Calmodulin Is a Phospholipase C-β Interacting Protein*

Phospholipase C-β3 (PLCβ3) is an important effector enzyme in G protein-coupled signaling pathways. Activation of PLCβ3 by Gα and Gβγ subunits has been fairly well characterized, but little is known about other protein interactions that may also regulate PLCβ3 function. A yeast two-hybrid screen of a mouse brain cDNA library with the amino terminus of PLCβ3 has yielded potential PLCβ3 interacting proteins including calmodulin (CaM). Physical interaction between CaM and PLCβ3 is supported by a positive secondary screen in yeast and the identification of a CaM binding site in the amino terminus of PLCβ3. Co-precipitation of in vitro translated and transcribed amino- and carboxyl-terminal PLCβ3 revealed CaM binding at a putative amino-terminal binding site. Direct physical interaction of PLCβ3 and PLCβ1 isoforms with CaM is supported by pull-down of both isoenzymes with CaM-Sepharose beads from 1321N1 cell lysates. CaM inhibitors reduced M1-muscarinic receptor stimulation of inositol phospholipid hydrolysis in 1321N1 astrocytoma cells consistent with a physiologic role for CaM in modulation of PLCβ activity. There was no effect of CaM kinase II inhibitors, KN-93 and KN-62, on M1-muscarinic receptor stimulation of inositol phosphate hydrolysis, consistent with a direct interaction between PLCβ isoforms and CaM.

lycerol. IP 3 mediates an increase in cytosolic Ca 2ϩ by releasing intracellular stores while diacylglycerol activates protein kinase C (1). Four isoforms of PLC␤, 1-4, have been identified in mammals. PLC␤ 2 and PLC␤ 4 have limited tissue distributions, whereas PLC␤ 3 and PLC␤ 1 are nearly ubiquitous in human tissues; PLC␤ 1 is dominant in brain and PLC␤ 3 is dominant in rat heart and smooth muscle (2)(3)(4)(5)(6)(7)(8)(9)(10). All four mammalian PLC␤ isoenzymes are activated by G␣ q -type G-protein subunits to various degrees, but only PLC␤ 2 and PLC␤ 3 are activated by G␤␥ dimers. PLC␤s, with some isoenzyme differences, may be purified from both cytosolic and particulate fractions of cells (11,12), suggesting that in addition to activation, translocation to substrate at the membrane or maintenance of the enzyme at the plasma membrane may be an additional means of regulation.
PLC␤ proteins, as a family, contain at least five distinct structural domains, identified by homology to the crystal structure of PLC␦ 1 (13). The amino terminus contains a pleckstrin homology (PH) domain, a known protein interaction and membrane-binding domain that may be also a region for G␤␥ subunits interaction with PLC␤ 2 and -␤ 3 isoforms (14,15). In all PLC␤ isoenzymes, the PH domain is followed sequentially by four EF-hand domains, which frequently bind Ca 2ϩ in other protein contexts but whose function in PLC␤s has yet to be characterized. The middle third of the PLC␤ proteins contains catalytic X and Y domains, the activity of which requires a Ca 2ϩ co-factor. The COOH-terminal portion of PLC␤ isozymes has recently been crystallized as a dimer from the turkey PLC␤ homologue and hypothesized to be the domain responsible for the dimerization of the full-length enzyme (16). This region also contains a site for activation by G␣ q subunits (17,18) in all PLC␤ isoenzymes and a 4-amino acid PDZ-binding domain in PLC␤ [1][2][3] (19,20). Whereas there is significant homology among mammalian PLC␤ isotypes (35-55% in full-length protein and 80 -90% in catalytic domains), the differences in G-protein selectivity for activation and subcellular localization among isotypes suggest that each has unique mechanisms of regulation.
Given the ubiquitous distribution of PLC␤ 3 and PLC␤ 1 and their differential regulation by G␤␥ G-protein subunits, isoenzyme-specific modifiers of PLC␤ 3 activity may be postulated. A yeast two-hybrid method was used to screen a mouse cDNA library and identify proteins that interact with the PH and/or the EF-hand domain of PLC␤ 3 . Several candidates have been identified, most intriguing being calmodulin (CaM). PLC␤ 3 is a Ca 2ϩ -sensitive protein whose activation leads to increases in cytosolic Ca 2ϩ levels. CaM acts as an intracellular Ca 2ϩ sensor and is a known regulator of other membrane-associated proteins and thus is a good candidate for interaction with and regulation of PLC␤ 3 . We have detected PLC␤ 3 and PLC␤ 1 expression in 1321N1 human astrocytomas, and using this cell line as a model system, determined that CaM is a regulator of G protein-coupled receptor-stimulated PLC␤ activity. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
For amplification and sequencing of positive clones, growth selectionand ␤-galactosidase-positive yeast were grown for 3 days on His Ϫ Leu Ϫ medium to allow for loss of bait plasmid. Plasmid DNA was isolated, amplified using pACT2 forward and reverse primers (Clontech), sequenced with the same primers, and analyzed by BLAST search (National Institutes of Health, National Library of Medicine) for similarity to known sequences. Previously identified sequences were assessed in a secondary yeast two-hybrid screen for interaction with PH-EF-␤3 in a different vector context. For secondary screening, positive clones in the pACT2 vector were co-transformed into the S. cerevisiae strain PL3␣ along with pBL1 (a histidine-selectable yeast expression plasmid) containing the PH-EF-PLC␤ 3 fragment subcloned in-frame with the estrogen receptor DNA-binding domain. PL3␣ yeast contain a URA3 reporter activity under the control of estrogen response elements. Transformants were selected for growth on Ura Ϫ His Ϫ Leu Ϫ medium.
Inositol Phosphate Assay-Inositol phosphate (IP) assays were performed with 1321N1 cells as previously reported (23). 1321N1 cells were routinely subcultured in DMEM, 10% fetal bovine serum and standard antibiotics (penicillin and streptomycin) in humidified 5% CO 2 at 37°C. For IP assays, 1321N1 cells were subcultured in 24-well plates at a density of 1 ϫ 10 5 cells/well and allowed to attach overnight. Cells were then labeled overnight with myo-[ 3 H]inositol, 1 Ci/0.5 ml/ well, prepared in sterile inositol-free, bicarbonate-buffered DMEM without additives. Following radiolabeling, cells were pre-treated for 30 min (KN-62, KN-92, and KN-93) or 10 min (W-13 and fluphenazine) in 20 mM Hepes-buffered DMEM, pH 7.4 (HDMEM), at 37°C in room air. After pre-treatment, 10 mM LiCl was added for 10 min followed by stimulation of muscarinic receptors with 1 mM carbachol for 20 min. Following stimulation, cells were lysed with 0.5 ml of cold 5% trichloroacetic acid and the soluble lysate was ether extracted three times. Inositol phosphates (IP, IP 2 , and IP 3 ) were purified by anion exchange chromatography with ammonium formate. Lipids were solubilized with 1 N NaOH and collected to quantitate [ 3 H]inositol phospholipids. Samples were harvested into scintillation vials and radioactivity was quantitated in a liquid scintillation counter. Percent (%) conversion was calculated as ( ). All assays were performed in triplicate and values reported reflect an average of at least three experiments Ϯ S.E. Student's t tests were performed to assess statistical significance where indicated.
Cell Fractionation and Immunoprecipitation-1321N1 cells were plated at 3 ϫ 10 6 cells/10-cm plate and grown for 3 days under standard conditions. On day 3, cells were washed twice with HDMEM and collected by scraping and pelleting at 500 ϫ g for 5 min. Cold lysis buffer (0.6 ml: 10 mM Tris, pH 7.4, 5 mM MgCl 2 , and protease inhibitors, 200 M benzamidine, 200 M PMSF, 2 M pepstatin A, and 2 M leupeptin) was added to each sample and incubated on ice for 10 min. Cells were lysed with 15 strokes of a Dounce homogenizer and a 50-l fraction was saved as crude lysate. The remainder of the lysate was centrifuged at 500 ϫ g, 4°C, for 5 min to pellet nuclei and intact cells. The low speed supernatant was centrifuged at 300,000 ϫ g, 4°C, for 25 min, separating soluble and particulate fractions. The particulate (membrane) fraction was resuspended in 0.6 ml of extraction buffer (50 mM Hepes, 2.5 mM EDTA, 150 mM NaCl, 1% Triton X-100, 1 mM dithiothreitol, and protease inhibitors) and incubated at 4°C for 1 h with inversion. After extraction, the membrane fraction was centrifuged at 16,000 ϫ g, 4°C, for 30 min to pellet detergent-insoluble particulates. Supernatant was saved as membrane extract.
For immunoprecipitation, anti-PLC␤ isozyme-selective polyclonal antibodies were added to soluble (cytosol) and membrane extract fractions at a 1:200 dilution and incubated at 4°C overnight with continuous inversion. Immune complexes were precipitated with 50 l of Protein A-Sepharose that had been washed and equilibrated in extraction buffer (v/v) according to the manufacturer's specifications. The resulting immunoprecipitates were separated by 7.5% SDS-PAGE. The separated proteins were transferred electrophoretically to nitrocellulose paper, incubated with PLC␤-selective polyclonal antibodies at a 1:1000 dilution followed by alkaline phosphatase-conjugated goat antirabbit antibodies at 1:5000 dilution, and visualized with alkaline phosphatase chemiluminescent substrate (ImmuneStar TM ) according to the manufacturer's specifications.
For preparation of cell lysates for CaM-Sepharose 4B binding assay, 1321N1 cells were plated at 3 ϫ 10 6 cells/10-cm plate and grown for 3 days at 37°C with 5% CO 2 in DMEM ϩ 10% fetal bovine serum and antibiotics. Plated cells were washed twice with HDMEM. Cells were scraped into tubes and pelleted at 500 ϫ g for 5 min. Cold lysis buffer (0.6 ml: 10 mM Tris, pH 7.4, 5 mM MgCl 2 , 2 mM CaCl, and protease inhibitors) was added to each sample and incubated on ice for 10 min. Cells were lysed with 15 strokes of a Dounce homogenizer. The lysate was centrifuged at 500 ϫ g, 4°C, for 5 min to pellet nuclei and intact cells prior to incubation with CaM-Sepharose 4B beads as described below.
Transcription and Translation of Radiolabeled Proteins-[ 35 S]Methionine-labeled amino-terminal and carboxyl-terminal PLC␤ 3 were prepared according to the manufacturer's instructions using the in vitro TNT TM quick coupled transcription/translation system (Promega, Madison, WI). Briefly, 1 g of pcDNA3 plasmid containing cDNA sequence for the amino-(amino acids 1-313, N-PLC␤ 3 ) or carboxyl-(amino acids 670 -1173, C-PLC␤ 3 ) terminal sequences of PLC␤ 3 was incubated with TNT TM reticulocyte lysate and 20 Ci of [ 35 S]methionine in a total volume of 50 l for 90 min at 30°C. Labeling efficiency and level of protein expression were evaluated by SDS-PAGE and autofluororadiography (described below).
CaM-Sepharose 4B Binding Assay-The CaM-Sepharose binding assay with [ 35 S]methionine-labeled proteins was performed according to the protocol of Chuang et al. (24) with the following modifications. CaM-Sepharose 4B resin was washed twice in 500 l of dH 2 O, 4°C, and equilibrated for 1 h in binding buffers (30 mM MOPS, pH 7.2, 100 mM KCl, 150 mM NaCl, 2 mM MgCl, 1 mM dithiothreitol, and protease inhibitors) made with calcium calibration buffers (K 2 EGTA and/or CaE-GTA in varying amounts) from Molecular Probes (Eugene, OR). Free Ca 2ϩ concentrations ranged from 0 to 39 M. Equivalent radioactive quantities of radiolabeled 35 S-N-PLC␤ 3 or 35 S-C-PLC␤ 3 were incubated with 50 l of pre-equilibrated calmodulin-Sepharose 4B resin in a total volume of 500 l overnight at 4°C with inversion. Unbound material was removed by washing three times with 200 l of binding buffer. SDS sample buffer was added to the final pellet and the calmodulin-Sepharose precipitants were resolved by 12.5% SDS-PAGE. For autofluororadiography the resultant gel was soaked in glacial acetic acid for 20 min, impregnated with 10% 2,4-diphenyloxazole scintillant for 15 min, and rinsed in ddH 2 O for 10 min before being dried to Whatman No. 1 filter paper. The gels were exposed overnight to x-ray film and the 35 S-radioactive bands were visualized. For cell lysates (prepared as described in the cell fractionation and immunoprecipitation section) 70 l of a 50% slurry of CaM-Sepharose 4B resin was equilibrated in extraction buffer (50 mM Hepes, 2.5 mM CaCl, 150 mM NaCl, 1% Triton X-100, 1 mM dithiothreitol, and protease inhibitors) and added to each lysate sample. Samples were incubated at 4°C with inversion for 1 h. After incubation the pelleted beads were washed three times in extraction buffer at 4°C; after the final wash, proteins were eluted from the beads with the addition of 50 l of 2ϫ SDS-PAGE loading buffer and boiling for 5 min. Proteins in the supernatant were separated by 7.5% SDS-PAGE and transferred electrophoretically to nitrocellulose paper followed by Western blotting as described above with PLC␤-selective polyclonal antibodies. 3 Interaction-To identify novel protein interactions with PLC␤ 3 , we utilized a yeast two-hybrid screening system. A mouse brain cDNA library was screened with the NH 2 -terminal region (PH and EF-hand domains) of PLC␤ 3 as bait. Of the positive clones identified in this screen, full-length mouse calmodulin was isolated at least three times and interaction was confirmed by a ␤-galactosidase assay (Table I) and a secondary growth screen as described under "Experimental Procedures" (Fig. 1).

Identification of CaM/PLC␤
The Calmodulin Binding Site-Binding site search and analysis site The Calmodulin Target Database 2 was used for computer analysis of the PLC␤ 3 protein sequence. Three putative calmodulin-binding sites were identified in the full-length PLC␤ 3 protein sequence by hydropathy, ␣-helical propensity, residue weight, residue charge, hydrophobic residue content, helical class, and occurrence of key residues. Two of these sites are located in the COOH-terminal region and one in the NH 2terminal region of PLC␤ 3 . The NH 2 -terminal CaM-binding site is shared by PLC␤ 1 and has 90% identity to the NH 2 -terminalbinding site in PLC␤ 3 ( Fig. 2A). When compared with other CaM-binding domains, conserved hydrophobic and positively charged amino acids were identified in the PLC␤ 3 and PLC␤ 1 sequences consistent with putative calmodulin-binding domains (Fig. 2B).
PLC␤ 3 and PLC␤ 1 Are Expressed in 1321N1 Cells-1321N1 cells express G q -coupled muscarinic receptors that are activated by carbachol to hydrolyze phospholipids (25). To establish which PLC␤ isoforms are expressed in 1321N1 cells to mediate G protein-activated inositol phospholipid hydrolysis, we performed immunoprecipitation and Western blot analysis using PLC␤ isoenzyme selective antibodies. We determined that PLC␤ 3 and PLC␤ 1 isoenzymes are expressed in 1321N1 cells, whereas PLC␤ 2 and PLC␤ 4 are not expressed at detectable levels (Fig. 3). Having confirmed the presence of PLC␤ 3 in 1321N1 cells, we investigated the possible interaction of the enzyme with calmodulin in vitro. N-PLC␤ 3 Physically Interacts with CaM-To verify that the CaM-binding domains of PLC␤ 3 physically interact with CaM, a co-precipitation assay was performed using 35 S-radiolabeled NH 2 -terminal (N-PLC␤ 3 ) or C-terminal (C-PLC␤ 3 ) regions of PLC␤ 3 (see ''Experimental Procedures'') and CaM-Sepharose beads. In vitro transcription and translation and 35 S radiola-beling of N-PLC␤ 3 and C-PLC␤ 3 were verified by SDS-PAGE and autofluororadiography (not shown). Co-precipitation assays were performed with 35 S-N-PLC␤ 3 (Fig. 4A) in the presence of binding buffers containing the indicated concentrations of free Ca 2ϩ (Fig. 4A). N-PLC␤ 3 protein bound most strongly in the buffer containing no free Ca 2ϩ and bound less effectively with increased Ca 2ϩ . Co-precipitation of C-PLC␤ 3 (Fig. 4B) with CaM-Sepharose in the presence of binding buffers containing the indicated concentrations of free Ca 2ϩ revealed that C-PLC␤ 3 did not bind CaM-Sepharose under any of the above conditions (Fig. 4B). Binding of N-PLC␤ 3 to CaM occurred in the absence of free Ca 2ϩ in the presence of varying NaCl concentrations up to 200 mM (data not shown).
Full-length PLC␤ 3 and PLC␤ 1 Are Isolated from Cell Lysates by CaM-Sepharose Beads-To determine whether the physical interaction between CaM and N-PLC␤ 3 identified by the in vitro co-precipitation assay occurred with the full-length protein, we used CaM-Sepharose beads to precipitate PLC␤ 3 or 2 calcium.uhnres.utoronto.ca/ctdb/pub_pages/general/index.htm.   3

with CaM
Table of ␤-galactosidase activity of representative growth-positive clones from the yeast two-hybrid assay. ␤-Galactosidase-positive clones were identified by activities at least 2-fold higher than background. Mitochondrial proteins are often considered to be false positives in yeast two-hybrid screens and were not pursued further. Calmodulin Antagonists Inhibit Activity of PLC␤ in Vivo-To ascertain the potential physiological consequences of the interaction between CaM and PLC␤, we treated 1321N1 cells with CaM inhibitors (W-13 and fluphenazine) and calmodulin-dependent kinase II (CaMKII) inhibitors (KN-93 and KN-62), and assayed for effects on PLC␤ activity by measuring subsequent carbachol-stimulated IP hydrolysis. CaM antagonists, W-13 and fluphenazine, reduced carbachol-stimulated IP hydrolysis in intact 1321N1 cells by 52 and 60%, respectively, compared with carbachol stimulation of vehicle pre-treated cells (Fig. 5A), suggesting that CaM is supportive of PLC␤ activity in 1321N1 cells.
To determine whether the inhibition of PLC␤ activity by W-13 was sensitive to stimulation of IP hydrolysis or changes in intracellular Ca 2ϩ concentration, we investigated effects of calmodulin inhibitor W-13 on IP hydrolysis either concurrent with carbachol activation or 10 min post-activation. When W-13 was added concurrently with carbachol, IP hydrolysis was inhibited by 50%, whereas only 25% inhibition occurred when W-13 was added 10 min post-activation (Fig. 5B).
To address whether CaM kinase may mediate the effects of CaM on PLC␤ activity, 1321N1 cells were treated with CaMKII inhibitors, 30 M KN-93 or 2 M KN-62 (26, 27). In 1321N1 cells pre-treated with CaMKII inhibitors for 30 min, there was no effect on basal or carbachol-stimulated IP hydrolysis (Fig. 6). CaMKII inhibitors had no effect on basal or carbachol-stimulated IP hydrolysis in 1321N1 cells at concentrations ranging from 1 nM to 100 M (data not shown). The lack of effect of CaMKII antagonists in this system supports a physiological role for direct interaction between PLC␤ enzymes and CaM without requiring activation of CaMKII.

DISCUSSION
The PLC␤ class of enzymes is a key component in G-proteinlinked receptor-mediated signaling cascades. In response to external cellular stimuli and subsequent activation of G protein-coupled receptors, PLC␤ is responsible for the production of IP 3 and DAG, two important second messengers in a variety of cellular functions. Examples of a few of the receptors that act through the G-protein/PLC␤ pathway include thromboxane A 2 , bradykinin, angiotensin, histamine, vasopressin, M1-type muscarinic cholinergic, and thyroid stimulating hormone. Efforts are being made to elucidate differences in signaling through PLC␤ pathways to identify tissue and/or receptor-specific targets for regulation. Novel putative regulators of PLC␤ isoenzymes identified recently include: NHERF2, which is postulated to interact with the 8-amino acid carboxyl terminus of PLC␤3 (19); de-polymerized tubulin, which interacts with PLC␤1 and interferes with G␣ q activation (28,29); and Rac1, which activates PLC␤2 (30 -32).
1321N1 cells are an established human astrocytoma cell line extensively used to study G protein-coupled inositol phospholipid hydrolysis. In 1321N1 astrocytoma cells, PLC␤ 3 segregates mostly with cellular particulate rather than soluble fractions, whereas the cellular distribution of PLC␤ 1 is such that it is primarily isolated from soluble fractions. 3 In addition to different subcellular locations of these enzymes in 1321N1 cells, previous work in other cell lines describe differences in activation among the isozymes (33,34) suggestive of unique mechanisms of regulation. Our laboratory has sought to identify novel regulators of PLC␤ 3 and/or PLC␤ 1 isozyme signaling.
Full-length CaM was identified as a possible PLC␤ 3 interacting protein by yeast two-hybrid screening of a mouse brain cDNA library. Sequence analysis of the full-length PLC␤ 3 protein reveals three putative CaM-binding sites, one in the amino terminus and two in the carboxyl tail of the protein. PLC␤ 1 has a CaM-binding site in the NH 2 -terminal with 90% identity to the PLC␤ 3 CaM-binding site. The best described CaM binding motif is an IQ motif (35); however, other non-IQ CaM-binding domains, both Ca 2ϩ -independent and Ca 2ϩ -dependent, have  been identified as short 16 -35-amino acid regions that segregate hydrophobic residues from basic and polar residues on an ␣-helical wheel projection (36). Non-IQ CaM-binding domains are found in other membrane-associated proteins including G protein-coupled receptor kinases (37), regulators of G-protein signaling (RGS) proteins (38), and myristoylated alanine-rich C kinase substrate protein (39,40), among others. The CaMbinding site identified in the NH 2 -terminal regions of PLC␤ 3 and PLC␤ 1 have a similar distribution of amphiphilic residues as these non-IQ motif CaM-binding proteins (Fig. 2B).
PLC␤ enzymes are Ca 2ϩ -sensitive proteins (12) that associate with the plasma membrane, both of which are properties associated with CaM-regulated proteins such as phosphatidylinositol 3-kinase (41), RGS proteins (38), and myristoylated alanine-rich C kinase substrate (42). Ca 2ϩ /CaM regulates many signaling molecules that are also sensitive to G␤␥ and phospholipids similarly to PLC␤ 3 and PLC␤ 1 . Anionic phospho-lipid-binding sites and CaM-binding sites share some similarities and may overlap in certain proteins. Ca 2ϩ /CaM attenuates the ability of PIP 3 to inhibit RGS4 intrinsic GTPase activity without activating RGS directly (38). Membrane association of G protein-coupled receptor kinases is required for activation and is enhanced by phosphatidylinositol bisphosphate and G␤␥ subunits (43,44); Ca 2ϩ /CaM inhibits several G protein-coupled receptor kinase subtypes as a consequence of reduced phospholipid binding/membrane association (24,37,45). The presence of a CaM-binding site in the NH 2 -terminal region of PLC␤, and the isolation of PLC␤ 3 and PLC␤ 1 from cell lysates with CaM-Sepharose beads, suggest that CaM is a PLC␤ regulatory protein that had not been previously recognized.
We demonstrated by co-precipitation assays that CaM binds PLC␤ 3 through the NH 2 -terminal region. These data suggest that N-PLC␤ 3 is a Ca 2ϩ -independent CaM-binding peptide (Fig. 4). Data to support physiological significance of a CaM/ PLC␤ interaction is shown by the isolation of both PLC␤ 3 and PLC␤ 1 isoforms from 1321N1 cell lysates by CaM-Sepharose (Fig. 5). Inhibition of CaM by W-13 was shown to reduce inositol phospholipid hydrolysis (Fig. 6, A and B) in vivo consistent with an inhibition of PLC␤ activity. Inhibition was seen when cells were treated with W-13 prior to activation when intracellular Ca 2ϩ levels are low, and when cells were treated concurrent with activation or 10 min post-activation when intracellular Ca 2ϩ levels are changing because of the activation of PLC␤. These data suggest that CaM is integral in the muscarinic receptor-activated inositol phospholipid hydrolysis pathway, and the association of PLC␤ with CaM may be independent of intracellular Ca 2ϩ levels similar to the Ca 2ϩ independence of CaM binding N-PLC␤ 3 (Fig. 4A). In neural tissues where concentrations of CaM and CaM-binding proteins are very high (46), many CaM-binding proteins are known to associate with CaM when intracellular Ca 2ϩ levels are very low and even in the presence of chelators (47). Neuromodulin binds CaM in the presence of EGTA and Ca 2ϩ disrupts binding (48), similar to the profile seen with N-PLC␤3 binding of CaM (Fig. 4A). Inducible nitric-oxide synthase is an example of a protein that binds CaM independent of Ca 2ϩ levels and is then activated by CaM maximally at 0.1 nM free Ca 2ϩ in vitro (49).
Data herein support a direct interaction between PLC␤ and CaM and maintain that regulation of PLC␤ activity by calmod- W-13 and fluphenazine pre-treatments significantly reduced total carbachol-stimulated IP hydrolysis as well as net carbachol-stimulated IP hydrolysis compared with vehicle pre-treated controls (*, p Ͻ 0.001). B, cells were treated with 1% Me 2 SO 10 min pre-stimulation (Vehicle), with 100 M W-13 10 min pre-stimulation (pre-treatment), with 100 M W-13 concurrent with stimulation (concurrent), or with 100 M W-13 10 min post-stimulation (post-treatment). W-13 treatments significantly reduced total carbachol-stimulated IP hydrolysis as well as net carbachol-stimulated IP hydrolysis compared with vehicle pre-treated controls (*, p Ͻ 0.001) by 42% with pre-treatment, 58% with concurrent treatment, and 32% with post-treatment. Experiments were performed in triplicate with n ϭ 3. ulin is not acting through CaMKII. Previous work by others has shown an inhibition of carbachol-stimulated PLC␤ 3 activity with CaMKII inhibitors in PLC␤ 3 -transfected COS cells, suggesting that CaMKII is an activator of PLC␤ 3 activity (50). We were not able to repeat this finding in 1321N1 cells endogenously expressing PLC␤ 3 and PLC␤ 1 (Fig. 7 and data not shown). Regulation and phosphorylation of PLC␤ 3 may occur through a different mechanism in 1321N1 cells distinct from exogenously expressing transfected cells.
We propose a model whereby CaM facilitates PLC␤ activity, possibly by increasing access to substrate. While cognizant that the in vivo activity of CaM inhibitors to inhibit inositol phospholipid hydrolysis may result from effects on any member of the pathway from M1 receptor to G-protein to PLC␤, the association of CaM in vitro with N-PLC␤ 3 and the co-precipitation of PLC␤ 3 and PLC␤ 1 by CaM-Sepharose suggest that PLC␤ isozymes are at least one target affected by CaM. We are currently working to determine the differential effects of CaM on PLC␤ 3 and PLC␤ 1 activity, membrane association, or stimulation by G-proteins.