TMBIM3/GRINA is a novel unfolded protein response (UPR) target gene that controls apoptosis through the modulation of ER calcium homeostasis

Abstract

Transmembrane BAX inhibitor motif-containing (TMBIM)-6, also known as BAX-inhibitor 1 (BI-1), is an anti-apoptotic protein that belongs to a putative family of highly conserved and poorly characterized genes. Here we report the function of TMBIM3/GRINA in the control of cell death by endoplasmic reticulum (ER) stress. Tmbim3 mRNA levels are strongly upregulated in cellular and animal models of ER stress, controlled by the PERK signaling branch of the unfolded protein response. TMBIM3/GRINA synergies with TMBIM6/BI-1 in the modulation of ER calcium homeostasis and apoptosis, associated with physical interactions with inositol trisphosphate receptors. Loss-of-function studies in D. melanogaster demonstrated that TMBIM3/GRINA and TMBIM6/BI-1 have synergistic activities against ER stress in vivo. Similarly, manipulation of TMBIM3/GRINA levels in zebrafish embryos revealed an essential role in the control of apoptosis during neuronal development and in experimental models of ER stress. These findings suggest the existence of a conserved group of functionally related cell death regulators across species beyond the BCL-2 family of proteins operating at the ER membrane.

Figure 1

TMBIM3 protects against ER stress-mediated apoptosis. (a) WT MEFs cells were transiently transfected with TMBIM3-MYC or TMBIM6-MYC expression vectors and their subcellular localization was evaluated by immunofluorescence (green) after co-staining with Brefeldine A-bodipy (red) using a confocal microscopy. Nucleus was stained with Hoechst (blue). (b) Left panel: MEFs cells were stably transduced with retroviral expression vectors for TMBIM3-MYC or empty vector (MOCK) and its expression assessed by western blot. Hsp90 levels were monitored as loading control. Right panel: cells were exposed to indicated concentrations of Tm or Thg for 24 h, and cell death was evaluated by propidium iodide (PI) staining and fluorescence-activated cell sorting (FACS) analysis. Mean and S.D. are presented of four independent experiments. Student's t-test was used to analyze statistical significance between control and TMBIM3-overexpressing cells (^*P<0.05, ^**P<0.01, ^***P<0.001). (c) Alternatively, cells were analyzed as described in (b) after treatment with 10 ng/ml TNF-α and 10 μM Actinomycin D (TNF-α), 40 μM etoposide (Eto), 0.1 μM stausorporine (Sts), nutrient starved, or exposed to serum withdrawal. Data represents mean and S.D. of four independent experiments. Student's t-test indicated no significant differences between MOCK or TMBIM3-expressing cells for each treatment. (d) Apoptosis was monitored in cells described in (b) after treatment with 1 μg/ml Tm or 1 μM Thg by quantifying the percentage of hypodoploid cells by FACS analysis. Mean and S.D. are presented of three independent experiments. Student's t-test was used to analyze statistical significance between control and TMBIM3-MYC-overexpressing cells (^*P<0.05). (e) MOCK and TMBIM3-MYC-expressing cells were exposed to a pulse of 2.5 μg/ml Tm for 4 h. Then Tm was washed out, and cells re-plated into 10 cm cell culture dish and cultured for up to 5 days. Total cell number was visualized by staining with crystal violet (left panel). Number of cells per area (1.5 mm²) was quantified over time. Mean and S.D. are presented of three independent experiments. Student's t-test was used to analyze statistical significance between control and TMBIM3-overexpressing cells (^*P<0.05). The color reproduction of this figure is available at Cell Death and Differentiation Journal online

Figure 2

Downregulation of TMBIM3 induces apoptosis of TMBIM6-deficient cells. (a) TMBIM6 WT and KO MEFs cells were transiently transduced with lentiviral vectors expressing two different shRNA constructs directed to tmbim3 mRNA (sh#1 and sh#2) or to the luciferase mRNA (shLuc) as control. After 4 days of transduction, cell morphology was visualized by phase contrast microscopy. Results are representative of three independent experiments. (b) Alternatively, cell viability of experiments presented in (a) was quantified by PI staining and FACS analysis. Mean and S.D. are presented of four independent experiments. Student's t-test was used to analyze statistical significance between WT and TMBIM6 KO cells (^*P<0.05). (c) In parallel to (a), nuclear morphology was visualized after Hoechst staining. Right panel: quantification of the percentage of cells with condensed and fragmented nucleus. Mean and S.D. are presented of three independent experiments. Student's t-test was used to analyze statistical significance between WT and TMBIM6 KO cells under different conditions (^**P<0.01, ^***P<0.001). (d) TMBIM6 WT (Left panel) and KO (right panel) cells were stably transduced with lentiviruses expressing shLuc or shRNA constructs against tmbim3 (shTMBIM3) construct #2, and then cells were exposed to indicated concentrations of Tm. After 24 h, cell viability was evaluated by PI staining and FACS analysis. Mean and S.D. are presented of three or four determinations. Statistically significant differences between TMBIM6KO shLuc and shTMBIM3 were detected by multiple comparisons using two-way ANOVA. (e) TMBIM6 WT and KO cells were stably transduced with retroviruses expressing TMBIM3-MYC or MOCK, and then cells were exposed to 1 μg/ml Tm or 1 μM Thg for 24 h. Cell viability/number was measured with the MTS assay. Average and S.D. of three determinations is presented representative of two independent experiments. (f) 293T cells were co-transfected with expression vectors for TMBIM6-human influenza hemagglutinin (HA) and TMBIM3-MYC or empty vector (pCDNA.3). After 48 h, TMBIM3-MYC was immunoprecipitated (IP) and then the co-precipitation of TMBIM6-HA was assessed by western blot (WB) analysis

Figure 3

Tmbim3 mRNA levels are regulated by the PERK signaling branch of the UPR. (a) MEFs cells were exposed to 1 μg/ml Tm for indicated time points, and tmbim3, tmbim6 and erdj4 (positive control) mRNA levels were evaluated by real-time PCR and normalized with actin levels. Mean and S.D. are presented of three independent experiments. (b) Right panel: PERK WT and KO cells were exposed to 1 μg/ml Tm for indicated time points, and tmbim3 mRNA levels were evaluated by real-time PCR. (c) In parallel, in the same cells described in (b), xbp-1 mRNA splicing was monitored by RT-PCR (upper panel). PCR fragments corresponding to spliced and unspliced forms are indicated. Bottom panels: CHOP or Hsp90 levels were monitored by western blot in the same experiment. (d) IRE1αKO MEFs cells were stably transduced with retroviruses expressing IRE1α-HA or MOCK, and tmbim3 mRNA levels were monitored by real-time PCR after treatment with Tm for indicated time points. (e) ATF6α WT and KO cells were treated with Tm for 24 h and tmbim3 mRNA levels were monitored as described in (a). As positive control, chop mRNA was monitored. Mean and S.E. is presented of three independent experiments. (f) PERK WT and KO cells were exposed to indicated concentrations of Tm for 24 h, and cell viability was evaluated by PI staining and FACS analysis. Mean and S.D. are presented of four independent experiments. Student's t-test was used to analyze statistical significant differences (^**P<0.01; ^***P<0.001). (g) MEFs cells stably transduced with MOCK or TMBIM3-MYC expression vector were incubated with 1 μg/ml Tm for 16 h. Then, the expression levels of the ER chaperones and foldases BIP, PDI, ERp74, ERp57, and EDEM was evaluated by western blot. Hsp90 levels were monitored as loading control. Note: The electrophoretic shift in EDEM1 is because of deglycosylation of the protein upon Tm treatment. (h) WT mice were i.p. injected with indicated concentrations of Tm. After 16 h of injection, kidneys were collected. Then, tmbim3, chop, actin, and xbp-1 mRNA splicing levels were monitored by RT-PCR using cDNA prepared from total mRNA. Each well of the gel represents independent animals. (i) atf4^+/+ or atf4^−/− mice were injected with 100 ng/g Tm. After 16 h, tmbim3 and actin mRNA levels were monitored by semiquantitative PCR in kidney tissue described in (h). Each well represents an independent animal. (j) In parallel, tmbim3 mRNA levels were measured by real-time PCR in the same experiment performed in (i). Mean and S.D. are presented of the analysis of three or four independent animals. Statistically significant differences detected by multiple comparisons using one-way ANOVA and Newman-Keuls Multiple Comparison Test (^***P<0.01)

Figure 4

TMBIM3 regulates ER calcium homeostasis and calcium-mediated cell death. MEFs cells were stably transduced with TMBIM3-MYC expression vectors and MOCK. Then, cells were loaded with the calcium indicator Fluo-4, and cytosolic calcium signals were monitored over time after stimulation in the absence of extracellular calcium with (a) 500 μM ATP, or (b) 5 mM H₂O_2. Left panel: a curve of calcium signal is presented representative of the analysis of at least 20 cells from three independent experiments. Right panel: the pick intensity of calcium signaling was quantified. Mean and S.E. are presented of three or four independent experiments. Student's t-test was used to analyze statistical significance (^***P<0.001). (c) Alternatively, calcium signaling was monitored in TMBIM6 KO cells stably transduced with a lentiviral vector to deliver shTmbim3 or shLuc after treating them with 500 μM ATP. (d) Mock and TMBIM3-MYC-expressing cells were exposed to 200 μM araquidonic acid (Araq) or 1 mM H₂O₂. After 6 h of treatment, cells were stained with PI (red) and Hoechst (Blue), and visualized with a fluorescent microscope. Data is representative of three independent experiments. (e) ER-calcium content was quantified in MOCK and TMBIM3 cells after transduction of cells with adenoviruses expressing a chimeric aequorin targeted to the ER. For transient expression of TMBIM3 and its control MOCK, cells were co-transfected with expression vectors for ER aequorin. ER calcium content was monitored as described in Materials and Methods. Mean and S.E. is presented of four independent experiments. No significant (n.s.) differences Student's t-test analysis. (f) The ER ⁴⁵Ca²⁺ content of MOCK and TMBIM3-expressing MEF cells loaded to steady state with ⁴⁵Ca²⁺ is plotted as a function of time, depicting the passive ER ⁴⁵Ca²⁺ leak from the ER. The figure shows a representative experiment of five independent experiments as mean and S.E.. No significant (n.s.) differences Student's t-test analysis

Figure 5

TMBIM3/GRINA interacts with IP3R modulating its activity under ER stress conditions. (a) Saponin-permeabilized MOCK and TMBIM3-MYC-expressing MEFs were exposed to IP3 at indicated concentrations. The ⁴⁵Ca²⁺ released from ER was measured, as described in Materials and Methods. Values were normalized using the Ca²⁺ ionophore A23187 and represent ⁴⁵Ca²⁺ release triggered by IP3 relative to ⁴⁵Ca²⁺ release triggered by A23187 (% of A23187-releasable ⁴⁵Ca²⁺). Mean and S.E is presented of three to five independent experiments. Student's t-test was used to analyze statistical significance of indicated comparisons (^***P<0.001). (b) TMBIM3-MYC was immunoprecipitated (IP) in MEFs cells stably expressing the protein or transduced with a MOCK expression vector and the co-precipitation of endogenous IP3R3. Results represent three independent experiments. (c) In addition, 293T cells were transfected with expression vector for TMBIM3-MYC. After 48 h, the co-precipitation of TMBIM3-MYC with endogenous IP3R3 was evaluated by western blot analysis. As control, protein expression is shown in total extracts. (d) Hela cells were transiently transfected with expression vector for TMBIM3-MYC and the co-precipitation of endogenous IP3R1 was evaluated as described in (c). (e) WT MEF cells were exposed to indicated concentrations of Tm for 16 h. Then, cells were loaded with the radiometric calcium indicator Fura-2, and cytosolic Ca²⁺ signals were monitored over time after ionomycin (1 μM) stimulation in the absence of extracellular Ca²⁺. The figure represents the mean of four independent experiments. (f) WT MEFs cells were exposed to 1 μg/ml of Tm for 16 h in presence or absence of 2-APB (10 μM) and the cell viability was evaluated by PI staining and FACS analysis. Mean and S.E. are presented of three independent experiments. Student's t-test was used to analyze statistical significance (^*P<0.01). (g) MOCK and TMBIM3-MYC-expressing cells were incubated in presence or absence of Tm (100 ng/ml) for 12 h. Then, the remaining ER Ca²⁺ was evaluated as described in (e). The pick intensity of Ca²⁺ signaling was quantified. Mean and S.E are presented of four to five independent experiments. Student's t-test was used to analyze statistical significance (^*P<0.001). (h) TMBIM6 KO cells stably transduced with lentiviruses expressing shLuc or shTMBIM3 construct #2 were exposed to 250 ng/ml Tm in the presence or absence of 10 μM 2-APB for 16 h. Cell viability was visualized after PI (bottom panel) and Hoechst staining (upper panel) using fluorescent microscopy (Scale bar, 60 μm). In parallel, (i) apoptotic nucleus was quantified by fluorescent microscopy. Mean and S.D. are presented of four independent experiments. Student's t-test was used to analyze statistical significance of indicated comparisons (^*P<0.05)

Figure 6

TMBIM3 and TMBIM6 have synergistic roles in the survival against ER stress in vivo in Drosophila melanogaster. (a) D. melanogaster TMBIM3 (dTMBIM3), dTMBIM6, or both genes (double RNAi) were knocked down by expressing specific RNAi constructs using the UAS-GAL4 system under the control of a tubulin driver. Then, basal adult fly viability was investigated. Data was normalized to 100% viability with the values obtained for each RNAi control fly line. Mean and S.D. are presented for three independent experiments. Student's t-test was used to analyze statistical significance between indicated comparisons. (b) Apoptosis was monitored using second instar larvae grown in food supplemented with 25 μg/ml Tm or 0.5% DMSO (solvent control) for 24 h. Intestine of fly larvae were then collected, and active-caspase 3 was detected by immunofluorescence (green). Nucleus was stained with DAPI (blue) to visualize total number of cells. Images represent the analysis of three independent experiments. (c) Life cycle was monitored using second instar larvae grown in food supplemented with 25 μg/ml Tm dissolved in DMSO or 0.5% DMSO as control in dTMBIM3, (d) dTMBIM6, or (e) both genes (double RNAi) were knocked down with RNAi. The number of individuals reaching the adult fly stage was evaluated over time starting from 100 viable larvae. Left panel: a representative experiment is presented following the survival of 100 larvae individuals over time. Right panel: quantification of the total adult flies obtained in a representative experiment after 15 days. Mean and S.E. are presented of three independent experiments. Student's t-test was used to analyze statistical significance (^*P<0.05; ^**P<0.01; ^***P<0.001)

Figure 7

Expression pattern and impact of knocking down TMBIM3 in zebrafish. (a) Pattern of expression of zTmbim3 mRNA as visualized by whole mount in situ hybridization at three developmental stages: 8 cells (i), 10 somites (ii) and 24 hpf (iii–v). Images correspond to lateral (i–ii) and dorsal (iv–v) views of embryos. (b) Effect of zTmbim3 knock down on cell death during zebrafish development. Acridine orange staining was assessed in 24 hpf embryos after injection of 1.5 ng of either control mismatch morpholino (Mis-MO) or zTmbim3-MO. Images of bright field (left panel) and fluorescence (right panel) correspond to lateral views of embryos, and are representative of three independent experiments. (c) Quantification of acridine orange staining (pixel intensity/area) in the spinal cord of zTmbim3-depleted embryos (dashed square depicted in panel (b)). Data represent mean and S.D. of the analysis of 20 embryos of three independent experiments. Student's t-test was used to analyze statistical significance (^**P<0.01). (d) Quantification of activated caspase-3 fluorescent immunoreactivity in the spinal cord of zTmbim3-depleted embryos as described in (c). Data represent mean and S.D. of the analysis of three embryos. Student's t-test was used to analyze statistical significance (^*P<0.05). (e) Panels show dorsal views of whole-mounted embryos, oriented anterior to the left (A) and posterior to the right (P). Zebrafish Tg(ngn1∷GFP) embryos were injected with 4.5 ng of either Mis-MO in one-cell stage. Embryos were then visualized in vivo at 24 hpf by confocal microscopy. Data represents the analysis of 20 embryos for each condition. GFP-labeled telencephalon (t), eye (e), third ventricle (3e), dorsal midbrain (m), interneurons (in) Rohon Beard sensory neurons (rb) over the dorsal longitudinal fasciculus (dlf). The boundaries of the spinal cord are delineated with a red arrow of felt double. Scale bars represent 20 μm. (f) One-cell stage zebrafish embryos were injected with 4.5 ng of either Mis-MO or zTmbim3-MO, in the presence or absence of 50–100 μg of either control (Gap43-cherry) or mTmbim3 mRNAs. Embryos were then visualized with bright field at 60–70% epiboly or 24 hpf. Data represents the analysis of 346 embryos. (g) Quantification of the percentage of surviving embryos at 24 hpf treated in (f). Data represent the average of two independent experiments (n=127 and 219 embryos analyzed per experiment, respectively)

Figure 8

TMBIM3 protects against ER stress in zebrafish in vivo. (a) Embryos at 24 hpf were exposed to 1 and 5 μM Thg for 4 h, and then zebrafish xbp-1 mRNA splicing was evaluated by RT-PCR. PCR fragments corresponding to spliced and unspliced forms are indicated. Actin levels were monitored as loading control. (b) In embryos described in (a), the mRNA levels of the zebrafish chop (zChop) and zTmbim3 were measured by real-time PCR, and normalized by actin expression. Data represents mean of the analysis of two independent experiments performed in triplicates. (c) One-cell stage embryos were injected with 50–100 pg of in vitro transcribed mouse tmbim3 (mTmbim3) mRNA. After 24 h, mRNA levels of the exogenous mTmbim3, and endogenous zTmbim3 were evaluated by RT-PCR as control for experiments presented in (d). (d) Cell death was monitored in 8 hpf embryos after exposition to indicated concentrations of Thg for 4 h. Then, embryos were stained with acridine orange. Insets show the corresponding bright field images of the embryos for each condition. Data represents three independent experiments of a total 3–5 embryos analyzed per determination. (e) Changes in tail morphology were monitored at 24 hpf after Thg treatments as described in (d). Arrows indicate alterations in tail morphology. Data is representative of three independent experiments of a total 10–15 embryos visualized per experiment. (f) Working model. Overall, our data suggest that TMBIM3 is a highly conserved anti-apoptotic protein that regulates cell survival under ER stress conditions. Activation of the PERK/ATF4 signaling branch of the UPR leads to the upregulation of tmbim3 mRNA under ER stress conditions. TMBIM3 negatively modulates the release of ER calcium by IP3R, decreasing the sensibility of cells to ER stress and calcium-mediated cell death. TMBIM3 has synergistic effects with TMBIM6, which may be associated with the formation of a protein complex containing TMBIM6