Immunohistochemical Localization ofthe InsP4 Receptor GTPase-ActivatingProtein GAP1IP4BP in the Rat BrainArmando P. Signore,1 Flavia O’Rourke,2 Xinghua Lu,2 Maurice B. Feinstein,2and Hermes H. Yeh1,2,3*1Program in Neuroscience, University of Connecticut Health Center, Farmington2Department of Pharmacology, University of Connecticut Health Center, Farmington3Department of Neurology, University of Connecticut Health Center, Farmington

The distribution of GAP1IP4BP, a GTPase-activatingprotein showing high affinity and stereospecificity forinositol 1,3,4,5-tetrakisphosphate (InsP4), was investi-gated by Western blot and immunohistochemistry ofrodent brain with polyclonal antibodies generatedagainst the carboxy-terminus of the cloned protein.GAP1IP4BP-like immunoreactivity was found through-out the brain, most notably in the pyriform cortex,neocortex, hippocampus, striatum, and cerebellar cor-tex. However, the most striking immunolabeling wasconsistently localized to area CA1 of the hippocampusand the central, medial, and intercalated nuclei of theamygdala. Western blot analysis of the correspondingbrain regions corroborated these immunohistochemi-cal observations. The regionally specific expression ofGAP1IP4BP provides the prerequisite neuroanatomicalsubstrate toward elucidating the functional role ofInsP4 and GAP1IP4BP in the central nervous system. J.Neurosci. Res. 55:321–328, 1999.r 1999 Wiley-Liss, Inc.

Key words: hippocampus; inositol-phosphate; metabo-tropic receptor

INTRODUCTIONThe activation of receptors in neural tissues that are

linked to phospholipase C increases intracellular levels ofinositol 1,4,5-trisphosphate (InsP3), resulting in release ofintracellular stores of Ca21 (Fisher and Agranoff, 1987).Inositol 1,3,4,5-tetrakisphosphate (InsP4), formed by theATP-dependent phosphorylation of InsP3 catalyzed byInsP3 3-kinase, has also been implicated in cellularcalcium signaling (Irvine and Moor, 1986; Berridge andIrvine, 1989; Cullen et al., 1990; Loomis-Husselbee et al.,1996; O’Rourke et al., 1996; Tsubokawa et al., 1996).InsP4 may have a synergistic effect with InsP3 (Cullen etal., 1990) or may exert a direct effect via its own receptor(Morris et al., 1987; Bird et al., 1991; Gawler et al., 1991;Kachintorn et al., 1993; Wilcox et al., 1993; Bird andPutney, 1996; Loomis-Husselbee et al., 1996; Tsubokawa

et al., 1996). Evidence for a direct effect of InsP4 includesthe demonstration that InsP4 opens calcium channels inneurons (De Waard et al., 1992; Tsubokawa et al., 1994;Tsubokawa et al., 1996), enhances calcium influx inendothelial cells (Lu¨ckhoff and Clapham, 1992), andinduces calcium efflux from inside-out platelet plasmamembrane vesicles (O’Rourke et al., 1996). InsP4 alsoappears to regulate influx of calcium subsequent tounloading of intracellular stores by InsP3 (Putney, 1986;Cullen et al., 1990; Gawler et al., 1991; Bird and Putney,1996).

High-affinity binding sites for radiolabeled InsP4

have been reported in whole brain (Theibert et al., 1991,1992; Challiss et al., 1991; Cullen and Irvine, 1992;Parent et al., 1994; Stricker et al., 1995; Smith andNahorski, 1996), specifically in the hippocampus (Parentet al., 1994) and cerebellum (Cullen and Irvine, 1992;Theibert et al., 1992). Several putative InsP4 bindingproteins of different properties have been isolated (Donie´and Reiser, 1991; Cullen et al., 1995a; Theibert et al.,1991, 1992), one of which is an Mr 5 104 kD protein thatwas purified from pig (Cullen et al., 1995a) and human(O’Rourke et al., 1996) platelet membranes. The humancDNA was cloned from a cDNA library from bonemarrow, sequenced, and identified as a member of theGAP1 family of GTPase-activating proteins (GAPs) anddesignated as GAP1IP4BP(Cullen et al., 1995b; Lockyer etal., 1997). The GAP1IP4BPwas shown to activate Ras andRap GTPase and to bind InsP4 with high affinity and

Contract grant sponsor: Public Health Services; Contract grant num-bers: NS24830, AA03510; Contract grant sponsor: Heart and LungSociety; Contract grant number: 18937; Contract grant sponsor:University of Connecticut Health Center Research Committee.

A.P. Signore and F. O’Rourke contributed equally to this article.

*Correspondence to: Hermes H. Yeh, Ph.D., Department of Pharmacol-ogy, MC 6125, University of Connecticut Health Center, 263 Farming-ton Avenue, Farmington, CT 06030. E-mail: hyeh@neuron.uchc.edu

Received 19 August 1998; Revised 28 October 1998; Accepted 30October 1998

Journal of Neuroscience Research 55:321–328 (1999)

r 1999 Wiley-Liss, Inc.

exquisite sensitivity. It contains a highly conservedGTPase-activating domain, domains for Ca21-dependentand Ca21-independent binding to phospholipids, and asingle pleckstrin homology (PH) domain with a Brutontyrosine kinase motif that binds both InsP4 and phospho-lipids (Fukuda et al., 1996; Lockyer et al., 1997).GAP1IP4BP enhances the GTPase activity of Rap and Rasin vitro, thereby terminating their activity (Cullen et al.,1995b; Yamamoto et al., 1995). The Ras GAP (GTPaseactivating protein) activity is inhibited by phospholipidsand stimulated by InsP4 (Cullen et al., 1995b), suggestingthat competition between InsP4 and phospholipid mayaffect membrane binding of this protein (Cullen et al.,1995b; Lockyer et al., 1977).

To understand the functions of GAP1IP4BP further,we transfected human erythroleukemia (HEL) cells witha vector to express antisense cDNA and obtained clonedcells with greatly diminished GAP1IP4BP protein (Lu etal., 1999). The GAP1IP4BP-deficient cells had substan-tially augmented Ca21 influx when stimulated by throm-bin, a Ca21 mobilizing mitogen, or when the intracellularpool of Ca21 was released by thapsigargin. The effect ofantisense knockout of GAP1IP4BPwas shown to be due tothe enhanced activity or expression of a hyperpolarizingCa21-activated K1 channel that increases Ca21 influx. Apossible link to Ras is suggested because aRas/Rafkinase-mediated pathway has been shown to increaseCa21-activated potassium current in fibroblasts (Huangand Rane, 1994). Thus, apart from any effects that may bemediated by InsP4, GAP1IP4BP itself appears to control anintracellular signal transduction pathway that can regulatethe synthesis and/or activity of a potassium channel thatmodulates Ca21 influx.

As a prerequisite for understanding the functions ofGAP1IP4BPin the nervous system, we describe the region-ally selective distribution of GAP1IP4BP-like immunoreac-tivity (GAP1IP4BP-LI) in the rat brain by using polyclonalantibodies generated against the carboxy-terminal se-quence of human GAP1IP4BPprotein (Cullen et al., 1995b;Loomis-Husselbee et al., 1996). Intense immunolabelingwas found in the limbic regions, in particular the CA1subfield of the hippocampus, amygdala, and olfactory-associated cortical structures. Prominent GAP1IP4BP-LIwas also detected in neocortex, striatum, and cerebellarcortex.

MATERIALS AND METHODSWestern Blot Analysis

Adult female Long-Evans rats were asphyxiatedwith carbon dioxide, and their brains were removed. Thecerebellum was detached, and the hippocampal CA1 andCA3 subfields and neocortex were microdissected frombrain slices. The dissected brain tissue was individually

homogenized in Dounce homogenizers on ice in asolution containing a cocktail of protease inhibitors anddetergent to solubilize intact protein: 20 mM sodiumbicarbonate, 1 mM EDTA, 10% CHAPS, 20 µM leupep-tin, 20 µM pepstatin, and 0.5 mM phenyl methyl sulfonylfluoride, pH 7.2. These steps must be carried out rapidlyto avoid generation of proteolytic fragments that reactwith the antisera. The detergent-solubilized samples wereassayed by using Coomassie blue reagent and wereadjusted to a protein concentration of 5.0 mg/ml withadditional buffer plus SDS at a final concentration of 2%,followed by immediate boiling for 10 min. Aliquots of 50µg total protein were loaded in individual wells forelectrophoresis on 7.5% SDS-PAGE gels followed byblotting onto an Immobilon membrane with a semi-dryblotting apparatus (Pharmacia, Piscataway, NJ), as previ-ously described (O’Rourke et al., 1996). All furtherincubations and washes were performed at room tempera-ture. Blots were incubated for 1 hr with blocking solutioncontaining 5% powdered non-fat milk in phosphatebuffered saline (PBS) with 0.1% Tween 20 (PBS-TWEEN). After the blocking step the blots were probedwith three anti-GAP1IP4BP antisera, one provided by Dr.P.J. Cullen (used at a dilution of 1:1,000) and twocommercially raised anti-GAP1IP4BP antisera (designated51–3 and 52–3) or preimmune sera (diluted to 1:500) inblocking buffer for an additional 2 hr. All antisera wereraised against the last 20 amino acid residues of thecarboxy-terminus sequence (GDKSFQSYIRQQSET-STHSI) of the human GAP1IP4BP. The blots were thenwashed twice with PBS-TWEEN for 15 min and threetimes for 5 min each and then incubated for an additional45 min with goat anti-rabbit peroxidase conjugatedsecondary antibody (Calbiochem, San Diego, CA) di-luted 1:5,000 in PBS-TWEEN, followed by washing asdescribed for the primary antibody. The washed immuno-blot was incubated with chemiluminescence reagents asdescribed in the kit (Kirkegard and Perry, Gaithersburg,MD) and exposed to X-ray film for 1 min. Bands on thefilms were quantified using a Molecular Dynamics(Sunnyvale, CA) laser densitometer and ImageQuantsoftware.

Cultures enriched in astrocytes were prepared ac-cording to modification of previously described methods(McCarthy and DeVellis, 1980). Cells derived fromneonatal rat cortical tissue were plated onto plasticculture plates and maintained in modified Eagle’s me-dium supplemented with 5% heat-inactivated fetal calfserum. After 5–7 days in culture, the plates were shakenat 200 rpm overnight to remove oligodendrocytes andmicroglia. The remaining cells, enriched in astrocytes,were suspended, plated onto new culture dishes, andallowed to grow to confluency. Cells in culture plates

322 Signore et al.

were solubilized and analyzed by Western blotting, asdescribed above.

ImmunohistochemistryAdult rats received an intraperitoneal injection of a

lethal dose of sodium pentobarbital prior to intra-aorticperfusion by using PBS with 1.0 U/ml heparin (pH 7.4)and then PBS containing 4% paraformaldehyde (FisherScientific, Springfield, NJ) to fix the tissue. The brainswere removed, immersed in PBS containing 30% (w/v)sucrose, and 25-µm saggital and coronal sections werecollected by using a cryostat (Model HM 505; Microm,Walldorf, Germany). The sections were stored in acryoprotective solution containing 30% (w/v) sucrose,30% (v/v) ethylene glycol, and 1% (w/v) polyvinylpyroli-dine in PBS. Sections were first incubated for 1 hr in ablocking solution consisting of 10% normal goat serumand 0.5% Triton X-100 in PBS. The GAP1IP4BPantiserumfrom Dr. P.J. Cullen was used at a dilution of 1:500, andpolyclonal antiserum lots 51–3 and 52–3 (AdvancedChemtech, Louisville, KY) were used at dilutions of1:100 to 1:500. Sections were incubated in the primaryantibodies at 4°C for 18–24 hr and then with biotinylatedgoat anti-rabbit secondary antibody (1:200 in PBS and1% goat serum) for 4–12 hr. An avidin-biotin–conjugatedhorseradish peroxidase kit (Vector, Burlingame, CA) wasused for detection, and after brief incubation in ammonia-PBS (PBS, pH 7.4, with ammonium hydroxide) thereaction product was visualized by using 0.14 mM3,38-diaminobenzidine (DAB), 0.003% H2O2, and 0.03%nickel ammonium sulfate in ammonia-PBS. Controlsections were processed in parallel but without theprimary antisera. Monoclonal anti-GFAP was used at adilution of 1:1,000 to identify astrocytes in brain sections.Bound anti-GFAP was detected with goat anti-mousesecondary antibody at 1:200 and visualized after peroxi-dase–anti-peroxidase and DAB reactions. A rodent atlas(Paxinos and Watson, 1997) was used to confirm anatomi-cal locations. Images were digitized in Photoshop 4.0(Adobe Systems, Mountain View, CA). Unless otherwisestated, all chemicals were obtained from Sigma (St.Louis, MO).

RESULTSWe first established the specificity and potency of

the GAP1IP4BPantiserum by Western blotting of detergent-solubilized brain tissue and cellular homogenates (Fig. 1).Human blood platelets, which contain high amounts ofGAP1IP4BP (O’Rourke et al., 1996), served as positivecontrol. We used a rabbit GAP1IP4BP antiserum obtainedfrom Dr. P.J. Cullen and two additional antisera (anti-GAP1IP4BP lots 51–3 and 52–3) that were commercially

generated against the identical 20–amino acid carboxy-terminal region of the GAP1IP4BP protein. The specificityof the first GAP1IP4BP antibody was demonstrated byCullen et al. (1995b) and Lockyer et al. (1997). OnWestern blots, each of the antisera detected a single bandat,Mr 5 100 kDa protein band in human blood platelets(Fig. 1, lane f), and in the striatum, hippocampus, basalforebrain, and cerebellum that were microdissected fromthe adult rat brain (Fig. 1, lanes a–c, e).

All three antisera were employed in parallel immu-nohistochemical experiments. Of the three antisera, theone obtained from Dr. P.J. Cullen consistently producedthe most intense immunostaining. Although lots 51–3 and52–3 of the commercially prepared GAP1IP4BP antiserawere suitable for Western blot analysis, they were not aseffective for immunohistochemical demonstration ofGAP1IP4BP-LI in brain sections. Even though all antiserawere raised against the identical carboxy-terminal portionof the GAP1IP4BP peptide, they may have preferentiallyrecognized different epitopes with different accessibilityto the protein in fixed tissue compared with the SDS/heat-denatured proteins on Western blots.

Figure 2 illustrates the localization of GAP1IP4BP-LIin cryosections from the rat brain by using the GAP1IP4BP

antiserum obtained from Dr. P.J. Cullen. A saggitalsection (Fig. 2A) and a series of coronal sections areshown arranged in a rostral-to-caudal sequence (Fig.2B–H). GAP1IP4BP-LI is present in gray matter through-out the brain, especially in the hippocampal CA1 subfield(Figs. 2A,E–G, 3A) and cerebellum (Figs. 2A, 5B). In

Fig. 1. Western blot of GAP1IP4BP protein in areas of brain,astrocytes, and platelets. Immunoblot using the GAP1IP4BP

antiserum visualizes a single band of protein at approximatelyMr 100 kDa. Lanea: Striatum. Laneb: Hippocampus. Lanec:Basal forebrain. Laned: Astrocytes. Lanee: Cerebellum. Lanef: Platelets. Positions of the molecular weight (kDa) standardsare shown on the left.

GAP1IP4BP in the Rat Brain 323

addition, distinct cellular immunostaining was found inthe amygdala (Figs. 2E, 3C), olfactory bulb (Fig. 2B),olfactory (Fig. 2C–G), and neocortices (Fig. 2, all sec-tions, and 5A). Control cryosections processed withoutthe primary antibody (Fig. 2I) or with preimmune serasubstituting for primary antibody displayed no immunore-activity (data not shown).

Pyramidal cells in the CA1 subfield of the hippocam-pus displayed the most intense GAP1IP4BP-LI (Fig. 3A),showing the soma as well as the apical dendrites extend-ing into the stratum radiatum (Fig. 3B, arrowhead).Immunostaining of pyramidal cells in the CA2–4 sub-fields and granule cells in the dentate gyrus was consider-ably weaker (Fig. 3A). By using the same antisera, thepresence of GAP1IP4BP-LI was further verified by Westernblotting of tissue obtained by microdissecting CA1 andCA3 hippocampal subfields, cerebellum, and cerebral

cortex (data not shown). Within the amygdaloid nuclearcomplex, the central, medial, and intercalated nucleishowed intense GAP1IP4BP-LI (Fig. 3C,D). At the cellularlevel, GAP1IP4BP-LI appeared to be concentrated alongthe cell membrane, outlining the somata of individualneurons in CA1 (Fig. 3B) and amygdala (Fig. 4).

Fiber tracts displayed only background levels ofimmunostaining (Figs. 2, 3A) and, at higher magnifica-tion, profiles displaying GAP1IP4BP-LI were sparse andlimited to small, punctate oligodendrocyte-like cell bod-ies. In sections processed immunohistochemically forglial fibrillary acidic protein, an astrocyte-specific marker,immunopositive cells were distinct from cells displayingGAP1IP4BP-LI, indicating that the cells containingGAP1IP4BP-LI were not astrocytes. Furthermore,GAP1IP4BP-LI was undetectable in Western blots ofcultured astrocytes (Fig. 1d).

Fig. 2. Distribution of GAP1IP4BP-like immunoreactivity (-LI)in rat brain.A: Sagittal section showing immunostaining ingray matter throughout the brain, particularly in the CA1subfield of the hippocampus (arrowhead) and the cerebellum. Inthe brainstem, the trigeminal nucleus is also heavily labeled.B–H: Coronal sections showing GAP1IP4BP-LI at differentlevels of the brain. In addition to the hippocampal CA1 subfield

(arrowheads in E–G), the amygdala (E, open arrow), habenula(E, filled arrow), and the pyriform cortex (C–G, open arrow-heads) are also heavily labeled. White tracts display only thelevel of background staining seen in control sections processedwithout the primary antibody.I: One such control section thathas been lightly counterstained with cresyl violet is illustrated.

324 Signore et al.

GAP1IP4BP-LI was also found in other brain regions,notably the neocortex and cerebellum. In the neocortex(Fig. 5A) comparable immunostaining was found amongall the layers, except layer I, which consistently displayedthe weakest immunoreactivity. In the cerebellar cortex,GAP1IP4BP-LI was evenly distributed throughout thePurkinje and granule cell layers (Fig. 5B). The stainingwas diffuse, but the contours of individual Purkinje cells,along with their proximal dendritic arbor, were discern-ible. Cell bodies situated along the entire thickness of themolecular layer, presumably those of basket/stellate cells,were also evident.

DISCUSSIONThis is the first immunohistochemical demonstra-

tion of the specific localization of the InsP4 receptor

protein GAP1IP4BP in the brain. The present study in-cludes a display of not only its occurrence but, moreimportantly, the regional selectivity of the pattern ofdistribution of GAP1IP4BP-LI. Most impressive is thedifferential expression in CA1 vis-a`-vis other subfields ofthe hippocampus. Staining was also strong in severalother nuclei, in particular the amygdala, pyriform cortex,and cerebellum. In all areas GAP1IP4BP-LI was concen-trated along the membrane around the soma as well asproximal dendrites, but not in the nucleus or axon ofindividual neurons. This membrane location is consistentwith the characterization of radiolabeled InsP4 bindingsites in isolated plasma membrane fractions of brain,platelets, and HEL cells (Theibert et al., 1991; Cullen andIrvine, 1992; Cullen et al., 1994; O’Rourke et al., 1996;Tsubokawa et al., 1996). It is also consistent with findingsthat InsP4 affects calcium flux across the plasma mem-

Fig. 3. Regionally specific expression of GAP1IP4BP-like immu-noreactivity (-LI) in the hippocampus and amygdaloid nuclei.A: In the hippocampus, GAP1IP4BP-LI is strong in the CA1subfield and weaker in CA2, CA3, and dentate gyrus (DG).White tracts (corpus callosum, CC) show virtually no staining.B: CA1 cells in the stratum pyramidale (SP) show staining of

somata, with dendrites extending into the stratum radiatum(SR) but not into the stratum oriens (SO).C: Cells in theamygdala (arrow; central, medial, and intercalated nuclei) arealso immunoreactive.D: GAP1IP4BP-LI is also primarily insomata and proximal processes of amygdaloid cells. Bars:(A,C) 1.0 mm; (B,D) 25 µm.

GAP1IP4BP in the Rat Brain 325

branes of CA1 hippocampal pyramidal neurons (Tsub-okawa et al., 1996), endothelial cells (Lu¨ckhoff andClapham, 1992), and platelets (O’Rourke et al., 1996).Furthermore, it is in agreement with the membrane-delimited localization of expressed GAP1IP4BP protein inCOS cells and HeLa cells (Lockyer et al., 1997) and HELcells (O’Rourke and Feinstein, unpublished). We alsoobserved some weaker GAP1IP4BP-LI within the cells,probably representing protein in the cytosol and/or ininternal membranes. In platelets, GAP1IP4BPwas found ininternal membrane fractions, and about 10% of the totalwas in the soluble fraction (O’Rourke et al., 1996).

The location of GAP1IP4BP-LI in the plasma mem-brane, which is probably attributable to its phospholipidand InsP4 binding PH domain (Lockyer et al., 1997), is insharp contrast to that of other brain-enriched GAPs, suchas GAP1m (Lockyer et al., 1997), neurofibromin (Gregoryet al., 1993), and p120GAP (Gibbs et al., 1988). Thissuggests unique regulatory roles for the different GAPsand a specific role for GAP1IP4BPrelated to some functionof the plasma membrane. Our review of the GenBank

Fig. 5. GAP1IP4BP-like immunoreactivity in the neocortex andcerebellum.A: Immunoreactive cell bodies can be seen through-out the neocortical layers.B: In the cerebellum, diffuse stainingis seen in the granule cell layer (GC), and somatic profilesof individual Purkinje cells are clearly evident (arrow). Bars,200 µm.

Fig. 4. GAP1IP4BP-like immunoreactivity (-LI) is localized, butnot limited to the somatic and proximal dendritic membrane.Photomicrograph of cells from the amygdala show GAP1IP4BP-LIoutlining pyramidal cells (arrows), suggesting that immunoreac-tivity is present in the plasma membrane. The cytoplasmicregion extending into the proximal portions of dendrites(arrowheads) is also immunoreactive. Bar, 25 µm.

326 Signore et al.

databases showed that GAP1IP4BP is the human counter-part of oligodendrocyte GAPIII (Baba et al., 1995) andbovine brain R-Ras GAP (Yamamoto et al., 1995). Eachof these GAP cDNAs codes for a very highly conserved834–amino acid protein with predicted molecular weightsof 95–96 kDa. The human, rat/mouse, and bovine formsare 95% identical and 97–98% conserved. Furthermore,the 20–amino acid carboxy-terminal peptide of humanGAP1IP4BP used to raise the antisera differs from rodentGAPIII by only one amino acid (T829/I), making it highlylikely that the latter is the GAP1IP4BP-LI observed in therat brain sections. Messenger RNA for GAPIII proteinwas most highly expressed in rat brain, including cortex,hippocampus, cerebellum, and oligodendrocytes, but notastrocytes (Baba et al., 1995), in agreement with ourmeasurements of expressed protein.

Receptors linked to phosphoinositide metabolism,the generation of InsP3 and InsP4, and the mobilization ofCa21 have been implicated in regulating neuronal plastic-ity (Henzi and MacDermott, 1992; Berridge, 1993;Putney and Bird, 1993; Tsubokawa et al., 1996). Calciumis involved in learning and memory (Molinari et al.,1996), and altered calcium homeostasis has been impli-cated in aging (Disterhoft et al., 1996; Hartmann et al.,1996), ischemia-induced cell death, a model of epilepto-genesis in the rat hippocampus (Faas et al., 1996), andneurodegeneration. In addition, proteins of the Ras super-family are involved in gene expression, cell proliferation,and apoptosis. Therefore, it is noteworthy that the pres-ence of GAP1IP4BP-LI is most evident in brain regions thatharbor a high degree of plasticity, i.e., structures of thelimbic system, neocortex, striatum, and cerebellum. Be-cause the neuroanatomical localization of a protein is akey step in ultimately establishing its functions in thebrain, the expression and pattern of distribution ofGAP1IP4BP-LI demonstrated in the present study providethe anatomical basis for guiding further functional stud-ies.

ACKNOWLEDGMENTS

We thank Drs. Douglas W. Sapp and Shao-Ming Lufor critical reading of the manuscript and for providinghelpful comments. We also thank Ms. Eileen Matthewsfor her help in conducting the Western blots and Dr. P.J.Cullen for his gift of the GAP1IP4BP antibody. This workwas supported by grants from Public Health Services(NS24830 and AA03510 to H.H.Y.), Heart and LungSociety (18937 to M.B.F.), and University of ConnecticutHealth Center Research Committee (to F.O.).

REFERENCES

Baba H, Fuss B, Urano J, Poullet P, Watson JB, Tamanoi F, MacklinWB. 1995. GapIII, a new brain-enriched member of theGTPase-activating protein family. J Neurosci Res 41:846–858.

Berridge MJ. 1993. Inositol trisphosphate signaling and calciumsignaling. Nature 361:315–325.

Berridge MJ, Irvine RF. 1989. Inositol trisphosphates and cell signal-ing. Nature 341:197–204.

Bird G St J, Putney JW. 1996. Effect of inositol 1,3,4,5-tetrakisphos-phate on inositol trisphosphate-activated Ca21 signaling inmouse lacrimal acinar cells. J Biol Chem 271:6766–6770.

Bird G St J, Rossier MF, Hughes SB, Shears DL, Putney JW. 1991.Activation of Ca21 entry into acinar cells by a non-phosphorylat-able inositol trisphosphate. Nature 352:162–165.

Challiss RAJ, Willcocks AL, Mulloy B, Potter BVL, Nahorski SR.1991. Characterization of inositol 1,4,5-trisphosphate and inosi-tol 1,3,4,5-tetrakisphosphate-binding sites in rat cerebellum.Biochem J 274:861–867.

Cullen PJ, Irvine RF. 1992. Inositol 1,3,4,5-tetrakisphosphate bindingsites in neuronal and non-neuronal tissues. Biochem J 288:149–154.

Cullen PJ, Dawson AP, Irvine RF. 1990. Synergistic control of Ca21

mobilization in permeabilized mouse L1210 lymphoma cells byinositol 2,4,5-trisphosphate and inositol 1,3,4,5-tetrakisphos-phate. Biochem J 271:549–553.

Cullen PJ, Patel Y, Kakkar VV, Irvine RF, Authi KS. 1994. Specificbinding sites for inositol 1,3,4,5-tetrakisphosphate are locatedpredominantly in the plasma membrane of human platelets.Biochem J 298:739–742.

Cullen PJ, Dawson AP, Irvine RF. 1995a. Purification and characteriza-tion of an Ins(1,3,4,5)P4 binding protein from pig platelets.Possible identification of a novel non-neuronal Ins(1,3,4,5)P4

receptor. Biochem J 305:139–143.Cullen PJ, Hsuan JJ, Truong O, Letcher AJ, Jackson TR, Dawson AP,

Irvine RF. 1995b. Identification of a specific Ins(1,3,4,5)P4-binding protein as a member of the GAP1 family. Nature376:527–530.

De Waard M, Seagar M, Feltz A, Couraud F. 1992. Inositol phosphateregulation of voltage-dependent calcium channels in cerebellargranule neurons. Neuron 9:497–503.

Disterhoft JF, Moyer JR, Thompson LT, Mogul DJ. 1996. Calcium-dependent afterhyperpolarization and learning in young andaging hippocampus. Life Sci 59:413–420.

Donie F, Reiser G. 1991. Purification of a high-affinity inositol1,3,4,5-tetrakisphosphate receptor from brain. Biochem J 275:453–457.

Faas GC, Wadman WJ, Vreugdenhil M. 1996. Calcium currents inpyramidal CA1 neurons in vitro after kindling epileptogenesisin the hippocampus of the rat. Neuroscience 75:57–67.

Fisher SK, Agranoff BW. 1987. Receptor activation and inositol lipidhydrolysis in neural tissues. J Neurochem 46:999–1016.

Fukuda M, Kabayama H, Kojima T, Mikoshiba K. 1996. Mutation ofthe pleckstrin homology domain of Bruton’s tyrosine kinase inimmunodeficiency impaired inositol 1,3,4,5-tetrakisphosphatebinding capacity. J Biol Chem 271:30303–30306.

Gawler DJ, Potter VL, Gigg R, Nahorski SR. 1991. Interactionsbetween inositol tris- and tetrakis-phosphates. Biochem J 276:163–167.

Gibbs JB, Sigal IS, Schaber MD, Scolnick EM, Allard WJ. 1988.Purification of ras GTPase activating protein from bovine brain.Proc Natl Acad Sci USA 85:5026–5030.

GAP1IP4BP in the Rat Brain 327

Gregory PE, Jacks T, Gutmann DH, Wood DL, Mitchell A, Jove R,Park S, Collins FS, Boguski M. 1993. Neurofibromatosis type 1gene product (neurofibromin) associates with microtubules.Somat Cell Mol Genet 19:265–274.

Hartmann H, Eckert A, Velbinger K, Muller WE. 1996. Region-specificdownregulation of free intracellular calcium in the aged ratbrain. Neurobiol Aging 17:557–563.

Henzi V, MacDermott AB. 1992. Characteristics and function of Ca21-and inositol 1,4,5-trisphosphate-releasable stores of Ca21 inneurons. Neuroscience 46:251–273.

Huang Y, Rane SG. 1994. Potassium channel induction by the Ras/Rafsignal transduction cascade. J Biol Chem 269:31183–31189.

Irvine RF, Moor RM. 1986. Microinjection of inositol 1,3,4,5-tetrakisphosphate activates sea urchin eggs by a mechanismdependent on extracellular Ca21. Biochem J 240:917–920.

Kachintorn U, Vajanaphanich M, Barrett KE, Traynor-Kaplan AE.1993. Elevation of inositol tetrakisphosphate parallels inhibitionof Ca21-dependent Cl2 secretion in T84 cells. Am J Physiol264:C671–C676.

Lockyer PJ, Bottomly JR, Reynolds JS, McNulty TJ, Venkateswarlu K,Potter BVL, Dempsey CE, Cullen PJ. 1997. Distinct subcellularlocalizations of the putative inositol 1,3,4,5-tetrakisphosphatereceptors InsP4 and GAP1m result from the PH domain directingplasma membrane targeting. Curr Biol 7:1007–1010.

Loomis-Husselbee JW, Cullen PJ, Dreikhausen UE, Irvine RF, DawsonAP. 1996. Synergistic effects of inositol 1,3,4,5-tetrakisphos-phate on inositol 1,4,5-trisphosphate-stimulated Ca21 release donot involve direct interaction of inositol 1,3,4,5-tetrakisphos-phate with inositol trisphosphate-binding sites. Biochem J314:811–816.

Lu X, Fein A, Feinstein MB, O’Rourke FA. 1999. Antisense knockoutof GAP1IP4BP in the HEL cell line leads to appearence ofintermediate conductance K(Ca) channels that hyperpolarize themembrane and enhance calcium currents. J Gen Physiol (inpress)

Luckhoff A, Clapham DE. 1992. Inositol 1,3,4,5-tetrakisphosphateactivates an endothelial Ca21-permeable channel. Nature 355:356–358.

McCarthy KD, DeVellis J. 1980. Preparation of separate astroglial andoligodendrocyte cell cultures from rat cerebral tissue. J CellBiol 85:890–902.

Molinari S, Billard JM, Battini R, Dutar P, Ferrari S, Lamour Y, PozziL, Baker WA, Killcross AS, Cox H, Robbins TW, Emson PC,Jouvenceau A. 1996. Deficits in memory and hippocampallong-term potentiation in mice with reduced calbindin D28Kexpression. Proc Natl Acad Sci USA 93:8028–8033.

Morris AP, Gallacher DV, Irvine RF, Peterson OH. 1987. Synergism ofinositol trisphosphate and tetrakisphosphate in activating Ca21-dependent K1 channels. Nature 330:653–655.

O’Rourke F, Matthews E, Feinstein MB. 1996. Isolation of InsP4 andInsP6 binding proteins from human platelets: InsP4 promotesCa21 efflux from inside-out membrane vesicles containing 104kDa GAP1IP4BPprotein. Biochem J 315:1027–1034.

Parent A, Poirier J, Baccichet A, Quirion R. 1994. Regulation of1,4,5-IP3, 1,3,4,5-IP4 and IP6 binding sites following entorhinalcortex lesions in rat brain. Neuroscience 61:565–573.

Paxinos G, Watson C. 1997. The rat brain in stereotaxic coordinates.3rd ed. San Diego: Academic Press.

Putney JW. 1986. A model for receptor-regulated calcium entry. CellCalcium 7:1–12.

Putney JW, Bird G St J. 1993. The inositol phosphate–calciumsignaling system in nonexcitable cells. Endocr Rev 14:610–631.

Smith SM, Nahorski SR. 1996. Characterization and distribution ofinositol polyphosphate and ryanodine receptors in the rat brain.J Neurochem 60:1605–1614.

Stricker R, Kalbacher H, Lottspeich F, Reiser G. 1995. Characteriza-tion of a high-affinity InsP4 (inositol 1,3,4,5-tetrakisphosphate)receptor from brain by an anti-peptide antiserum. FEBS Lett370:236–240.

Theibert AB, Estevez VA, Ferris CD, Danoff SK, Barrow RK,Prestwich GD, Snyder SH. 1991. Inositol 1,3,4,5-tetrakisphos-phate and inositol hexakisphosphate receptor proteins: isolationand characterization from rat brain. Proc Natl Acad Sci USA88:3165–3169.

Theibert AB, Estevez VA, Mourey RJ, Marecek JF, Barrow RK,Prestwich GD, Snyder SH. 1992. Photoaffinity labeling andcharacterization of isolated inositol 1,3,4,5-tetrakisphosphate-and hexakisphosphate-binding proteins. J Biol Chem 267:9071–9079.

Tsubokawa H, Oguro K, Robinson HPC, Masuzawa T, Rhee TSG,Takenawa T, Kawai N. 1994. Inositol 1,3,4,5-tetrakisphosphateas a mediator of neuronal death in ischemic hippocampus.Neuroscience 59:291–297.

Tsubokawa H, Oguro K, Robinson HPC, Masuzawa T, Kawai N. 1996.Intracellular inositol 1,3,4,5-tetrakisphosphate enhances thecalcium current in hippocampal CA1 neurons of the gerbil afterischemia. J Physiol (Lond) 497:67–78.

Wilcox RA, Whitham EW, Liu C, Potter BVL, Nahorski SR. 1993.Myo-inositol 1,3,4,5-tetrakisphosphate can independently mobi-lize intracellular calcium via the inositol 1,4,5-trisphosphatereceptor: studies with myo-inositol 1,4,5-trisphosphate-3-phosphorothioate and myo-inositol hexakisphosphate. FEBSLett 336:267–271.

Yamamoto T, Matsui T, Nakafuku M, Iwamatsu A, Kaibuchi K. 1995.A novel GTPase-activating protein for R-Ras*. J Biol Chem270:30557–30561.

328 Signore et al.