80K-H Interacts with Inositol 1,4,5-Trisphosphate (IP3) Receptors and Regulates IP3-induced Calcium Release Activity*

Inositol 1,4,5-trisphosphate receptors (IP3Rs) are intracellular channel proteins that mediate calcium (Ca2+) release from the endoplasmic reticulum, and they are involved in many biological processes (e.g. fertilization, secretion, and synaptic plasticity). Recent reports show that IP3R activity is strictly regulated by several interacting molecules (e.g. IP3R binding protein released with inositol 1,4,5-trisphosphate, huntingtin, presenilin, DANGER, and cytochrome c), and perturbation of this regulation causes intracellular Ca2+ elevation leading to several diseases (e.g. Huntington disease and Alzheimer disease). In this study, we identified protein kinase C substrate 80K-H (80K-H) to be a novel molecule interacting with the COOH-terminal tail of IP3Rs by yeast two-hybrid screening. 80K-H directly interacted with IP3R type 1 (IP3R1) in vitro and co-immunoprecipitated with IP3R1 in cell lysates. Immunocytochemical and immunohistochemical staining revealed that 80K-H colocalized with IP3R1 in COS-7 cells and in hippocampal neurons. We also showed that the purified recombinant 80K-H protein directly enhanced IP3-induced Ca2+ release activity by a Ca2+ release assay using mouse cerebellar microsomes. Furthermore 80K-H was found to regulate ATP-induced Ca2+ release in living cells. Thus, our findings suggest that 80K-H is a novel regulator of IP3R activity, and it may contribute to neuronal functions.

Modulation of the cytosolic free calcium (Ca 2ϩ ) concentration is a highly versatile signaling system involved in the regulation of numerous processes such as fertilization, muscle contraction, secretion, cell growth, differentiation, apoptosis, and synaptic plasticity (1). The inositol 1,4,5-trisphosphate receptor (IP 3 R) 2 is an intracellular IP 3 -gated Ca 2ϩ release channel on the endoplasmic reticulum (ER) that plays a critical role in the generation of complex cytosolic free Ca 2ϩ concentration patterns, e.g. Ca 2ϩ waves and oscillations (1)(2)(3)(4). Three distinct types of IP 3 Rs (types 1-3) have been cloned in mammals, and each type shows distinct properties in terms of their IP 3 sensitivity (5), modulation by cytoplasmic Ca 2ϩ concentration (6), and unique tissue distribution (7). Among them, the type 1 IP 3 R (IP 3 R1) is highly expressed in the central nervous system, particularly in the cerebellum (8,9). Mouse IP 3 R1 is composed of 2749 amino acids (10), and it is divided into five functionally distinct regions: the IP 3 -binding core and suppressor domain near the NH 2 terminus (11,12), the middle modulatory domain for various intracellular modulators (Ca 2ϩ , calmodulin, and ATP) and for phosphorylation by several protein kinases (13)(14)(15)(16)(17), the channel-forming domain with six membranespanning regions, and a gatekeeper domain, a cytoplasmic COOH-terminal tail (CTT) that interacts with several proteins (18,19).
The protein kinase C substrate 80K-H (80K-H) was originally identified as a substrate for protein kinase C. It plays a role as a regulatory subunit of ␣-glucosidase 2 (20) and as a signaling complex of fibroblast growth factor signal transduction (21). 80K-H was also identified as a molecule that interacts with the epithelial Ca 2ϩ channel transient receptor potential cation channel V5 (TRPV5), and it is known to regulate the channel activity (22). In addition, 80K-H interacts with protein kinase C and munc18c to induce glucose transporter 4 translocation to the plasma membrane (23). Moreover 80K-H was also identified as a genetic background of autosomal dominant polycystic liver disease (24,25).
In this study, we identified 80K-H to be a molecule that binds to the CTT of IP 3 Rs by yeast two-hybrid screening. 80K-H interacted with IP 3 R1 in vitro and in cell lysates. We also found that 80K-H colocalized with IP 3 R1 on the ER in COS-7 cells and hippocampal neurons. In addition, we showed that the purified recombinant 80K-H protein enhanced IP 3 -induced calcium release activity in a Ca 2ϩ release assay using mouse cerebellar microsomes. Moreover 80K-H regulated ATP-induced Ca 2ϩ release in living cells. We conclude that 80K-H regulates IP 3induced calcium release by interacting with the cytoplasmic CTT of IP 3 Rs.

EXPERIMENTAL PROCEDURES
Animals-All animal experiments were performed in accordance with the RIKEN guidelines for animal experiments. Every effort was made to minimize the number of animals used.
Yeast Two-hybrid Assay-Yeast two-hybrid assay was performed as described previously (28) using the GAL4-based MATCHMAKER two-hybrid system III (Clontech). Briefly the cDNA libraries were prepared from mouse brain and inserted into pGADT7. To search for proteins interacting with the COOH-terminal cytoplasmic tail of IP 3 R1 (IP 3 R1/CTT), the cDNA libraries were screened using pGBKT7-IP 3 R1/CTT as a bait in AH109 yeast. Positive clones were tested further for specificity by co-transformation into yeast either with pGBKT7-IP 3 R1/CTT or with pGBKT7 alone. DNA from positive clones was isolated, amplified in Escherichia coli strain HB101, and sequenced.
Transfection-Transfection was performed as described previously (32). Briefly COS-7 cells (1 ϫ 10 5 cells) were plated in a 35-mm tissue culture dish with 2.0 ml of medium (in some cases, on glass coverslips). After 24 h, the culture medium was replaced with 2.0 ml of fresh medium. A plasmid DNA (2.0 g) or short interfering RNA (20 M, Stealth siRNA, Invitrogen) was diluted with 100 l of Opti-MEM (Invitrogen). Then FuGENE HD (for DNA, 4.0 l; for siRNA, 6.0 l; Roche Diagnostics) was directly added into the Opti-MEM/DNA mixture and vigorously mixed well by tapping. After incubation for 15 min at room temperature (RT) for complex formation, the mixture was added to the cells in a dropwise manner.
Co-immunoprecipitation and Pulldown Assay-For immunoprecipitation of exogenous proteins, COS-7 cells expressing HA-80K-H and EGFP-IP 3 R1 were washed with PBS and were solubilized in TNE buffer (10 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1.0% Nonidet P40 (Nakarai Tesque), 0.1% SDS). The homogenate was centrifuged at 20,000 ϫ g for 10 min. The supernatant was incubated with the appropriate antibodies and protein G-Sepharose 4B Fast Flow (GE Healthcare) for 2 h at 4°C. The beads were then washed four times with TNE buffer, and proteins were eluted by boiling in SDS-PAGE sampling buffer.
For immunoprecipitation of endogenous proteins, HeLa cells were washed once with PBS and were solubilized in 0.5% Triton X-100, immunoprecipitation (IP) buffer (5.0 mM EDTA, 5.0 mM EGTA, 1.0 mM Na 3 VO 4 , 10 mM Na 4 P 2 O 7 , 50 mM NaF in PBS). The homogenate was centrifuged at 100,000 ϫ g for 30 min. The supernatant was precleared with protein G-Sepharose 4B Fast Flow overnight at 4°C. The precleared supernatant was then incubated with the appropriate antibodies and protein G-Sepharose 4B Fast Flow for overnight at 4°C. The beads were then washed three times with IP buffer, and the proteins were eluted by boiling in SDS-PAGE sampling buffer.
For pulldown binding assays, recombinant MBP fusion proteins and GST fusion proteins were expressed in E. coli BL21 and purified with amylose resin (New England Biolabs) or glutathione-Sepharose 4B (GE Healthcare). Recombinant GSTproteins and MBP-proteins were mixed in PBS containing 100 mM NaCl and 1.0% Triton X-100 for 2 h at 4°C. The beads were then washed four times with TNE buffer, and the precipitates were eluted by boiling in SDS-PAGE sampling buffer.
Western Blotting Analysis-Proteins were separated by SDS-PAGE and transferred to a polyvinylidene difluoride membrane. The membranes were blocked with 5.0% skim milk in PBS containing 0.05% Tween 20 (PBST) for 1 h and probed with the primary antibody for 1 h at RT. After washing with PBST, the membranes were incubated with an appropriate horseradish peroxidase-conjugated secondary antibody, and signals were detected with Immobilon Western Detection Reagents (Millipore).
Immunostaining-The fixed brains were sectioned parasagittally at 8.0-m thickness with a cryostat (CM1850, Leica, Frankfurt, Germany). Transfected COS-7 cells, hippocampal primary neurons grown on glass coverslips, and brain sections were washed once with PBS, fixed with 4.0% formaldehyde in PBS for 10 min, permeabilized with Triton X-100 (0.2% for COS-7 cells and brain sections and 0.1% for hippocampal primary neurons) in PBS for 5 min, and blocked with 1.0% skim milk with or without 1.0% normal goat serum (Vector Laboratories) in PBS for 60 min at RT. The cells were then stained with the indicated primary antibodies (for COS-7 cells and hippocampal neurons, a mouse anti-GFP antibody (1.0 g/ml) and commercially available rabbit anti-80K-H antibody (1.0 g/ml); for brain sections, a rat anti-IP 3 R1/4C11 and affinity-purified rabbit anti-80-KH antibody (1.0 g/ml)) for 60 min at RT. Following three washes with PBS for 15 min in total, appropriate secondary antibodies (for COS-7 cells and hippocampal neurons, Alexa 594-conjugated goat anti-rabbit IgG and Alexa 488conjugated goat anti-mouse IgG; for brain sections, Alexa Fluor 488-conjugated anti-rabbit and Alexa Fluor 594-conjugated anti-rat secondary antibodies (Invitrogen)) were applied for 60 min at RT. After washing with PBS, the coverslips were mounted with Vectashield containing 4,6-diamidino-2-phenylindole (Vector Laboratories) and observed under confocal fluorescence microscopy (FV1000, Olympus, Tokyo, Japan).
Intracellular Ca 2ϩ Imaging-Intracellular Ca 2ϩ imaging was performed as described previously (33). Before imaging Ca 2ϩ signals, fluorescence images of monomeric red fluorescent protein (mRFP) were acquired to identify the transfected cells and saved in the computer.
Measuring Ca 2ϩ Release from Cerebellar Microsomes-IP 3 -induced Ca 2ϩ release from cerebellar microsomes was measured as described previously (34). Briefly the membrane fractions were incubated with 460 nM purified recombinant MBP-80K-H or MBP expressed in E. coli for 15min on ice and suspended at a concentration of 50 g/ml protein in buffer containing 1.0 g/ml oligomycin (Sigma), 2.0 mM MgCl 2 , 25 g/ml creatine kinase (Roche Diagnostics), 10 mM creatine phosphate (Sigma), and 2.0 M Fura-2 (Dojindo). After loading Ca 2ϩ into microsomes by activating Ca 2ϩ -ATPase using 1.0 mM ATP (Sigma), Fura-2 was alternately excited at 340 and 380 nm, and the fluorescence changes at 510 Ϯ 10 nm were detected in response to 10, 20, 40, 80, 1000, or 5000 nM IP 3 application in the presence of 300 -400 nM Ca 2ϩ . After IP 3 stimulation, Ca 2ϩ store size was measured with an ionophore (BrA23187, Sigma). At the end of each set of experiments, maximum and minimum values of Fura-2 fluorescence were obtained in the presence of 2.0 mM CaCl 2 and 10 mM EGTA, respectively. Signals were recorded every 0.01 s with MacLab (version 3.6, ADInstruments) at 30°C. Fits of the Hill equation to Ca 2ϩ release activity were performed by Igor Pro (version 4.04, WaveMetrics) software. 3 Rs-To identify the proteins that interact with the CTT of IP 3 Rs, we screened a mouse brain cDNA library by yeast two-hybrid screening using the CTT of IP 3 R1 as a bait (Fig. 1A). We screened ϳ1.2 ϫ 10 6 yeast transformants and obtained 170 positive clones by the nutritional selection assay. Among them, we found three clones that encoded the sequences corresponding to various lengths of the COOH-terminal fragment of the mouse protein kinase C substrate 80K-H ( Fig. 1B). The shortest fragment was the COOH-terminal region (amino acid residues 341-528) of 80K-H. To specifically identify the region of 80K-H that binds to IP 3 R1, we constructed several 80K-H deletion mutants (Fig. 1C, F0 -F4) and examined their interaction with the CTT of IP 3 R1 in yeast. As shown in Fig. 1C, the F0, F1, and F2 fragments but not the F3 or F4 fragments of 80K-H bound to the COOH-terminal of IP 3 R1, indicating that amino acid residues 365-386 of 80K-H are necessary for interaction with IP 3 R1.

Identification of 80K-H as a Protein Binding to the Cytoplasmic CTT of IP
We next examined whether other types of IP 3 Rs, namely IP 3 R2 and IP 3 R3, interact with 80K-H using IP 3 R2/CTT or IP 3 R3/CTT in yeast. As shown in Fig. 1D, we found that both IP 3 R2 and IP 3 R3 bound to the fragments of 80K-H (F0, F1, and F2) similarly to IP 3 R1.
80K-H Directly Interacts with IP 3 R1 in Vitro-To further verify the direct interaction between 80K-H and IP 3 R1, we performed a pulldown assay with the 80K-H deletion mutants using purified GST-IP 3 R1/CTT and various MBP fusion proteins ( Fig. 2A, F1-F10). Consistent with the results of the yeast two-hybrid assay shown in Fig. 1, GST-IP 3 R1/CTT was observed to bind to the F1 and F2 fragments of 80K-H but not to the F3 or F4 fragments or MBP alone ( Fig. 2A). In addition, we also found that the F7 and F9 fragments but not the F5, F6, F8, or F10 fragments of 80K-H bound to GST-IP 3 R1/CTT. These results indicate that 80K-H directly binds to the cytoplasmic CTT of IP 3 R1 in vitro, and the binding region, i.e. the amino acid residues 365-418 of 80K-H (F9 fragment), is sufficient for binding to IP 3 R1. We do not know why F7 and F9 fragments of 80K-H were strongly bound to IP 3 R1/CTT compared with F1 and F2 fragments, but the amino acid residues 419 -528 may function as a suppressor domain to decide the binding affinity for IP 3 R1.
To determine the region of IP 3 R1 that is responsible for interaction with 80K-H, we performed pulldown assays using recombinant proteins. First we investigate the possibility whether 80K-H also interacts with the NH 2 -terminal region of IP 3 R1. As shown in Fig. 2B, the NH 2 -terminal region of IP 3 R1 did not interact with 80K-H. To narrow down the region of IP 3 R1/ CTT that is responsible for interaction with 80K-H, we performed a pulldown assay using various GST-IP 3 R1/CTT fusion proteins and MBP-80K-H. As shown in Fig. 2C, MBP-80K-H/F9 bound to the ⌬1, ⌬2, and ⌬3 fragments but not to the ⌬4 or ⌬5 fragments or GST alone; this indicated that the amino acid residues 2637-2651 of IP 3 R1 are necessary for binding to 80K-H. We examined the homology of the COOH-terminal region of three types of IP 3 Rs and found that the binding region of IP 3 R1 to 80K-H was highly conserved among the three isoforms of IP 3 R (Fig. 2D). These results, together with the yeast two-hybrid data indicating that 80K-H binds to the CTT of all three IP 3 R types (Fig. 1D), suggest that 80K-H interacts with the corresponding region of IP 3 R2 and IP 3 R3.

80K-H Interacts with IP 3 R1 in Cell Lysates and Colocalizes on the ER in Intact
Cells-To examine whether 80K-H binds to IP 3 R1 in cell lysates, we performed the co-immunoprecipitation assay using COS-7 cells transiently expressing both HA-80K-H and GFP-IP 3 R1. As shown in Fig. 3A, when we immunoprecipitated GFP-IP 3 R1 with anti-GFP antibody, HA-80K-H was co-immunoprecipitated. Conversely GFP-IP 3 R1 was found in the immunoprecipitates with the HA-80K-H antibody. No binding was observed when control IgG was used. We also demonstrated the endogenous IP 3 R1-80K-H interaction in HeLa cells, which mainly express IP 3 R1 (Fig. 3B) (35). Moreover endogenous 80K-H was observed as a meshwork structure with GFP-IP 3 R1 (Fig. 3C) in COS-7 cells. These results indicated that 80K-H colocalized with IP 3 R1 on the ER in intact cells.

80K-H Enhances ATP-induced Ca 2ϩ
Release-To explore the functional effect of 80K-H binding to IP 3 Rs, we examined the effect of 80K-H overexpression on ATP-induced Ca 2ϩ release in COS-7 cells. We transiently expressed both mRFP and 80K-H or mRFP alone in COS-7 cells. 80K-H overexpression did not affect the expression level of IP 3 Rs and several other Ca 2ϩ -related proteins (Fig. 4A). We stimulated the cells with various concentrations of ATP (0.33, 3.0, and 10 M) in the presence or absence (0.5 mM EGTA) of 2.0 mM Ca 2ϩ and imaged the Ca 2ϩ signals (Fig. 4B). In the cells expressing

80K-H Regulates Ca 2؉ Release via IP 3 Rs
80K-H, we found that the peak amplitude of Ca 2ϩ transients induced by various concentrations of ATP was significantly increased compared with that in control cells expressing only mRFP (Fig. 4B). This enhancement was also detected in the absence of extracellular Ca 2ϩ , indicating that overexpression of 80K-H enhances Ca 2ϩ release from the intracellular Ca 2ϩ stores (Fig. 4C). We also obtained similar results in HeLa cells transiently expressing 80K-H (data not shown). When we measured and compared the amount of Ca 2ϩ stored in the ER of COS-7 cells overexpressing either 80K-H or mRFP by using a sarco-ER calcium ATPase inhibitor (cyclopiazonic acid), no significant difference was observed in the Ca 2ϩ stores (data not shown). These results suggest that overexpression of 80K-H enhanced the Ca 2ϩ release activity of IP 3 Rs in vivo.
Knockdown of 80K-H Attenuates ATP-induced Ca 2ϩ Release-We next investigated the knockdown effect of 80K-H on ATPinduced Ca 2ϩ release in COS-7 cells using siRNA for 80K-H. First we checked the specificity of siRNA for 80K-H in COS-7 cells. As shown in Fig. 5A, siRNA targeting 80K-H (si-80K-H) specifically decreased the endogenous 80K-H expression without interfering with the expression of other proteins (IP 3 Rs, PLC␤2, protein kinase C ␣, and ␤-actin). Control siRNA did not affect the expression levels of 80K-H, IP 3 Rs, PLC␤2, protein kinase C ␣, and ␤-actin, and the levels were similar to those in nontransfected cells (Fig. 5A) and non-siRNA control (data not shown). Consistent with this result, the immunosignals of 80K-H almost disappeared in si-80K-H-transfected cells compared with those in the nontransfected cells and control siRNAtreated cells (Fig. 5B). These results revealed that si-80K-H specifically led to the knockdown of 80K-H expression in COS-7 cells. Then we stimulated the si-80K-H transfected COS-7 cells with various concentrations of ATP (0.33, 3.0, and 10 M) and examined the Ca 2ϩ signals. The average peak amplitude of Ca 2ϩ release decreased in si-80K-H-treated cells compared with the control siRNA-treated cells at all ATP concentrations (Fig. 5, C and D). No significant difference was observed in the amount of Ca 2ϩ stored between the si-80K-H-transfected cells and control siRNA-transfected cells (data not shown). Thus, we concluded that 80K-H knockdown decreased ATP-induced Ca 2ϩ release in COS-7 cells.

80K-H Directly Enhanced IP 3 -induced Ca 2ϩ
Release from the Cerebellar Microsomes-To confirm the direct effect of 80K-H on IP 3 R channel activity, we performed an in vitro calcium release assay using recombinant MBP-80K-H and microsomes prepared from mice cerebella (36). After preincubation of microsomes with MBP-80K-H or control MBP, Ca 2ϩ release activity was examined by addition of various concentrations of IP 3 . As shown in Fig. 6, MBP-80K-H significantly enhanced Ca 2ϩ release in response to 20 -1000 nM IP 3 compared with MBP alone. The xhalf values, which are the apparent IP 3 concentrations inducing half of the maximum Ca 2ϩ release, obtained from the Hill equation fitting in this assay were 49.46 Ϯ 1.17 (MBP) and 37.93 Ϯ 1.09 nM (MBP-80K-H); this indicated that MBP-80K-H increased the apparent IP 3 sensitivity by ϳ1.3 times compared with that of MBP alone. These results demonstrated that 80K-H directly binds to IP 3 R1 and enhances IP 3 -induced Ca 2ϩ release activity.

80K-H Colocalized with IP 3 R1 in Hippocampal
Neurons-Finally we stained the mouse brain section with affinity-purified anti-80K-H antibody and anti-IP 3 R1 antibody to characterize the 80K-H expression pattern in the adult mouse brain. As shown in Fig. 7A, 80K-H was highly expressed in the pyramidal cell layer of the hippocampal CA1-CA3 region and extensively colocalized with IP 3 R1 in the CA1 hippocampal neurons. We also immunostained primary cultured hippocampal neurons to further clarify the intracellular distribution of 80K-H. As shown in Fig. 7B, 80K-H predominantly colocalized with IP 3 R1 in the cell body. Higher magnification images of hippocampal dendrites revealed that 80K-H showed a punctate distribution and partially colocalized with IP 3 R1 in the dendrites. Although IP 3 R1 is highly expressed in the cerebellar Purkinje cells, 80K-H was barely detected in Purkinje cells (data not shown).

DISCUSSION
In this study, we identified a novel IP 3 R-interacting protein, 80K-H, using the IP 3 R1 COOH-terminal domain as a bait. We showed that (i) 80K-H directly binds to IP 3 R1/CTT, (ii) 80K-H colocalizes with IP 3 R1 in intact cells and hippocampal neurons, (iii) 80K-H directly enhances IP 3 R activity in vitro and in vivo, (iv) 80K-H binds to all three types of IP 3 Rs, and (v) the 80K-H binding region of IP 3 R1 (amino acids 2637-2651 of IP 3 R1) was highly conserved among the three types of IP 3 R. On the basis of these results, we conclude that 80K-H directly binds to IP 3 Rs and enhances Ca 2ϩ release activity in vivo.
Recently several molecules that regulate Ca 2ϩ release activity by binding to the COOH terminus of IP 3 Rs have been reported, namely cytochrome c (18,37), polyglutamine expanded huntingtin (Htt exp ) and huntingtin-associated protein-1A (19,38), and Bcl-X L (39,40). Interestingly the binding region for these molecules within the IP 3 Rs overlapped with the 80K-H binding region, and their interaction generally enhanced IP 3 R activity. Thus, the corresponding domain could play an important role in the activation of the channel activity through protein-protein interaction. However, the molecular mechanisms underlying IP 3 R activation by these interacting molecules are unclear. Our research challenge in the future is to investigate the IP 3 R gating mechanism and the effect of interacting molecules on IP 3 R gating.
Previously by cDNA microarray analysis, 80K-H was identified as a transcript that was down-regulated in the kidneys of 1,25-dihydroxyvitamin D 3 -deficient mice that suffered from hypocalcemia (41). Interestingly the recovery from severe hypocalcemia in these mice by 1,25-dihydroxyvitamin D 3 or dietary Ca 2ϩ supplementation was shown to be accompanied by an increase in 80K-H mRNA expression. In addition, rats with chronic hypocalcemia induced by vitamin D depletion show significant decreases in the resting Ca 2ϩ FIGURE 6. 80K-H directly enhanced IP 3 -induced Ca 2؉ release from the cerebellar microsomes. A, relationship between IP 3 concentration and IP 3 R activity with control MBP (gray) or MBP-80K-H (black). Results are shown as the mean Ϯ S.E. of six independent experiments. Peak amplitude of Ca 2ϩ release (ordinate) was normalized by the store size. IP 3 concentration is shown on the abscissa as logarithm to the base 10 (nM). *, p Ͻ 0.05 compared with MBP (Student's t test). B, results in A were fitted by the Hill equation using the Igor Pro software. The fitting parameters were as follows: base, apparent basal Ca 2ϩ release; max, apparent maximum Ca 2ϩ release; rate, index of curve kinetics; x, IP 3 concentration; and xhalf, apparent IP 3 concentration inducing half of the maximum Ca 2ϩ release. concentration, IP 3 -sensitive Ca 2ϩ pool, and IP 3 -mobilizable Ca 2ϩ release in response to several classes of Ca 2ϩ -mobilizing agonists in hepatocytes (42). These reports together with our data, which established that 80K-H enhanced IP 3 R activity, suggest that the decreased IP 3 -induced Ca 2ϩ release due to the down-regulation of 80K-H is one of the causes of intracellular Ca 2ϩ perturbations in the case of hypocalcemia.
Recently the PRKCSH gene encoding 80K-H was also identified as a genetic background of autosomal dominant polycystic liver disease (24,25). Autosomal dominant polycystic liver disease is characterized by the formation of multiple liver cysts that arise from proliferating intrahepatic epithelial cells of the bile ducts (cholangiocytes) (43). Proliferation of these bile duct cells is also reported in a cholestatic liver disease (cholestasis) in which the loss of IP 3 Rs in cholangiocytes was reported (44). Because abnormal bile duct cell proliferation in a cholestasis model rat is inhibited by chronic infusion of gastrin that increases intracellular IP 3 levels, we believe that IP 3 -mediated Ca 2ϩ signals might be involved in the regulation of bile duct cell proliferation (45). It would be interesting to examine whether the decreased IP 3 R-mediated Ca 2ϩ signaling caused by the loss or mutation of 80K-H is responsible for the excessive cholangiocyte proliferation in autosomal dominant polycystic liver disease.
In summary, we identified 80K-H to be a novel IP 3 R-interacting protein that enhanced the Ca 2ϩ release activity of IP 3 Rs. Because 80K-H is coexpressed with IP 3 R1 in hippocampal neurons and IP 3 Rs play an important role in synaptic plasticity in hippocampal neurons (46,47), 80K-H may contribute to synaptic plasticity in the hippocampal neurons by regulating the IP 3 R1 activity. Further studies using 80K-H knock-out mice would help us to reveal the relationship between 80K-H and IP 3 R-mediated Ca 2ϩ signaling in brain function and understand the pathological considerations.