Protein phosphatase-1 is a novel regulator of the interaction between IRBIT and the inositol 1,4,5-trisphosphate receptor

Abstract

IRBIT is an IP3R [IP3 (inositol 1,4,5-trisphosphate) receptor]-binding protein that competes with IP3 for binding to the IP3R. Phosphorylation of IRBIT is essential for the interaction with the IP3R. The unique N-terminal region of IRBIT, residues 1-104 for mouse IRBIT, contains a PEST (Pro-Glu-Ser-Thr) domain with many putative phosphorylation sites. In the present study, we have identified a well-conserved PP1 (protein phosphatase-1)-binding site preceeding this PEST domain which enabled the binding of PP1 to IRBIT both in vitro and in vivo. IRBIT emerged as a mediator of its own dephosphorylation by associated PP1 and, hence, as a novel substrate specifier for PP1. Moreover, IRBIT-associated PP1 specifically dephosphorylated Ser68 of IRBIT. Phosphorylation of Ser68 was required for subsequent phosphorylation of Ser71 and Ser74, but the latter two sites were not targeted by PP1. We found that phosphorylation of Ser71 and Ser74 were sufficient to enable inhibition of IP3 binding to the IP3R by IRBIT. Finally, we have shown that mutational inactivation of the docking site for PP1 on IRBIT increased the affinity of IRBIT for the IP3R. This pinpoints PP1 as a key player in the regulation of IP3R-controlled Ca2+ signals.

Figure 1. Conservation and identification of the PP1 docking site in IRBIT that allows in vitro binding to PP1

(A) Sequence alignment of the unique N-terminal region of IRBIT from H. sapiens (GenBank® accession number NM_006621), M. musculus (GenBank® accession number AB092504), X. laevis (GenBank® accession number BC081269), D. rerio (GenBank® accession number NP_958450) and D. melanogaster (GenBank® accession number NM_139489). The PP1 consensus docking site, and one PKD and four CK1 (broken line) in silico phosphorylation consensus sequences, are indicated. {P} in the PP1 consensus docking site indicates that this residue can be any residue except proline [20]. Note that the in vitro CK1 phosphorylation cascade differs from the in silico-predicted cascade shown here (see the Results section). Neither the PP1 consensus docking site nor the PKD consensus docking site are conserved in D. melanogaster. (B) Purified GST, GST–IRBIT, GST–IRBIT^1–104, GST–IRBIT⁷⁴-⁵³⁰, GST–IRBIT^{1–104(I42A/F44A)} and GST–IRBIT^{1–104(K40R/Q41R/I42V/Q43R)} were subjected to the trypsin-releasable phosphorylase a phosphatase assay, as described in the Materials and methods section. Values are means±S.E.M. of the relative phosphorylase a phosphatase activities (n=3).

Figure 2. IRBIT and PP1 interact in vivo

(A) COS-1 cells were co-transfected with pEGFP-PP1α and either the pTRACER-IRBIT constructs [for overexpression of untagged IRBIT or IRBIT^(I42A/F44A)] or the empty vector (for endogenous IRBIT). IRBIT was immunoprecipitated from the cell lysates, and the trypsin-releasable phosphorylase a phosphatase activity of PP1 was measured in the immunoprecipitates. The PP1 activities of the immunoprecipitates in the absence of IRBIT antibodies were subtracted, and results are expressed as means±S.D. of the resulting specific PP1 activity (n=3). (B) HeLa cells were transiently co-transfected with pEGFP-PP1γ1 and pEXPR-IBA103-IRBIT (wild-type or I42A/F44A mutant) or empty vector pEXPR-IBA103 as a control. After 24 h, cells were analysed for GFP fluorescence. Scale bar, 20 μm.

Figure 3. Phosphorylation of Ser⁶⁸ enables phosphorylation by CK1 and inhibition of IP₃ binding, and PP1 specifically dephosphorylates Ser⁶⁸

(A) GST–IRBIT^1–104 [wild-type (WT)] and GST–IRBIT^{1–104(S68A)} (S68A) or GST–IRBIT^{1–104(S68D)} (S68D) mutants were phosphorylated with PKD and/or CK1 (see the Materials and methods section). The autoradiogram (indicated as ³²P) and Coomassie Blue staining (indicated as CB) from a typical gel are shown (representative of three). (B) GST–IRBIT^1–104 was phosphorylated with PKD and/or CK1 (see the Materials and methods section). Subsequently, buffer or PP1 (indicated as-PP1 or +PP1 respectively) was added and the sample was incubated for another 10 min at 30 °C. The autoradiogram (indicated as ³²P) or Coomassie Blue staining (indicated as CB) from a typical gel are shown (representative of three). (C) The inhibition of [³H]IP₃ binding to 1.5 μg of purified IP₃R1^1–581 by 250 nM GST–IRBIT phosphorylated by PKD (+PKD) and PKD and CK1 (+PKD +CK1). Addition of non-phosphorylated IRBIT was used as a control (0% inhibition). A sample that was first phosphorylated by PKD and CK1 and subsequently incubated with PP1 (see the Materials and methods section) was also used (+PKD+CK1+PP1). Binding was measured at pH 7.4 in the presence of 1 mM EDTA and 1.5 nM [³H]IP₃. Values are expressed as means±S.E.M. for three independent experiments, each assayed five times. Individual groups were compared by an unpaired Student's t test (P<0.01).

Figure 4. IRBIT inhibits the activity of PP1 towards phosphorylase a phosphatase as a substrate

Various concentrations of purified GST–IRBIT were incubated with PP1 in the presence of 50 μM synthetic decapeptide NIPP1¹⁹⁷-²⁰⁶ (RVTF; ●) or the variant NIPP1^{197–206(V201A/F203A)} (RATA; ○; see the text for more details), or buffer (□). The spontaneous phosphorylase a phosphatase activity is shown as means±S.E.M. (n=3). Addition of buffer was used as a control.

Figure 5. IRBIT targets Ser⁶⁸ for dephosphorylation by associated PP1

(A) GST–IRBIT and GST–IRBIT^(I42A/F44A) were phosphorylated by PKD. Subsequently, PP1 was added, and samples were taken at the indicated time points. (B) GST–IRBIT was phosphorylated by PKD in the presence of the synthetic decapeptide NIPP1¹⁹⁷-²⁰⁶ (RVTF) or the variant NIPP1^{197–206(V201A/F203A)} (RATA). Subsequently, PP1 was added, and samples were taken at the indicated time points. For both panels, the autoradiogram (indicated as ³²P) or Coomassie Blue staining (indicated as CB) from a typical gel is shown (n=3).

Figure 6. Binding of PP1 to IRBIT decreases the IRBIT–IP₃R interaction

COS-1 cells were transiently transfected with pEXPR-IBA103-IRBIT or pEXPR-IBA103-IRBIT^(I42A/F44A). Cells were lysed, and the cleared lysate was used as a reference (indicated as INPUT). The GST pull-down assay with purified GST or GST–IP₃R1^1–604 was performed as described in the Materials and methods section. The amount of sample used in lanes 2, 3, 5 and 6 was 5-fold higher than in lanes 1 and 4. IRBIT present in the eluates was detected using the HRP-conjugated Strep-Tactin antibody. Experiments were repeated four times and a representative blot is shown.

Figure 7. Proposed functions of PP1 in IP₃R signalling

Proposed model in which IP₃ competes with IRBIT for binding to the N-terminus of IP₃R (residues 1–604). Binding of IP₃ is favoured as it has a higher affinity for the IP₃R. The binding of IP₃ induces Ca²⁺ release from the ER (endoplasmic reticulum). Ca²⁺ can then activate the phosphorylation cascade on Ser⁶⁸, Ser⁷¹ and Ser⁷⁴. We propose that PKD, CaMKI/II/IV, AMPK and/or MK-2 are probable candidates to phosphorylate Ser⁶⁸ in response to Ca²⁺ release. PP1 is bound to the N-terminal KQIQF motif on IRBIT, and this binding of PP1 allows for efficient dephosphorylation of Ser⁶⁸. PP1 also directly interacts with the C-terminus of the IP₃R (residues 2590–2749) and indirectly via AKAP9 to the regulatory domain of the IP₃R.