TMBIM6/BI-1 contributes to cancer progression through assembly with mTORC2 and AKT activation

Transmembrane B cell lymphoma 2-associated X protein inhibitor motif-containing (TMBIM) 6, a Ca2+ channel-like protein, is highly up-regulated in several cancer types. Here, we show that TMBIM6 is closely associated with survival in patients with cervical, breast, lung, and prostate cancer. TMBIM6 deletion or knockdown suppresses primary tumor growth. Further, mTORC2 activation is up-regulated by TMBIM6 and stimulates glycolysis, protein synthesis, and the expression of lipid synthesis genes and glycosylated proteins. Moreover, ER-leaky Ca2+ from TMBIM6, a unique characteristic, is shown to affect mTORC2 assembly and its association with ribosomes. In addition, we identify that the BIA compound, a potentialTMBIM6 antagonist, prevents TMBIM6 binding to mTORC2, decreases mTORC2 activity, and also regulates TMBIM6-leaky Ca2+, further suppressing tumor formation and progression in cancer xenograft models. This previously unknown signaling cascade in which mTORC2 activity is enhanced via the interaction with TMBIM6 provides potential therapeutic targets for various malignancies.

In their manuscript, Kim et al identify the gene TMBIM6 as being important in tumorigenesis through its impact on mTORC 2:ribosome assembly and therefore mTORC 2 activity. The authors characterize transcriptional, metabolic, and transformation changes that occur upon TMBIM6 deletion or its suppression by a putative inhibitor, BIA. These experiments provide evidence for an interesting novel novel of this protein in regulating mTORC 2 activity.
The most convincing experiment is the lack of rescue with the D213R mutant combined with rescue of the wild-type protein. The use of this system should be expanded, as described below. While a variety of assays are performed in support of their hypotheses, often times critical controls are missing, as described below.
Major C omments 1. Viability data reported in Figure 2 and metabolic and gene expression data presented in Figure 4 needs to be rescued by re-expression of the TMBIM6 cDNA, as opposed to the D213R mutant. 2. The authors should demonstrate that BIA does not have anti-proliferative or anti-cell migration activity in cell lines lacking TMBIM6, and does not impact AKT signaling in these lines. If a loss of cell proliferation is observed in KO cells, that result would indicate an off-target effect of the compound at that dose.
Minor C omments 1. The authors should show data validating their IHC technique provided in Figure 1. For example, by staining WT and KO HT-1080 cells using the same fixation protocol as used for tissues. 2. Survival data provided in Figure 1 are all from microarray using the same probe. The authors should report similar public data, readily available, from more recent studies that use RNASeq methods. 3. Figure 1G appears to be from a single replicate. The authors should provide additional replicates of this transcriptomic data. Primary transcriptomic data should be provided in a supplementary table and deposited in a public database. 4. Authors should indicate whether the enriched pathways reported in Figure 1H are significantly enriched. 5. Authors need to address whether the re-expression of TMBIM6 in Figure 3C is at physiologic levels and include blots where WT, KO, and HA-TMBIM are on the same membrane. 6. Supplementary Figure 3 is incorrectly referenced in the text as supplementary figure 2 at page 7-8 7. Authors claim that Supplementary Figure 3E shows dose dependent increase in pAKT, but total AKT levels seem to increase as well. 8. The statement "TMBIM6 is required for AKT activity and signaling" (Page 7) is overstated based on the data provided. 9. PLA assays (Figure 3 and 5) require a negative control for an uninvolved cytoplasmic protein. 10. The authors should describe how the BIA compound was identified.
Reviewer #2 (Remarks to the Author) (Expertise: mTOR signalling, ribosomes) : TMBIM6 is a calcium channel like protein and is upregulated in many cancer types. Suppression of TMBIM6 promotes cell death and decreases tumor growth. In the current study, the authors now elucidate a mechanism as to how TMBIM6 could promote tumor growth. By knocking out TMBIM6 in HT1080 cells via C RISPR, they found overall defects in metabolism and decreased mTORC 2 signaling. By gel filtration, they found that rictor and GbL fractionated at lower molecular weights in the KO suggesting that mTORC 2 dissociates from a large complex in the absence of TMBIM6. They then conducted studies to determine association of mTORC 2 with ribosomes by proximity ligation assay (PLA) and immunoprecipitation. They found that the mTORC 2/ribosome (via expression of RPL19) association is disrupted in TMBIM6 KO. TMBIM6 also cofractionated and associated with mTORC 2 components and RPL19. Knockdown of rictor specifically abolished interaction of TMBIM6 with mTOR and SIN1. Mutagenesis of TMBIM6 revealed that the N-terminus and the cytosolic loop residues are required for interaction with rictor and Akt activation. Since TMBIM6 is a calcium channel-like protein, they then analyzed if calcium release from TMBIM6 could affect the interaction of mTORC 2 with ribosomes. By PLA, they found that there was decreased rictor/mTOR and rictor/rpl19 interaction in the TMBIM6 mutant (D213A) that affects calcium release. Although rictor association with TMBIM6 was similar in the WT and mutant TMBIM6, there was a slight decrease in mTOR and strong decrease in rpl19 and rpl16 association. By screening a chemical library, they identified B1A as a TMBIM6 antagonist that decreases tumor cell proliferation at rather high concentrations. Akt phosphorylation, total expression of mTORC 2 components and rpl19 are also decreased by this drug as well as release of calcium from the ER. B1A also regressed tumor growth in vivo. Based on these results, the authors propose that TMBIM6 promotes cancer progression via association with mTORC 2 and Akt activation.
Overall, the results are interesting and support a role for TMBIM6 inhibition in preventing tumor growth. The role of TMBIM6 in promoting mTORC 2/ribosome assembly is also interesting but would need further clarifications as detailed below.
1. Is the association of TMBIM6 with mTOR unique to mTORC 2? Authors should also blot for raptor. 2. In Figure 4, the authors found decreased expression of proteins involved in glycosylation. They should also include the analysis of metabolites of the hexosamine pathway. mTORC 2 has been shown to control this metabolic pathway (authors should cite PMID: 27570073 by Moloughney et al Mol C ell on page 9). 3. Akt phosphorylation should also be analyzed in Fig 5I. 4. While the PLA and immunoprecipitation assays to show association of mTORC 2 and rpl19, the authors should more carefully analyze how TMBIM6 promotes association of mTORC 2 with the ribosomes and whether this association is occurring in translating ribosomes. Polysome purification and analysis of mTORC 2 association or cofractionation with TMBIM6 should be conducted. 5. The IC 50 of B1A should be assessed. In Fig 7, it seems that high concentrations are needed to prevent cell proliferation of all the cell lines examined. 6. B1A ( Fig 7B) seems to decrease total protein expression of mTOR and rpl19, as well as TMBIM6. There seems to be no strong effect on disruption of the complexes, other than slight decrease in TMBIM6. The authors should quantitate this figure more carefully to reflect comparative amount of total proteins. 7. Expression of total mTOR and rictor proteins should be included in Fig 7C . 8. The authors state in the Discussion that "This effect of TMBIM6 on cell growth differs from its classical role in ER stress." This has really not been addressed in the current study. Previous studies have already shown a role for mTORC 2 in negative regulation of the ER calcium channel regulator Mid1 (in yeast) as a negative regulator of starvation response (PMID 27899413 Vlahakis et al JC B 2016). They should also cite this work to support the relationship of calcium signaling to mTORC 2. Furthermore, they should verify that the disruption of TMBIM6 does not cause ER stress. The overall decrease in metabolic pathways and possible decrease in total ribosomal proteins (see comment 6) suggest that translation could also be decreased (which would relieve ER stress at least partially).
Reviewer #3 (Remarks to the Author) (Expertise: stress, ER) : NC OMMS-19-31425-T C omments to the authors, In this work, Kim and colleagues show that TMBIM6 -a member of the highly conserved TMBIM family of proteins -interacts with components of the mTORC 2 complex to regulate cancer progression. Indeed, they first show that TMBIM6 is overexpressed in several different types of carcinomas and its expression correlates with poorer patient overall survival. Using loss-of-function approaches, the authors show that TMBIM6 promotes cell migration and invasion in vitro and tumor growth in vivo. These experiments are well performed and highlight the known role of TMBIM6 as a protumorigenic protein. In fact, work by this same group has previously shown that TMBIM6 overexpression promotes metastasis through the regulation of cell migration and invasion (PMID: 20118983). At the molecular level they show that TMBIM6 is required for the assembly of the mTORC 2 complex, leading to improved glucose homeostasis and promoting cancer progression. In short, TMBIM interacts and tether the mTORC 2 complex and the ribosomes to the ER where it would promote its assembly through TMBIM6 calcium leak channel function. Finally, the authors described a putative TMBIM6 antagonist with anti-tumor properties.
Overall, the in vitro and in vivo results are very interesting and well-presented. However, I have some concerns regarding the proposed role of TMBIM6 as an mTORC 2 regulator and its connection to cancer progression. Some key control experiments are currently missing, making the interpretation of the data difficult. The statistical analysis used throughout the manuscript should also be revised.
A revised version of this manuscript should still be suitable for publication in Nature C ommunications.
Major concerns: 1. Model of choice and generality: In Figure 1, the authors showed that TMBIM6 is overexpressed in several different cancer types, including cervical, endometrial and vulvar, breast, lung and prostate cancer. They also show that high TMBIM6 expression correlates with poorer prognosis in patients with squamous cell carcinoma and endocervical adenocarcinoma, esophageal carcinoma, skin cutaneous melanoma, head and neck squamous carcinoma and brain lower grade glioma, most of them of epithelial origins ( Figure 1). However, they perform most of the loss-of-function experiments (migration, invasion, and tumor growth and mTORC 2 biochemical experiments) in HT1080 cell lines, which is a fribrosarcoma cell line of mesenchymal origin. Two comments on this: (1) the authors should show, is it were, TMBIM6 expression and survival data on fibrosarcoma, since it will correlate better with the observations performed in HT1080 TMBIM6 KO cells and (2) the authors should repeat some key experiments (mTORC 2 assembly, AKT phosphorylation and tumor growth) in an additional TMBIM KO cell line cancer model directly related to the cancers shown in figure 1. Most of the experiments connecting TMBIM6 with mTORC 2 were performed in only one cancer cell line.
2. Is the TMBIM6 KO HT1080 cell line a pool or was derived from a single clone? In the case of it being a clonal population, the authors should repeat some key experiments (i.e. AKT phosphorylation and the assembly of the mTORC 2 complex) with additional clones. Throughout the manuscript, the authors compare the HT1080 TMBIM6 KO cells with wild-type controls when they should have been compared with a matched C RISPR/Cas9 scramble or mock control. I am worried about C RISPR/C as9 off-target effects. physical association between mTORC 2 and the ribosomes". However, the authors did not rule out the possibility that TMBIM6 expression may regulate the concentration of proteins forming the mTORC 2 complex (the authors also showed that TMBIM6 KO reduces global protein synthesis). The same seems to be the case with mTOR and RIC TOR co-localization with PDI (Supplementary Figure 4): TMBIM6 KO cells seem to exhibit decreased levels of these proteins, hence, decreased co-localization. At the very least, the authors should compare the endogenous protein levels of mTOR, RIC TOR, RPL19 and RPS16 between HT1080 WT and TMBIM6 KO cell lines. If willing, they should also check the mRNA levels of these genes. Is it reduced assembly, reduced protein expression or both? 4. To assess the putative role of calcium released from TMBIM6 on mTORC 2 assembly, the authors use a TMBIM6-GC aMP3 construct that they say "… is based on the finding that leaky calcium but not ER lumen [calcium] is detected upon binding of C a2+ to the cytosol of this protein ( Supplementary Fig. 6B)". To what side of TMBIM6 was GC aMP3 attached? The latest structural models based on the crystallization of the bacterial homolog BsYetJ suggest that TMBIM proteins are composed by seven transmembrane domains with the N-terminus facing the cytosol and the C -terminus facing the intraluminal space (see for example: PMID:24904158 and PMID:30930064). The authors should comment and cite these structural works and incorporate them into their models presented in Supplementary Figure 6C , where they only show the 6 transmembrane models for TMBIM6. How are the authors completely positive that they are not measuring ER calcium content? 5. In Figure 6, the authors reconstituted TMBIM6 KO cells with either TMBIM6-HA or the channel mutant TMBIM D213A and assessed calcium leak ( Figure 6A), mTORC 2 assembly ( Figure 6B) and AKT phosphorylation ( Figures 6D and 6E). However, to correctly interpret these experiments, a comparison of the total reconstituted levels of TMBIM6 and TMBIM6 D213A proteins in KO cells is necessary. The authors should also check the ER localization of the WT and the D213A mutant proteins by immunofluorescence. Differential partial reconstitution of these proteins may account for the differences in mTORC 2 assembly and AKT phosphorylation.
6. Specificity of BI: from the data presented in Figures 7 and 8, and Supplementary Figure 7, the authors suggest that BIA acts as a TMBIM6 antagonist, blocking ER calcium release, leading to mTORC 2 disassembly and inhibition of TMBIM6-associated tumorigenicity. Although it is clear that BIA reduces cell proliferation, migration and invasion, it is not so clear that these effects dependent on BIA's function as a TMBIM antagonist. Does BIA work when TMBIM6 is knocked out? There are also several key control experiments missing from figures 7 and 8 that preclude the interpretation of these experiments. For example, total mTOR and RIC TOR levels are missing from the IP shown in Figure 7C . The authors should also report the effect of BIA on cell viability under the conditions reported in Figure 7A. The data supporting the role of BIA as an inhibitor of the interaction between TMBIM6 and mTORC 2 is currently not convincing enough. The authors could perform PLA as in previous experiments to strengthen this data. 7. Overstatements and unsupported evidence: In page 18 the authors state that "BIA is a newly identified TMBIM6 antagonist that disrupts the TMBIM6-mTORC 2 interaction, leading to the inhibition of tumor growth even in cases that are resistant to mTOR inhibitors". However, the authors have not provided evidence suggesting this. They have only shown that BIA decreases tumor growth in vitro and in vivo and that it works in combination with other mTORC 2 inhibitors to kill HT1080 cells. They have not shown that these effects are a direct consequence of BIA's putative role as a TMBIM6 antagonist or as a disruptor of TMBIM6-mTORC 2 interaction. Throughout the manuscript there are several instances where the authors jump from a phenotypic observation to a molecular mechanism whose causal connection to the observation has not been directly demonstrated.
8. Statistics: The authors should review the statistical procedures used throughout the manuscript. When there are two independent variables (e.g. genotype and time) the authors should use a Two-Way ANOVA followed by post hoc test instead of multiple two-tailed unpaired Student's t-tests .  Figures where Two-Way ANOVA should be used include Figures 1E, 2A, 2D, 2E, 2H, 3B, 3F, 4A, 4E,  4H, 8E and 8G. When there is one independent variable but more than two groups are compared, the authors should use a One-Way ANOVA followed by post hoc test. Examples of these include Figures 2I, 2K, 6D and 6E. Finally, the authors should specify p-values as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001.
Minor comments: 1. For in vitro migration and invasion experiments shown in Figures 2B and C , the authors should also express the data as the percentage of migrating and invading cells and not only as fold changes compared to WT. Figure 3C the authors should show the blot results of the second mTORC 2 target NDRG1 (pT346) in the HT1080 TMBIM6 KO cells reconstituted with TMBIM6-HA.

In
3. In page 7, the authors describe the effects of insulin referencing supplementary figure 2D and 2E. This is not correct; it should read Supplementary Figure 3D and 3E respectively.
4. There are many misplaced conclusions throughout the manuscript. For example, in page 11 the authors say that "…TMBIM6 interacts with mTORC 2 and ribosomes and that this interaction is important for the kinase activity of mTORC 2". However, the effects of TMBIM6-mTORC 2 interaction in AKT phosphorylation are explored in the following paragraph ( Figure 5H). 5. How does calcium leak from the ER increase mTORC 2 assembly efficiency? The authors should discuss this in more detail.
Reviewer #4 (Remarks to the Author) (expertise: cancer, zebrafish model) : This is a continuous work of the authors´ previous claims that TMBIM6/BI-1 promotes tumor growth and progression (ref 9). In this study, the authors provide new evidence of potential signaling events of the TMBIM6/BI-1-mTORC 2-AKT axis in different cancer types. First, they show that in several cancer types TMBIM6/BI-1 expression is elevated, which is largely confirmed by GC TA analysis. Using in vitro KO technology, they show that deletion of TMBIM6/BI-1 in cancer cells retards cancer cell proliferation, migration, and tumor formation. They then defined signaling pathways that potential involved in these tumorigenic activities. Finally, they screened a chemical library to identify a potential inhibitor BIA, which inhibits tumor growth in mice and in fish.

C omments:
1) This work covers almost all respects of cancer development, including cancer cell proliferation, migration, survival, tumor formation, and metastasis. From the provided data, it is hard to believe that TMBIM6/BI-1 is a master regulator of cancer development. In particular, the characterized signaling events do not support TMBIM6/BI-1 as a master regulator. The authors should focus on a particular signaling and activity to obtain an in-depth mechanistic insight, but not a wikipediaassocaited cancer development.
2) The TC GA data and their own experimental findings do not completely match. What about lung cancer, pancreatic cancer, and prostate cancer in TC GA? If TMBIM6/BI-1 is also highly expressed in these cancers, but not associated with poor survival, what does this mean? I am almost certain that some cancer types express high levels of TMBIM6/BI-1, but lack clinical correlation. The authors should not only choose the results favor to your findings.
3) The zebrafish cancer model is useless in this experimental setting. Are BIA-treated cancer cells dead in the fish body? How do they discriminate living cancer cells from dead cancer cells? How do they know that BIA is active in fish body? 4) BIA looks like a non-specific chemical compound as most chemical compounds do. These data can only be used as indirect supportive data.
5) The manuscript needs language editing.

Reviewers' comments:
Reviewer #1 (Remarks to the Author) (Expertise: mTOR, cancer metabolism): In their manuscript, Kim et al identify the gene TMBIM6 as being important in tumorigenesis through its impact on mTORC2:ribosome assembly and therefore mTORC2 activity. The authors characterize transcriptional, metabolic, and transformation changes that occur upon TMBIM6 deletion or its suppression by a putative inhibitor, BIA. These experiments provide evidence for an interesting novel novel of this protein in regulating mTORC2 activity.
Response: We thank the reviewer for his/her insightful comments and criticisms.
The most convincing experiment is the lack of rescue with the D213R mutant combined with rescue of the wild-type protein. The use of this system should be expanded, as described below. While a variety of assays are performed in support of their hypotheses, often times critical controls are missing, as described below.
Major Comments 1. Viability data reported in Figure 2 and metabolic and gene expression data presented in Figure 4 needs to be rescued by re-expression of the TMBIM6 cDNA, as opposed to the D213R mutant. Response: We agree with the reviewer's comments. In accordance with the reviewer's recommendation, we performed cell proliferation, metabolic, and gene expression analysis by re-expression of the TMBIM6 and TMBIM6-D213A. As shown in the revised Fig. 6, the inhibition of cell proliferation was rescued by re-expression of TMBIM6, not TMBIM6 D213A. Consistently, the expression of genes related to glycolysis, pentose phosphate pathway (PPP), GSH biosynthesis, and de novo lipogenesis was rescued by re-expression of TMBIM6, but not by that of TMBIM6 D213A. Furthermore, the levels of metabolites from glycolysis, tricarboxylic acid cycle, PPP, and hexosamine biosynthesis pathway (HBP) in TMBIM6 KO cells were rescued in the re-expressing condition of TMBIM6, not TMBIM6 D213A. Since the presence of D213A starts from the original Fig. 6, the reviewer's comment was answered in the revised Figure 6   Data represent mean ± SD. Statistical differences were detected with one-Way ANOVA followed by Tukey's test (D), and two-Way ANOVA followed by Bonferroni's test (B, C).

Manuscripts
To investigate the importance of Ca 2+ leakage through TMBIM6, we stably rescued TMBIM6 expression of TMBIM6 or TMBIM6 D213A in TMBIM6 KO HT1080 cells, and then determined cell proliferation rates. TMBIM6 and D213A were expressed at a comparable level in the cells ( Supplementary Fig. 8A). We found that cell proliferation was restored in TMBIM6-rescued KO cells, not D213A (Fig. 6F). Furthermore, the expression of genes related to the glycolysis and PPP was recovered in TMBIM6-rescued KO cells, not D213A, recovering glucose consumption and lactate production ( Fig. 6G-I). We next analyzed the expression of genes related to GSH biosynthesis, and de novo lipogenesis. Consistently, the expression of genes was restored in TMBIM6-rescued KO cells, not D213A (Supplementary Fig. 8B-C). By mass spectrometry, the levels of metabolites from glycolysis, tricarboxylic acid cycle, PPP, and HBP have restored in TMBIM6 rescued cells compared with TMBIM6 D213A mutant rescued cells (Fig. 6J, Supplementary Fig. 8D). Moreover, the patterns of ribosome profiling were same in all the cells, that is consistent to Supplementary  Fig. 8E; TMBIM6 is independent of ribosome maturation. These results suggest that TMBIM6 regulates metabolic pathways through its characteristics "Ca 2+ leakage-associated mTORC2 activation".
2. The authors should demonstrate that BIA does not have anti-proliferative or anti-cell migration activity in cell lines lacking TMBIM6, and does not impact AKT signaling in these lines. If a loss of cell proliferation is observed in KO cells, that result would indicate an offtarget effect of the compound at that dose. Response: In accordance with the reviewer's suggestion, we performed cell proliferation, migration assay, and immunoblotting for AKT signaling in BIA-treated TMBIM6 KO cell lines. As shown in Fig Figure 8 (n = 9 and 11 for control and treatment groups for HT1080 cells (I), and n = 6 mice per group for MDA-MB-231 cells (J)). Data represent mean ± SD. Statistical differences were detected with one-Way ANOVA followed by Tukey's test (F), and two-Way ANOVA followed by Bonferroni's test (B, C, D, H).

Manuscripts
To clarify whether anti-proliferation effect by BIA at that dose is an off-target effect, we examined cell proliferation in TMBIM6 KO HT1080 cells. The cell proliferation rate and AKT phosphorylation in TMBIM6 KO HT1080 cells were same in the presence or absence of BIA (Fig. 7A, Supplementary Fig. 10E) with the exceptions of high concentrations "20 and 30 µM" (Supplementary Fig. 10F), suggesting that BIA has on-target effect on TMBIM6 up to 10 µM.
Cell migration and invasion are representative in vitro markers for cancer characteristics. BIA treatment decreased cell migration in HT1080, MCF7, MDA-MB-231, and SKBR3 cells (Fig. 8A), not TMBIM6 KO HT1080 cells ( Supplementary Fig. 10G). Consistently, the wound healing assay showed that BIA significantly inhibited cell migration compared with DMSO-treatment ( Supplementary Fig. 10H). Moreover, cell invasion in MDA-MB-231 and HT1080 cells was also decreased in the BIA-treated cells (Fig. 8B). We also evaluated the effect of BIA on cell proliferation in a three-dimensional (3D) culture, which more accurately recapitulates in vivo tumor growth. Under the BIA treatment condition, the numbers and the size of spheroids formed by the 3D cultured cells were significantly decreased, not showing multi-acinar structures (Fig. 8C).

Minor Comments
1. The authors should show data validating their IHC technique provided in Figure 1. For example, by staining WT and KO HT-1080 cells using the same fixation protocol as used for tissues. Response: In accordance with the reviewer's suggestion, we performed IHC in WT and KO HT1080 cells using the same protocol as used for tissues. In the revised manuscript, we have included a new image in Figure 1F, showing that TMBIM6 expression is only detected in the TMBIM6 WT cells, and not in TMBIM6 KO cells. The updated Fig. 1F is given below.

Figure 1. TMBIM6 expression increased in cancer patient samples. (A-E) TMBIM6
expression was analyzed using the Gene Expression Omnibus database from NCBI. fibrosarcoma (GSE2719), cervix (GSE63678), breast (GSE31448), lung GSE19804, and prostate (GSE69223) datasets are presented. (F) Representative immunohistochemical staining of TMBIM6 on tissue microarrays containing fibrosarcoma, cervix, breast, lung, and prostate tissue and adjacent normal tissues. TMBIM6 WT and KO HT1080 cells were used as a control for validation of the method. Right; quantification data of TMBIM6 expression.
2. Survival data provided in Figure 1 are all from microarray using the same probe. The authors should report similar public data, readily available, from more recent studies that use RNASeq methods. Response: We performed an analysis of survival data from GEPIA2 tool (http://gepia2.cancer-pku.cn) and OncoLnc (http://www.oncolnc.org.). GEPIA2 (Gene Expression Profiling Interactive Analysis 2) is developed by Zefang Tang and colleagues for analyzing the RNA sequencing expression data of 9,736 tumors and 8,587 normal samples from the TCGA and the GTEx projects (Tang et al., 2019). OncoLnc contains survival data for 8,647 patients from 21 cancer studies performed by The Cancer Genome Atlas (TCGA), along with RNA-SEQ expression for mRNAs from TCGA (Anaya, 2016). Both tools allow researchers to study a specific gene and facilitate quick investigation in a single click. The updated Figure 1G and Supplementary Fig. 1A and corresponding text are given below.
3. Figure 1G appears to be from a single replicate. The authors should provide additional replicates of this transcriptomic data. Primary transcriptomic data should be provided in a supplementary table and deposited in a public database. Response: We added additional replicates transcriptome data, and detail primary transcriptomic data and showed them as supplementary table 1 as per the reviewer's suggestion. The expressed genes related to apoptotic process, proliferation, and metabolic pathways were greatly changed in TMBIM6 KO HT1080 cells compared with WT cells. The updated Figure 1H is given below. Figure 1G in the original version is changed into Figure  1H in the revised version.   Figure 1I is given below. Figure 1H in the original version is changed into Figure 1I in the revised version. 5. Authors need to address whether the re-expression of TMBIM6 in Figure 3C is at physiologic levels and include blots where WT, KO, and HA-TMBIM are on the same membrane. Response: In accordance with the reviewer's suggestion, we performed immuno-blotting in TMBIM6 KO HT1080 cells with re-expression of TMBIM6 using pLenti-vector for expression of physiologic levels, and confirmed the expression of TMBIM6 by using qRT-PCR. In the revised manuscript, we have included a new image in Figure 3C, showing that TMBIM6 expression is the same between TMBIM6 WT cells and the re-expression of TMBIM6 in KO cells. The updated Figure 3C and corresponding text are given below.  Figure 3E shows dose dependent increase in pAKT, but total AKT levels seem to increase as well. Response: We carefully repeated immunoblotting for pAKT and AKT. We confirmed the dose-dependent increase in pAKT levels without change in total AKT levels. The updated Supplementary Figure 3E is given below. Supplementary Fig. 3E in the original version is changed into Supplementary Fig. 3G in the revised version.

Authors claim that Supplementary
8. The statement "TMBIM6 is required for AKT activity and signaling" (Page 7) is overstated based on the data provided. Response: This is a very reasonable concern. In this study, we demonstrated TMBIM6 mediates mTORC2 activation through its binding to ribosome, thereby inducing AKT activation as mTORC2 substrate. In the revised version, the statement was updated "TMBIM6 is one of the essential genes for mTORC2 signaling, which regulates AKT activity". 9. PLA assays (Figure 3 and 5) require a negative control for an uninvolved cytoplasmic protein.
Response: In accordance with the reviewer's suggestion, we performed PLA assays using ribosomal protein S6 kinase beta-1 (S6K1) as an uninvolved cytoplasmic protein. The updated Figure 3F and 5D, and their corresponding text are given below.   10. The authors should describe how the BIA compound was identified. Response: To search for novel small molecules, TMBIM6 antagonists, we first performed high-throughput screening (HTS) from materials of the Korea Chemical Bank, and elicited chalcone scaffold. Next, total of 44 substituents through modification of R1 and R2 position of chalcone scaffold were synthesized, and were tested for cell viability. BIA was developed as a tool compound, and confirmed dissociation between mTORC2 and RPL19, decreasing AKT phosphorylation, and inhibiting Ca 2+ release. The updated supplementary Figure 9 and its corresponding text are given below. Manuscripts TMBIM6 antagonist reduces mTORC2 activity, inhibiting TMBIM6-associated tumorigenicity Initially, to identify novel small molecule TMBIM6 antagonists, we performed highthroughput screening (HTS) from materials of the Korea Chemical Bank and elicited chalcone scaffold ( Supplementary Fig. 9A). From the optimization of R1 and R2 position with diverse substituents, BIA was developed as a tool compound dependent cell viability ( Supplementary Fig. 9B-D).
Reviewer #2 (Remarks to the Author) (Expertise: mTOR signalling, ribosomes) : TMBIM6 is a calcium channel like protein and is upregulated in many cancer types. Suppression of TMBIM6 promotes cell death and decreases tumor growth. In the current study, the authors now elucidate a mechanism as to how TMBIM6 could promote tumor growth. By knocking out TMBIM6 in HT1080 cells via CRISPR, they found overall defects in metabolism and decreased mTORC2 signaling. By gel filtration, they found that rictor and GbL fractionated at lower molecular weights in the KO suggesting that mTORC2 dissociates from a large complex in the absence of TMBIM6. They then conducted studies to determine association of mTORC2 with ribosomes by proximity ligation assay (PLA) and immunoprecipitation. They found that the mTORC2/ribosome (via expression of RPL19) association is disrupted in TMBIM6 KO. TMBIM6 also cofractionated and associated with mTORC2 components and RPL19. Knockdown of rictor specifically abolished interaction of TMBIM6 with mTOR and SIN1. Mutagenesis of TMBIM6 revealed that the N-terminus and the cytosolic loop residues are required for interaction with rictor and Akt activation. Since TMBIM6 is a calcium channel-like protein, they then analyzed if calcium release from TMBIM6 could affect the interaction of mTORC2 with ribosomes. By PLA, they found that there was decreased rictor/mTOR and rictor/rpl19 interaction in the TMBIM6 mutant (D213A) that affects calcium release. Although rictor association with TMBIM6 was similar in the WT and mutant TMBIM6, there was a slight decrease in mTOR and strong decrease in rpl19 and rpl16 association. By screening a chemical library, they identified B1A as a TMBIM6 antagonist that decreases tumor cell proliferation at rather high concentrations. Akt phosphorylation, total expression of mTORC2 components and rpl19 are also decreased by this drug as well as release of calcium from the ER. B1A also regressed tumor growth in vivo. Based on these results, the authors propose that TMBIM6 promotes cancer progression via association with mTORC2 and Akt activation.
Response: We thank the reviewer for his/her enthusiasm and insightful summary for this study.
Overall, the results are interesting and support a role for TMBIM6 inhibition in preventing tumor growth. The role of TMBIM6 in promoting mTORC2/ribosome assembly is also interesting but would need further clarifications as detailed below.
1. Is the association of TMBIM6 with mTOR unique to mTORC2? Authors should also blot for raptor. Response: We appreciate the reviewers' comments. To identify whether TMBIM6 is associated with mTOR unique to mTORC2, not mTORC1, we performed immunoblot from immunoprecipitation lysates and PLA assay between TMBIM6 and RAPTOR, mTORC1 subunit, or between TMBIM6 and RICTOR, mTORC2 subunit. As shown in Figure 5C and 5D, TMBIM6 is associated with mTORC2, not mTORC1. The updated Figure 5C and 5D, and their corresponding text are given below.

Manuscripts
We also detected an association between TMBIM6 and endogenous mTORC2 or ribosomes (60S RPL19 and 40S RPS16) in TMBIM6-HA-overexpressing HeLa cells, but not with raptor as mTORC1 subunit (Fig. 5C). The localization of TMBIM6-HA in close proximity to mTORC2 components and ribosomes was observed by confocal microscopy (Fig. 5D). Figure 4, the authors found decreased expression of proteins involved in glycosylation. They should also include the analysis of metabolites of the hexosamine pathway. mTORC2 has been shown to control this metabolic pathway (authors should cite PMID: 27570073 by Moloughney et al Mol Cell on page 9). Response: We appreciate the reviewer's comments. In accordance with the reviewer's suggestion, we performed an analysis of metabolites of hexosamine biosynthesis pathway (HBP) in TMBIM6 WT and KO HT1080 cells. In the revised manuscript, we have included new data in Figure 4D, showing that metabolites of HBP such as Uridine diphosphate N-Acetylglucosamine (UDP-GlcNAc), N-Acetyl-D-Glucosamine 6-Phosphate (GlcNAc-6-P), glucosamine-6-phosphate, and N-Acetyl-D-Glucosamine were decreased in TMBIM6 KO cells compared to TMBIM6 WT HT1080 cells. The updated Fig. 4D and its corresponding text are given below.

Figure 4
Manuscripts mTORC2 activation by TMBIM6 regulates cellular metabolism mTORC2 regulates cellular bioenergetics by modulating glycolytic gene expression, aerobic glycolysis, glutathione (GSH) biosynthesis, hexosamine biosynthesis pathway (HBP), and glycosylation [36][37][38][39] . In this study, TMBIM6 KO cells showed downregulation of glycolytic genes (Fig. 4A), resulting in reduced glucose consumption and lactate production ( Fig. 4B-C). The expression of genes related to the pentose phosphate pathway (PPP) was also decreased in TMBIM6 KO, which was reversed in TMBIM6-overexpressing HeLa cells ( Fig.  4A and Supplementary Fig. 6). An MS analysis showed that the levels of metabolites from glycolysis, tricarboxylic acid cycle, PPP, and HBP were decreased in TMBIM6 KO cells relative to those in WT cells (Fig. 4D) indicating that metabolic pathways are dysregulated in the absence of TMBIM6, where is linked to the inhibition of mTORC2 activity.
3. Akt phosphorylation should also be analyzed in Fig 5I. Response: In accordance with the reviewer's comments, we performed immunoblotting in TMBIM6 WT and C-terminal 40 amino acids-deleted HT1080 cells. The phosphorylation of AKT Ser473 was reduced in the mutant cells. The updated Fig 5I is given below.

Manuscripts
Under the expression of TMBIM6 with deletion of 40 C-terminal amino acids, the association with RPL19 was abrogated, whereas the interaction with RICTOR or mTOR was not altered (Fig. 5I). The phosphorylation of AKT Ser473 was also decreased in RPL19-non-associated TMBIM6 mutants (Fig. 5H, I).

Manuscripts
To identify whether reducing mTORC2 activity in TMBIM6 KO cells is related to impairment of ribosome maturation, we performed fractionation in a sucrose gradient assay to separate polysomes from 80S, 60S, and 40S ribosomes. As shown in Supplementary Fig.  4A, the pattern of ribosome profiling was same between TMBIM6 WT and KO cells, indicating TMBIM6 is not related with ribosome maturation. In addition, mTOR, RICTOR, and SIN1 were found in both the polysomal and ribosomal fractions in TMBIM6 WT HT1080 cells (Supplementary Fig. 4B). However, mTORC2 components were relatively less detected in both fractions from TMBIM6 KO HT1080 cells compared to those from WT cells. In TMBIM6-rescued cells, TMBIM6 was co-purified with polysome and ribosome fractions ( Supplementary Fig. 4C). Since mTORC2 physically interacts with translating (mRNAbound) and non-translating 80S ribosomes, and TMBIM6 binds to the mTORC2, we next determined whether TMBIM6 is copurified with mTORC2 at mRNA-bound ribosomes. In mRNA bound ribosomes purified by pull-down of poly(A) mRNA with oligo(dT) cellulose, TMBIM6 was copurified with mTOR, RICTOR, and RPL19 ( Supplementary Fig. 4D). Collectively, these results suggest that TMBIM6 regulates the assembly of mTORC2 components and promotes the physical association between mTORC2 and ribosomes.
5. The IC50 of B1A should be assessed. In Fig 7, it seems that high concentrations are needed to prevent cell proliferation of all the cell lines examined. Response: In accordance with the reviewer's comments, we calculated IC50 of BIA using GraphPad 8.0 software. In Fig 7A, the IC50 values at 3 days were 1.7 ± 0.1 μM for HT1080, 2.6 ± 0.4 μM for MCF cells, 2.6 ± 0.5 μM for MDA-MB-231 cells, and 2.4 ± 0.4 μM for SKBR3 cells.

Manuscripts
Next, the proliferation of all three cell lines containing HT1080, MCF7, and MDA-MB-231 cells was inhibited by treatment with 5 µM BIA (Fig. 7A). The IC50 values at 3 days were 1.7 ± 0.1 μM for HT1080, 2.6 ± 0.4 μM for MCF cells, 2.6 ± 0.5 μM for MDA-MB-231 cells, and 2.4 ± 0.4 μM for SKBR3 cells. We also checked cell viability at three days, and confirmed inhibition by BIA at all cell lines ( Supplementary Fig. 10B). We also checked cell viability at three days, and confirmed inhibition by BIA at all cell lines ( Supplementary Fig.  10B). Moreover, HT1080 cells stably overexpressing TMBIM6 showed high sensitivity to BIA (Supplementary Fig. 10C).
6. B1A (Fig 7B) seems to decrease total protein expression of mTOR and rpl19, as well as TMBIM6. There seems to be no strong effect on disruption of the complexes, other than slight decrease in TMBIM6. The authors should quantitate this figure more carefully to reflect comparative amount of total proteins. Response: We appreciate the reviewer's comments. The total expressions of TMBIM6, mTOR, RICTOR, and RPL19 were not changed under the BIA, which was confirmed by immunoblot assay from size-fractionation samples. It is indicated that the BIA-induced effect is not related to the mTOR and rpl19 protein expression but more related to the regulatory effect on the complex formation between mTOR/RICTOR and RPL19. The updated Fig 7B is given below. Fig 7C. Response: We appreciate the reviewer comments. The updated Fig 7C is given below.

Expression of total mTOR and rictor proteins should be included in
8. The authors state in the Discussion that "This effect of TMBIM6 on cell growth differs from its classical role in ER stress." This has really not been addressed in the current study. Previous studies have already shown a role for mTORC2 in negative regulation of the ER calcium channel regulator Mid1 (in yeast) as a negative regulator of starvation response (PMID 27899413 Vlahakis et al JCB 2016). They should also cite this work to support the relationship of calcium signaling to mTORC2. Furthermore, they should verify that the disruption of TMBIM6 does not cause ER stress. The overall decrease in metabolic pathways and possible decrease in total ribosomal proteins (see comment 6) suggest that translation could also be decreased (which would relieve ER stress at least partially). BI-1 Ko condition is not related with ER stress, document it. They concern the possibility "The KO of BI-1 reduces ER stress, a possibility Response: We appreciate the reviewer's thoughtful comments, and respond to each comment individually.
About the issue "the correlation between mTORC2 and Ca 2+ ", our data support that leaky calcium from TMBIM6 affects directly mTORC2 activation through inducing the physical association of RICTOR with core proteins of the ribosome, RPL16, and RPS proteins. In a previous study about a role for mTORC2 in negative regulation of the ER calcium channel regulator Mid1 (in yeast) (Vlahakis et al. JCB 2016), the impairment of mTORC2-Ypk1 signaling activates calcineurin via Mid1, and the resultant autophagy flux disturbance by inhibition of GAAC response under conditions of amino acid starvation. However, we also reported that TMBIM6 rather enhances autophagy flux through Ca 2+ leaky characteristics (Kim et al., Autophagy 2020). Due to the TMBIM6 Ca 2+ leaky channel characteristics, mTORC2 binding affinity, and the recruitment of mTORC2 to the local Ca 2+ enriched area is the main thing to explain the correlation between mTORC2 and Ca2+.
In relation ER stress, the absence of TMBIM6 increases ER folding impairment and ER stress, enhancing UPR signaling (Chae et al., 2004;Lee et al., 2011). Most of these reports have examined the characteristics of TMBIM6 only in stress condition "in the presence of ER stress inducer". However, the presence or absence of TMBIM6 does not affect ER stress response under non-stressed conditions or in resting conditions (Chae et al., 2004;Lee et al., 2011). Therefore, TMBIM6 is not a simple ER stress regulator but rather an AKT activator enhancing cell proliferation, especially in cancer conditions.
In the revised discussion, we have updated as following.
Manuscripts mTORC2 was found to be associated with the ER through its direct binding to TMBIM6. We determined that TMBIM6 serves as a signaling scaffold that recruits mTORC2 to the ER and thus promotes cell survival. This effect of TMBIM6 on cell growth differs from its classical role in ER stress. TMBIM6 regulates ER stress-induced cell death 1,50 . The overall decrease of metabolic pathways and protein synthesis in the absence of TMBIM6 should have relieved ER stress at least partially. However, the absence of TMBIM6 increases ER stress, enhancing UPR signaling 1,50 . Most of these reports have examined the characteristics of TMBIM6 only in stress conditions "in the presence of ER stress inducer". However, the presence or absence of TMBIM6 does not affect ER stress response under non-stressed conditions or in resting conditions 1,50 . At least in a resting condition, TMBIM6 is not a simple ER stress regulator but rather a core protein enhancing protein recruitment and assembly, including mTORC2, ultimately affecting AKT activation and cell proliferation, especially in cancer condition.
Similar to mTORC1 studies 52-54 , we found that mTORC2 activity is sensitive to inhibition by BAPTA-AM ( Figure 6A-B, supplementary Fig. 7A), suggesting that intracellular calcium is required for mTORC2 activation. In a recent study about the correlation of Ca 2+ channel and mTORC2, mTORC2 negatively regulates Mid1, an ER/plasma membrane (PM)localized calcium channel regulatory protein. Decreased signaling of TORC2 and its downstream target protein kinase, Ypk1, induced Mid1 activation 55 . The mTORC2 signaling regulates the Ca 2+ -associated Mid as downstream effector under amino acid starvation, whereas the Ca 2+ -leaky TMBIM6 first affects mTORC2 activation through the releasable Ca 2+ from TMBIM6 under basal condition. Our point is not general cytosolic Ca 2+ but the local Ca 2+ from the Ca 2+ leaky protein, TMBIM6, which is also abrogated in the presence of BAPTA-AM, not BAPTA, a non-cell permeable Ca 2+ chelating agent. The precise mechanism of regulation of mTORC2 by calcium through TMBIM6 Ca 2+ leaky characteristics is distinct in our model. First, we demonstrated that the ER leaky calcium plays a unique and critical role in mTORC2 activation and the resultant AKT in mammalian cells. Second, D213A mutant had no effect on the interaction between TMBIM6 and RICTOR, ruling out the involvement of the local Ca 2+ in the direct binding with the mTORC2 subunit protein. However, we also found that D213A mutant had a strong controlling effect on the interaction between mTOR and RICTOR or between RICTOR and the ribosomal proteins, RPL19 and RPS16 suggesting the involvement of the local Ca 2+ in the assembly of mTORC2 and further the recruitment of ribosomal proteins also.

NCOMMS-19-31425-T
Comments to the authors, In this work, Kim and colleagues show that TMBIM6 -a member of the highly conserved TMBIM family of proteins -interacts with components of the mTORC2 complex to regulate cancer progression. Indeed, they first show that TMBIM6 is overexpressed in several different types of carcinomas and its expression correlates with poorer patient overall survival. Using loss-of-function approaches, the authors show that TMBIM6 promotes cell migration and invasion in vitro and tumor growth in vivo. These experiments are well performed and highlight the known role of TMBIM6 as a protumorigenic protein. In fact, work by this same group has previously shown that TMBIM6 overexpression promotes metastasis through the regulation of cell migration and invasion (PMID: 20118983). At the molecular level they show that TMBIM6 is required for the assembly of the mTORC2 complex, leading to improved glucose homeostasis and promoting cancer progression. In short, TMBIM interacts and tether the mTORC2 complex and the ribosomes to the ER where it would promote its assembly through TMBIM6 calcium leak channel function. Finally, the authors described a putative TMBIM6 antagonist with anti-tumor properties. Overall, the in vitro and in vivo results are very interesting and well-presented. However, I have some concerns regarding the proposed role of TMBIM6 as an mTORC2 regulator and its connection to cancer progression. Some key control experiments are currently missing, making the interpretation of the data difficult. The statistical analysis used throughout the manuscript should also be revised. A revised version of this manuscript should still be suitable for publication in Nature Communications.
Response: We appreciate that the reviewer found the significance of this study, and thank for his/her insightful comments and criticisms Major concerns: 1. Model of choice and generality: In Figure 1, the authors showed that TMBIM6 is overexpressed in several different cancer types, including cervical, endometrial and vulvar, breast, lung and prostate cancer. They also show that high TMBIM6 expression correlates with poorer prognosis in patients with squamous cell carcinoma and endocervical adenocarcinoma, esophageal carcinoma, skin cutaneous melanoma, head and neck squamous carcinoma and brain lower grade glioma, most of them of epithelial origins (Figure 1). However, they perform most of the loss-of-function experiments (migration, invasion, and tumor growth and mTORC2 biochemical experiments) in HT1080 cell lines, which is a fribrosarcoma cell line of mesenchymal origin. Two comments on this: (1) the authors should show, is it were, TMBIM6 expression and survival data on fibrosarcoma, since it will correlate better with the observations performed in HT1080 TMBIM6 KO cells and (2) the authors should repeat some key experiments (mTORC2 assembly, AKT phosphorylation and tumor growth) in an additional TMBIM KO cell line cancer model directly related to the cancers shown in figure 1. Most of the experiments connecting TMBIM6 with mTORC2 were performed in only one cancer cell line. Response: We appreciate the reviewer's comments, and responded to each comment individually.
(1) the authors should show, is it were, TMBIM6 expression and survival data on fibrosarcoma, since it will correlate better with the observations performed in HT1080 TMBIM6 KO cells Response: As per the reviewer's first comments, we searched TMBIM6 expression data in NCBI Gene Expression Omnibus database (NCBI/GEO), and confirmed that TMBIM6 expression was increased in sarcoma samples (GSE2719). Since fibrosarcoma is part of a wider family of sarcomas, we used sarcoma for survival analysis. Survival analysis of sarcomas (SARC) was represented in Fig. 1G, showing the correlation of the high expression of TMBIM6 with poor prognosis. Figure 1 (2) the authors should repeat some key experiments (mTORC2 assembly, AKT phosphorylation and tumor growth) in an additional TMBIM KO cell line cancer model directly related to the cancers shown in figure 1. Response: To solve the second comments, we generated scrambled and two TMBIM6 KO HeLa cell lines directly related to cervical cancer shown in Figure 1 using CRISPR/Cas9 systems. Next, we performed several experiments, including cell proliferation, PLA for mTORC2 assembly, immunoblotting for detection of AKT phosphorylation, and in vivo xenograft for tumor growth. TMBIM6 KO HeLa cells showed that cell proliferation, AKT phosphorylation, and tumor growth were decreased. The updated data and its corresponding text are given below.  (I) mRNA levels of indicated genes in TMBIM6 KO and WT HT1080 cells, as determined by qRT-PCR (n = 3 independent experiments). (J) Immunoblot analysis of anti-RICTOR IP and WCL of TMBIM6 KO and WT MEFs. WCL, whole cell lysates. Data represent mean ± SD. Statistical differences were detected with two-tailed unpaired Student's t-tests (D), and two-Way ANOVA followed by Bonferroni's test (H, I).

TMBIM6 depletion suppresses the tumorigenicity of cancer cells
To validate the above results, we performed cell proliferation, migration, and invasion assay. TMBIM6 KO HT1080, HeLa cells, and MEFs both exhibited slow growth relative to WT cells ( Fig. 2A), which was restored in TMBIM6 KO cells with re-expressing TMBIM6 (Supplementary Fig. 2A-B). Cell migration and invasion-as indices of cancer progressionwere inhibited in cells lacking TMBIM6 (Fig. 2B-C, Supplementary Fig 2C-D). To investigate the role of TMBIM6 in the growth of tumor cells in animals, we subcutaneously injected TMBIM6 WT and KO HT1080 cells into the left and right flanks, to immunecompromised mice and monitored tumor growth during 27 days ( Supplementary Fig. 2E). As shown in Fig. 2D, tumor formation in TMBIM6 KO was significantly reduced compared with that in WT cells over the same period. The weight of tumors originating from TMBIM6 KO HT1080 cells was decreased more than 6-folds than that of those originating from WT cells at the end of the experiment (Fig. 2E and F). Immunohistochemistry analysis of Ki-67 expression showed a significant decrease in tumors derived from TMBIM6 KO cells compared with that in those from WT cells (Fig. 2G). Consistently, we also found that tumor formation and weight, and the expressions of Ki-67 was apparently reduced in TMBIM6 KO HeLa cells compared with that in WT cells (Fig. 2H-K, Supplementary Fig 2F). In addition, TMBIM6 knockdown by injection of self-assembled micelle inhibitory RNA (SAMiRNA), a stable siRNA silencing platform for efficient in vivo targeting of genes, reduced tumor formation as well as Ki-67 expression when compared with those in the control groups ( Supplementary Fig. 2G-L). Taken together, these in vitro and in vivo experiments demonstrate that TMBIM6 regulates tumor growth.
2. Is the TMBIM6 KO HT1080 cell line a pool or was derived from a single clone? In the case of it being a clonal population, the authors should repeat some key experiments (i.e. AKT phosphorylation and the assembly of the mTORC2 complex) with additional clones. Throughout the manuscript, the authors compare the HT1080 TMBIM6 KO cells with wildtype controls when they should have been compared with a matched CRISPR/Cas9 scramble or mock control. I am worried about CRISPR/Cas9 off-target effects. Response: We generated two TMBIM6 KO and scrambled HT1080 cell lines as a single clone. Since the number of gene editing sequences at genome is different on allele by CRISPR/Cas9 system, we isolated individual clones by the limiting dilution method after transfection. Therefore, as reviewer's comments, we performed cell proliferation, cell migration, cell invasion, immunoblotting for AKT phosphorylation, PLA for the assembly of the mTORC2 complex, glucose consumption, and lactate production assays in additional clones (clone #2). All results obtained from TMBIM6 KO#2 clones showed similar results as those from TMBIM6 KO#1 clones. Moreover, AKT phosphorylation from scramble clone designated similar results those from WT cells. The updated data and corresponding text are given below. (C) Immunofluorescence using cells overexpressing TMBIM6 tagged with the N-terminal (HA-TMBIM6) and C-terminal (TMBIM6-HA) HA tag after permeabilization by digitonin or triton X-100.

Manuscripts
TMBIM6 is composed of six or seven transmembrane regions with mostly α-helical structures, which C-terminus of TMBIM6 resides in the cytosol by TMHMM or in ER intraluminal space by the bacterial homolog BsYetJ 40-44 (Additional Supplementary Item 2A). Although BsYetJ is a bacterial protein related to hTMBIM6, it has only 23.77% amino acid identity by Blastp (Additional Supplementary Item 2B). To further understand TMBIM6 topology, we performed immunofluorescence using cells overexpressing TMBIM6 tagged with the N-terminal (HA-TMBIM6) and C-terminal (TMBIM6-HA) HA tag, and selective membrane permeabilization reagents, by modified previous reports 42 . Digitonin makes epitopes in the cytosol, but not in the ER lumen, accessible to antibodies. In contrast, treatment with Triton X-100, permeabilizes all membranes and leads to staining of both luminal and cytosolic epitopes. The protein disulfide isomerase (PDI) retained in the ER lumen was used as a negative control. In the presence of digitonin, both HA-TMBIM6 and TMBIM6-HA represented similar fluorescence intensity, whereas PDI fluorescence was not detected (Additional Supplementary Item 2C). On the other hand, HA-TMBIM6, TMBIM6-HA, and PDI fluorescence was increased in the Triton X-100-induced cell membrane permeabilization conditions. These results suggest that N-terminal and C-terminal of TMBIM6 is cytosolic exposed in six-transmembrane structure condition, although we cannot exclude a possibility that topology of TMBIM6 might be altered by the fusion with HA itself as previously mentioned 42 . 5. In Figure 6, the authors reconstituted TMBIM6 KO cells with either TMBIM6-HA or the channel mutant TMBIM D213A and assessed calcium leak ( Figure 6A), mTORC2 assembly ( Figure 6B) and AKT phosphorylation (Figures 6D and 6E). However, to correctly interpret these experiments, a comparison of the total reconstituted levels of TMBIM6 and TMBIM6 D213A proteins in KO cells is necessary. The authors should also check the ER localization of the WT and the D213A mutant proteins by immunofluorescence. Differential partial reconstitution of these proteins may account for the differences in mTORC2 assembly and AKT phosphorylation. Response: In our recent report, we documented the ER localization of TMBIM6 WT and TMBIM D213A (Kim et al., 2020). Moreover, the mutant with C-terminal nine amino acids of the protein was replaced by alanines (BI-1 C9A ) was also localized at ER, which was consistent with a previous report (Lisbona et al., 2009). Consider that TMBIM6 has six or seven transmembrane domains, and the D213A mutant would not affect TMBIM6 topology. However, in accordance to reviewer's suggestions, we performed immunofluorescence in TMBIM6-HA WT or TMBIM6-HA D213A-transiently overexpressed TMBIM6 KO cells. As shown in supplementary Fig. 7C, both of TMBIM6 WT and D213A expression were observed in the ER. Thus, the differences about mTORC2 assembly ( Figure 6B) and AKT phosphorylation between the TMBIM6 and the mutant D213A ( Figures 6D and 6E) are suggested to be related with the characteristics of TMBIM6, "Ca 2+ leakage", not differential reconstitution. The updated data and its corresponding text are given below. Supplementary Fig. 7 Ca 2+ regulates mTORC2 activation. (A) PLA between the indicated proteins (red dots) in HT1080 cells treated with BAPTA-AM (10 μM), BAPTA (10 μM), and EGTA-AM (10 μM). Scale bar, 15 μm. Right, quantification of red dots (n = 3 independent experiments). (B) Illustration of TMBIM6-GCaMP3 by a genetically-encoded Ca 2+ indicator (GCaMP3) fused directly to the C-terminus of TMBIM6 (TMBIM6-GCaMP3). (C) TMBIM6 and D213A expression-rescued KO cells were stained for calnexin (CANX, ER marker). (D) The scheme of TMBIM6 characteristics; TMBIM6-leaky Ca 2+ and the interaction with mTORC2 and ribosome complex. Data represent mean ± SD. Statistical difference was detected with one-Way ANOVA followed by Tukey's test (A).

Manuscripts
To identify whether differential retention of TMBIM6 WT or D213A mutant in ER affects mTORC2 assembly and AKT phosphorylation, we performed immunofluorescence in HT1080 cells transiently transfected with WT TMBIM6-HA and D213A mutant. As shown Supplementary Fig. 7C, TMBIM6 WT and D213A were retained in ER at a comparable level. Thus, TMBIM6-associated AKT activation is based upon the following characteristics of protein interactions: the binding of TMBIM6 with RICTOR was independent of Ca 2+ leakage, whereas the interaction of TMBIM6 with mTOR or ribosomal subunits including RPL19 was dependent on local Ca 2+ leakage. The Ca 2+ -relevant TMBIM6 state was schematically described (Supplementary Fig. 7D).
6. Specificity of BI: from the data presented in Figures 7 and 8, and Supplementary Figure 7, the authors suggest that BIA acts as a TMBIM6 antagonist, blocking ER calcium release, leading to mTORC2 disassembly and inhibition of TMBIM6-associated tumorigenicity. Although it is clear that BIA reduces cell proliferation, migration and invasion, it is not so clear that these effects dependent on BIA's function as a TMBIM antagonist. Does BIA work when TMBIM6 is knocked out? There are also several key control experiments missing from figures 7 and 8 that preclude the interpretation of these experiments. For example, total mTOR and RICTOR levels are missing from the IP shown in Figure 7C. The authors should also report the effect of BIA on cell viability under the conditions reported in Figure 7A. The data supporting the role of BIA as an inhibitor of the interaction between TMBIM6 and mTORC2 is currently not convincing enough. The authors could perform PLA as in previous experiments to strengthen this data. Response: We appreciate the reviewer's comments, and responded to each comments individually.
-Does BIA work when TMBIM6 is knocked out? As per the reviewer 1's comments, we performed cell proliferation, migration assay, and immunoblotting about AKT signaling in BIA-treated TMBIM6 KO cell lines. As shown in Fig. 7A, and Supplementary Fig. 10E-G, the proliferation, migration, and the status of AKT phosphorylation of TMBIM6 KO HT1080 cells were not significantly affected even in the presence of BIA.
-Total mTOR and RICTOR levels are missing from the IP shown in Figure 7C. In the revised version, we added image of blot for total mTOR and RICTOR, and updated them in the Figure 7C.
-The authors should also report the effect of BIA on cell viability under the conditions reported in Figure 7A. In the revised version, we have added cell viability data in supplementary Fig. 10B under the same condition as reported in the original Figure 7A.
-The data supporting the role of BIA as an inhibitor of the interaction between TMBIM6 and mTORC2 is currently not convincing enough. The authors could perform PLA as in previous experiments to strengthen this data. As per the reviewer's suggestions, we performed PLA assay to elucidate the BIA effect "inhibition of the interaction between TMBIM6 and mTORC2 or between TMBIM6 and ribosome". As shown in Figure 7E and Supplementary Fig. 10D, the interaction of TMBIM6 with mTORC2 components was decreased in the BIA-treated cells. The updated data and corresponding text are given below.

Manuscripts
To identify whether BIA decreases the binding between TMBIM6 and mTORC2, we performed a gel filtration assay in TMBIM6-overexpressing HT1080 cells after treatment with BIA during 24 h. As shown in Fig. 7B, TMBIM6-HA was co-eluted with mTORC2 components (mTOR and RICTOR) and ribosomes (RPL19), whereas the co-elution pattern was delayed in the BIA-treated HT1080 cells, indicating that BIA induces the dissociation of TMBIM6 from mTORC2 and ribosome. Interaction between RICTOR and mTOR, or RICTOR and RPL119, or RICTOR and RPS16 by PLA assay was decreased in HT1080 cells with BIA compared to control cells ( Supplementary Fig. 10D). In addition, BIA decreased the binding of TMBIM6 to mTORC2 and inhibited the phosphorylation of AKT (Fig. 7C). Consistently, the phosphorylation of AKT was fully decreased in three breast cancer cell lines, MCF7, MDA-MB-231, and SKBR3 cells (Fig. 7D). To elucidate whether BIA impaired interaction between TMBIM6 and mTORC2 or ribosome, we performed the PLA assay. As shown in Figure 7E, interactions of mTOR, RICTOR, or RPL19 with TMBIM6 were decreased in BIA-treated cells compared to control cells.
7. Overstatements and unsupported evidence: In page 18 the authors state that "BIA is a newly identified TMBIM6 antagonist that disrupts the TMBIM6-mTORC2 interaction, leading to the inhibition of tumor growth even in cases that are resistant to mTOR inhibitors". However, the authors have not provided evidence suggesting this. They have only shown that BIA decreases tumor growth in vitro and in vivo and that it works in combination with other mTORC2 inhibitors to kill HT1080 cells. They have not shown that these effects are a direct consequence of BIA's putative role as a TMBIM6 antagonist or as a disruptor of TMBIM6-mTORC2 interaction. Throughout the manuscript there are several instances where the authors jump from a phenotypic observation to a molecular mechanism whose causal connection to the observation has not been directly demonstrated. Response: To show these effects are a direct consequence of BIA's putative role as a TMBIM6 antagonist or as a disruptor of TMBIM6-mTORC2 interaction, we added the other data, including PLA analysis. In this study, BIA has emerged as a potential candidate as an antagonist regulating TMBIM6-mTORC2 interaction and its related tumor growth even in cases that are resistant to mTOR inhibitors. Besides, we carefully mentioned these effects and mechanisms of BIA, not jumping to the conclusion from a phenotypic observation throughout this revised manuscript. In the revised version, we mentioned about the point as following, Supplementary Fig. 11 BIA decreases cell survival. (A) The images of crystal violet staining in HT1080, PANC-1, Capan-1, and MIA PaCa-2 cells after treatment with 10 μM BIA and mTOR inhibitors. Right; quantification of cell viability normalized to control cells. (B) PLA between the indicated proteins (red dots) in BIA or mTOR inhibitors-treated PANC-1 cells. Right, quantification of red dots (n = 5 independent experiments). Scale bar, 20 μm. Data represent mean ± SD. Statistical differences were detected with one-Way ANOVA followed by Tukey's test (A), and two-Way ANOVA followed by Bonferroni's test (B).

Manuscripts
To determine whether BIA is effective against HT1080, PANC-1 pancreatic cancer cells resistant to mTOR inhibitor, and other pancreatic cancer cells including Capan-1 and MIA PaCa-2 cells, we compared the growth of cells treated with BIA to that of cells treated with anti-mTOR inhibitors such as AZD8055, INK128, Omipalisib, OSI-027, and Voxtalisib. Cell viability was reduced to a greater extent by the treatment with BIA as compared to the other agents ( Supplementary Fig. 11A). Especially, BIA almost abrogated live cells in PANC-1 cells, which have 30 ∼ 40% cell viability by the other mTOR inhibitors, including AZD8055, INK128, Omipalisib, OSI-027, and Voxtalisib. In PLA assay, BIA diminished association between RICTOR and mTOR or between RICTOR and RPL19, but the other anti-mTOR inhibitors did not affect any association in PANC-1 cells ( Supplementary Fig.  11B), suggesting that BIA has potential as an effective anticancer agent controlling cancer cell survival although the experiment is only in vitro state.
To this end, BIA has emerged as a potential candidate as an antagonist regulating TMBIM6-mTORC2 interaction and its related tumor growth even in cases that are resistant to mTOR inhibitors. Specifically, the characteristics of BIA indicate the dissociation between RICTOR and TMBIM6 through the regulation of TMBIM6-leaky Ca 2+ and the resultant inhibition of AKT activation impeding cancer formation.
8. Statistics: The authors should review the statistical procedures used throughout the manuscript. When there are two independent variables (e.g. genotype and time) the authors should use a Two-Way ANOVA followed by post hoc test instead of multiple two-tailed unpaired Student's t-tests . Figures where Two-Way ANOVA should be used include Figures  1E, 2A, 2D, 2E, 2H, 3B, 3F, 4A, 4E, 4H, 8E and 8G. When there is one independent variable but more than two groups are compared, the authors should use a One-Way ANOVA followed by post hoc test. Examples of these include Figures 2I, 2K, 6D and 6E. Finally, the authors should specify p-values as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001. Response: In accordance with the reviewer's recommendation, we performed the statistical reanalysis. One-way analysis of variance (ANOVA) with Tukey post hoc test, Two-Way ANOVA followed by Bonferroni post hoc test, and Student's unpaired t-test were performed using Prism v.8 software. Also, *, p < 0.05; **, p < 0.01; ***, p < 0.001, were considered statistically significant. In each case, the statistical test used is indicated, and the number of experiments is stated in the legend of each figure.
Minor comments: Figures 2B and C, the authors should also express the data as the percentage of migrating and invading cells and not only as fold changes compared to WT. Response: In accordance with the reviewer's suggestion, we represent two types of data "fold and percentage of migrating and invading cells in Figures Figure 3C the authors should show the blot results of the second mTORC2 target NDRG1 (pT346) in the HT1080 TMBIM6 KO cells reconstituted with TMBIM6-HA. Response: In addition to reviewer 1's comments, we performed immunoblotting of NDRG1 T346 in TMBIM6 KO HT1080 cells with re-expression of TMBIM6. The updated Figure 3C and corresponding text are given below.

Manuscripts
Immunofluorescence analysis revealed and confirmed that the phosphorylation of AKT was decreased in TMBIM6 KO as compared to that in wild-type (WT) cells ( Supplementary Fig.  3D). Consistently, overexpressing TMBIM6 in HeLa cells increased mTORC2 activity ( Supplementary Fig. 3E). Reintroducing TMBIM6 into TMBIM6 KO HT1080 cells restored AKT Ser473 and NDRG1 Ser939 phosphorylation comparably (Fig. 3C).
3. In page 7, the authors describe the effects of insulin referencing supplementary figure 2D and 2E. This is not correct; it should read Supplementary Figure 3D and 3E respectively. Response: We apologized incorrect referencing Supplementary Figure 2D and 2E in the original text. In the revised version, we changed the original supplementary Fig. 3D and Supplementary Fig. 3E to Supplementary Fig. 3F and 3G, respectively. 4. There are many misplaced conclusions throughout the manuscript. For example, in page 11 the authors say that "…TMBIM6 interacts with mTORC2 and ribosomes and that this interaction is important for the kinase activity of mTORC2". However, the effects of TMBIM6-mTORC2 interaction in AKT phosphorylation are explored in the following paragraph ( Figure 5H). Response: As per the reviewer's suggestion, that specific sentence was rearranged in the following Figure 5H.

Manuscripts
The phosphorylation of AKT Ser473 was also decreased in RPL19-non-associated TMBIM6 mutants (Fig. 5H, I). Taken together, these data indicate that TMBIM6 interacts with mTORC2 and ribosomes and that this interaction is important for the kinase activity of mTORC2.
5. How does calcium leak from the ER increase mTORC2 assembly efficiency? The authors should discuss this in more detail. Response: This part is very important part for this study. D213A mutant role and the presence of BAPTA-AM control the mTORC2 assembly. In addition, the PLA between mTOR and rictor and between mTOR and RPL19 and RPS16 was shown as a positive evidence showing that calcium leak from the ER increased mTORC2 assembly efficiency. According to the reviewer's comment, we have updated the following part in the revised discussion.

Manuscripts
Similar to mTORC1 studies 52-54 , we found that mTORC2 activity is sensitive to inhibition by BAPTA-AM ( Figure 6A-B, supplementary Fig. 7A), suggesting that intracellular calcium is required for mTORC2 activation. In a recent study about the correlation of Ca 2+ channel and mTORC2, mTORC2 negatively regulates Mid1, an ER/plasma membrane (PM)-localized calcium channel regulatory protein. Decreased signaling of TORC2 and its downstream target protein kinase, Ypk1, induced Mid1 activation 55 . The mTORC2 signaling regulates the Ca 2+associated Mid as downstream effector under amino acid starvation, whereas the Ca 2+ -leaky TMBIM6 first affects mTORC2 activation through the releasable Ca 2+ from TMBIM6 under basal condition. Our point is not general cytosolic Ca 2+ but the local Ca 2+ from the Ca 2+ leaky protein, TMBIM6, which is also abrogated in the presence of BAPTA-AM, not BAPTA, a not cell permeable Ca 2+ chelating agent.
The precise mechanism of regulation of mTORC2 by calcium through TMBIM6 Ca 2+ leaky characteristics is distinct in our model. First, we demonstrated that the ER leaky calcium plays a unique and critical role in mTORC2 activation and the resultant AKT in mammalian cells. Second, D213A mutant had no effect on the interaction between TMBIM6 and RICTOR, ruling out the involvement of the local Ca 2+ in the direct binding with the mTORC2 subunit protein. However, we also found that D213A mutant had a strong controlling effect on the interaction between mTOR and RICTOR or between RICTOR and the ribosomal proteins, RPL19 and RPS16 suggesting the involvement of the local Ca 2+ in the assembly of mTORC2 and further the recruitment of ribosomal proteins also.
Reviewer #4 (Remarks to the Author) (expertise: cancer, zebrafish model) : This is a continuous work of the authors´ previous claims that TMBIM6/BI-1 promotes tumor growth and progression (ref 9). In this study, the authors provide new evidence of potential signaling events of the TMBIM6/BI-1-mTORC2-AKT axis in different cancer types. First, they show that in several cancer types TMBIM6/BI-1 expression is elevated, which is largely confirmed by GCTA analysis. Using in vitro KO technology, they show that deletion of TMBIM6/BI-1 in cancer cells retards cancer cell proliferation, migration, and tumor formation. They then defined signaling pathways that potential involved in these tumorigenic activities. Finally, they screened a chemical library to identify a potential inhibitor BIA, which inhibits tumor growth in mice and in fish.

Comments:
1) This work covers almost all respects of cancer development, including cancer cell proliferation, migration, survival, tumor formation, and metastasis. From the provided data, it is hard to believe that TMBIM6/BI-1 is a master regulator of cancer development. In particular, the characterized signaling events do not support TMBIM6/BI-1 as a master regulator. The authors should focus on a particular signaling and activity to obtain an indepth mechanistic insight, but not a wikipedia-assocaited cancer development. Response: Various oncogenic pathways are related to mTOR signaling. Among the mTOR complexes, mTORC1 function is hyperactivated in up to 70% of all human tumors (Forbes et al., 2011;Xie et al., 2016). mTORC2 is linked to PI3K signaling, also highly activated in many tumor cells. The mechanism for mTORC1 activation has been well established, but that for mTORC2 activation is not well understood. In this study, it is aimed that the role of TMBIM6 is ultimately linked to mTORC2 and AKT, a general cancer master signaling. However, we agree with the reviewer's comment. We don't want to show the role of TMBIM6 as a master regulator in cancer progression. Through the developed mTOR inhibitors and the other anti-cancer agents' application to the resistant cancer models (Supplementary Figure 11), newly targeted molecule/protein-protein interaction would be a strong target toward the resistant cancer cases, a main theme in this study. The unique point is that TMBIM6 is a Ca 2+ regulator controlling local area surrounding ER, so that the binding affinity of some proteins such as mTORC2 etc. can be increased. Due to the local Ca 2+ environment, the interaction of TMBIM6 with mTORC2 and the resultant AKT activation contributes to the cancer development, a summary in this study. In this revised version, we did our best to focus on particular signaling and activity to obtain an indepth mechanistic insight; local Ca 2+ -based mTORC2 and ribosome recruitment and the resultant AKT activation. We appreciate returning comments.
2) The TCGA data and their own experimental findings do not completely match. What about lung cancer, pancreatic cancer, and prostate cancer in TCGA? If TMBIM6/BI-1 is also highly expressed in these cancers, but not associated with poor survival, what does this mean? I am almost certain that some cancer types express high levels of TMBIM6/BI-1, but lack clinical correlation. The authors should not only choose the results favor to your findings. Response: According to the reviewer's comments, we performed survival analysis about lung cancer, pancreatic cancer, and prostate cancer. TMBIM6 is highly expressed in lung cancer (502.54 TPM for tumor, 330.77 TPM for normal), pancreatic cancer (379.38 TMP for tumor, 115.42 TPM for normal), and prostate cancers (505.34 TPM for tumor, 258.46 TPM for normal) using GEPIA2. In survival analysis, lung and pancreatic cancer were also associated with poor survival. In prostate cancer, the lists of cancers in OncoLnc database were not included, so we could not examine the correlation between the TMBIM6 expressions and prostate cancer survival. Also, we were not confirmed in GEPIA2 analysis. However, the group with high expression of TMBIM6 showed poor survival in either altered or and unaltered group using all datasets in cBioPortal, suggesting that high expression of TMBIM6 may affect patient survival. We need to further study for prostate cancer.
In the revised version, we added the data in Figure 1G   Supplementary Fig. 1 3) The zebrafish cancer model is useless in this experimental setting. Are BIA-treated cancer cells dead in the fish body? How do they discriminate living cancer cells from dead cancer cells? How do they know that BIA is active in fish body? Response: Zebrafish model was applied to utilize the transparent embryo to visualize the migration pattern or behavior of the injected cancer cells. Later, BIA treatment prevented the migration of injected cancer cells to other parts, whereas in control tumor cells migrated away from the primary site of injection. Our objective from this experiment was to visualize the migration behavior to confirm the efficacy of BIA in preventing migration of tumor cells, thereby effective in the prevention of spread. Moreover, developing embryos injected with cancer cells were grown in water containing 2.0 μM BIA, which itself makes sure that the activity of BIA. Several other experiments in this study sufficiently prove the cause thus, reviewers might have thought that zebrafish experiments as unnecessary. However, to confirm migration behaviors of cancer cells with visualization zebrafish model is the possible solution. Hence, zebrafish model was used in the study. 4) BIA looks like a non-specific chemical compound as most chemical compounds do. These data can only be used as indirect supportive data. Response: To show the specificity, we added the application of BIA to the TMBIM6 KO cells in the revised version. Compared with the WT cells, the BIA had no effect at the TMBIM6 KO cells, indicating the specificity of BIA as an inhibitor of TMBIM6. As a screening approach for the development of TMBIM6 antagonist, BIA has emerged as a potential candidate as an antagonist regulating TMBIM6-mTORC2 interaction and its related tumor growth even in cases that are resistant to mTOR inhibitors. To show these effects are a direct consequence of BIA's putative role as a TMBIM6 antagonist or as a disruptor of TMBIM6-mTORC2 interaction, we added the other data, including PLA analysis.
5) The manuscript needs language editing. Response: As per the reviewer's advice, we get editing help from someone with full professional proficiency in English.
Additional changes made in the revised manuscript are as follows: 1) In method section, the R1 primer sequence for confirming TMBIM6 KO cells by CRISPR/Cas9 genome editing was incorrect, so that we have been changed to R1, 5'-TCAATCCTGCCTCTCCTGAT-3'.
2) We apologize missing explanation about some method, and the updated text is as below.
Live cell imaging Live cell imaging with BI-GCaMP3 and G-CEPIAer was performed using LSM 880 microscopy. Briefly, 2 × 10 5 HT1080 cells stably expressing GCaMP3-ML1 or G-CEPIAer were cultured in a 35-mm confocal dish. Changes in fluorescence levels were monitored for 20 min upon addition of BIA in Ca 2+ -free external solution containing 145 mM NaCl, 5 mM KCl, 3 mM MgCl 2 , 10 mM glucose, 1 mM EGTA, and 20 mM HEPES (pH 7.4). The intensity of fluorescence was measured using ZEN software.

Human phospho-kinase array
Human Phospho-kinase arrays were performed according to manufacturer's instructions (ARY003B, R&D Systems, Minneapolis, MN, USA). The quantification of pixels was performed using Fiji ImageJ software.
3) We apologize missing to explanation about phospho-kinase profiling in Figure 3A. The updated text is below.

4)
We have added four authors who helped us to conduct experiments that are necessary for the revision. New coauthors, Suvarna H Pagire, screened the new small-molecule inhibitor with Jin Hee Ahn. Hyun Ju Yoo performed the metabolite analysis such as glycolysis, TCA, PPP, and HBP. Jaeseok Han and Duckgue Lee performed the ribosome profiling assay. Kyung-Woon Kim performed the T4 phage display screening. 5) In revised manuscripts, we added some sentences to support our explanation.
-In above our results, since TMBIM6 regulates mTORC2 activation through ER Ca 2+ release, we measured TMBIM6-GCaMP3 green fluorescence for real-time by application of 10 μM BIA to identify whether BIA inhibits Ca 2+ release from TMBIM6.
-To investigate the role of TMBIM6 in the growth of tumor cells in animals, we subcutaneously injected TMBIM6 WT and KO HT1080 cells into the left and right flanks, to immune-compromised mice and monitored tumor growth during 27 days ( Supplementary Fig.  2E-F).