© The Rockefeller University Press, 0022-1295/1999//243/ $5.00
Journal of General Physiology, Volume 114, Number 2, 1999
Location of the Permeation Pathway in the Recombinant Type 1 Inositol 1,4,5-Trisphosphate Receptor
Josefina Ramos-Francoa,
Daniel Galvana,
Gregory A. Mignerya, and
Michael Filla
a From the Department of Physiology, Loyola University Chicago, Maywood, Illinois 60153-5500
Loyola University Chicago, Department of Physiology, 2160 S. First Avenue, Maywood, IL 60153-5500.Fax: 708-216-5158;
mfill{at}luc.edu
 |
ABSTRACT
|
|---|
The inositol 1,4,5-trisphosphate receptor (InsP3R) forms ligand-regulated intracellular Ca2+ release channels in the endoplasmic reticulum of all mammalian cells. The InsP3R has been suggested to have six transmembrane regions (TMRs) near its carboxyl terminus. A TMR-deletion mutation strategy was applied to define the location of the InsP3R pore. Mutant InsP3Rs were expressed in COS-1 cells and single channel function was defined in planar lipid bilayers. Mutants having the fifth and sixth TMR (and the interceding lumenal loop), but missing all other TMRs, formed channels with permeation properties similar to wild-type channels (gCs = 284; gCa = 60 pS; PCa/PCs = 6.3). These mutant channels bound InsP3, but ligand occupancy did not regulate the constitutively open pore (Po > 0.80). We propose that a region of 191 amino acids (including the fifth and sixth TMR, residues 2398–2589) near the COOH terminus of the protein forms the InsP3R pore. Further, we have produced a constitutively open InsP3R pore mutant that is ideal for future site-directed mutagenesis studies of the structure–function relationships that define Ca2+ permeation through the InsP3R channel.
Key Words: intracellular Ca2+ release • Ca2+ signaling • recombinant channel
© 1999 The Rockefeller University Press
|
Introduction
|
|---|
The inositol 1,4,5-trisphosphate receptor (InsP3R)1 gene family encodes a highly homologous group of proteins localized to the endoplasmic reticulum (ER). The InsP3R gene family has three members (types 1, 2, and 3) that are ubiquitously expressed in metazoans (Newton et al. 1994
; Nakanishi et al. 1996; De Smedt et al. 1997). The InsP3R proteins tetramerize to form ion channels that are responsible for the regulated release of Ca2+ from intracellular Ca2+ stores (reviewed in Bezprozvanny and Ehrlich 1995; Joseph 1996). The single channel properties of the three InsP3R Ca2+ channels (types 1, 2, and 3) have been defined by reconstituting the native receptor complex into planar lipid bilayers (Bezprozvanny et al. 1991; Hagar et al. 1998; Ramos-Franco et al. 1998a). Recently, the single channel properties of a recombinant type 1 InsP3R channel and a splice variant have been defined (Kaznacheyeva et al. 1998; Ramos-Franco et al. 1998b). The recombinant and native InsP3R channels have nearly identical functional properties. The capacity to define single channel behavior of recombinant InsP3Rs established a foundation from which new molecular/biophysical approaches can be used to define the structure–function properties of the InsP3R channel family.
A first step towards understanding structure–function relationships in any protein is to locate the primary sequences that associated with its general features, such as ligand binding and/or transmembrane regions (TMRs). Analysis of the recombinant InsP3R revealed that this protein is composed of three domains: the NH2-terminal InsP3-binding domain, the COOH-terminal channel domain, and the central coupling domain (Mignery and Südhof 1990; Südhof et al. 1991). The originally proposed channel domain contains all the putative TMRs. This interpretation is consistent with several lines of evidence. For example, deletion of the channel-domain generates soluble monomeric InsP3-binding proteins. The green fluorescent protein–-tagged channel-domain, after deletion of the InsP3-binding and coupling domains, oligomerizes and is localized to the ER (Sayers et al. 1997). Currently, it is thought that the InsP3R's channel domain has six TMRs. A six-TMR model is consistent with immunogold electron microscopy studies showing that the NH2 and COOH termini are both localized in the cytoplasm (Mignery et al. 1989). It is also consistent with glycosylation data that demonstrates that the loop between the fifth and sixth TMR is lumenal (Michikawa et al. 1994). Moreover, the six-TMR model was experimentally confirmed by differential permeabilization combined with immunohistochemistry (Galvan, D., E. Borrego-Diaz, P.J. Perez, and G.A. Mignery, manuscript submitted for publication).
The InsP3R protein is thought to tetramerize to form functional Ca2+ release channels (Mignery et al. 1989). It is clear that important determinants of InsP3R tetramerization are associated with the TMRs (Mignery and Südhof 1990; Galvan, D., E. Borrego-Diaz, P.J. Perez, and G.A. Mignery, submittedmanuscript for publication). Galvan and co-workers constructed several TMR-deletion mutants from the full-length type 1 InsP3R cDNA to define important determinants of InsP3R tetramerization. At least two TMRs are required for the initiation of InsP3R channel assembly (Galvan, D., E. Borrego-Diaz, P.J. Perez, and G.A. Mignery, manuscript submitted for publication). Tetramerization is also stabilized in the presence of additional TMRs and in the presence of the COOH terminus. Further, two studies have implicated the fifth and sixth TMRs as a particularly strong determinant of InsP3R tetramerization (Joseph et al. 1997; Galvan, D., E. Borrego-Diaz, P.J. Perez, and G.A. Mignery, submittedmanuscript for publication). This region of the InsP3R has three notable attributes. First, the sixth TMR is a point of very high sequence homology with the RyR channel (Mignery et al. 1989). Second, the fifth and sixth TMRs contain a putative leucine zipper motif that could be important for stable tetramerization and/or pore formation (Galvan, D., E. Borrego-Diaz, P.J. Perez, and G.A. Mignery, manuscript submitted for publication). Third, the lumenal 5–6 loop has been proposed to be analogous to the H-loop of voltage-activated Ca2+, Na+, and K+ channels (Mignery and Südhof 1993).
It is reasonable to hypothesize that the fifth and sixth TMR and the interceding loop are the most likely region of the InsP3R to form the trans-ER ion permeation pathway. To experimentally test this hypothesis, single channel function of different type 1 InsP3R TMR-deletion mutants was defined in planar bilayer studies. Our data indicate that the fifth and sixth TMRs and the interceding loop contain important structural determinants of the InsP3R channel's permeation pathway that govern its conduction and selectivity. Our data also suggest that the 1–4 TMR region of the pInsP3R contains important sequence and/or structural elements that regulate gating of the pore.
|
Materials and Methods
|
|---|
Materials
[3H]InsP3 (21 Ci/mmol) was obtained from Du Pont-New England Nuclear. Unlabeled InsP3 was purchased from LC Laboratories Inc., and heparin was from Sigma Chemical Co. Ryanodine was purchased from Calbiochem Corp. Lipids, L-
-phosphatidylcholine, L--phosphatidylethanolamine, and L--phosphatidylserine were obtained from Avanti Polar Lipids.
Plasmid Construction and Expression
The full-length type 1 plasmid (pInsP3R-T1) was assembled from overlapping cDNA clones isolated from a rat brain library as previously described (Mignery et al. 1990). This plasmid is identical to the pInsP3R plasmid we used previously (Ramos-Franco et al. 1998b). Two TMR-deletion plasmids were also used in this study. The expression plasmid pInsP3R
1-4 encoded a protein missing residues 2211–2416. The expression plasmid pInsP3R5-6 encoded a protein missing residues 2398–2589. Construction strategies of these two expression plasmids is described in detail elsewhere (Galvan, D., E. Borrego-Diaz, P.J. Perez, and G.A. Mignery, manuscript submitted for publication). Galvan and co-workers denoted the plasmids as TMR5-6+C and TMR1-4+C, respectively. Here, the plasmids (pInsP3R-T1, pInsP3R1-4, and pInsP3R5-6) were transiently transfected into COS-1 cells. COS-1 cells were transfected with each plasmid or sheared salmon sperm (SS) DNA using the DEAE-dextran method as described by Gorman 1985. The sheared SS DNA was used to mock transfect COS-1 cells and served as a negative control. Cells were incubated at 37°C, 5% CO2 for 48–72 h before harvesting for biochemical and functional analysis. Typical transfection efficiencies were routinely 60% or greater, as determined by indirect immunofluorescence or via green fluorescent reporter chimeras.
CHAPS Solubilization and Gradient Sedimentation
COS-1 cells transfected with either pInsP3R-T1, pInsP3R1-4 and pInsP3R5-6, or the SS DNA were harvested 48–72 h after transfection, and microsomes were prepared as described previously (Mignery et al. 1990). COS-1 cells were washed with PBS, harvested by scrapping into 50 mM Tris-HCl, pH 8.3, 1 mM EDTA, 1 mM 2-mercaptoethanol, 1 mM PMSF, and lysed by 40 passages through a 27-gauge needle. Membranes were pelleted by a 20-min centrifugation (289,000 g), resuspended in buffer, and either used immediately or frozen at –80°C. Microsomal fractions were solubilized in 50 mM Tris-HCl, pH 8.3, 1 mM EDTA, 1 mM 2-mercaptoethanol, 1 mM PMSF, 1.8% CHAPS {3-[(3-cholamido-propyl)dimethylammonio]-1-propanesulfonate} on ice for 1 h. Insoluble fractions were eliminated by a 10-min centrifugation at 289,000 g, and the supernatant containing solubilized receptor was fractionated through 5–20% sucrose (wt/vol) gradients as previously described (Mignery et al. 1989). COS-1 cell microsomes and sucrose gradient fractions were analyzed by 5% sodium dodecyl sulfate PAGE as described previously (Mignery et al. 1990), followed by immunoblotting with the InsP3R antibody and detected using chemiluminescence reagents (Amersham Life Sciences, Inc.). Gradient fractions containing recombinant receptor protein were reconstituted into proteoliposomes as previously described (Ramos-Franco et al. 1998a).
InsP3 Binding Assays
[3H]InsP3 ligand binding assays were performed as previously described (Mignery et al. 1990). Binding assays were performed using 50 µg of membrane protein in 100 µl of 50 mM Tris-HCl, pH 8.3, 1 mM EDTA, 1 mM 2-mercaptoethanol, 1 mM PMSF containing 9.52 nM [3H]InsP3 ± 1 µM unlabeled InsP3. Samples were incubated on ice for 10 min, and then the radioactivity of the membrane pellets was determined by scintillation spectrometry. All assays were performed in quadruplicate and replicated three times.
Single Channel Assay
Planar lipid bilayers were formed across a 150-µm diameter aperture in the wall of a Delrin partition as described (Ramos-Franco et al. 1998b). Lipid bilayer–forming solution contained a 7:3 mixture of phosphatidylethanolamine and phosphatidylcholine dissolved in decane (50 mg/ml). Proteoliposomes were added to the solution on one side of the bilayer (defined as the cis chamber). The other side was defined as the trans chamber (virtual ground). Standard solutions contained 220 mM CsCH3SO3 cis (20 mM trans), 20 mM HEPES, pH 7.4, and 1 mM EGTA {[Ca2+]FREE = 250 nM; Ca2+ added as Ca(CH3SO3)2}. The [Ca2+]FREE was verified using a Ca2+ electrode. The Ca2+ electrodes were comprised of the Ca ligand ETH 129 in a polyvinylchloride membrane at the end of small (2 mm) polyethylene tube. These Ca2+ minielectrodes were made and used as described previously (Baudet et al. 1994). A custom amplifier was used to optimize single-channel recording. Acquisition software (pClamp; Axon Instruments), an IBM compatible 486 computer, and a 12 bit A/D-D/A converter (Axon Instruments) were used. Single channel data were digitized at 5–10 kHz and filtered at 2 kHz. Ligands (1 µM InsP3, 50 µg/ml heparin, 10 µM ryanodine) were added symmetrically to reconstituted single channels. Open probability and unitary current amplitude was defined from Gaussian fitting of total amplitude histograms. Selectivity was defined under bionic conditions. The trans solution contained 30 mM Ca2+ and the cis solution contained 30 mM Cs+ as charge carrier. Unitary current was recorded during the application of a voltage ramp protocol (–50 to +50 mV over 3 s). To calculate the Ca2+/Cs+ selectivity ratio (PCa/PCs), the apparent reversal potential (Erev) was measured from constant field equation:
 |
where R, T, and F have their usual meanings. Note that [Ca]trans/[Ca]cis was equal to one.
|
results
|
|---|
Expression and Isolation of Type 1 InsP3R TMR-deletion Mutants
The full-length (pInsP3R-T1) and TMR-deletion (pInsP3R1-4 and pInsP3R5-6) mutants were transfected into COS-1 cells using the DEAE-dextran method (Gorman 1985). A schematic of the full-length and deletion mutants used in this study is shown in Fig. 1 A. The putative pore region of the InsP3R includes the 150 amino acids bounded by the fifth and sixth TMRs. A sequence aligned between the type 1 InsP3R and ryanodine receptor 2 (RyR2) proteins over this region is illustrated in Fig. 1 B. The fifth and sixth InsP3R TMRs are boxed, and identical residues in the InsP3R and RyR sequences are shaded. Residues marked by a star indicate those conserved between all three InsP3R isoforms and RyR2. These expression vectors were under the control of the cytomegalovirus promoter (Mignery et al. 1990), and these plasmids expressed immunoreactive InsP3R protein. Microsomes prepared from COS-1 cells transfected with sheared SS DNA revealed no immunoreactive endogenous receptor protein (Fig. 1 C). Extended exposures of the SS DNA Western blots revealed only low levels of immunoreactive protein (data not shown). Microsomes (10 µg protein) from cells expressing the pInsP3R-T1, pInsP3R1-4, and pInsP3R5-6 plasmids were Western blotted with antibodies directed against the NH2 and COOH termini of the receptor. Blots performed with the COOH-terminus antibody are shown in Fig. 1 C. These data indicate that the expressed InsP3R proteins were of the expected size and targeted to the correct microsomal fraction. It is possible that over expression of recombinant InsP3R may induce elevated expression of endogenous receptor. This possibility was thoroughly examined and dispelled in a previous study (Ramos-Franco et al. 1998b). The absence of full-length receptor in the TMR-deletion mutant lanes of the Western blot indicates that there was no substantial upregulation of endogenous type 1 receptor in this study (Fig. 1 C). The abundance of recombinant InsP3R protein (compared with the endogenous receptor) was comparable with that observed in our previous recombinant InsP3R studies (Ramos-Franco et al. 1998b). Thus, proteoliposomes prepared from transfected COS-1 cells contain predominantly recombinant protein. These proteoliposomes can then be reconstituted into planar lipid bilayers to define the single channel properties of the mutant InsP3R channels. This strategy to define the function of recombinant InsP3R channels has been successfully applied by two laboratories (Kaznacheyeva et al. 1998; Ramos-Franco et al. 1998b).
InsP3 Binding of the Type 1 InsP3R TMR-deletion Mutants
Equilibrium InsP3 binding assays were performed using microsomal proteins from transfected COS-1 cells (Table ). The full-length recombinant receptor (pInsP3R-T1) and both TMR-deletion mutants (pInsP3R1-4 and pInsP3R5-6) bind significant amounts of InsP3. The SS DNA control microsomes did not bind InsP3 at significant levels above nonspecific background. The amount of InsP3 bound was normalized to the relative protein expression of each InsP3R construct by densitometry. These results are consistent with previous studies in which microsomes of transfected COS-1 cells contained abundant amounts of immunoreactive receptor protein and bound significant amounts of [3H]InsP3 (Mignery et al. 1990). In each case, the level of InsP3 binding was reduced in the presence of heparin or unlabeled InsP3. These data indicate that the expressed TMR-deletion mutant proteins are functional in terms of InsP3 binding.
Single Channel Properties of the Type 1 InsP3R TMR-deletion Mutants
The TMR-deletion mutant InsP3R receptor proteins were incorporated into proteoliposomes for fusion into planar lipid bilayer studies. Microsomes from COS-1 cells transfected with either pInsP3R-T1, pInsP3R1-4, pInsP3R5-6, or control SS DNA were solubilized in CHAPS detergent and sedimented over 5–20% sucrose density gradients. The tetrameric receptor complex (i.e., the channel complex) migrates to a position on the gradient beyond the majority of other proteins (Mignery and Südhof 1990). Its position in the gradient was detected by Western immunoblotting and fractions containing the highest levels of recombinant receptor reconstituted into L--phosphatidylcholine and L--phosphatidylserine containing liposomes as described previously (Perez et al. 1997; Ramos-Franco et al. 1998b).
No detectable InsP3/heparin-sensitive Cs+ conducting channels were incorporated into the bilayer after fusion of proteoliposomes containing gradient receptor fractions from nontransfected COS-1 cells, control (SS DNA)-transfected cells, or pInsP3R5-6–transfected cells. Incorporation of proteoliposomes containing the pInsP3R1-4 construct resulted in the appearance of a high conductance (
300 pS) ion channel with high open probability (>0.80). Sample single channel activity from the pInsP3R1-4 channel is shown in Fig. 2. The pInsP3R1-4 channel was nearly always open with frequent and usually brief (1 ms) flickers to the close state. Long closed events (>20 ms) were rare. Single channel activity was observed in the presence (Fig. 2 A) and absence (Fig. 2 B) of InsP3. Single channel activity was also not impacted by the addition of 10 µM ryanodine or 50 µg/ml heparin (Fig. 2 C). Corresponding total amplitude histograms under each experimental condition are also presented in Fig. 2. The channel was open most of the time with frequent but brief transitions to the closed state. Thus, these data suggest that the pore formed by the pInsP3R1-4 protein was not modulated by agents (i.e., InsP3 and heparin) that modulate function of wild-type InsP3R channels. Additionally, the pInsP3R1-4 pore was constitutively open (i.e., high Po, n = 6). Under optimal experimental conditions, the Po of full-length type 1 InsP3R channels is relatively low (Bezprozvanny et al. 1991; Ramos-Franco et al. 1998a,Ramos-Franco et al. 1998b). The high Po and absence of channel regulation by InsP3 and heparin indicates that the channel activity observed is not due to endogenous InsP3R channels. The absence of ryanodine action indicates that it is not due to endogenous RyR channels.
View larger version (51K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2 Single channel currents through the pore formed by the InsP3R TMR-deletion mutant pInsP3R1-4. This deletion mutant is missing the first through fourth TMRs. Representative single channel records obtained with 220 mM CsCH3SO3 as the current carrier. The membrane potential was 0 mV. The open and closed current levels are indicated. The data shown are representative of data collected on six different single channels. Single channel activity monitored in the presence (A) and absence (B) of InsP3, and in the presence of heparin (C). The associated amplitude histograms were obtained from several minutes of recording under each experimental condition. Solid lines represent the best fits of the two Gaussian distributions. Note the consistently high Po and similar unitary currents under all experimental conditions.
| |
The permeation properties of the pInsP3R1-4 pore were also defined. Stationary single channel activity was recorded for extended periods (5 min) at several different membrane potentials. The unitary current amplitude (Cs+ charge carrier) was measured as a function of membrane potential. Sample single channel records at different membrane potential (0, 20, and 40 mV) are shown in Fig. 3 A. A sustained high Po was a fundamental feature the pInsP3R1-4 pore at all membrane potentials tested. The activity of this mutant channel was voltage dependent. For example, the Po increased from 85 to 95% when the membrane potential was changed from 0 to 40 mV. This modest voltage dependency of channel activity appears to be a persistent and consistent feature unique to the mutant InsP3R channel.
View larger version (36K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3 The conductance and selectivity of the pInsP3R1-4 pore. (A) Sample single channel activity at three different voltages (0, 20, and 40 mV) recorded with Cs+ as a charge carrier. The open and closed current levels are indicated. Calibration bars represent 1 s and 20 pA. (B) Current–voltage relationship in asymmetrical (220/20 mM) CsCH3SO3 solutions. Each point represents mean (±SEM) of several (three to nine) determinations on different channels. Points without error bars had error levels within the area of the symbol. (C) Current–voltage relationship in asymmetrical Ca2+ solutions [0/30 mM Ca(CH3SO3)2]. Each point represents mean (±SEM) of several (three to seven) determinations on different channels. (D) Current–voltage relationship measured under biionic conditions (30 mM Ca2+ trans, 30 mM Cs+ cis). Each point represents mean (±SEM) of several (4–10) determinations on different channels. Arrow indicates the extrapolated reversal potential.
| |
The unitary Cs+ current carried by the pInsP3R1-4 pore was Ohmic with a slope conductance of 284 pS (n = 9). Unitary Cs+ current reversed at –22 mV, indicating that the pInsP3R1-4 pore was cation selective (Fig. 3 B). The unitary Ca2+ current was also Ohmic at relatively large negative membrane potentials, with a slope conductance of 60 pS (Fig. 3 C; n = 7). The selectivity of the channel was probed under biionic conditions (Fig. 3 D). In brief, 30 mM Ca2+ was applied to one side of the channel and 30 mM Cs+ was applied to the other. The selectivity between Ca2+ and Cs+ can then be calculated from the reversal potential (see METHODS). The reversal potential was near +40 mV (n = 10), indicating the pInsP3R1-4 pore was Ca2+ selective (PCa/PCs 6.3). Thus, the pInsP3R1-4 protein forms a high conductance and Ca2+ selective pore.
|
Discussion
|
|---|
The principal functional attribute of the InsP3R is its capacity to operate as an intracellular Ca2+ release channel. The permeation and InsP3 regulation of the native type 1 InsP3R pore have been defined in bilayer studies (Bezprozvanny and Ehrlich 1994). The permeation and regulatory properties of the native and recombinant InsP3R channels are comparable (Kaznacheyeva et al. 1998; Ramos-Franco et al. 1998a,Ramos-Franco et al. 1998b). The InsP3R is a high conductance, poorly selective Ca2+ channel. It is activated by InsP3 (1 µM) and blocked by heparin (Bezprozvanny and Ehrlich 1994; Hagar et al. 1998; Ramos-Franco et al. 1998a). In the presence of 1 µM InsP3 (250 nM Ca2+), the native type 1 InsP3R channel has a relatively low open probability (0.15; Bezprozvanny et al. 1991). The InsP3R channels are permeable to a variety of monovalent (e.g., K+, Na+, and Cs+) and divalent cations (e.g., Ca2+, Ba2+, and Mg2+). The main conductance is near 300 pS for monovalent ions and 60–80 pS for divalent cations (Hagar et al. 1998; Ramos-Franco et al. 1998a). The channel is also remarkable for its relatively poor selectivity. The estimated permeability ratio (divalent/monovalent) of the InsP3R channel pore is near 6 (Bezprozvanny and Ehrlich 1994). Surface membrane channels (e.g., L-type Ca2+ channel) typically have PDIVALENT/PMONOVALENT > 1,000 (Lee and Tsien 1982). The high conductance and poor selectivity of the InsP3R channel is similar to that of the RyR Ca2+ release channel (Tinker and Williams 1992; Bezprozvanny and Ehrlich 1994). This is interesting because the transmembrane regions of the InsP3R and RyR share significant (40%) primary cDNA sequence homology (Mignery et al. 1989). Thus, the structural determinants defining the ion permeation pathway may be similar in the InsP3R and RyR channels.
The original analysis of the InsP3R cDNA suggested the existence of a channel-forming domain near the COOH terminus of the protein (Mignery et al. 1989). This suggestion was based on hydropathy and sequence homology to the RyR protein. It is also clear that the InsP3R protein oligomerizes (i.e., tetramerizes) to form the functional Ca2+ release channel entity (Mignery et al. 1989; Sayers et al. 1997) and that the TMRs are involved in targeting and stabilizing the oligomer (Mignery and Südhof 1990; Galvan, D., E. Borrego-Diaz, P.J. Perez, and G.A. Mignery, submittedmanuscript for publication). Channel assembly is thought to be a multideterminant process involving interplay between the TMRs and the COOH terminus. Two studies have suggested that the fifth and sixth TMRs are key elements that stabilize the InsP3R tetramer (Joseph et al. 1997; Galvan, D., E. Borrego-Diaz, P.J. Perez, and G.A. Mignery, submittedmanuscript for publication). The loop that links the fifth and sixth TMRs has been proposed to be analogous to the H loop of voltage-activated Ca2+, Na2+, and K+ channels (Mignery and Südhof 1993). A similar suggestion has been made for the corresponding sequence of the RyR protein (Balshaw et al. 1999). Balshaw et al. 1999 proposed that the region of the RyR protein bounded by its two most COOH-terminal TMRs contains a pore-forming segment analogous to the H loop. Point mutations in this region of RyR1 are known to modify channel function (Gao et al. 1999; Lynch et al. 1999). The fifth and sixth TMRs of the InsP3R may also contain a putative leucine zipper motif. The presence of a leucine zipper could confer a degree of structural rigidity that may be important in stabilizing a pore through coiled-coil interactions (Simmerman et al. 1996). Thus, it is reasonable to hypothesize that the fifth and sixth TMRs (and the interceding lumenal loop) are the most likely region of the InsP3R to form the Ca2+-selective pore.
Deletion of the sequence bounded by the fifth and sixth TMRs (i.e., the pInsP3R5-6 mutant) did not form detectable Ca2+ channels. However, this mutant protein did occasionally (15% of attempts) induce a very small, sustained nonspecific leak current. The leak current reversed at 0 mV and no clear opening or closing events were observed. Thus, the leak current was not attributed to the opening and closing of an ion channel. It is more likely that this leak current was due to destabilization of the bilayer after incorporation of integral non–channel-forming protein. A similar leak current was observed in a previous study with our pInsP3RT1ALT construct (Ramos-Franco et al. 1998b). This type 1 InsP3R construct codes a truncated protein missing the 310 amino-terminal amino acids of the InsP3 binding domain. This is interesting because the pInsP3RT1ALT construct contains all six TMRs. The implication is that the pInsP3RT1ALT mutant formed a constitutively closed channel, while the pInsP3R5-6 mutant formed a constitutively open Ca2+ release channel.
Deletion of the first four TMRs (i.e., the pInsP3R1-4 mutant) did form high conductance fast gating ion channels. Control experiments with SS cDNA–transfected cells indicated that the appearance of these channels was not due to some endogenous COS-1 cell protein or factor. The activity of the pInsP3R1-4 channel was not modified by the addition of InsP3 or heparin. This is interesting because the protein binds InsP3, and this binding is blocked by heparin. In the absence of pharmacological tools, channel identity was thus confirmed by its permeation profile. The pInsP3R1-4 channel was permeable to both monovalent (i.e., Cs+) and divalent (Ca2+) cations. The Cs+ and Ca2+ conductances were 280 and 60 pS, respectively. The channel was cation selective, with a Ca2+/Cs+ permeability ratio of 6.3. These values match those described for the wild-type InsP3R channel (Bezprozvanny and Ehrlich 1994). This suggests that amino acid residues 2398–2589 (i.e., fifth and sixth TMR and interceding loop) contains key determinants of the InsP3R's permeation pathway.
The absence of InsP3 regulation despite InsP3 binding suggests that the deleted sequence (TMRs 1–4, residues 2211–2416) may couple binding to channel gating. Alternatively, the missing sequence may annul ligand regulation by sterically limiting pore structure. For example, removing surrounding TMRs could energetically restrict molecular motions in pore structure needed for normal ligand regulation. Such restricted molecular motion could be manifested as a constitutively open pore.
In summary, this study has localized the InsP3R pore to a region of 191 amino acids near the COOH terminus of the protein. This region includes the fifth and sixth TMR and interceding loop. We suggest that a putative leucine zipper may infer the structural integrity needed to form a stable pore. A sequence alignment between the RyR and the InsP3R pore-forming region reveals potential "hot spots" for future mutagenesis studies. These hot spots include the valine/isoleucine residues (i.e., the β-branched amino acids) in the fifth and sixth TMR and the cluster of conserved glycines in the fifth to sixth TMR loop.
1used in this paper: ER, endoplasmic reticulum; InsP3R, inositol 1,4,5-trisphosphate receptor; RyR, ryanodine receptor; SS, salmon sperm; TMR, transmembrane region
|
ACKNOWLEDGMENTS
|
|---|
We thank Dr. Thomas Südhof for the kind gift of the InsP3R cDNA.
This work was supported by National Institutes of Health grants R29-MH53367 and R01-HL58851 (G.A. Mignery), and R01-HL570832 (M. Fill). M. Fill is an American Heart Association established investigator.
Submitted: 17 May 1999
Revised: 23 June 1999
Accepted: 24 June 1999
|
REFERENCES
|
|---|
Baudet S., Hove-Madsen L. & Bers D.M.. How to make and use calcium-specific mini- and microelectrodes, Methods Cell Biol., 40, 1994, 9–114.
Balshaw D., Gao L. & Meissner G.. Luminal loop of the ryanodine receptora pore-forming segment?, Proc. Natl. Acad. Sci. USA, 96, 1999, 3345–3347.[Free Full Text]
Bezprozvanny I., Watras J. & Ehrlich B.E.. Bell-shaped calcium-response curves of Ins(1,4,5)P3 and calcium-gated channels from endoplasmic reticulum of cerebellum, Nature., 351, 1991, 751–754.[Medline]
Bezprozvanny I. & Ehrlich B.E.. InsP3 receptorfunctional properties and regulation, Peracchia C.. In Handbook of Membrane Channels, 1994, 511–526, Academic Press, NY, NY.
Bezprozvanny I. & Ehrlich B.E.. The inositol 1,4,5-trisphosphate (InsP3) receptor, J. Membr. Biol, 145, 1995, 205–216.[Medline]
De Smedt H., Missiaen L., Parys J.B., Henning R.H., Sienaert I., Vanlingen S., Gijsens A., Himpens B. & Casteels R.. Isoform diversity of the inositol 1,4,5-trisphosphate receptor in cell types of mouse origin, Biochem. J., 322, 1997, 575–583.[Medline]
Gao L., Tripathy A., Xe L., Pasek D., Balshaw D., Xin C. & Meissner G.. Mutation of charged amino acids in a putative lumenal loop of the skeletal muscle Ca2+ release channel results in the loss of high affinity ryanodine binding, Biophys. J., 76, 1999, A303.
Gorman, C., 1985. DNA Cloning. Vol. II. D.M. Glover, editor. IRL Press, Oxford, UK. 143–190.
Hagar R.E., Burgstahler A.D., Nathanson M.H. & Ehrlich B.E.. Type-III InsP3 receptor channel stays open in the presence of increased calcium, Nature., 396, 1998, 81–84.[Medline]
Joseph S.K.. The inositol trisphosphate receptor family, Cell Signal, 8, 1996, 1–7.[Medline]
Joseph S.K., Boehning D., Pierson S. & Nicchitta C.V.. Membrane insertion, glycosylation, and oligomerization of inositol trisphosphate receptors in a cell-free translation system, J.Biol.Chem, 272, 1997, 1579–1588.[Abstract/Free Full Text]
Kaznacheyeva E., Lupu V.D. & Bezprozvanny I.. Single channel properties of inositol trisphosphate receptor heterologously expressed in HEK-293 cells, J. Gen. Physiol., 111, 1998, 847–856.[Abstract/Free Full Text]
Lee K.S. & Tsien R.W.. Reversal of current through Ca2+ channels in dialyzed single heart cells, Nature., 297, 1982, 498–501.[Medline]
Lynch P.J., Tong J., Lehane M., Mallet A., Giblin L., Heffron J.J.A., Vaughan P., Zafra G., MacLennan D.H. & McCarthy T.V.. A mutation in the transmembrane/lumenal domain of the ryanodine receptor is associated with abnormal Ca2+ release channel function and severe central core disease, Proc. Natl. Acad. Sci. USA., 96, 1999, 4164–4169.[Abstract/Free Full Text]
Michikawa T., Hamanaka H., Otsu H., Yamamoto A., Miyawaki A., Furuichi T., Tashiro Y. & Micoshiba K.. Transmembrane topology and sites of glycosylation of the inositol trisphosphate receptor, J. Biol. Chem., 269, 1994, 9184–9189.[Abstract/Free Full Text]
Mignery G.A., Südhof T.C., Takei K. & DeCamilli P.. Putative receptor for inositol 1,4,5-trisphosphate similar to ryanodine receptor, Nature., 342, 1989, 192–195.[Medline]
Mignery G.A. & Südhof T.C.. The ligand-binding site and transduction mechanism in the inositol-1,4,5-trisphosphate receptor, EMBO (Eur. Mol. Biol. Organ.) J., 9, 1990, 3893–3898.[Medline]
Mignery G.A., Newton C.L., Archer B.T. III & Südhof T.C.. Structure and expression of the rat inositol-1,4,5-trisphosphate receptor, J. Biol. Chem, 265, 1990, 12679–12685.[Abstract/Free Full Text]
Mignery G.A. & Südhof T.C.. Molecular analysis of inositol 1,4,5-trisphosphate receptors, Methods Neurosci., 18, 1993, 247–265.
Nakanishi S., Fujii A., Nakade S. & Mikoshiba K.. Immunohistochemical localization of inositol 1,4,5-trisphosphate receptors in non-neural tissues, with special reference to epithelia, the reproductive system and muscular tissues, Cell Tissue Res., 285, 1996, 235–251.[Medline]
Newton C.L., Mignery G.A. & Südhof T.C.. 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.[Abstract/Free Full Text]
Perez P.J., Ramos-Franco J., Fill M. & Mignery G.A.. Identification and functional reconstitution of the type-2 InsP3 receptor from ventricular cardiac myocytes, J. Biol. Chem, 272, 1997, 23961–23969.[Abstract/Free Full Text]
Ramos-Franco J., Fill M. & Mignery G.A.. Isoform specific function of single inositol 1,4,5-trisphosphate receptor channels, Biophys. J., 75, 1998, 834–839a.[Medline]
Ramos-Franco J., Fill M. & Mignery G.A.. Single channel function of recombinant type-1 inositol 1,4,5-trisphosphate receptor ligand binding domain splice variants, Biophys. J., 75, 1998, 2783–2793b.[Medline]
Sayers L.G., Miyawaki A., Muto A., Takeshita H., Yamamoto A., Michikawa T., Furuichi T. & Mikoshiba K.. Intracellular targeting and homotetramer targeting of a truncated inositol 1,4,5-trisphosphate receptor-green fluorescent protein chimera in Xenopus laevis oocytesevidence for the involvement of the membrane spanning domain in endoplasmic reticulum targeting and homotetramer complex formation, Biochem. J., 323, 1997, 273–280.[Medline]
Simmerman H.K.B., Kobayashi Y.M., Autry J.M. & Jones L.R.. A leucine zipper stabilizes the pentameric membrane domain of phospholamban and forms a coiled-coil pore structure, J. Biol. Chem, 271, 1996, 5941–5946.[Abstract/Free Full Text]
Südhof T.C., Newton C.L., Archer B.T., Ushkaryov Y.A. & Mignery G.A.. The structure of a novel InsP3 receptor, EMBO (Eur. Mol. Biol. Organ.) J., 10, 1991, 3199–3206.[Medline]
Tinker A. & Williams A.J.. Divalent cation conduction in the ryanodine receptor of sheep cardiac muscle sarcoplasmic reticulum, J. Gen. Physiol, 100, 1992, 479–493.[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
Top
ABSTRACT
Introduction
