Relationship between Inositol 1, 4, 5-Trisphosphate Receptor Isoforms ...

Relationship between Inositol 1,4,5-Trisphosphate Receptor Isoforms and Subcellular Ca21 Signaling Patterns in Nonpigmented Ciliary Epithelia Keiji Hirata,1 Michael H. Nathanson,2 Angela D. Burgstahler,2 Keisuke Okazaki,1 Elisabetta Mattei,1 and Marvin L. Sears1 PURPOSE. Subcellular Ca21 signaling patterns, such as Ca21 waves, gradients, and oscillations, are an important aspect of cell regulation, but the molecular basis for these signaling patterns is not understood. Because Ca21 release patterns differ among isoforms of the inositol 1,4,5-trisphosphate (InsP3) receptor, the relationship between the distribution of these isoforms and subcellular Ca21 signaling patterns in nonpigmented epithelial (NPE) cells was investigated. METHODS. The distributions of the types I, II, and III InsP3 receptors were determined in NPE cells by immunofluorescence, and subcellular Ca21 signaling patterns in these cells were examined by confocal line scanning microscopy. RESULTS. The type I InsP3 receptor was concentrated at the basal pole of NPE cells, whereas the type III receptor was localized to the apical pole. The type II InsP3 receptor was not expressed in detectable amounts. Acetylcholine induced increases in Ca21 that were mediated by InsP3, and these Ca21 increases began as Ca21 waves that were initiated at the apical pole, in the region of the type III InsP3 receptor. Acetylcholine occasionally induced sustained or repetitive Ca21 increases that were prominent at the basal pole, in the region of the type I InsP3 receptor, but only subtle or absent apically. CONCLUSIONS. Because the type I InsP3 receptor is thought to be responsible for repetitive Ca21 release events, and the type III InsP3 receptor instead is suited to initiate Ca21 signals, the subcellular distribution of these two isoforms corresponds to the Ca21 signaling patterns observed in this cell type. Differential subcellular expression of InsP3 receptor isoforms may be an important molecular mechanism by which NPE cells organize their Ca21 signals in space and time. (Invest Ophthalmol Vis Sci. 1999;40:2046 –2053) patial and temporal patterns of cytosolic Ca21 (Cai21) signals are highly organized in many cell types and play an important role in regulating cell function.1,2 For example, spatial patterns of Cai21 signals, such as Cai21 waves and gradients, direct functions such as secretion3,4 and cell migration,5 and temporal Cai21 signaling patterns such as oscillations direct functions such as gene expression.6,7 The molecular basis for the subcellular organization of Cai21 signals is not completely understood, though. Cai21 signaling in epithelia and other nonexcitable cells generally is mediated by Ca21 release via the inositol 1,4,5-

S

From the Departments of 1Ophthalmology and Visual Sciences, Medicine and Cell Biology, Yale University School of Medicine, New Haven, Connecticut. Supported by NIH Grants EY-08879, EY-00785, and DK-45710; the E. Matilda Ziegler Foundation; an Established Investigator Grant from the American Heart Association; a Pilot and Feasibility Grant from the Cystic Fibrosis Foundation; and the Morphology Core Facility of the Yale Liver Center (NIH DK-34989). Submitted for publication January 21, 1999; revised April 9, 1999; accepted April 28, 1999. Proprietary interest category: N. Reprint requests: Marvin L. Sears, Department of Ophthalmology and Visual Science, Yale University School of Medicine, 330 Cedar Street, New Haven, CT 06520-8061. 2

2046

trisphosphate (InsP3) receptor.1,2 Three isoforms of this receptor have been identified, and many cell types express more than one of these isoforms.8 –11 The function of the type I InsP3 receptor has been characterized in greatest detail12; the receptor functions as a Ca21 channel in the presence of InsP3, but the open probability of that channel exhibits a bell-shaped dependence on the concentration of Cai21.13,14 This Ca21 dependence of the type I InsP3 receptor is thought to be important for the formation of certain types of Cai21 signaling patterns, such as Cai21 oscillations.15,16 The type III InsP3 receptor also is an InsP3-gated Ca21 channel17; however, unlike the type I receptor, the type III receptor functions purely as a positive feedback Ca21 channel.17 It has been proposed that this characteristic of the type III InsP3 receptor would enable this isoform to act preferentially as a trigger for Ca21 release.17 This difference in dependence on Cai21 thus suggests that the subcellular distribution of these two isoforms could provide a mechanism by which subcellular Cai21 signals are organized. The goal of the present study was to examine the relationship between subcellular Cai21 signaling patterns and the subcellular distribution of these two InsP3 receptor isoforms in one type of polarized epithelial cell, the nonpigmented epithelium (NPE) of the ciliary epithelial bilayer of the eye. Investigative Ophthalmology & Visual Science, August 1999, Vol. 40, No. 9 Copyright © Association for Research in Vision and Ophthalmology

IOVS, August 1999, Vol. 40, No. 9

MATERIALS

AND

Ca21 Waves and InsP3 Receptor Isoforms in NPE Cells

METHODS

Animals and Materials Male albino New Zealand rabbits weighing 2 to 3 kg obtained from Millbrook Farms (Amherst, MA) were used for all experiments. Acetylcholine (ACh), atropine, U73122, heparin (average molecular weight 6000), de-N-sulfated heparin, trypsin, and sulforhodamine 101 (Texas red) were obtained from Sigma Chemical (St. Louis, MO). Fluo-3 in acetoxy-methoxylated (AM) form and Pluronic F-127 were obtained from Molecular Probes (Eugene, OR). All other chemicals were of the highest quality commercially available.

Antibodies The type I InsP3 receptor was labeled using affinity-purified rabbit polyclonal antibody T210, directed against the 19 Cterminal amino acids of the mouse type I InsP3 receptor.18,19 The type II InsP3 receptor was labeled using affinity-purified rabbit polyclonal antibody CT2 directed against the C-terminal portion of the rat type II InsP3 receptor.9 The type III InsP3 receptor was labeled using a commercially obtained mouse monoclonal antibody directed against the N-terminal portion of the human type III InsP3 receptor (mab InsP3R-3; Transduction Laboratories, Lexington, KY).17,20 The M3 muscarinic ACh receptor was labeled using a commercially obtained mouse monoclonal antibody21 raised against affinity-purified calf forebrain receptor (M35; Argene, North Massapequa, NY).

Preparation of Isolated Ciliary Epithelium Isolated ciliary bilayer epithelium was prepared as described previously,22,23 with slight modification. Briefly, rabbits were anesthetized with an intramuscular injection of ketamine hydrochloride and xylazine, then euthanatized by intravenous injection of pentobarbital sodium and phenytoin sodium. The eyes were enucleated promptly, then the anterior segments were isolated after careful removal of the lens. From the isolated anterior segment of the eye, ciliary processes were separated from the iris and cut into 10 to 20 strips, each 2 to 3 mm in length. In selected experiments, the NPE layer was mechanically separated from the bilayer,24 then individual NPE cells were obtained by trypsin digestion of the monolayer in EDTA as described previously,25 maintained in HEPES-buffered M199 solution (Sigma) containing 10% fetal calf serum, and examined in short-term culture. All procedures conformed with NIH recommendations and the ARVO Statement on the Use of Animals in Ophthalmic and Vision Research.

Confocal Immunofluorescence Histochemistry Sections of rabbit ciliary epithelia were labeled with isoformspecific antibodies to determine the subcellular distributions of the types I, II, and III InsP3 receptors. Specimens were colabeled with rhodamine-phalloidin (Molecular Probes) because this stain facilitates identification of the apical and basolateral poles of epithelial cells.26,27 Additional sections were labeled with the M3 muscarinic ACh receptor antibody M35 plus rhodamine phalloidin, to determine the distribution of that receptor on NPE cells. Immunochemistry was performed on 4-mm-thick frozen sections of rabbit ciliary epithelia. Tissue was fixed by perfusion with 4% paraformaldehyde in 0.12 M sodium phosphate buffer (pH 7.4), cryopreserved overnight in 15% sucrose, and

2047

frozen in isopentane/liquid nitrogen. After quenching with 50 mM NH4Cl and 16% goat serum in phosphate-buffered saline with Triton X-100, the sections were labeled overnight with a 1:10 dilution of antibody T210, a 1:250 dilution of antibody CT2, a 1:50 dilution of InsP3R-3 antibody, or a 1:200 dilution of antibody M35, then washed and incubated with either fluorescein isothiocyanate (FITC)– conjugated (Sigma) or Alexa 488 – conjugated (Molecular Probes) anti–mouse or anti–rabbit secondary antibody, along with rhodamine– conjugated phalloidin.26 Negative controls were labeled with preimmune serum rather than with anti–InsP3 receptor antibodies but were otherwise processed as noted above. Specimens were examined with a Bio–Rad MRC-600 Confocal Microscope equipped with a krypton/argon mixed gas laser (Richmond, CA). To ensure specificity of InsP3 receptor staining, images were obtained using confocal machine settings (i.e., aperture, gain, and black level) at which no fluorescence was detectable in negative control samples labeled with preimmune serum. Doublelabeled specimens were serially excited at 488 nm and observed at .515 nm to detect FITC, then excited at 568 nm and observed at .585 nm to detect rhodamine. This approach eliminated bleed-through of FITC fluorescence into the rhodamine channel.28

Confocal Microscopic Measurements of Cytosolic Ca21 Isolated ciliary epithelial bilayers were prepared as described above, then loaded with fluo-3/AM (50 mM) and Pluronic F-127 for 1 hour at room temperature in Hanks’ balanced salt solution containing 10% fetal calf serum. Specimens were then placed between two glass coverslips in a gravity-driven perifusion chamber on the stage of a Zeiss Axiovert microscope (Thornwood, NY), and perifused at room temperature at a rate of 1 to 2 ml/min. Nonpigmented epithelial cells within the tissue were observed through a 363 1.40 NA objective using either a Bio–Rad MRC-600 or a Bio–Rad MRC-1024 laser scanning confocal imaging system. An argon laser was used to excite the dye at 488 nm, and emission signals above 515 nm were collected. Optical sections 1 to 2 mm in thickness were obtained. Neither autofluorescence nor other background signals were detectable at the machine settings used, and there was no change in size, shape, or location of cells during the experiments. In most experiments, two-dimensional images consisting of 768 3 512 pixels (0.26 mm/pixel) were recorded at a rate of 1 frame/s on an optical disc recorder and analyzed subsequently, using the mean pixel values of preselected areas to monitor intensity changes. Increases in Cai21 were expressed as (F/F0) 3 100%.22,29 In selected experiments, tissues instead were examined using the line scanning mode of the confocal microscope, to increase temporal resolution (to 10 or 200 msec). In this mode, fluorescence is determined at each point along a single line across the image, rather than at each point across the entire image.22,30 Line scans were displayed as images consisting of 768 3 512 pixels, with a spatial resolution of 0.26 mm/pixel (in the “x” direction) and a temporal resolution of 10 or 200 msec/pixel (in the “y” direction). Velocities of Cai21 waves in individual cells were determined from the rate at which initial fluorescence increases moved along the scan line.30

2048

Hirata et al.


FIGURE 1. Subcellular localization of the types I, II, and III InsP3 receptors in NPE cells, visualized by confocal immunofluorescence histochemistry. Ciliary epithelial bilayers in this figure are oriented so that the top layer of cells is the NPE and the bottom layer is the pigmented epithelium (PE). The apical membranes of the NPE and PE are in contact. (A) Rhodamine– conjugated phalloidin labeling. Scale bar, 10 mm. (B) Same tissue segment, with type I InsP3 receptor labeling. (C) Superimposure of (A) and (B) shows that the type I InsP3 receptor is concentrated at the basal pole of the NPE, as well as at the basal pole of the PE. (D) Negative control for the type I InsP3 receptor, labeled with preimmune serum and counterstained with FITC-conjugated anti-rabbit secondary antibody (green) plus rhodamine phalloidin (red). (E) A separate tissue section labeled with rhodamine-conjugated phalloidin. (F) Same tissue segment, with type III InsP3 receptor labeling. (G) Superimposure of (E) and (F) shows the type III InsP3 receptor is concentrated at the apical pole. (H) Negative control for the type III InsP3 receptor. (I) A separate tissue section labeled with rhodamine-conjugated phalloidin. (J) Same tissue segment, with type II InsP3 receptor labeling. No (green) labeling is seen. (K) Superimposure of (I) and (J). (L) Negative control for the type II InsP3 receptor.

Microinjection Studies In selected studies, individual isolated NPE cells were stimulated with ACh to confirm their responsiveness to this agonist, then either heparin (1 mg/ml) or de-N-sulfated heparin (1 mg/ml) was delivered into the cells by microinjection, and the cells were restimulated with ACh. Cells were loaded with fluo-3/AM, then examined by confocal video microscopy as described above. Micropipettes with an internal diameter of ,0.5 mm were made from glass capillary tubes using a Narishige PD-5 micropipette puller. A series 5171 Eppendorf micromanipulator was used for positioning, and an Eppendorf series 5242 microinjector was used for pressure-microinjections.31 Micropipettes were loaded with heparin or its de-N-sulfated analogue dissolved in an intracellular-like buffer (150 mM KCl plus 1 mM HEPES), and Texas red was coinjected as a marker of successful microinjection.31

InsP3 Measurement Segments of ciliary epithelia were isolated as described above, then equilibrated for 15 minutes in Ringer’s solution consisting of NaCl (120 mM), KCl (2.8 mM), CaCl2 (2 mM), MgCl2 (2 mM), HEPES (10 mM), and glucose (10 mM). Either ACh (10 mM) or buffer was added for 2, 5, or 10 seconds, then stimulation was stopped by adding 20% perchloric acid and placing the samples on ice. The acid extracts were centrifuged, then the supernatants were removed and neutralized with KOH, HEPES, and EDTA, and InsP3 was measured in the neutralized extracts

using a radiobinding assay (Amersham). The pellets were solubilized in NaOH for protein determination using a BCA-Protein Assay kit (Pierce). Results were expressed as picomoles of InsP3 per milligram of protein.

RESULTS

AND

DISCUSSION

Localization of InsP3 Receptor Isoforms in NPE Cells The subcellular distribution of the types I, II, and III InsP3 receptors was investigated by confocal immunofluorescence histochemistry. Ciliary epithelial bilayers were labeled with either antibody T210, CT2, or mab InsP3R-3 and colabeled with rhodamine– conjugated phalloidin to identify the apical and basolateral margins of the NPE cells (Fig. 1). T210 labeling was limited to the basolateral pole of NPE cells and was found on the basolateral pole of pigmented epithelial cells as well (Figs. 1B, 1C). In contrast, the type III InsP3 receptor antibody labeled the apical pole of NPE cells (Figs. 1F, 1G). No such basal or apical labeling was seen in the NPE in tissue stained instead with preimmune serum (Figs. 1D, 1H). Unlike antibody T210 or mab InsP3R-3, antibody CT2 did not label NPE cells (Figs. 1J, 1K), even though this antibody has been used by others20,32 and by us (unpublished observation) to label the type II InsP3 receptor in other epithelia. Thus, like other cell types,8,9,11 including other epithelia,10,20,32 multiple InsP3 re-



2049

FIGURE 2. The phospholipase C inhibitor U73122 (10 mM) inhibits ACh (10 mM)-induced Cai21 signals in NPE cells. NPE cells within ciliary bilayers were monitored using timelapse confocal microscopy as they were sequentially stimulated with ACh, then ACh 1 U73122, and then ACh again. Result is representative of that seen in four separate groups of NPE cells from three separate experimental preparations.

ceptor isoforms are expressed in NPE cells. In particular, our findings suggest that NPE cells express types I and III but not the type II InsP3 receptor. Apical localization of the InsP3 receptor, especially the type III isoform, has also been shown in other epithelia, including pancreatic and salivary acinar cells.20,28,32 However, in those epithelia the type I and type II isoforms are predominantly expressed in the apical region as well.20,32 The finding that type I and type III InsP3 receptors are concentrated in different regions of the NPE cell, whereas the type II receptor is minimally expressed, thus suggests that this may provide a novel system in which to compare the function of InsP3R-I and InsP3R-III when the two are coexpressed in a single cell.

ACh-Induced Ca21 Signaling in NPE cells Is Mediated by InsP3 To determine the relationship between the distribution of InsP3 receptor isoforms and Cai21 signaling patterns, we tried to identify an agonist that increases Cai21 via InsP3 in NPE cells. Acetylcholine increases Cai21 in NPE cells,22,33–35 so we examined whether this increase is mediated by InsP3. Acetylcholine (10 mM) increased fluo-3 fluorescence by 175% 6 25% (mean 6 SEM) in these cells but by only 10% 6 1% when cells were stimulated in the presence of 10 mM atropine (n 5 10 experiments; P , 0.0001 by paired t-test). In separate studies, ACh increased fluo-3 fluorescence by 126% 6 18% in the presence of 1.26 mM extracellular Ca21 and by 110% 6 14% in Ca21-free medium (n 5 10 experiments; P . 0.05). These findings demonstrate that ACh increases Cai21 in NPE cells via stimulation of muscarinic receptors, leading to release of Ca21 from intracellular stores. It has previously been shown that carbachol stimulates production of inositol polyphosphates, including InsP3, in NPE cells,36 so we examined the time course of InsP3 production. Acetylcholine (10 mM) induced a net

increase of 0.3, 6.8, and 9.9 pmol InsP3/mg protein after 2, 5, and 10 seconds of stimulation, respectively. These values correspond to increases of 1%, 20%, and 32% relative to InsP3 content of unstimulated controls. To demonstrate a causal link between ACh-induced InsP3 production and Cai21 signaling in NPE cells, we examined the effects of the phospholipase C inhibitor U73122.37 Ciliary epithelial bilayers were sequentially stimulated, first with ACh (10 mM), then with ACh 1 U73122 (10 mM), and then with ACh again. Fluo-3 fluorescence was monitored in groups of at least 10 adjacent NPE cells, and we found that the ACh-induced increase in Cai21 was reversibly inhibited by U73122 (Fig. 2). To investigate whether this Ca21 release is mediated by activation of the InsP3 receptor, cells were microinjected with either heparin (1 mg/ml), which is a high-affinity competitive antagonist for the InsP3 receptor,38 or de-N-sulfated heparin (1 mg/ml), which neither inhibits InsP3 binding to its receptor nor blocks InsP3-induced Ca21 release from microsomes.39 As an extra control, only cells that responded to ACh were subsequently injected with heparin or its deN-sulfated analogue, then each of those cells were restimulated with ACh after microinjection. Ten of 11 cells did not respond to ACh after injection with heparin (Figs. 3A, 3B, 3C); fluorescence increased by 125% 6 30% in these cells when stimulated before heparin injection, but by only 9% 6 1% after injection (P , 0.005 by paired t-test). In contrast, 6 of 7 cells responded to ACh after injection with de-N-sulfated heparin (Fig. 3D); fluorescence increased by 108% 6 31% in these cells when stimulated before injection, and by 102% 6 31% after injection (P 5 0.31). Taken together, these studies demonstrate that ACh increases Cai21 in NPE cells by stimulation of muscarinic receptors, which then leads to phospholipase C–mediated mobilization of intracellular Ca21 stores by activation of InsP3 receptors.

2050

Hirata et al.


Subcellular Organization of Ca21 Signals in NPE Cells

FIGURE 3. Heparin but not de-N-sulfated heparin blocks ACh-induced Cai21 signals in isolated NPE cells, as revealed by double-channel time-lapse confocal microscopy. (A) Confocal image of an isolated NPE cell microinjected with heparin (1 mg/ml), plus free Texas red as a marker of successful injection. (B) Simultaneous image of the same cell (arrow) and its neighbor obtained before stimulation with ACh, which shows loading of both cells with the Ca21 dye fluo-3. (C) Subsequent to stimulation with ACh, an increase in fluo-3 fluorescence is seen in a neighboring cell but not in the cell microinjected with heparin. (D) Acetylcholine-induced increases in fluo-3 fluorescence are blocked in cells microinjected with heparin (n 5 10) but not in cells microinjected with de-N-sulfated heparin (n 5 6). Values are mean 6 SEM (*P , 0.005).

To determine the polarity of muscarinic receptors on NPE cells, ciliary epithelial bilayers were labeled by confocal immunofluorescence histochemistry. Ciliary bilayers were labeled with monoclonal antibody M35 directed against the M3 subtype of the muscarinic receptor, because this subtype often links to InsP3-mediated Cai21 signaling in epithelia,21,40 and because previous pharmacological studies suggest this subtype is present on NPE cells.36 Tissue specimens were colabeled with rhodamine– conjugated phalloidin to identify the apical and basolateral margins of the NPE cells (Fig. 4). M35 labeling was limited to the basal pole of NPE cells. The M3 receptor directly couples to G proteins that activate phospholipase C–b,1,41 which suggests that stimulation with ACh would preferentially generate InsP3 in the basolateral region rather than apically.

To observe the subcellular organization of ACh-induced Cai21 signals, NPE cells within intact ciliary epithelial bilayers were examined using confocal line scanning microscopy.22,30 This approach permitted examination of NPE cells in a system in which their structural polarity was maintained,22 and in which spatial and temporal resolutions were maximized while photobleaching was minimized.42 To determine the site of initiation of ACh-induced Cai21 signals, images were collected every 10 msec. Cai21 signals always began in the apical region, then traveled as a wave from the apical to the basal pole in each of 6 NPE cells (Fig. 5). The wave speed was no different in Ca21-containing versus Ca21-free medium (23.2 6 1.6 versus 24.4 6 1.9 mm/sec, respectively; P 5 0.60). These findings demonstrate that Cai21 waves are initiated in the apical region of NPE cells, then propagate to the basal region purely via release of Ca21 from intracellular stores. This polarized apicalto-basal pattern of Cai21 wave propagation is similar to the pattern observed in other epithelia, including pancreatic,30,43– 45 lacrimal,46 and salivary20 acinar cells and hepatocytes.26 InsP3 receptor isoforms have been localized in both pancreatic and salivary acinar cells, and in each of these types of acinar cell, each type of isoform present is concentrated apically.20,32 Therefore, from these previous studies it has not been possible to determine whether one of these isoforms would preferentially behave as a trigger for Ca21 release. Because type I and type III InsP3 receptors are spatially separated in NPE cells, these cells provide a novel system in which to investigate this question. Although the type I InsP3 receptor is in the same region as the M3 ACh receptor, where increases in InsP3 likely originate, Cai21 signals nonetheless began in the region of the type III receptor instead. This finding suggests that the type III isoform may have a much lower threshold than the type I isoform for InsP3-mediated Ca21 release. This similarly suggests that the type III InsP3 receptor serves to initiate Cai21 signals in cells that coexpress the type I and type III isoforms. Furthermore, this finding supports the hypothesis that the role of the type III InsP3 receptor is to act as a trigger for cellular Ca21 release.17 To observe subcellular Cai21 signaling patterns that occur after the initial Cai21 wave, NPE cells were examined for 100 seconds rather than 5 seconds. Confocal line scanning microscopy was used here as well, but line scans were collected at a frequency of 200 msec rather than 10 msec. At the lowest ACh

FIGURE 4. Subcellular localization of the M3 muscarinic ACh receptor in NPE cells, visualized by confocal immunofluorescence histochemistry. (A) The ciliary epithelial bilayer of the eye, labeled with rhodamine-conjugated phalloidin. The NPE (top layer) and PE (bottom layer) are oriented so that their apical membranes are in contact. Scale bar, 10 mm. (B) Same tissue segment, labeled with antibody M35 directed against the M3 muscarinic receptor. (C) Superimposure of (A) and (B), revealing that the M3 receptor is concentrated at the basal pole of the NPE. (D) Negative control for the M3 receptor, stained only with Alexa 488 – conjugated anti-rabbit secondary antibody (green) plus rhodamine phalloidin (red). No nonspecific antibody labeling is observed.



FIGURE 5. Acetylcholine-induced increases in Cai21 begin as apicalto-basal Cai21 waves in NPE cells. (A) Confocal image of a segment of the isolated ciliary bilayer loaded with fluo-3. The confocal line scan in (B) was performed along the white horizontal line across this image, which runs along the apical-to-basal pole of an NPE cell. Pseudocolor scale is shown at bottom. (B) Line scan collected during stimulation with 10 mM ACh. Fluorescence intensity along the x axis reflects distance (across the scan line) and along the y axis reflects time (between serial scans). Line scans were obtained every 10 msec for a total of 5.12 seconds ( from top to bottom). The increase in fluorescence begins apically, then spreads to the opposite (basal) pole. Results are representative of those seen in 6 preparations. (C) Graphical representation of the fluorescence intensity over time at an apical and a basal point in the NPE cell that was scanned. The increases in Cai21 (arrows) within the NPE cell occur 300 msec apart, and the two points are separated by a distance of 6.11 mm, which corresponds to a wave speed of 20.4 mm/s.

2051

concentration perifused (0.1 mM), an increase in Cai21 was detected in only 23% (n 5 6 of 26) of NPE cells. In contrast, a Cai21 increase was detected in 66% (n 5 65 of 98) of cells stimulated with higher ACh concentrations (0.5, 1, 5, or 10 mM). Repetitive Cai21 spikes, persistent Cai21 gradients across the cytosol, or both were detected in 36 of the 71 cells that responded to ACh (Fig. 6). Among these cells, repetitive Cai21 increases were either of greater amplitude, more sustained, or only present in the basolateral region, whereas persistent Cai21 gradients were manifested as prolonged increases in Cai21 in the basolateral region relative to the apical region (Fig. 6). Localized increases in Cai21, including localized Cai21 oscillations, also occur in pancreatic acinar cells,44,45 but those Cai21 increases are restricted to a region in which both the type I and type III InsP3 receptors are expressed.28,45,47 Localized Cai21 increases have been reported in the presynaptic region of neurons48 and in the subplasmalemma of neuroendocrine cells3 as well, but those increases are thought to occur by Ca21 influx rather than localized release of intracellular Ca21 stores. The current work provides evidence that localized persistent or repetitive increases in Cai21 may be driven preferentially by Ca21 released from the type I rather than the type III InsP3 receptor. This differential signaling pattern by distinct Ca21 storage pools supports the hypothesis that the role of the type III InsP3 receptor is to initiate cellular Cai21 signals,17 whereas the type I InsP3 receptor instead drives Cai21 oscillations and other longer-term Cai21 signaling patterns. Although these findings suggest specific and complementary roles for the types I and III InsP3 receptors, whether the type II InsP3 receptor also plays a distinctive role in Cai21 signaling is not addressed here. In B cells that normally coexpress all three InsP3 receptor isoforms, B-cell receptor stimulation results in InsP3-mediated Cai21 signaling when expression of one or even two of the isoforms is disrupted.11 This finding suggests that each isoform may provide a redundant Cai21 signaling mechanism in this cell type.11 In B cells that have been genetically engineered to express only a single isoform of the InsP3 receptor, Cai21 signaling patterns are different for each isoform.49 In addition, the function of the

FIGURE 6. Examples of distinct apical and basolateral Cai21 signaling patterns in NPE cells. Cells were stimulated with ACh while examined by confocal line scanning microscopy, using a collection rate of 200 msec per line (note that apical and basal signals appear to begin simultaneously because of the expanded time scale). (A) Periodic Cai21 spikes with a frequency of ;0.1 s21 are seen in the basolateral but not the apical region. (B) An increase in Cai21 persists for .1 minute in the basolateral region, but Cai21 is elevated apically for ,20 seconds. (C) There is a sustained ;45% increase in fluo-3 fluorescence, with superimposed Cai21 spikes (frequency, ;0.1 second-1) in the basolateral region, whereas apically there is only a ;15% increase with no superimposed Cai21 spikes.

2052

Hirata et al.

type II receptor has recently been described at the single channel level, and it differs from the function of the type I InsP3 receptor.50 Together, these findings suggest that each InsP3 receptor isoform contributes to cellular Cai21 signaling but in a unique way. Although several types of epithelium express all three InsP3 receptor isoforms,10,20,32 colocalization of the various isoforms in those cell types had made it difficult to determine their relative contribution to Cai21 signaling in epithelia until now. What is the functional significance of the current findings? The ability to generate Cai21 gradients, waves, and oscillations may be critical for secretion to occur in polarized epithelia. For example, apical increases in Cai21 direct exocytosis,4,51 because localized intense increases in Cai21 in the apical region induce targeting of vesicles to the apical membrane.4 Apical increases in Cai21 also can direct the movement of subapical actin, which may mechanically facilitate secretion.26,52–54 Apical-to-basal Cai21 waves direct vectorial movement of electrolytes such as Cl2 and Na1.4,43 Finally, repetitive increases in Cai21 (i.e., Cai21 oscillations) direct repetitive membrane fusion and exocytic events.51,55 The current work provides evidence that in cells coexpressing the type I and type III InsP3 receptors, the type III receptor is responsible for initiating Cai21 signals, whereas repetitive or sustained increases in Cai21 may instead be driven by the type I receptor. The NPE is unusual among epithelia because it transports fluid and electrolytes from the apical to the basal pole, and secretion occurs basolaterally rather than apically.24,25,56 Therefore, it may be preferable for NPE cells to generate sustained or repetitive Cai21 signals basolaterally rather than apically. Although it can be speculated that the subcellular distribution of InsP3 receptor isoforms organizes subcellular Cai21 signaling patterns in all cells, the novel functional requirements of NPE cells may provide a unique cell model in which to investigate this hypothesis.

Acknowledgments The authors thank Barbara E. Ehrlich for useful discussions. We also thank Alden Mead for help with isolation of ciliary epithelial bilayers, Pietro DeCamilli and Kohji Takei for generously providing InsP3R-1 antibody T210, and Richard Wojcikiewicz for generously providing InsP3R-2 antibody CT2.

References 1. Berridge MJ. Inositol trisphosphate and calcium signalling. Nature. 1993;361:315–325. 2. Clapham DE. Calcium signaling. Cell. 1995;80:259 –268. 3. Cheek TR, Jackson TR, O’Sullivan AJ, Moreton RB, Berridge MJ, Burgoyne RD. Simultaneous measurements of cytosolic calcium and secretion in single bovine adrenal chromaffin cells by fluorescent imaging of fura-2 in cocultured cells. J Cell Biol. 1989;109: 1219 –1227. 4. Ito K, Miyashita Y, Kasai H. Micromolar and submicromolar Ca21 spikes regulating distinct cellular functions in pancreatic acinar cells. EMBO J. 1997;16:242–251. 5. Hahn K, DeBiasio R, Taylor DL. Patterns of elevated free calcium and calmodulin activation in living cells. Nature. 1992;359:736 – 738. 6. Dolmetsch RE, Xu K, Lewis RS. Calcium oscillations increase the efficiency and specificity of gene expression. Nature. 1998;392: 933–936. 7. Li W, Llopis J, Whitney M, Zlokarnik G, Tsien RY. Cell-permeant caged InsP3 ester shows that Ca21 spike frequency can optimize gene expression. Nature. 1998;392:936 –941.

IOVS, August 1999, Vol. 40, No. 9 8. Newton CL, Mignery GA, Su ¨ dhof TC. Co-expression in vertebrate tissues and cell lines of multiple inositol 1,4,5-trisphosphate (InsP3) receptors with distinct affinities for InsP3. J Biol Chem. 1994;269:28613–28619. 9. Wojcikiewicz RJH. Type I, II, and III inositol 1,4,5-trisphosphate receptors are unequally susceptible to down-regulation and are expressed in markedly different proportions in different cell types. J Biol Chem. 1995;270:11678 –11683. 10. Sugiyama T, Yamamoto–Hino M, Wasano K, Mikoshiba K, Hasegawa M. Subtype-specific expression patterns of inositol 1,4,5trisphosphate receptors in rat airway epithelial cells. J Histochem Cytochem. 1996;44:1237–1242. 11. Sugawara H, Kurosaki M, Takata M, Kurosaki T. Genetic evidence for involvement of type 1, type 2 and type 3 inositol 1,4,5-trisphosphate receptors in signal transduction through the B-cell antigen receptor. EMBO J. 1997;16:3078 –3088. 12. Mignery GA, Newton CL, Archer BT, Sudhof TC. Structure and expression of the rat inositol 1,4,5-trisphosphate receptor. J Biol Chem. 1990;265:12679 –12685. 13. Bezprozvanny I, Watras J, Ehrlich BE. Bell-shaped calcium-response curves of Ins(1,4,5)P3- and calcium-gated channels from endoplasmic reticulum of cerebellum. Nature. 1991;351:751–754. 14. Finch EA, Turner TJ, Goldin SM. Calcium as a coagonist of inositol 1,4,5-trisphosphate-induced calcium release. Science. 1991;252: 443– 446. 15. Harootunian AT, Kao JPY, Paranjape S, Tsien RY. Generation of calcium oscillations in fibroblasts by positive feedback between calcium and IP3. Science. 1991;251:75–78. 16. Dupont G, Goldbeter A. One-pool model for Ca21 oscillations involving Ca21 and inositol 1,4,5-trisphosphate as co-agonists for Ca21 release. Cell Calcium. 1993;14:311–322. 17. Hagar RE, Burgstahler AD, Nathanson MH, Ehrlich BE. Type III InsP3 receptor channel stays open in the presence of increased calcium. Nature. 1998;396:81– 84. 18. Takei K, Stukenbrok H, Metcalf A, et al. Ca21 stores in Purkinje neurons: endoplasmic reticulum subcompartments demonstrated by the heterogeneous distribution of the InsP3 receptor, Ca21ATPase, and calsequestrin. J Neurosci. 1992;12:489 –505. 19. Mignery GA, Sudhof TC, Takei K, De Camilli P. Putative receptor for inositol 1,4,5-trisphosphate similar to ryanodine receptor. Nature. 1989;342:192–195. 20. Lee MG, Xu X, Zeng WZ, et al. Polarized expression of Ca21 channels in pancreatic and salivary gland cells : correlation with initiation and propagation of [Ca21]i waves. J Biol Chem. 1997; 272:15765–15770. 21. Alvaro D, Alpini G, Jezequel AM, et al. Role and mechanisms of action of acetylcholine in the regulation of rat cholangiocyte secretory function. J Clin Invest. 1997;100:1349 –1362. 22. Hirata K, Nathanson MH, Sears ML. Novel paracrine signaling mechanism in the ocular ciliary epithelium. Proc Natl Acad Sci USA. 1998;95:8381– 8386. 23. Sears ML, Yamada E, Cummins D, Mori N, Mead A, Murakami M. The isolated ciliary bilayer is useful for studies of aqueous humor formation. Trans Am Ophthalmol Soc. 1991;89:131–152. 24. Chen S, Sears M. A low conductance chloride channel in the basolateral membranes of the non-pigmented ciliary epithelium of the rabbit eye. Curr Eye Res. 1997;16:710 –718. 25. Edelman JL, Sachs G, Adorante JS. Ion transport asymmetry and functional coupling in bovine pigmented and nonpigmented ciliary epithelial cells. Am J Physiol Cell Physiol. 1994;266:C1210 – C1221. 26. Nathanson MH, Burgstahler AD, Fallon MB. Multi-step mechanism of polarized Ca21 wave patterns in hepatocytes. Am J Physiol Gastrointest Liver Physiol. 1994;267:G338 –G349. 27. Fallon MB, Gorelick FS, Anderson JM, Mennone A, Saluja A, Steer ML. Effect of cerulein hyperstimulation on the paracellular barrier of rat exocrine pancreas. Gastroenterology. 1995;108: 1863–1872. 28. Nathanson MH, Fallon MB, Padfield PJ, Maranto AR. Localization of the type 3 inositol 1,4,5-trisphosphate receptor in the Ca21 wave trigger zone of pancreatic acinar cells. J Biol Chem. 1994;269: 4693– 4696.



29. Schlosser SF, Burgstahler AD, Nathanson MH. Isolated rat hepatocytes can signal to other hepatocytes and bile duct cells by release of nucleotides. Proc Natl Acad Sci USA. 1996;93:9948 – 9953. 30. Nathanson MH, Padfield PJ, O’Sullivan AJ, Burgstahler AD, Jamieson JD. Mechanism of Ca21 wave propagation in pancreatic acinar cells. J Biol Chem. 1992;267:18118 –18121. 31. Nathanson MH, Burgstahler AD, Mennone A, Boyer JL. Characterization of cytosolic Ca21 signaling in rat bile duct epithelia. Am J Physiol Gastrointest Liver Physiol. 1996;271:G86 –G96. 32. Yule DI, Ernst SA, Ohnishi H, Wojcikiewicz RJH. Evidence that zymogen granules are not a physiologically relevant calcium pool: defining the distribution of inositol 1,4,5-trisphosphate receptors in pancreatic acinar cells. J Biol Chem. 1997;272: 9093–9098. 33. Farahbakhsh NA, Cilluffo MC. Synergistic effect of adrenergic and muscarinic receptor activation on [Ca21]i in rabbit ciliary body epithelium. J Physiol (Lond). 1994;477:215–221. 34. Schutte M, Diadori A, Wang C, Wolosin JM. Comparative adrenocholinergic control of intracellular Ca21 in the layers of the ciliary body epithelium. Invest Ophthalmol Vis Sci. 1996;37:212–220. 35. Ohuchi T, Yoshimura N, Tanihara H, Kuriyama S, Ito S, Honda Y. Ca21 mobilization in nontransformed ciliary nonpigmented epithelial cells. Invest Ophthalmol Vis Sci. 1992;33:1696 –1705. 36. Wax MB, Coca–Prados M. Receptor-mediated phosphoinositide hydrolysis in human ocular ciliary epithelial cells. Invest Ophthalmol Vis Sci. 1989;30:1675–1679. 37. Yule DI, Williams JA. U73122 inhibits Ca21 oscillations in response to cholecystokinin and carbachol but not to JMV-180 in rat pancreatic acinar cells. J Biol Chem. 1992;267:13830 –13835. 38. Ghosh TK, Eis PS, Mullaney JM, Ebert CL, Gill DL. Competitive, reversible, and potent antagonism of inositol 1,4,5-trisphosphateactivated calcium release by heparin. J Biol Chem. 1988;263: 11075–11079. 39. Tones MA, Bootman MD, Higgins BF, Lane DA, Pay GF, Lindahl U. The effect of heparin on the inositol 1,4,5-trisphosphate receptor in rat liver microsomes. FEBS Lett. 1989;252:105–108. 40. Caulfield MP. Muscarinic receptors: characterization, coupling and function. Pharmacol Ther. 1993;58:319 –379. 41. Wu D, Jiang H, Katz A, Simon MI. Identification of critical regions on phospholipase C-b1 required for activation by G-proteins. J Biol Chem. 1993;268:3704 –3709. 42. Nathanson MH, Burgstahler AD. Subcellular distribution of cytosolic Ca21 in isolated rat hepatocyte couplets: evaluation using confocal microscopy. Cell Calcium. 1992;13:89 –98.

2053

43. Kasai H, Augustine GJ. Cytosolic Ca21 gradients triggering unidirectional fluid secretion from exocrine pancreas. Nature. 1990; 348:735–738. 44. Thorn P, Lawrie AM, Smith PM, Gallacher DV, Petersen OH. Local and global cytosolic Ca21 oscillations in exocrine cells evoked by agonists and inositol trisphosphate. Cell. 1993;74:661– 668. 45. Kasai H, Li YX, Miyashita Y. Subcellular distribution of Ca21 release channels underlying Ca21 waves and oscillations in exocrine pancreas. Cell. 1993;74:669 – 677. 46. Toescu EC, Lawrie AM, Petersen OH, Gallacher DV. Spatial and temporal distribution of agonist-evoked cytoplasmic Ca21 signals in exocrine acinar cells analysed by digital image microscopy. EMBO J. 1992;11:1623–1629. 47. Thorn P, Moreton R, Berridge M. Multiple, coordinated Ca21release events underlie the inositol trisphosphate-induced local Ca21 spikes in mouse pancreatic acinar cells. EMBO J. 1996;15: 999 –1003. 48. Llinas R, Sugimori M, Silver RB. Microdomains of high calcium concentration in a presynaptic terminal. Science. 1992;256:677– 679. 49. Miyakawa T, Maeda A, Yamazawa T, Hirose K, Kurosaki T, Iino M. Encoding of Ca21 signals by differential expression of IP3 receptor subtypes. EMBO J. 1999;18:1303–1308. 50. Ramos–Franco J, Fill M, Mignery GA. Isoform-specific function of single inositol 1,4,5-trisphosphate receptor channels. Biophys J. 1998;75:834 – 839. 51. Maruyama Y, Inooka G, Li YX, Miyashita Y, Kasai H. Agonistinduced localized Ca21 spikes directly triggering exocytotic secretion in exocrine pancreas. EMBO J. 1993;12:3017–3022. 52. Watanabe S, Smith CR, Phillips MJ. Coordination of the contractile activity of bile canaliculi: evidence from calcium microinjection of triplet hepatocytes. Lab Invest. 1985;53:275–279. 53. Watanabe N, Tsukada N, Smith CR, Phillips MJ. Motility of bile canaliculi in the living animal: implications for bile flow. J Cell Biol. 1991;113:1069 –1080. 54. Nathanson MH, Burgstahler AD, Mennone A, Fallon MB, Gonzalez CB, Saez JC. Ca21 waves are organized among hepatocytes in the intact organ. Am J Physiol Gastrointest Liver Physiol. 1995;269: G167–G171. 55. Tse A, Tse FW, Almers W, Hille B. Rhythmic exocytosis stimulated by GnRH-induced calcium oscillations in rat gonadotropes. Science. 1993;260:82– 84. 56. Sears J, Sears M. Circadian rhythms in aqueous humor formation. In: M. Civan, ed. From Secretion to Glaucoma. Academic Press, Current Topics in Membranes vol. 45: p. 203–232, 1998.