The Journal of Neuroscience, March 1, 1996, 76(5):1645-1658 Insulin Receptor in Aplysia Neurons: Characterization, Molecular Cloning, and Modulation of Ion Currents Elizabeth A. Jonas,2,a Ronald J. Knox,2 Leonard K. Kaczmarek,p James H. Schwat-tz,l and David H. Solomonl,” ‘Center for Neurobiology and Behavior, Columbia University College of Physicians and Surgeons, New York, New York 10032, and 2Department of Pharmacology, Yale University School of Medicine, New Haven, Connecticut 06520 We have isolated the cDNA for a tyrosine kinase receptor that can be detected within 15-30 min. These effects were not seen is expressed in the nervous system of Aplysia californica and with insulin-like growth factor-l. In voltage-clamped neurons, that is similar to the vertebrate insulin receptor. Binding studies insulin produces an increase in the amplitude of the voltage- and immunocytochemical staining show that the receptor is dependent Ca2+ current that can be blocked by preincubation abundant in the bag cell neurons. Application of vertebrate with herbimycin A, an inhibitor of tyrosine kinases. Insulin also insulin to clusters of bag cell neurons stimulates the phosphor- enhances a delayed K’ current. We suggest that insulin-like ylation of the receptor on tyrosine residues, and exposure of peptides regulate the excitability of the bag cell neurons. isolated bag cell neurons to insulin produces an increase in Key words: insulin receptor; bag cell neurons; tyrosine kinase; height and a decrease in duration of the action potentials that Aplysia; calcium channels; potassium channels Insulin is a peptide growth factor that stimulates growth and (Schwartz et al., 1992),suggesting that insulin may act through the differentiation of various kinds of developing cells. For many blood streamas a hormone.In addition, insulincan be synthesized growth factors, the mechanismof transductioninvolves binding of by neurons and is releasedby depolarization, suggestingthat it the peptide to a membrane-spanningreceptor activating a ty- may act as a neurotransmitter or as a local peptide hormone rosine kinase in the cytoplasmicdomain of the receptor (Rosen, (Uvnas-Wallenstein,1981; Boyd et al., 1985).Several electrophys- 1987; Cantley et al., 1991). Many receptor and nonreceptor ty- iological actions of insulin have been described, including in- rosine kinasesare expressedabundantly in the mature nervous creasedelectrical coupling betweencultured sympatheticneurons system(Lai and Lemke, 1991), and their functions in long-term (Wolinsky et al., 1985) and inhibition of spontaneousfiring of rat programs of cell growth and differentiation are well described hippocampalneurons (Palovcik et al., 1984).Moreover, identified (Schlessingerand Ullrich, 1992). In neurons,changes induced by Aplysia neurons are rapidly hyperpolarized in responseto verte- protein tyrosine kinasesinfluence clustering of receptors during brate insulin (Shapiro et al., 1991). development (Huganir and Greengard, 1990) and extension of Insulin-like peptides have also been describedin nervous sys- axonal growth cones (Goldberg and Burmeister, 1992; Maness tems of many invertebrates (Thorpe and Duve, 1984; Steiner et and Cox, 1992). It is unclear, however, whether receptor tyrosine al., 1985), notably in the pond snail, Lymnaea stagnalis. These kinasesparticipate in short-term changesof neuronal excitability. molluscaninsulin-like peptides (MIPS) are similar in sequenceto Nevertheless,recent evidence suggeststhat tyrosine kinasesreg- mammalianinsulin and are believed to regulate reproduction and ulate ion channelsacutely (Huang et al., 1993;Wilson and Kacz- growth (Smit et al., 1988, 1991). Immunocytochemicalevidence marek, 1993; Wang and Salter, 1994) and that nonreceptor ty- indicates that many neurons in the central nervous systemsof rosine protein kinases are needed to induce long-term Lymnaea and Aplysia contain these peptides (Van Minnen and potentiation (O’Dell et al., 1991) and spatial learning (Grant et Schallig, 1990). al., 1992). We report the cDNA cloning and biochemicalcharacterization Whether insulin and insulin-like peptidesact in synaptic trans- of an insulin receptor from Aplysia califomica and show that it is missionis still uncertain. Havrankova et al. (1978) and Baskin et expressedabundantly in the bag cell neurons,a group of cellsthat al. (1987) detected receptors for insulin in mature vertebrate play a key role in the onset of reproductive behaviors(Conn and brain. An uptake mechanismfrom plasma has been described Kaczmarek, 1989). We also provide functional evidence that bag cell neuronsexpress a receptor similar to the mammalianinsulin Received Nov. 11, 1995; revised Dec. 7, 1995; accepted Dec. 8, 1995. receptor. Exposure of bag cell neurons to mammalianinsulin This work was supported by National Institutes of Health Grants NS29809 (J.H.S.), NS18492 (L.K.K.), and AGO0486 (E.A.J.), and by Research Scientist Award triggers autophosphorylationon tyrosine residuesof the bag cell MH00921 (J.H.S.). D.H.S. was supported by a Medical Research Council of Canada neuron receptor and enhancesvoltage-dependent Ca2+ and K+ Fellowship. We thank Alice Elste for her excellent technical assistance, John Koester currents, leading to changesin action potentials. for reading this manuscript, and Ken Kirschner for preparing this manuscript. Correspondence should be addressed to Dr. James H. Schwartz, Center for Neurobiology and Behavior, Columbia University College of Physicians and Sur- MATERIALS AND METHODS geons, 722 West 168th Street, New York, NY 10032. aE.A.J. and D.H.S. contributed equally to the work. Immunocytochemistry. Anti-peptide antibodies against the insulin-binding Dr. Solomon’s current address: Department of Pharmacology, Columbia Univer- pocket of the human insulin receptor 01 subunit (UBI, Lake Placid, NY) sity College of Physicians and Surgeons, 630 West 168th Street, New York, NY were diluted 1:20 for immunocytochemistry and 1:200 for immunoblot 10032. analysis. Tissues were fixed in Bouin’s fixative modified for Aplysia [2% Copyright 0 1996 Society for Neuroscience 0270-6474/96/161645-14$05.00/O paraformaldehyde, 15% picric acid, 1% glacial acetic acid in 0.1 M sodium 1646 J. Neurosci., March 1, 1996, 76(5):1645-1658 Jonas et al. l Insulin Receptor in Bag Cell Neurons phosphate, pH 7.4, containing 30% (w/v) sucrose]. Cryostat serial sections for 10 min at 16°C. The clusters were then homogenized in 400 ~1 of (10 pm) were mounted on gelatin-coated glass slides, rinsed with PBS buffer (1% Lubrol, 50 mM Tris-HCI, pH 7.4, 10 FM Na-orthovanadate, 30 containing 0.25% saponin, dehydrated and bleached in 0.3% H,O,, and mM Na-pyrophosphate, 50 mM NaF, 20 pM ZnCl,, 0.25 mM PMSF, and 10 then rehydrated. Sections were blocked with 2% normal goat serum and pg/ml each of leupeptin, antipain, pepstatin, and aprotinin) in glass-glass incubated in a 1:20 dilution of rabbit anti-insulin receptor antibody (150 tissue grinders for 3 min on ice. &ml) overnight at 4°C. Controls used were normal rabbit IgG (150 or 36 Samples were then centrifuged at 100,000 X g for 30 min. The super- &ml). Slides were rinsed again in PBS/saponin, incubated in rhodamine- natants were collected and transferred to I .5 ml Eppendorf tubes. Tubes conjugated goat anti-rabbit IgG, and rinsed and coverslipped with Aqua- were incubated at 4°C with rotation for 2 hr after addition of 1 pg of PolyMount (Polysciences, Warrington, PA). Sections were viewed by anti-insulin receptor p subunit monoclonal antibody (AB-3, Oncogene phase microscopy and epifluorescence with a Leitz microscope (filter Science, Manhasset, NY). After 20 ~1 protein G-Sepharose beads [50% pack N-2) and photographed with high-speed Kodak Tri-X film (Ro- (v/v) in lysis buffer; Pharmacia, Piscataway, NJ] were added and the chester, NY). samples were rotated at 4°C for 2 hr, the immunoprecipitates were Intact bag cell clusters and bag cell neurons in culture were examined collected by centrifugation at 3000 X g for 1 min and washed twice in 500 with a Bio-Rad MRC 1000 confocal apparatus (Bio-Rad, Cupertino, CA). ~1 of homogenization buffer, once in 500 ~1 of wash buffer (100 mM NaCI, Immunocytochemistry was performed as described above. Images were 50 mM Tris-HCI, pH 7.4, and 10 pM Na-orthovanadate), and once in 500 processed with Bio-Rad’s COMOS software package; 1.0 pm slices in the ~1 of kinase reaction mix (5 mM MnCI,, 5 mM MgCI,, 10 mM Tris-HCl, pH z-axis were processed and analyzed. 7.4, and 10 WM Na-orthovanadate). Phosphorylation reactions were per- Immunoblot analysis. Tissues were homogenized in 50 mM Tris-HCl, formed in 10 ~1 of kinase start mix (kinase reaction mix plus 10 &i of pH 7.5, 10 IIIM MgCl,, 1 mM EGTA, 5 mM 2-mercaptoethanol, 0.1 mM [$*P]ATP, 6000 Ciimmol; DuPont NEN, Boston, MA) for 3 min at room phenolmethyl sulfonylfluoride (PMSF; Sigma, St. Louis, MO), 50 Fg/ml temperature. The reaction was stopped by adding 100 ~1 of 100 mM NaCI, aprotinin, 5 mM benzamidine, and 0.1 mM leupeptin (400 ~1) with 50 mM Tris-HCI, pH 7.4, 50 mM EDTA, and 1 mM ATP. Immunopre- glass-glass homogenizers (Micrometric, Tampa, FL). The homogenate cipitates were sedimented, and SDS sample buffer was added after the was centrifuged at 1000 X g to remove debris and then at 100,000 X g for supernatant was aspirated. The samples were boiled and then applied to 30 min in a TL-100 (Beckman, Palo Alto, CA). After SDS-PAGE of 100 SDS-acrvlamide gels. For the immunocomnlex assay. I?*PlATP 16000 pg samples [as measured by the Bradford (1976) method], fractionated Ci/mmolj was used. For immunoblotting, t’he SDS ge&ec&ophoresed proteins were transferred to nitrocellulose (0.45 pm, Bio-Rad). Complete proteins were transferred to nitrocellulose (Bio-Rad, Hercules, CA). transfer was shown by staining with Amido black (Harlow and Lane, Nonspecific binding was blocked by incubating the blot in TBSN buffer 1988). The membrane was blocked in TBSN (25 mM Tris-HCI, pH 8, 130 with 3% BSA for 1 hr at room temperature.