Protein 4.1N does not interact with the inositol 1,4,5-trisphosphate receptor in an epithelial cell line

Protein 4.1N does not interact with the inositol 1,4,5-trisphosphate receptor in an epithelial cell line

Cell Calcium 38 (2005) 469–480 Protein 4.1N does not interact with the inositol 1,4,5-trisphosphate receptor in an epithelial cell line Sona Sehgal a...

826KB Sizes 0 Downloads 81 Views

Cell Calcium 38 (2005) 469–480

Protein 4.1N does not interact with the inositol 1,4,5-trisphosphate receptor in an epithelial cell line Sona Sehgal a , Mateus T. Guerra b , Emma A. Kruglov b , Jun Wang b , Michael H. Nathanson b,∗ b

a Department of Pediatrics, Yale University School of Medicine, New Haven, CT, USA Department of Medicine, Section of Digestive Diseases, Room TAC S241D, Yale University School of Medicine, 333 Cedar St., New Haven, CT 06520-8019, USA

Received 17 June 2005; accepted 17 June 2005 Available online 24 August 2005

Abstract Cytosolic Ca2+ regulates a variety of cell functions, and the spatial patterns of Ca2+ signals are responsible in part for the versatility of this second messenger. The subcellular distribution of the inositol 1,4,5-trisphosphate receptor (IP3R) is thought to regulate Ca2+ -signaling patterns but little is known about how the distribution of the IP3R itself is regulated. Here we examined the relationship between the IP3R and the cytoskeletal linker protein 4.1N in the polarized WIF-B cell line because protein 4.1N regulates targeting of the type I IP3R in neurons, but WIF-B cells do not express this cytoskeletal protein. WIF-B cells expressed all three isoforms of the IP3R, and each isoform was distributed throughout the cell. These cells did not express the ryanodine receptor. Photorelease of microinjected, caged IP3 induced a rapid rise in cytosolic Ca2+ , but the increase began uniformly throughout the cell rather than at a specific initiation site. Expression of protein 4.1N was not associated with redistribution of the IP3R or changes in Ca2+ -signaling patterns. These findings are consistent with the hypothesis that the subcellular distribution of IP3R isoforms regulates the formation of Ca2+ waves, and the finding that interactions between protein 4.1N and the IP3R vary among cell types may provide an additional, tissue-specific mechanism to shape the pattern of Ca2+ waves. © 2005 Elsevier Ltd. All rights reserved. Keywords: Cytosolic Ca2+ ; Inositol 1,4,5-trisphosphate; Protein 4.1N; Hepatocyte

1. Introduction Cytosolic Ca2+ is a second messenger that controls many cellular events [1]. In polarized epithelia such as hepatocytes, cytosolic Ca2+ simultaneously regulates such processes as secretion, glucose metabolism, and mitochondrial redox potential [2]. Ca2+ also regulates gene transcription [3] and apoptosis [4–6]. The ability of Ca2+ to simultaneously control such varied processes within a cell is thought to depend in part on its ability to form spatial and temporal patterns such as gradients, waves and oscillations [1,2]. Ca2+ signals in a number of polarized epithelia, including both hepatocytes and cholangiocytes, are mediated entirely by ∗

Corresponding author. Tel.: +1 203 785 7312; fax: +1 203 785 4306. E-mail address: [email protected] (M.H. Nathanson).

0143-4160/$ – see front matter © 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.ceca.2005.06.038

the inositol 1,4,5-trisphosphate (IP3) receptor (IP3R) [7,8]. There are three isoforms of the IP3R [9–11], and the expression and distribution of each isoform varies among tissue types and might account for tissue-specific Ca2+ -signaling patterns [12–16]. For example, hepatocytes express the types I and II but not the type III isoform of the IP3R [7,17]. The type II IP3R accounts for the majority of the IP3R in hepatocytes and is concentrated near the apical membrane. In contrast, cholangiocytes express all three IP3R isoforms. The type III IP3R predominates in these cells, but is concentrated subapically as well [7]. Ca2+ waves begin in the apical region in both hepatocytes and cholangiocytes, and in each case this has been attributed to the increased IP3R concentration in the region of this initiation site [7,8]. However, this structure–function relationship has not been proven.

470

S. Sehgal et al. / Cell Calcium 38 (2005) 469–480

Study of the Ca2+ -signaling machinery in hepatocytes has been limited because they lose certain features of polarity soon after isolation [18,19], which limits the assessment of polarized signal formation. WIF-B cells are a hybrid rat hepatoma and human fibroblast cell line that preserves many morphological and physiological features of hepatocytes [20–23]. The plasma membrane proteins in WIF-B cells are distributed in a polarized fashion [23], and these cells exhibit polarized functions of hepatocytes such as secretion of bile acids into canalicular structures [20,24]. Furthermore, the apical and basolateral regions of the WIF-B cell are aligned in a single focal plane which makes this cell model particularly amenable to observe sites of initiation and spread of Ca2+ signals across the cell. Although WIF-B cells have been used as a model for a number of aspects of epithelial cell biology and physiology, Ca2+ -signaling in these cells has not yet been examined. The goal of this work was to define the Ca2+ -signaling machinery in WIF-B cells to understand the extent to which they could serve as a model for Ca2+ signaling in hepatocytes and other polarized epithelia.

2. Materials and methods 2.1. Cells and reagents Fluo-3 (cell-impermeant), TO-PRO-3, and rhodamineconjugated phalloidin were purchased from Molecular Probes (Eugene, OR, USA). NPE-caged IP3 was from Calbiochem (LaJolla, CA, USA). Modified F12 medium was from Sigma Chemicals (St. Louis, MO, USA). The chemiluminescence kit and protein A/Sepharose CL-4B beads were from Amersham (Arlington Heights, IL, USA). A phenolfree total RNA isolation kit was from Ambion (Austin, TX, USA). WIF-B cells were a generous gift from Dr. Ann Hubbard (Johns Hopkins). The cells were grown at 37 ◦ C with 7% CO2 /air in modified F12 medium supplemented with 5% FBS and HAT (10 ␮M hypoxanthine, 4 nM aminopterin, 1.6 ␮M thymidine) as described [23,24]. Cells were grown on glass coverslips and used between day 10 and 14 when they had maximal density and polarity [24]. Primary rat hepatocytes were isolated by collagenase perfusion as described previously [8] and were used within 2 h of isolation.

were kindly provided by Richard Wojcikiewicz (SUNY Syracuse). Commercially available monoclonal antibodies from Transduction Laboratories (Lexington, KY, USA) were used to label the N-terminal region of the type III IP3R [14] and amino acid 510–626 of protein 4.1N [27]. The secondary antibodies for immunoblots were peroxidase-conjugated antirabbit or anti-mouse IgG from Sigma. Secondary antibodies for immunofluorescence were Alexa 488, Alexa 647 or Alexa 555 anti-rabbit or anti-mouse IgG from Molecular Probes (Eugene, OR, USA). 2.3. Immunoblot analysis The protein concentration of cell homogenates was determined as previously described [28]. Proteins collected from WIF-B cells were separated by sodium dodecyl sulfatepolyacrylamide gel electrophoresis using a 5% gel and were subsequently transferred to protein nitrocellulose membranes. The membranes were blocked for 1 h and incubated at 4 ◦ C overnight with the primary antibody. The antibody against IP3R isoform type I was used at a dilution of 1:10,000, the antibody against IP3R isoform type II was used at a dilution of 1:100, the antibody against IP3R isoform type III was used at a dilution of 1:1000 and the antibody against protein 4.1N was used at a dilution of 1:5000. The membranes were washed, incubated for 1 h with peroxidaseconjugated anti-rabbit (1:5000) or anti-mouse (1:4000) secondary IgG antibody, and revealed by enhanced chemiluminescence. For quantitative analyses, immunoblotting was performed on samples of WIF-B cells and positive controls containing known concentrations of IP3R isoforms. For each immunoblot, type I, II and III IP3R immunoreactivity was then quantified using NIH Image J analysis software. IP3R isoform protein expression in WIF-B cells and in known controls was compared directly to determine the relative abundance of each isoform in WIF-B cells. The ratio of WIF-B cell band intensity divided by control cell band intensity was multiplied by the IP3R isoform concentration in the control cells or tissue to obtain the IP3R isoform concentration in WIF-B cells. Cerebellum was used as a standard for type I IP3R (129.6 ng/10 ␮g protein), AR4-2J cells were used to control for type II IP3R (9.8 ng/10 ␮g protein), and RIN cells were used for type III IP3R (14.1 ng/10 ␮g protein) [17]. The relationship between band intensity and the amount of IP3R loaded onto the gels was linear over the range used.

2.2. Antibodies

2.4. Immunoprecipitation

Each IP3R isoform was labeled with isoform-specific antibodies that have been characterized previously [7,8]. Type I IP3R antibodies were affinity-purified specific rabbit polyclonal antiserum directed against the 19 C-terminal residues of mouse type I IP3R [14,25,26] and were kindly provided by Barbara Ehrlich (Yale). Type II IP3R antibodies were from affinity-purified specific rabbit polyclonal antiserum directed against the 18 C-terminal residues of rat type II IP3R [17] and

The protein A/Sepharose beads were washed three times in PBS pH 8.0 and then blocked with 5% milk at 4 ◦ C overnight. The beads were washed with PBS and resuspended in a 1:1 ratio with fresh PBS pH 8.0. One milligram of freshly isolated solubilized protein from cerebrum was incubated with 25 ␮l slurry of protein A/Sepharose beads and 5–10 ␮l of primary antibody (directed against protein 4.1N or type I IP3R) at 4 ◦ C for 4 h. The complex was washed three times with lysis

S. Sehgal et al. / Cell Calcium 38 (2005) 469–480

buffer and finally spun down with STE buffer (100 mmol NaCl, 10 mmol Tris–HCl pH 8.0 and 1 mmol EDTA pH 8.0). Forty microlitres of lysis buffer was then added along with 10 ␮l of 5× Laemmli sample buffer. The complex was boiled for 5 min and then separated by SDS–PAGE. It was subsequently transferred to protein nitrocellulose membranes and immunoblotted with appropriate antibodies for protein 4.1N or for the type I IP3R. 2.5. Immunofluorescence Confocal immunofluorescence histochemistry was performed as described previously [7,8,29] using WIF-B cells or primary hepatocytes in culture. The cells were fixed with cold methanol and permeabilized with Triton X-100. After blocking steps, cells were labeled with primary antibody, rinsed with phosphate-buffered saline, and incubated with Alexa 488, 647 or 555-conjugated secondary antibody. WIFB cells were also labeled with a cytoskeletal marker to stain the plasma membrane. For negative control images, cells were incubated with secondary antibodies, and primary IP3R or protein 4.1N antibodies were omitted. Specimens were examined with a Zeiss LSM 510 Laser Scanning Confocal Microscope equipped with a krypton/argon mixed gas laser (Thornwood, NY). To ensure specificity of 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 secondary antibodies alone. Images were collected with a Plan Apochromat 63×, 1.20 NA water immersion objective lens and at a confocal pinhole adjusted to obtain a 1-␮m depth of focus. Double or triple labeled specimens were simultaneously observed at excitation 488 nm and emission 505–550 nm, excitation 543 nm and emission >585 nm and excitation 647 nm and emission LP 650 nm as appropriate. 2.6. Reverse-transcription and polymerase chain reaction RT-PCR was performed as described previously [7,30] and was used to test for RNA for each of the three isoforms of the ryanodine receptor (RyR). RNA was extracted from confluent WIF-B cells and rat cerebrum using a phenolfree total RNA isolation kit from Ambion. cDNA was then synthesized and RT-PCR performed using superscript firststrand synthesis system for RT-PCR (Invitrogen, Carlsbad, CA, USA). PCR was amplified with a PTC-100 automated thermocycler (MJ Research, Inc. Watertown, MA, USA) using 2 ␮l of the first-strand cDNA reaction, 150 pmol of each primer, 50 ␮mol/l deoxynucleoside triphosphates, 2.5 U ampliTaq DNA polymerase in 10 mmol/l Tris–HCl (pH 8.3) containing 50 mmol/l KCl and 2.5 mmol/l MgCl2 in a total volume of 100 ␮l. After a hot start (2 min at 94 ◦ C), the samples were subjected to 30 cycles of 30 s at 94 ◦ C, 30 s at 63 ◦ C and 30 s at 72 ◦ C. This was followed by a final extension at 72 ◦ C for 10 min. Primers were designed to rec-

471

ognize regions unique to each RyR isoform. The primer for the type I isoform of the RyR was designed to amplify a 436 base pair (bp) product, that for the type II isoform of RyR was designed to amplify a 353 bp product and that for the type III isoform of RyR was designed to amplify 319 bp product. The forward primer for the type I isoform of RyR was 5 -AGGACATGGTGGTGATGTTGCT-3 , and the reverse primer was 5 -ATGCTCTGACAGGTTGGTGAGC3 . The forward primer for the type II isoform of RyR was 5 -CGGAACAATAGGCAAGCAGATG-3 , and the reverse primer was 5 -GCATGTGCTCTGAGAGGTTGGT-3 . The forward primer for the type III isoform of RyR was 5 -ACCTGTCCTTCCTGAGGTTTGC-3 , and the reverse primer was 5 -TCAGGAGCACATCGGCCTAATA-3 . The resulting polymerase chain reaction (PCR) product was analyzed by agarose gel electrophoresis. 2.7. DNA construction DsRed-tagged protein 4.1N was generated using a cDNA clone of human protein 4.1N in pCMV5-HA vector, which was kindly provided by Dr. Ilya Bezprozvanny (University of Texas—Southwestern Dallas), plus the cDNA fragment coding the red fluorescent protein pDsRedExpress-C1 (BD Biosciences, Palo Alto, CA, USA). Protein 4.1N was amplified by PCR using forward primer 5 CGGGGTACCATGGAGGAGAAGGACTACAGTG-3 and reverse primer 5 -CGCGGATCCTCAGGATTCCTGTGGCTTCTTG-3 . Protein 4.1N and DsRed were sequentially digested with KpnI and BamHI. The digested products were ligated and the plasmid was propagated in E. coli DH5␣ competent cells. The ligated product was verified by sequencing. 2.8. Cell transfection WIF-B cells were transfected by direct intranuclear injection of the cDNA construct for 4.1N-pDsRed-Express-C1 or of co-injection of the constructs for protein 4.1N and pDsRed-Express-C1. Primary hepatocytes were also transfected by intranuclear co-injection of cDNA for protein 4.1N and pDsRed-Express-C1. cDNA for protein 4.1N-pDsRedExpress-C1 was purified via QIAquick PCR purification kit to obtain a concentration of 200 ng/␮l, which was centrifuged and then loaded in a microinjector needle. Microinjection was performed using an Eppendorf Micromanipulator 5171 and Transjector 5246 (Brinkmann Instruments, Westbury, NY, USA) coupled to a Zeiss Axiovert S100TV inverted microscope. Microinjections used an injection pressure Pi = 100 hPa, compensation pressure Pc = 28 hPa and injection time ti = 0.3 s. Eppendorf femtotips were used for penetration. After injection the cells were maintained in a 37 ◦ C incubator with 5%/7% CO2 for 90–120 min. Successfully transfected cells were then examined by confocal microscopy as described above and identified by DsRed fluorescence.

472

S. Sehgal et al. / Cell Calcium 38 (2005) 469–480

2.9. Ca2+ measurements Cytosolic Ca2+ was monitored in WIF-B cells using timelapse confocal microscopy [7,8,31]. Transfected WIF-B cells expressing either the protein 4.1N-DsRed construct or DsRed alone were first identified, and then microinjected with NPEcaged IP3 (Calbiochem, La Jolla, CA, USA). Non-transfected WIF-B cells in the same field were injected as controls. The microinjection pipettes were loaded with NPE-cage IP3 along with fluo-3 (Molecular Probes, Eugene, OR, USA) to verify successful injection and to monitor subsequent Ca2+ signals. After injection the cells were allowed to recover for 5 min before flash photolysis studies were performed. Caged IP3 was released by UV flash photolysis using a custom built mercury arc lamp (50 W) coupled to a 1 mm quartz fiber optic cable through a high speed shutter as previously described [14]. The cells were transferred to a chamber on the stage of a Zeiss Axiovert microscope, perfused with HEPES-buffered solution and observed with a Bio-Rad MRC 1024 confocal imaging system (Hercules, CA). Cells were observed with a spatial resolution of .32 ␮m and a temporal resolution of 500–600 ms. Fluorescence signals were expressed as percent of initial (baseline) fluorescence [14,31]. 3. Results 3.1. Expression and subcellular distribution of Ca2+ release channels in WIF-B cells We examined whether WIF-B cells express either of the two principal intracellular Ca2+ release channels, the IP3R and the RyR. There are three isoforms of the IP3R, and immunoblotting using isoform-specific antibodies revealed that WIF-B cells possess all three of these isoforms (Fig. 1a). In order to determine the relative abundance of each isoform, densitometric analysis was performed using cells or tissues with known amounts of each isoform as standards [17]. This analysis showed the type III IP3R was the predominant isoform, accounting for 48% of the total amount of IP3R, followed by type II IP3R at 36%, and type I IP3R constituted the rest at 16% (Fig. 1b). The subcellular distribution of each IP3R isoform was examined by confocal immunofluorescence (Fig. 2). Specimens were double labeled with either the actin stain rhodamine phalloidin, the basolateral membrane marker CE9, or the nuclear stain TO-PRO-3. Each of the three isoforms was distributed throughout the cytosol (Fig. 2a–c). Unlike what has been observed in hepatocytes [8] and a number of other polarized epithelia [7,32,33], in WIF-B cells none of the IP3R isoforms were concentrated in or excluded from either the apical or the basolateral region. Expression of RyR in WIF-B cells was analyzed by RT-PCR using separate sets of primers to recognize each of the three RyR isoforms (Fig. 3). Cerebrum expresses each RyR isoform and so was used as a positive control. PCR products confirmed expression of each RyR isoform in cerebrum, whereas none of the

Fig. 1. WIF-B cells express all three isoforms of the IP3R. (a) Western analysis using isoform-specific IP3R antibodies identifies bands of the appropriate sizes in increasing concentrations of WIF-B protein lysate (25, 35 and 50 ␮g, left to right) loaded in the first three lanes. Each antibody also identifies the specific type of IP3R isoform in lysates from positive control: cerebellum (2 ␮g) for type I IP3R, AR4-2J cells (50 ␮g) for type II IP3R and RIN cells (50 ␮g) for type III IP3R. (b) Quantitative densitometric analysis of IP3R isoforms in WIF-B cells. The absolute and then relative amount of each IP3R subtype was assessed from the immunoblots based on the known amounts in each positive control. The type III IP3R isoform was predominant at 48%, followed by type II IP3R at 36% and type I IP3R at 16%. Results are representative of duplicate experiments.

RyR isoforms were identified in WIF-B cells (Fig. 3). The RNA-positive control GAPDH was found in both cerebrum and WIF-B cells. Thus, like hepatocytes and a number of other epithelia, WIF-B cells express multiple IP3R isoforms but do not express RyR. However, unlike what is observed in most polarized epithelia, no IP3R isoform is distributed in a polarized fashion in WIF-B cells. 3.2. Subcellular targeting of IP3R in WIF-B cells and primary hepatocytes Little is known about the factors that regulate subcellular targeting of the IP3R. Protein 4.1N is a cytoskeletal linker protein that associates with the type I but not the type II or III IP3R. The biochemical basis for this association has been investigated in cerebral neurons [34] and we used confocal immunofluorescence to directly confirm that

S. Sehgal et al. / Cell Calcium 38 (2005) 469–480

473

Fig. 2. Subcellular distribution of IP3R isoforms in WIF-B cells, determined by confocal immunofluorescence. (a) Type I IP3R: cells were double labeled with rhodamine-phalloidin to stain submembranous actin (red) and a polyclonal antibody to stain the type I IP3R (green). Note that the type I IP3R is distributed throughout the cytoplasm. Scale bar, 20 ␮g. (b) Type II IP3R: here, cells were double labeled with to-pro-3 to stain the nucleus (blue) and a polyclonal antibody to stain the type II IP3R (green). Like the type I IP3R, the type II receptor is distributed throughout the cytoplasm. (c) Type III IP3R: cells were double labeled with a polyclonal antibody to stain the basolateral membrane marker CE-9 (red) and a monoclonal antibody to stain the type III IP3R (green). Like the types I and II IP3R, the type III receptor is distributed throughout the cytoplasm. For each figure, the corresponding transmission image is shown in the lower right panel.

the type I IP3R and protein 4.1N co-localize in cerebrum (Fig. 4a and b). We further confirmed this association by coimmunoprecipitating endogenous protein 4.1N and the type I IP3R from cerebrum (Fig. 4c). The interaction between these two proteins is thought to be responsible for targeting of the type I IP3R to the region of the plasma membrane

[34,35]. Neither WIF-B cells nor primary rat hepatocytes were found to express protein 4.1N (Fig. 5), so we examined whether exogenous expression of protein 4.1N could alter the distribution of type I IP3R in WIF-B cells or primary hepatocytes. WIF-B cells were injected with cDNA encoding a protein 4.1N-DsRed fusion construct, and then

474

S. Sehgal et al. / Cell Calcium 38 (2005) 469–480

Fig. 3. WIF-B cells do not express the ryanodine receptor (RyR). Using isoform-specific primers, RT-PCR was used to test for the presence of each of the three RyR isoforms in WIF-B cells. No PCR products were identified for any of the three isoforms, while single products for each isoform were identified in rat cerebellum (positive control). A single band is seen for glyceraldehydes-3-phosphate dehydrogenase (GAPDH) in both WIF-B cells and rat cerebellum (RNA control).

DsRed fluorescence was examined within 2 h by confocal microscopy (Fig. 6). Some of the expressed protein reached the plasma membrane (Fig. 6a) but some was distributed in a punctuate pattern throughout the cell (Fig. 6b). This is in contrast to the fluorescence pattern observed in cells expressing DsRed alone, which was diffusely distributed (Fig. 6c). Confocal immunofluorescence examination of the type I IP3R in WIF-B cells revealed that the receptor did not redistribute to the region of the plasma membrane (Fig. 7). To investigate whether the protein 4.1N-pDsRed chimera had impaired ability to target appropriately or interact with the type I IP3R, we next co-injected individual cDNAs for protein 4.1N and DsRed in WIF-B cells and primary hepatocytes. Both of these cell types were examined by confocal immunofluorescence within 4 h of injection to determine the distribution of protein 4.1N and type I IP3R. In both WIF-B cells and primary hepatocytes, nearly all protein 4.1N was detected at the plasma membrane but the type I IP3R, as before, did not redistribute (Fig. 8a and b). These findings suggest that the type I IP3R does not bind to protein 4.1N in either WIF-B cells or primary hepatocytes, perhaps because additional scaffolding or accessory proteins are needed that are not expressed in these cell types. 3.3. IP3R and Ca2+ -signaling patterns Ca2+ -signaling patterns are thought to be determined in part by the IP3R isoforms that are expressed within a cell,

plus the subcellular distribution of these isoforms [8,29]. For example, in rat hepatocytes the type II IP3R is the predominant isoform, and Ca2+ signals originate in the canalicular region, where this isoform is most concentrated [8]. To understand the relationship between IP3R isoforms and Ca2+ signaling patterns in WIF-B cells, the cells were microinjected with NPE-caged IP3 and then Ca2+ signals were monitored by time-lapse confocal microscopy as the IP3 was photoreleased. Eighteen uncaging events were observed in five separate cells (Fig. 9a). The resulting Ca2+ signal began in the apical and the basolateral region one time each, and began in both regions simultaneously in the other 16 instances. Thus, Ca2+ signals do not consistently start in either the apical or basolateral region of WIF-B cells (Fig. 9b). This correlates with the diffuse distribution of each isoform of the IP3R and the fact that no isoform is localized to any specific region in WIF-B cells. Ca2+ -signaling patterns also were monitored in cells transfected to express protein 4.1N. Once again, cells were microinjected with caged IP3 and then Ca2+ signals were monitored by confocal microscopy as the IP3 was photoreleased. Ten uncaging events were observed in three separate cells. The resulting Ca2+ signal began in the apical region once, and began simultaneously throughout the cell in the other nine instances (Fig. 9c). Thus, expression of protein 4.1N does not alter the pattern of IP3-induced Ca2+ signals in WIF-B cells. This observation is concordant with the observation that the distribution of the IP3R was not affected by expression of protein 4.1N in WIF-B cells.

4. Discussion WIF-B cells are a well-characterized model of hepatocytes that has been particularly useful because this cell line maintains structural and functional polarity [21,24]. There is a paucity of cell culture models that preserve the polarized Ca2+ -signaling patterns observed in primary hepatocytes and other epithelia, so we examined the machinery and mechanisms of Ca2+ -signaling in WIF-B cells. Both similarities and differences were observed between WIF-B cells and primary polarized epithelia. WIF-B cells expressed all three isoforms of the IP3R. This is similar to what has been observed in epithelia such as cholangiocytes [7] and pancreatic acinar cells [36,37] but different from hepatocytes, which express only the type I and type II isoforms [8]. Interestingly, we found that the type III IP3R is the predominant isoform in WIF-B cells, even though this isoform is not expressed in primary hepatocytes. This may reflect the observation that expression of the type III isoform increases in some cells in culture [38]. Like hepatocytes and cholangiocytes, WIF-B cells do not express RyR. However, RyR are found in some polarized epithelia, including pancreatic acinar cells [30] and the T84 cell line [39]. Moreover, RyR are responsible for the spread of Ca2+ waves into the basolateral region in pancreatic acinar cells [40]. An important difference between WIF-B cells and most other polarized epithelia [41] is that none of

S. Sehgal et al. / Cell Calcium 38 (2005) 469–480

475

Fig. 4. Protein 4.1N interacts with the type I IP3R in rat cerebrum. (a) Protein 4.1N colocalizes with the type I IP3R in rat cerebrum, as determined by confocal immunofluorescence. (b) Higher magnification image of the same tissue section demonstrates colocalization of protein 4.1N and type I IP3R in a single neuron. In both figures (A) shows type I IP3R (red) staining of cell bodies of neurons; (B) shows protein 4.1N (green) outlining the cell membranes of the neurons and is seen within other cellular processes as well; (C) shows co-localization (yellow) of the type I IP3R and protein 4.1N near the neuron cell membrane and elsewhere, and (D) shows transmission image of the section of cerebrum stained for immunofluorescence. (c) Co-immunoprecipitation confirms the association of protein 4.1N and type I IP3R. Whole cerebral protein, cerebral protein immunoprecipitated with protein 4.1N, and cerebral protein immunoprecipitated with type I IP3R were blotted for protein 4.1N. Two adjacent bands corresponding to the two isoforms of protein 4.1N were observed in each case. Immunoblotting for type I IP3R similarly revealed a single band in each column (not shown).

the IP3R isoforms are concentrated in the apical region in WIF-B cells. Since apical IP3Rs are thought to establish a trigger zone that initiates Ca2+ signals in epithelia [8,32,42], this structural difference may be responsible for another key difference between WIF-B cells and other epithelia, which is that Ca2+ signals do not routinely begin in the apical region of WIF-B cells. Consistent with this explanation, primitive hepatocytes from the little skate Raja erinacea also lack both a subcellular region in which IP3Rs are concentrated and a discrete initiation site for Ca2+ signals [43]. Thus, the current

work reveals an important limitation of the WIF-B cell as a model for the manner in which Ca2+ signals are formed in polarized epithelia. The current work may have important implications for WIF-B cells as a model for epithelial secretion as well. Apical IP3Rs are thought to be responsible not only for the pattern of Ca2+ signals in epithelia, but for two features of epithelial Ca2+ signals that are necessary for efficient regulation of secretion. First, apical localization of the trigger zone may assure that Ca2+ signals will be formed as polarized,

476

S. Sehgal et al. / Cell Calcium 38 (2005) 469–480

Fig. 5. WIF-B cells do not express protein 4.1N. Western analysis using a specific monoclonal antibody does not detect this protein in protein lysate from rat hepatocytes and 10-day-old WIF-B cells (50 ␮g each). In contrast, two bands are identified in lysate from rat cerebrum (50 ␮g) corresponding to the two isoforms of this protein that are known to be present.

apical-to-basal Ca2+ waves [32]. This in turn directs vectorial fluid and electrolyte secretion [44]. Although WIF-B cells are capable of canalicular secretion [24], their lack of organized, polarized Ca2+ waves thus may mean that fluid and electrolyte secretion is less efficient in WIF-B cells than in primary hepatocytes. This situation may be analogous to what is observed in skate liver, in which the lack of polarized Ca2+ waves has been related to bile secretion which is much less efficient than what is observed in mammalian liver [43]. Second, an increased concentration of the IP3R in the apical region may permit local Ca2+ concentrations to reach 5–10 ␮M [45]. This relatively high-Ca2+ concentration, which could be toxic elsewhere in the cytoplasm, is likely necessary to direct vesicle targeting and fusion and exocytosis at the apical membrane [45,46]. Thus, the current work may reveal previously unrecognized limitations of the WIF-B cell as a model for the manner in which secretion is regulated in polarized epithelia. IP3R isoforms often are distributed heterogeneously within cells and in patterns that differ among cells types. However, very little is known about the molecular mechanisms that regulate IP3R targeting. The principal factor that

Fig. 6. Expression of DsRed-tagged protein 4.1N in WIF-B cells. cDNA for the fusion construct or for DsRed alone was microinjected into the nuclei of WIF-B cells, then cell fluorescence was examined within 2 h by confocal microscopy. Scale bar, 20 ␮m. (A) WIF-B cells expressing protein 4.1N tagged to DsRed. The fusion construct is predominantly expressed near the cell membrane. (B) Higher magnification of a different field of WIF-B cells expressing the fusion construct. Here it can be appreciated that fluorescence is found not only in the vicinity of the plasma membrane (arrow), but in a punctuate pattern throughout the cell as well (arrowheads). (C) WIF-B cells expressing DsRed alone. DsRed was diffusely distributed throughout the cells, in contrast to what was observed in cells expressing the DsRed-protein 4.1N fusion.

Fig. 7. The distribution of type I IP3R is not altered in WIF-B cells transfected with protein 4.1N-DsRed. (A) WIF-B cells expressing protein 4.1N tagged to DsRed (red). Protein 4.1N localizes mostly to the plasma membrane. (B) Type I IP3R (green) is identified simultaneously by confocal immunofluorescence. The distribution is similar in transfected and non-transfected WIF-B cells. (C) Merged image further illustrates that protein 4.1N colocalizes with type I IP3R within some intracellular regions but the overall distribution of type I IP3R is not altered in transfected cells.

S. Sehgal et al. / Cell Calcium 38 (2005) 469–480

477

Fig. 8. Protein 4.1N does not alter the distribution of type I IP3R in WIF-B cells or hepatocytes. (a) The distribution of type I IP3R is not altered in WIF-B cells transfected with protein 4.1N. Cells were co-transfected with separate cDNAs for protein 4.1N and DsRed and then examined for the distribution of these plus the type I IP3R by confocal immunofluorescence. (A) Immunofluorescence reveals that protein 4.1N (green) is expressed in the region of the cell membrane. (B) The type I IP3R (blue) is homogenously distributed in the cytosol, regardless of whether cells express protein 4.1N. (C) Merged image. (b) The distribution of type I IP3R is not altered in primary rat hepatocytes transfected with protein 4.1N. (A) Protein 4.1N (green) is expressed in the region of the cell membrane. (B) The type I IP3R (blue) is homogenously distributed in the cytosol. (C) Merged image.

has been identified to date is protein 4.1N, which plays a role in targeting the type I IP3R in particular, in both neurons and MDCK cells [34,35,47]. Protein 4.1N was cloned and identified as a brain homologue of the erythrocyte membrane cytoskeletal protein 4.1R [47]. Protein 4.1N is found in almost all parts of the brain and is distributed in the neuronal cell body, dendrites and axon [47]. It is thought to confer stability and plasticity to the neuronal cell membrane through its interaction with spectrin and ankyrin. The carboxy-terminal domain (CTD) binds to several proteins, including the nuclear mitotic apparatus protein and the GluR1 glutamate receptor [48,49]. Protein 4.1N also associates with the type-I IP3R in neurons but not with type II or type III IP3R, as determined by yeast two-hybrid screening [34]. The association with the type I IP3R has subsequently been confirmed by protein pulldown and immunofluorescence approaches in MDCK cells, in which evidence suggests that protein 4.1N is responsible for the translocation of type I IP3R to the basolateral membrane. In these cells, the CTD of protein 4.1N comprising of amino acids 767–879 binds to the last 14 amino acid

residues of type I IP3R [35]. In the current study, confocal immunofluorescence in rat cerebrum showed that protein 4.1N and the type I IP3R co-localize in the cell body as well as in neuronal processes, further corroborating the association of these two proteins in brain. We therefore tried to target the type I IP3R to the region of the plasma membrane in WIF-B cells, in order to see if this would create a trigger zone in these cells. A DsRed-tagged 4.1N localized in part to the plasma membrane, but also was distributed in part in a punctuate pattern in the cytoplasm, which differs from the distribution of protein 4.1N that has been reported in other cells. To determine whether targeting of this fusion protein was altered compared to native 4.1N, we alternatively co-expressed untagged protein 4.1N along with DsRed as a marker of transfection. Indeed, native 4.1N was concentrated almost exclusively at or near the plasma membrane, suggesting that the DsRed chimera was in part mis-targeted. This may reflect the fact that DsRed can sometimes form multimers, which in turn could interfere with targeting signals or motifs. It is less likely that the DsRed moiety interfered

478

S. Sehgal et al. / Cell Calcium 38 (2005) 469–480

Fig. 9. Ca2+ -signaling in WIF-B cells. Cells were microinjected with caged IP3, and then observed by time-lapse confocal microscopy as the IP3 was photoreleased. (a) Serial confocal images before and 0.8 and 1.2 s after photorelease of IP3 in a representative cell. The cell was co-injected with the Ca2+ dye fluo-3 and images were pseudocolored according to the color scale shown. Fluorescence was monitored in specific apical (A) and basolateral (B) regions identified by their location relative to canaliculi in transmission images (not shown). (b) IP3 does not cause a preferential increase in Ca2+ in the apical or basolateral region. Tracings correspond to the fluorescence increases observed in the regions indicated in the previous panel. Signals began simultaneously in the apical and basolateral region 16 out of 18 times in five separate cells. (c) IP3 does not cause a preferential increase in Ca2+ in either the apical or basolateral region in cells expressing protein 4.1N. Tracing is representative of what was observed 9 out of 10 times in three separate cells expressing protein 4.1N that were co-transfected with DsRed and then identified by DsRed fluorescence.

with the IP3R-binding site of protein 4.1N, since the DsRed was fused to the N-terminus, whereas the type I IP3R binds to the C-terminus of 4.1N [34]. However, expression of protein 4.1N in WIF-B cells did not lead to re-targeting of the type I IP3R to the plasma membrane, even when 4.1N was localized correctly. Not surprisingly, subcellular Ca2+ signals also were not altered in WIF-B cells expressing protein 4.1N. Thus, the current findings are consistent with the hypothesis that the subcellular distribution of IP3R isoforms regulates the formation of Ca2+ waves. We also tested the ability of protein 4.1N to redistribute type I IP3R in primary hepatocytes to see if this lack of association of the two proteins was a feature unique to WIF-B cells since they are a hybrid cell line. Similar to WIF-B cells, primary hepatocytes expressed protein 4.1N mostly at the plasma membrane and this did not alter the distribution of type I IP3R in primary hepatocytes. Thus, the current work also suggests that targeting of the type I IP3R isoform is regulated differently in hepatocytes than in neurons or MDCK cells. In addition, the basis for apical targeting of the type II or type III IP3R in hepatocytes or cholangiocytes, respectively, remains unknown. Further work will be needed to demonstrate more directly that these tissue-specific differences in the Ca2+ -signaling machinery regulates Ca2+ -signaling and its downstream effects in the hepatocyte.

Acknowledgements This work was supported by NIH grants DK45710, DK57751, DK34989, DK07356, and TW01451.

References [1] M.J. Berridge, M.D. Bootman, H.L. Roderick, Calcium signalling: dynamics, homeostasis and remodelling, Nat. Rev. Mol. Cell. Biol. 4 (2003) 517–529. [2] M.F. Leite, M.H. Nathanson, Calcium signaling in the liver, in: The Liver: Biology and Pathobiology, Lippincott Williams & Wilkins, 2001, pp. 539–554. [3] R.E. Dolmetsch, K. Xu, R.S. Lewis, Calcium oscillations increase the efficiency and specificity of gene expression, Nature 392 (1998) 933–936. [4] I. Yokoyama, S. Hayashi, T. Kobayashi, M. Negita, Y. Namii, A. Katayama, T. Nagasaka, C. Koike, Y. Tachi, A. Hayakawa, H. Tajiri, H. Takagi, Hepatocyte apoptosis and cytosolic calcium dynamics in ischemic injury, Transplant. Proc. 29 (1997) 3514. [5] M.C. Bassik, L. Scorrano, S.A. Oakes, T. Pozzan, S.J. Korsmeyer, Phosphorylation of BCL-2 regulates ER Ca(2+) homeostasis and apoptosis, EMBO J. 23 (2004) 1207–1216. [6] W. Li, J. Llopis, M. Whitney, G. Zlokarnik, R.Y. Tsien, Cellpermeant caged InsP3 ester shows that Ca2+ spike frequency can optimize gene expression, Nature 392 (1998) 936–941.

S. Sehgal et al. / Cell Calcium 38 (2005) 469–480 [7] K. Hirata, J.F. Dufour, K. Shibao, R. Knickelbein, A.F. O’Neill, H.P. Bode, D. Cassio, M.V. St-Pierre, N.F. Larusso, M.F. Leite, M.H. Nathanson, Regulation of Ca(2+) signaling in rat bile duct epithelia by inositol 1,4,5-trisphosphate receptor isoforms, Hepatology 36 (2002) 284–296. [8] K. Hirata, T. Pusl, A.F. O’Neill, J.A. Dranoff, M.H. Nathanson, The type II inositol 1,4,5-trisphosphate receptor can trigger Ca2+ waves in rat hepatocytes, Gastroenterology 122 (2002) 1088–1100. [9] T. Furuichi, S. Yoshikawa, A. Miyawaki, K. Wada, N. Maeda, K. Mikoshiba, Primary structure and functional expression of the inositol 1,4,5-trisphosphate-binding protein P400, Nature 342 (1989) 32–38. [10] A.R. Maranto, Primary structure, ligand binding, and localization of the human type 3 inositol 1,4,5-trisphosphate receptor expressed in intestinal epithelium, J. Biol. Chem. 269 (1994) 1222–1230. [11] T.C. Sudhof, C.L. Newton, B.T. Archer 3rd, Y.A. Ushkaryov, G.A. Mignery, Structure of a novel InsP3 receptor, EMBO J. 10 (1991) 3199–3206. [12] C.L. Newton, G.A. Mignery, T.C. Sudhof, 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. 269 (1994) 28613–28619. [13] T. Sugiyama, M. Yamamoto-Hino, K. Wasano, K. Mikoshiba, M. Hasegawa, Subtype-specific expression patterns of inositol 1,4,5trisphosphate receptors in rat airway epithelial cells, J. Histochem. Cytochem. 44 (1996) 1237–1242. [14] R.E. Hagar, A.D. Burgstahler, M.H. Nathanson, B.E. Ehrlich, Type III InsP3 receptor channel stays open in the presence of increased calcium, Nature 396 (1998) 81–84. [15] J. Ramos-Franco, D. Bare, S. Caenepeel, A. Nani, M. Fill, G. Mignery, Single-channel function of recombinant type 2 inositol 1,4,5-trisphosphate receptor, Biophys. J. 79 (2000) 1388–1399. [16] M. Hattori, A.Z. Suzuki, T. Higo, H. Miyauchi, T. Michikawa, T. Nakamura, T. Inoue, K. Mikoshiba, Distinct roles of inositol 1,4,5trisphosphate receptor types 1 and 3 in Ca2+ signaling, J. Biol. Chem. 279 (2004) 11967–11975. [17] R.J. Wojcikiewicz, 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. 270 (1995) 11678–11683. [18] A. Gautam, O.C. Ng, J.L. Boyer, Isolated rat hepatocyte couplets in short-term culture: structural characteristics and plasma membrane reorganization, Hepatology 7 (1987) 216–223. [19] E. Knop, A. Bader, K. Boker, R. Pichlmayr, K.F. Sewing, Ultrastructural and functional differentiation of hepatocytes under long-term culture conditions, Anat. Rec. 242 (1995) 337–349. [20] Y. Sai, A.T. Nies, I.M. Arias, Bile acid secretion and direct targeting of mdr1-green fluorescent protein from Golgi to the canalicular membrane in polarized WIF-B cells, J. Cell. Sci. 112 (1999) 4535–4545. [21] V. Bender, P. Bravo, C. Decaens, D. Cassio, The structural and functional polarity of the hepatic human/rat hybrid WIF-B is a stable and dominant trait, Hepatology 30 (1999) 1002–1010. [22] C.S. Schaffert, S.L. Todero, B.L. McVicker, P.L. Tuma, M.F. Sorrell, D.J. Tuma, WIF-B cells as a model for alcohol-induced hepatocyte injury, Biochem. Pharmacol. 67 (2004) 2167–2174. [23] G. Ihrke, E.B. Neufeld, T. Meads, M.R. Shanks, D. Cassio, M. Laurent, T.A. Schroer, R.E. Pagano, A.L. Hubbard, WIF-B cells: an in vitro model for studies of hepatocyte polarity, J. Cell. Biol. 123 (1993) 1761–1775. [24] C. Decaens, P. Rodriguez, C. Bouchaud, D. Cassio, Establishment of hepatic cell polarity in the rat hepatoma-human fibroblast hybrid WIF-B9. A biphasic phenomenon going from a simple epithelial polarized phenotype to an hepatic polarized one, J. Cell. Sci. 109 (1996) 1623–1635. [25] K. Takei, H. Stukenbrok, A. Metcalf, G.A. Mignery, T.C. Sudhof, P. Volpe, P. De Camilli, Ca2+ stores in Purkinje neurons: endoplasmic reticulum subcompartments demonstrated by the heterogeneous dis-

[26]

[27]

[28] [29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

479

tribution of the InsP3 receptor, Ca(2+)-ATPase, and calsequestrin, J. Neurosci. 12 (1992) 489–505. G.A. Mignery, T.C. Sudhof, K. Takei, P. De Camilli, Putative receptor for inositol 1,4,5-trisphosphate similar to ryanodine receptor, Nature 342 (1989) 192–195. A.V. Binda, N. Kabbani, R. Lin, R. Levenson, D2 and D3 dopamine receptor cell surface localization mediated by interaction with protein 4.1N, Mol. Pharmacol. 62 (2002) 507–513. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. K. Shibao, K. Hirata, M.E. Robert, M.H. Nathanson, Loss of inositol 1,4,5-trisphosphate receptors from bile duct epithelia is a common event in cholestasis, Gastroenterology 125 (2003) 1175–1187. M.F. Leite, J.A. Dranoff, L. Gao, M.H. Nathanson, Expression and subcellular localization of the ryanodine receptor in rat pancreatic acinar cells, Biochem. J. 337 (1999) 305–309. M.F. Leite, E.C. Thrower, W. Echevarria, P. Koulen, K. Hirata, A.M. Bennett, B.E. Ehrlich, M.H. Nathanson, Nuclear and cytosolic calcium are regulated independently, Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 2975–2980. M.H. Nathanson, M.B. Fallon, P.J. Padfield, A.R. Maranto, Localization of the type 3 inositol 1,4,5-trisphosphate receptor in the Ca2+ wave trigger zone of pancreatic acinar cells, J. Biol. Chem. 269 (1994) 4693–4696. K. Hirata, M.H. Nathanson, A.D. Burgstahler, K. Okazaki, E. Mattei, M.L. Sears, Relationship between inositol 1,4,5-trisphosphate receptor isoforms and subcellular Ca2+ signaling patterns in nonpigmented ciliary epithelia, Invest. Ophthalmol. Vis. Sci. 40 (1999) 2046–2053. A. Maximov, T.S. Tang, I. Bezprozvanny, Association of the type 1 inositol (1,4,5)-trisphosphate receptor with 4.1N protein in neurons, Mol. Cell. Neurosci. 22 (2003) 271–283. S. Zhang, A. Mizutani, C. Hisatsune, T. Higo, H. Bannai, T. Nakayama, M. Hattori, K. Mikoshiba, Protein 4.1N is required for translocation of inositol 1,4,5-trisphosphate receptor type 1 to the basolateral membrane domain in polarized Madin-Darby canine kidney cells, J. Biol. Chem. 278 (2003) 4048–4056. D.I. Yule, S.A. Ernst, H. Ohnishi, R.J. Wojcikiewicz, 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. 272 (1997) 9093–9098. M.G. Lee, X. Xu, W. Zeng, J. Diaz, T.H. Kuo, F. Wuytack, L. Racymaekers, S. Muallem, Polarized expression of Ca2+ pumps in pancreatic and salivary gland cells. Role in initiation and propagation of [Ca2+ ]i waves, J. Biol. Chem. 272 (1997) 15771–15776. O. Blondel, J. Takeda, H. Janssen, S. Seino, G.I. Bell, Sequence and functional characterization of a third inositol trisphosphate receptor subtype, IP3R-3, expressed in pancreatic islets, kidney, gastrointestinal tract, and other tissues, J. Biol. Chem. 268 (1993) 11356–11363. V. Verma, C. Carter, S. Keable, D. Bennett, P. Thorn, Identification and function of type-2 and type-3 ryanodine receptors in gut epithelial cells, Biochem. J. 319 (1996) 449–454. M.F. Leite, A.D. Burgstahler, M.H. Nathanson, Ca2+ waves require sequential activation of inositol trisphosphate receptors and ryanodine receptors in pancreatic acini, Gastroenterology 122 (2002) 415–427. T. Pusl, M.H. Nathanson, The role of inositol 1,4,5-trisphosphate receptors in the regulation of bile secretion in health and disease, Biochem. Biophys. Res. Commun. 322 (2004) 1318–1325. P. Thorn, A.M. Lawrie, P.M. Smith, D.V. Gallacher, O.H. Petersen, Ca2+ oscillations in pancreatic acinar cells: spatiotemporal relationships and functional implications, Cell Calcium 14 (1993) 746–757. M.H. Nathanson, A.F. O’Neill, A.D. Burgstahler, Primitive organization of cytosolic Ca(2+) signals in hepatocytes from the little skate Raja erinacea, J. Exp. Biol. 202 (1999) 3049–3056. H. Kasai, G.J. Augustine, Cytosolic Ca2+ gradients triggering unidirectional fluid secretion from exocrine pancreas, Nature 348 (1990) 735–738.

480

S. Sehgal et al. / Cell Calcium 38 (2005) 469–480

[45] K. Ito, Y. Miyashita, H. Kasai, Micromolar and submicromolar Ca2+ spikes regulating distinct cellular functions in pancreatic acinar cells, EMBO J. 16 (1997) 242–251. [46] M. Missler, W. Zhang, A. Rohlmann, G. Kattenstroth, R.E. Hammer, K. Gottmann, T.C. Sudhof, Alpha-neurexins couple Ca2+ channels to synaptic vesicle exocytosis, Nature 423 (2003) 939– 948. [47] L.D. Walensky, S. Blackshaw, D. Liao, C.C. Watkins, H.U. Weier, M. Parra, R.L. Huganir, J.G. Conboy, N. Mohandas, S.H. Sny-

der, A novel neuron-enriched homolog of the erythrocyte membrane cytoskeletal protein 4.1, J. Neurosci. 19 (1999) 6457–6467. [48] S.N. Mattagajasingh, S.C. Huang, J.S. Hartenstein, M. Snyder, V.T. Marchesi, E.J. Benz, A nonerythroid isoform of protein 4.1R interacts with the nuclear mitotic apparatus (NuMA) protein, J. Cell. Biol. 145 (1999) 29–43. [49] L. Shen, F. Liang, L.D. Walensky, R.L. Huganir, Regulation of AMPA receptor GluR1 subunit surface expression by a 4.1N-linked actin cytoskeletal association, J. Neurosci. 20 (2000) 7932–7940.