Fragmented inositol 1,4,5-trisphosphate receptors retain tetrameric architecture and form functional Ca2+ release channels

Abstract

Inositol 1,4,5-trisphosphate receptor isoforms are a family of ubiquitously expressed ligand-gated channels encoded by three individual genes. The proteins are localized to membranes of intracellular Ca(2+) stores and play pivotal roles in Ca(2+) homeostasis. Previous studies have demonstrated that IP3R1 is cleaved by the intracellular proteases calpain and caspase both in vivo and in vitro. However, the resultant cleavage products are poorly defined, and the functional consequences of these proteolytic events are not fully understood. We demonstrate that IP3R1 is cleaved during staurosporine-induced apoptosis, yielding N-terminal fragments encompassing the ligand-binding domain and the majority of the central modulatory domain together with a C-terminal fragment containing the channel domain and cytosolic tail. Notably, these fragments remain associated with the membrane after initiation of apoptotic cleavage. Furthermore, when recombinant IP3R1 fragments, corresponding to those predicted to be generated by caspase or calpain cleavage, are stably coexpressed in cells, they physically associate and form functional channels. These data provide novel insights regarding the regulation of IP3R1 during proteolysis and provide direct evidence that polypeptide continuity is not required for IP3R activation and Ca(2+) release.

FIGURE 1.

Schematic diagrams of the IP₃R structure. A, representation of the IP₃R1 basic domain structure with functional domains: the suppressor domain (aa residues 1–226), the IP₃ binding core (aa residues 227–577), the modulatory domain (residues 578–2275), the channel domain (aa residues 2276–2589), and the cytosolic tail (aa residues 2589–2749). Also shown are the fragments that result from limited tryptic digestion of IP₃R1, with residues constituting the boundaries between fragments indicated by the numbers below the schematic. The location of antibody epitopes used in this study are also denoted: CS centered around aa residue 40, SC within aa residues 1894–1973, UCD within aa residues 2680–2749, and CT-1 within aa residues 2731–2749. B and C, representations of the N-terminal and C-terminal fragments expected to result from caspase and calpain cleavage of IP₃R1, respectively. D, the recombinant N-terminal and C-terminal fragments of IP₃R2 with α-IP₃R2 antibody epitope locations: NT-2 within aa residues 320–338 and CT-2 within aa residues 2686–2701.

FIGURE 2.

Characterizing IP₃R1 fragmentation pattern during staurosporine-induced apoptosis. A, DT40 cells stably expressing IP₃R1 were treated either as a control or incubated with 2 μm staurosporine for the indicated times. Cells were then harvested and lysed, and equivalent amounts of proteins were fractionated and processed for immunoblot analyses with the CS antibody. The blot was reprobed with an antibody recognizing PKD as a loading control (lower panel). B, DT40 cells stably expressing IP₃R1 were treated either as a control or incubated with 2 μm staurosporine for the indicated times. Cells were then harvested and lysed, and equivalent amounts of proteins were fractionated and processed for immunoblot analyses with the CS antibody. Lysates from DT40–3ko cells, HEK293 cells expressing an IP₃R1 truncation fragment (1–1891), or IP₃R1 truncation mutant fragI-III (residues 1–1582) were used as migration controls. The right panel is a longer exposure of the portion of the same gel. C, a duplicate gel with the same lysates as in B but probed with C-terminal UCD antibody. The right panel is a longer exposure of the portion of the same gel. B and C, bands were visualized using chemiluminescence. Representative experiments are shown. The asterisks denote nonspecific bands seen in 3ko lysates.

FIGURE 3.

Association of fragmented IP₃R1 with membranes. DT40 cells stably expressing IP₃R1 were either treated as a control or incubated with 2 μm staurosporine for 12 h. Cells were then either harvested and lysed (total) or homogenized, and cytosol (cyto) and membrane (mem) fractions were prepared by ultracentrifugation. Equivalent amounts of proteins were fractionated and processed for immunoblot analyses with CS in A, CT-1 in B, and α-GAPDH and α-SERCA in C. A–C, bands were visualized using Licor infrared imaging. D, lysates from DT40 treated as a control or incubated with 2 μm staurosporine for 12 h were prepared using CHAPS lysis buffer. Lysates were then resolved on 3–12% Native PAGE^TM Novex as described under “Materials and Methods.” Proteins were immunoblotted with CS (left panel) or CT-1 (right panel). Molecular weight markers are on the basis of NativeMark^TM unstained protein standard (Invitrogen). Representative experiments are shown. The asterisks denote nonspecific bands.

FIGURE 4.

Establishment of stable DT40 cells lines expressing IP₃R constructs. A, whole cell lysates were prepared from either DT40–3ko (3ko) cells or DT40 cells stably expressing rat IP₃R1 (R1), R1_caspsol, R1_caspmem, or R1_caspsol/mem. Lysates were fractionated on 5% gel and probed with CS (left panel) or UCD (right panel). B, whole cell lysates were prepared from either 3ko or DT40 cells stably expressing rat IP₃R1 (R1), R1_calpsol, R1_calpmem, or R1_calpsol/mem. Lysate were fractionated on 5% gel and probed with CS (left panel) or UCD (right panel). C, whole cell lysates from 3ko cells, DT40 cells stably expressing mouse IP₃R2 (R2), or complementary IP₃R2 fragments (R2-frags) were processed on 5% gel and probed with NT-2 (left panel) or CT-2 (right panel). The arrowheads point to the position of the expressed constructs. The arrow indicates a band that likely represents a degradation or an alternative translation product. The asterisks denote nonspecific bands.

FIGURE 5.

Interaction of recombinant IP₃R fragments. A, DT40 cells expressing R1_caspsol/mem, R1_calpsol/mem, or R2-frags were lysed, and proteins were subjected to mock or fragment-specific immunoprecipitation (IP) followed by Western blot analysis. A, R1_caspsol/mem lysates were immunoprecipitated with CT-1 and probed with either CS specific to the N-terminal fragment (upper panel) or C-terminal fragment-specific antibody (CT-1) (lower panel). B, R1_calpsol/mem lysates were immunoprecipitated with N-terminal fragment antibody (SC) and probed with either CS (upper panel) or CT-1 (lower panel). C, R2-frags lysates were immunoprecipitated with C-terminal antibody (CT-2) and probed with either N-terminal fragment-specific antibody (NT-2) (upper panel) or C-terminal specific CT-2 (lower panel).

FIGURE 6.

Fragmented IP₃Rs migrate as tetrameric complexes on native gels. Whole cell lysates were prepared from either DT40–3ko (3ko) cells or DT40 cells stably expressing rat IP₃R1 (R1), R1_caspsol, R1_caspmem, or R1_caspsol/mem using CHAPS lysis buffer. Lysates were then resolved on 3–12% native gel as described in Fig. 3D. Proteins were immunoblotted with CS in A or CT-1 in B. Parallel aliquots were fractionated on denaturing 5% SDS-PAGE and probed with CS in C or CT-1 in D. The asterisks denote nonspecific bands.

FIGURE 7.

Robust IP₃R-mediated Ca²⁺ release activity in cells expressing protease-generated IP₃R fragments. DT40–3ko (3ko), or DT40 cells stably the expressing various IP₃R constructs as indicated were loaded with the Ca²⁺ indicator Fura-2AM and stimulated with 500 nm trypsin to induce IP₃ formation. Ca²⁺ release was measured as a change in the 340/380 fluorescence ratio. Shown are averaged Ca²⁺ traces of 3ko and IP₃R1 expressing cells in A; R1_caspsol, R1_caspmem and R1_caspsol/mem in B; R1_calpsol, R1_calpmem, and R1_calpsol/mem in C; and IP₃R2 (R2) and R2-frags in D. E, histograms depicting the average change over the basal 340/380 fluorescence ratio resulting from trypsin stimulation in cells expressing the indicated constructs. Experiments were repeated at least three times with more than 40 cells imaged in each run. Data are presented as mean ± S.E.

FIGURE 8.

Stable expression of IP₃R membrane fragments does not significantly alter ER store content. DT40 cells expressing the indicated constructs were exposed to cyclopiazonic acid (CPA) following perfusion in Ca²⁺-free buffer (1 mm EGTA) following the protocol shown in A. The 340/380 ratio (average of the initial 10 s of recording) corresponding to basal [Ca²⁺] was measured prior to removal of extracellular Ca²⁺. Pooled data for the basal ratio are shown in B. C, pooled data of the maximum ratio achieved following cyclopiazonic acid treatment as an indicator of store content. Four experimental runs were conducted for each construct, in which > 30 cells were analyzed per run.

FIGURE 9.

IP₃R Ca²⁺ release activity is restored in cells expressing complementary IP₃R fragments. DT40 cells stably expressing various IP₃R truncation constructs were transiently transfected with complementary constructs. 24 h post-transfection, cells were loaded with Fura-2AM and stimulated with 500 nm trypsin. Ca²⁺ release was measured as a change in the 340/380 fluorescence ratio. Representative traces are shown of DT40 cells stably expressing R1(I+II) and transiently transfected with R1(III+IV+V) in A, DT40 cells stably expressing R1(I+II+III) and transiently transfected with R1(IV+V) in B, DT40 cells stably expressing R1_calpmem and transiently transfected with R2sol fragment (residues 1–1869) in C, and DT40 cells stably expressing R1_calpmem and transiently transfected with R1_caspsol in D.

FIGURE 10.

Expression of complementary IP₃R fragments supports apoptosis. DT40 cells expressing the indicated constructs were incubated for 17 h in the presence or absence of 15 μg/ml IgM, and the IgM-mediated change in cell viability was monitored using the CellTiter A_queous One assay (Promega) as described under “Materials and Methods.” IgM incubation of cells expressing either IP₃R1 or complementary caspase membrane and soluble fractions resulted in a comparable loss of cell viability, which was significantly different from DT40–3ko cells (p > 0.05, Student's t test). The graph shows mean ± S.E. for three experiments performed in triplicate.