade(s) modulated by the taste input, thus resulting in augmentation of the cellular effect of taste alone in the insular cortex to a level that markedly modulates protein tyrosine phosphorylation. In this context it is pertinent to note that LiCl and CuSO4 may act in vivo via different anatomical pathways (27), and yet the effect on protein tyrosine phosphorylation in cortex is apparently similar. The functional role of such US-mediated amplification of the effect of CS in cortex in the realization of the internal representation of taste-malaise association is not yet clear. The insular cortex is clearly

1160

Proc. NatL Acad ScL USA 92

Neurobiology: Rosenblum et al.

CTA

cn)D Un C:

cu

(1995)

Water

Xa)

* . *

co

CLI

kDa

, ^-20~00

:-Q E._

-0

97

CL -

69

FIG. 2. Effect of CTA training on the tyrosine phosphorylation of pl8Ogc in a partially isolated membrane fraction. (Left) The level of PY in pl8Ogc as determined by immunoblots of membranes prepared from the insular cortex of rats subjected to CIA training compared to that of controls. (Right) A representative immunoblot. *, P < 0.05 for the difference between CIA and water (n = 3).

required for normal learning of the novel taste (14, 28), but association of the taste with malaise can take place in the absence of normal cortical activity (28). The methodology used in this paper is based on excision and homogenization of cortical areas, followed by detection of the level of protein tyrosine phosphorylation in the homogenate by immunoblot analysis with aPY antibodies. This methodology does not permit us to determine whether the effect of protein tyrosine phosphorylation occurs only in the taste area of the insular cortex. Taste neurons in the insular cortex are located in close proximity to somatosensory neurons and are difficult to separate from the latter even in electrophysiological analysis (21, 22). However, since a decisive difference between the control and experimental groups in our study is the nature of the taste input, it is tempting to assume that the changes indeed take place in the taste and not somatosensory area. Since the enhancement detected in the homogenate is rather marked in a heterogeneous cortical area, it should be expected to be much stronger in the taste area, or in subpopulation(s) of cells in it. Further delineation of the neuronal tissue modulated by the taste and CTA input must await fine immunohistochemical analysis to be performed with specific anti-pl8Ogc, anti-pllSgc, and anti-plOOgc antibodies when they become available. Immunohistochemistry performed with the polyclonal aPY antibodies detect specific labeling in neural cells in the cortex (KR., unpublished data). The set of tyrosine-phosphorylated proteins detected by us may represent a specific interrelated cascade of molecular events triggered by the sensory input; alternatively, the modulation of protein tyrosine phosphorylation might result from a general activation of tyrosine kinases and phosphatases due to accelerated blood supply and enhanced metabolic activity in the activated insular cortex. However, since only part of the proteins recognized by the aPY antibodies are modulated by CTA training, an effect due to global metabolic enhancement of a cortical area is unlikely. We therefore suggest that the transient modulation of protein tyrosine phosphorylation specifically subserves mechanisms of neuronal plasticity, as suggested in long-term potentiation (9). What might be the identity and function of the tyrosinephosphorylated proteins modulated in the insular cortex? Comparison to available data on protein tyrosine phosphorylation in the brain might be of value here. The activity of protein tyrosine kinase in the adult nervous system is high compared to that in other tissues, and tyrosine-phosphorylated proteins are mainly localized in neurons and synapses (10, 29-33). The tyrosine phosphorylation of some of these proteins has been reported to be controlled by growth factors, depolarization, dopaminergic innervation, glutamatergic agonists, Ca2+ ionophores, and phorbol esters (34-37). Of special interest in the present context is a protein of 180 kDa reported

by several groups to be a major phosphoprotein of neuronal tissue in vivo (32, 38) and in vitro (34). A protein termed ppl80 is present in neurons but not glia and is abundant in the adult cortex, hippocampus, and striatum but not in the adult cerebellum, spinal cord, pituitary gland, and other tissues (34). A glycoprotein termed PSD-GP180 is a constituent of the postsynaptic density and is also a substrate for Ca2+/ calmodulin protein kinase (38). Further characterization will determine whether ppl80, PSD-GP180, and pl80gc are the same molecular species. It should be noted that the cellular localization of the aforementioned proteins, and specifically the subcellular localization of PSD-GP180 and pl80gc, is in consonance with a potential role in neuronal plasticity. The skillful technical assistance of Shoshi Hazvi is gratefully acknowledged. This research was supported by a grant from the Whitehall Foundation. 1. Schlessinger, J. & Ullrich, A. (1992) Neuron 9, 383-391. 2. Kaplan, D. R., Hempstead, B. L., Martin-Zanca, D., Chao, M. V. & Parada, L. F. (1991) Science 252, 554-558. 3. Lamballe, F., Klein, R. & Barbacid, M. (1991) Cell 66, 967-979. 4. Lo, D. C. (1992) Curr. Opin. Biol. 2, 336-340. 5. Atashi, J. R., Klinz, S. G., Ingraham, C. A., Matten, W. T., Schachner, M. & Maness, P. F. (1992) Neuron 8, 831-842. 6. Hopfield, J. F., Tank, D. W., Greengard, P. & Huganir, R. L. (1988) Nature (London) 336, 677-680. 7. Raymond, L.A., Blackstone, C. D. & Huganir, R. L. (1993) Trends Neurosci. 16, 147-153. 8. Houslay, M. D. (1991) Eur. J. Biochem. 195, 9-27. 9. O'Dell, T. J., Kandel, E. R. & Grant, S. G. N. (1991) Nature (London) 353, 558-560. 10. Maness, P. F. (1992) Dev. Neurosci. 14, 257-270. 11. Rose, S. P. R. (1991) Trends Neurosci. 14, 390-397. 12. Bailey, C. H. & Kandel, E. R. (1993) Annu. Rev. PhysioL 55, 397-426. 13. Grant, S. G. N., O'Dell, T. J., Karl, K. A., Stein, P. L., Soriano, P. & Kandel, E. R. (1992) Science 258, 1903-1910. 14. Rosenblum, K., Meiri, N. & Dudai, Y. (1993) Behav. Neural Biol. 59, 49-56. 15. Revusky, S. A. & Garcia, J. (1970) Psychol. Learn. Motiv. 4,1-84. 16. Bures, J., Buresova, 0. & Krivanek, J. (1988) Brain and Behavior: ParadigmsforResearch in Neural Mechanisms (Wiley, New York). 17. Kiefer, S. W., Leach, L. R. & Braun, J. J. (1984) Behav. Neurosci. 98, 590-608. 18. Fernandez-Ruiz, J., Escober, M. L., Pina, A. L., Diaz-Cintra, S., Cintra-McGlone, F. L. & Bermudez-Rattoni, F. (1991) Behav. Neural Biol. 55, 179-193. 19. Heffetz, D. & Zick, Y. (1989) J. Biol. Chem. 264, 10126-10132. 20. Slotnick, B. M. (1990) in Comparative Perception, eds. Stebbins, W. & Berkley, M. (Wiley, New York), Vol. 1. 21. Cagan, R. H. (1989) Neural Mechanisms in Taste (CRC, Boca Raton, FL). 22. Kosar, E., Grill, H. J. & Norgren, R. (1986) Brain Res. 379, 329-341.

Neurobiology: Rosenblum et al. 23. Rao, A. & Steward, 0. (1991) J. Neurosci. 11, 2881-2895. 24. Laemmli, U. K. (1970) Nature (London) 227, 680-685. 25. Burnette, W. W. (1981) Anal. Biochem. 112, 195-203. 26. Li, P. P., Sibony, D., Green, M. A. & Warsh, J. J. (1993) J. Neurochem. 61, 1722-1730. 27. Gamzu, E., Vincent, G. & Boff, E. (1985) Ann. N.Y Acad. Sci. 443, 231-249. 28. Gallo, M., Roldan, G. & Bures, J. (1992) Behav. Brain Res. 52, 91-97. 29. Ellis, P. D., Bissoon, N. & Gurd, J. W. (1988) J. Neurochem. 51, 611-620. 30. Hirano, A. A., Greengard, P. & Huganir, R. L. (1988) J. Neurochem. 50, 1447-1455. 31. Pang, D. T., Wang, J. K. T., Valtora, F., Benfenati, F. & Greengard, P. (1988) Proc. Natl. Acad. Sci. USA 85, 762-766.

Proc. Natl. Acad Sci. USA 92 (1995)

1161

32. Cudmore, S. B. & Gurd, J. W. (1991) J. Neurochem. 57, 12401248. 33. Wagner, K. R., Mei, L. & Huganir, R. L. (1991) Curr. Opin. Neurobiol. 1, 65-73. 34. Girault, J.-A., Chamak, B., Bertuzzi, G., Tixier, H., Wang, J. K. T., Pang, D. T. & Greengard, P. (1992) J. Neurochem. 58, 518-528. 35. Girault, J. A., Siciliano, J. C., Robel, L. & Herve, D. (1992) Proc. Natl. Acad. Sci. USA 89, 2769-2773. 36. Woodrow, S., Bissoon, N. & Gurd, J. W. (1992)J. Neurochem. 59, 857-862. 37. Siciliano, J. C., Gelman, M. & Girault, J.-A. (1994) J. Neurochem. 62, 950-959. 38. Gurd, J. W. & Bissoon, N. (1990) J. Neurosci. Res. 25, 336-344.