Fig 2: Y306F RtcB mutant rescues a defective XBP1 mRNA splicing.(A) HEK cells were co-transfected with 1 μg of wt IRE1α and 1 μg of WT. or mutant Flag-RtcB plasmid. 24 h later, cell lysis and anti-Flag immunoprecipitation were performed. The immunoprecipitates were blotted for IRE1α, and the input samples for IRE1α, Flag, and actin. The arrowhead denotes the Flag-RtcB protein. (B) Docked complexes of RtcB WT and pY306 variants toward the IRE1 tetramer/XBP1 complex. (Top panels) RtcB WT (green) binds at the loop-binding RNase area of IRE1 (blue) and is perfectly oriented to initiate the ligation of XBP1 (red) upon its cleavage by IRE1. (Lower panels) The interaction area of the phosphorylated RtcB is shifted toward the middle of XBP1 at the IRE1 dimer–dimer interface region and will not be able to initiate the ligation reactions of the spliced ends at the two XBP1 loops (pY306 of RtcB is represented in yellow). The red arrow indicates RtcB position shift on the IRE1 tetramer when Y306 is phosphorylated. (C) In vitro reconstitution of XBP1 mRNA splicing using in vitro transcribed XBP1 mRNA which was incubated with or without 250 ng of the recombinant IRE1 cytosolic domain and with WT or Y306F RtcB-Flag immunoprecipitated from cells. The retrotranscribed XBP1 DNA from the assay was analyzed by PCR using primers recognizing the XBP1 mRNA. (D) Immunoprecipitated RtcB levels were detected by immunoprecipitating the cell lysates using anti-Flag antibody–conjugated beads and immunoblotting of the immunoprecipitates with anti-Flag antibodies. (C, D, E) XBP1s (C) was then normalized to RtcB protein levels (D) to obtain the RtcB-specific activity. (F) PTP1B+/+ (WT) and PTP1B−/− (KO) MEFs untransfected (CTL) or transfected with 2 μg of Flag-RtcB-WT or Flag-RtcB-Y306F plasmid were treated 24 h post-transfection with 50 ng/ml Tunicamycin for 0, 2, 4, 8, 16, and 24 h. Their cDNA was analyzed by PCR using primers recognizing the XBP1 mRNA. (F, G) Quantification of gels in (F). The graph shows the comparison of XBP1 mRNA splicing in CTL cells and cells expressing either Flag-RtcB-WT or Flag-RtcB-Y306F. The composite comparison of the results obtained for ctl (circles) or rescues with either wt RtcB (triangles) or Y306F RtcB (squares) is also shown. (Bar graph) The slope of each curve was calculated, and for each condition, the slope of the WT cells was subtracted from the one of the KO cells, called as Δslope. The calculation of the Δslope was also corrected by the expression levels of Flag-RtcB-WT or Flag-RtcB-Y306F as determined by Western blotting. Data information: Data values presented in (E) represent two independent experiments. Data values in (G) are the mean ± SEM of three independent experiments. The unpaired t test was applied for the statistical analyses (ns, nonsignificant, *P < 0.05, **P < 0.01).Source data are available online for this figure.

Fig 3: SiRNA screening and in vitro IRE1α cleavage assay identify PTP1B as both XBP1 splicing regulator and regulated IRE1-dependent decay target.(A) siRNA-based screening assay: HEK293T cells were transfected with siRNA sequences against genes encoding ER proteins. They were subsequently transfected with an XBP1s-luciferase reporter (Fig S1A), and after 48 h and the induction of ER stress, the cells were tested for the intensity of light signal after the addition of luciferin. (A, B) Luminescence quantification of the screening assay described in (A). (A, B, C) Venn diagram of the gene list resulting from the assay described in (A, B), a list of possible regulated IRE1-dependent decay substrates obtained from an in vitro IRE1α cleavage assay (Fig S1B) and a genome-wide siRNA-based screening assay (described in Yang et al [2018]). (C, D) A schematic network of the genes in the intersections of the lists as described in (C). (E) PTP1B+/+ (WT) and PTP1B−/− (KO) MEFs were treated for 0 and 6 h with 10 μg/ml TM. RNA was isolated, and the resulting samples were analyzed with qPCR for spliced and total XBP1 mRNA levels. The bar graph presents the tunicamycin-induced fold change in XBP1 mRNA splicing between PTP1B+/+ (WT, blue) and PTP1B−/− (KO, red). Data information: Data values are the mean ± SEM of four independent experiments. The unpaired t test was applied for the statistical analyses comparing the values (mean ± SEM) for the 0 and 6 h time points. **P = 0.0074; *P = 0.0187. Total XBP1 mRNA qPCR primers amplifying the region before the cleavage site or the region spanning the cleavage site were used (Fig S1C). (F, G) PTP1B mRNA expression levels in U87 cells expressing or not a dominant negative form of IRE1α at 0 h time point (F) and during DTT treatment (1 mM) with a 2-h Actinomycin D (5 μg/ml) pretreatment (G). EV, empty vector; IRE1 DN, dominant negative (cytosolic-deficient) form of IRE1α.

Fig 4: RtcB is a substrate of the tyrosine kinase c-Abl and the tyrosine phosphatase PTP1B.(A) HEK293T cells were left non-transfected (CTL) or transfected with 1 μg of the wt RtcB-Flag and 1 μg of either the wt PTP1B plasmid or C215S mutant one. Immunoprecipitation (IP) was carried out in the cell lysates with the PTP1B antibody, the immunoprecipitates were immunoblotted for RtcB, and the membrane was re-probed with PTP1B. Inputs probed for RtcB, Flag, PTP1B, and VCP are shown. (B) PTP1B+/+ or PTP1B−/− MEFs were left untransfected (CTL) or transfected with 2 μg wt RtcB-Flag and treated with 15 μM bpV(phen) for 2 h. IP was performed in the cell lysates using Flag Ab, the immunoprecipitates were immunoblotted for phosphotyrosine, and the membrane was re-probed with Flag. Input samples probed for Flag, RtcB, PTP1B, pY-HRP, and VCP are shown. Black arrowheads indicate the RtcB-Flag protein, and white arrowheads indicate an unspecific band at 55 kD. (C) Samples from a scaled up in vitro kinase reaction containing recombinant human c-ABL and RtcB were analyzed using mass spectrometry. Fragment spectra corresponding to three different phospho-peptides containing tyrosine residues are depicted (y ions are shown in blue and b ions in red). (D) HEK293T cells transfected with the wt RtcB-Flag were treated 24 h post-transfection with 10 μM of tyrosine kinase inhibitors afatinib, crizotininb, dasatinib, and imatinib for 8 h and 15 μM bpV(phen) for 2 h. IP was performed in the cell lysates using Flag antibody, and the immunoprecipitates were first immunoblotted for RtcB-pY475 and then re-probed with Flag Ab. Input samples probed for pY-HRP, Flag, and VCP are shown. (D, E) The levels of phosphorylated RtcB (D) were normalized to RtcB protein levels (D). (F) HEK293T cells transfected with the Flag-RtcB-WT were treated 24 h post-transfection with 1 μg/ml TM for 6 h or transfected with c-ABL siRNA for 2 d. IP was carried out in the cell lysates using Flag antibody, and the immunoprecipitates were first immunoblotted for pY and then re-probed with Flag antibodies. Input samples probed for pY-HRP, Flag, c-ABL, XBP1s, and actin are shown. (F, G) The levels of phosphorylated RtcB (F) were normalized to RtcB protein levels (F). Data information: The blots shown are representative of three or more independent experiments. Data shown in the graphs correspond to the mean ± SEM of n = 3 independent experiments. One-way ANOVA was applied for the statistical analyses (**P < 0.01).Source data are available online for this figure.

Fig 5: Impact of RtcB tyrosine phosphorylation on XBP1 mRNA splicing and interaction with c-ABL.(A) PTP1B+/+ (wt) and PTP1B−/− (KO) MEFs were treated with 10, 50, or 100 ng/ml tunicamycin for 0, 2, 4, 8, 16, and 24 h. Their cDNA was analyzed by PCR using primers recognizing the XBP1 mRNA. (A, B) Quantification of the gels in (A). (C) PTP1B+/+ (wt) and PTP1B−/− (KO) MEFs untransfected (CTL) or transfected with 2 μg of wt or Y306F RtcB-Flag plasmids were treated 24 h post-transfection with 10 μg/ml tunicamycin for 6 h. The resulting RNA was analyzed using RT-qPCR for XBP1s and XBP1 total, thereby allowing XBP1 mRNA splicing using the ratio of both values. XBP1 mRNA splicing is represented as the ratio of the XBP1 splicing in the WT toward WT MEFs that equals 1 for every condition, and this ratio in the KO toward the WT cells for each condition. (D) Protein samples of the untreated cells in Fig 6F were analyzed for the expression of the RtcB-Flag protein, PTP1B, and VCP (loading control) using immunoblotting. (E) Protein complex after docking of c-ABL to the Y306 site of RtcB and 200-ns molecular dynamics simulation. Ribbon-like structures are shown. (F) Zoom-in on the interacting residues D325 and D444 of c-ABL (in green) and K279, R283, and K357 of RtcB (in red). (G) cDNA samples from Fig 7A were analyzed for endogenous RTCB mRNA levels using suitable primers. U6 snRNA was used as the loading control. (H) Protein samples from Fig 7A were analyzed for expression of the RtcB-Flag protein and calnexin which served as a loading control. The blots shown here are representative of three or more independent experiments. Data represented are the mean ± SEM of n ≥ 3 independent experiments. The unpaired t test was applied for the statistical analyses (ns, nonsignificant, *P < 0.05, ****P < 0.0001). (B) Error bars for (B) are not depicted.Source data are available online for this figure.