Oxysterol-binding protein-related protein 5 (ORP5) promotes cell proliferation by activation of mTORC1 signaling

Oxysterol-binding protein (OSBP) and OSBP-related proteins (ORPs) constitute a large family of proteins that mainly function in lipid transport and sensing. ORP5 is an endoplasmic reticulum (ER)-anchored protein implicated in lipid transfer at the contact sites between the ER and other membranes. Recent studies indicate that ORP5 is also involved in cancer cell invasion and tumor progression. However, the molecular mechanism underlying ORP5’s involvement in cancer is unclear. Here, we report that ORP5 promotes cell proliferation and motility of HeLa cells, an effect that depends on its functional OSBP-related domain (ORD). We also found that ORP5 depletion or substitutions of key residues located within ORP5-ORD and responsible for interactions with lipids interfered with cell proliferation, migration and invasion. with the protein mechanistic target of rapamycin


INTRODUCTION
Alterations in cell signaling pathways that control normal cell growth are very common in cancer. In particular, deregulation of the mechanistic target of rapamycin (mTOR) signaling is known to play an important part in cancer development (1,2). mTOR is an evolutionarily conserved protein kinase that is present in two distinct complexes, each containing several other proteins, termed mTOR complex 1 (mTORC1) and mTORC2. mTORC1 is a central regulator of cell proliferation. It senses nutrients, especially amino acids, as well as energy levels and stress, and regulates cell proliferation in response to these signals. In this regard, activated mTORC1 phosphorylates and activates its major downstream effector ribosomal protein S6 kinase (S6K). Once activated, S6K phosphorylates the ribosomal S6 protein to control fundamental cellular processes including protein synthesis, lipid metabolism, and cell proliferation (3,4).
Oxysterol-binding protein (OSBP)related protein 5 (ORP5) belongs to a large family of lipid transfer proteins. In mammalian cells, the OSBP/ORPs protein family consists of twelve members with variant sequence identities, giving rise to six subfamily groups (5,6). OSBP/ORPs share a characteristic feature of a conserved carboxyl-terminal OSBP related domain (ORD) or ligand-binding domain. The N-termini of these proteins often possess a FFAT motif (diphenylalanine in an acidic tract) and a pleckstrin homology (PH) domain. The FFAT motif and the PH domain are responsible for targeting OSBP/ORPs to endoplasmic reticulum (ER) membranes and non-ER organelle membranes, respectively (7)(8)(9). These two membrane-targeting determinants enable OSBP and some ORPs to function at ERassociated membrane contact sites, where they contribute to the intracellular exchange of lipids (10). For example, OSBP associates with the Golgi through its PH domain and the ER through FFAT/VAPs interaction (9). Tethering the ER and Golgi, OSBP has been shown to transport cholesterol from the ER to the Golgi by the ORD, where it exchanges cholesterol with PI(4)P (phosphatidylinositol-4-phosphate) and transfers it back to the ER (11).
Within the OSBP/ORPs family, ORP5 and ORP8, which lack the FFAT motif, are the only members that have a transmembrane domain anchoring them to the ER (12,13). ER-anchored ORP5 has been identified as a phosphatidylserine (PS) transporter through its ORD (14,15). In a counter-transport process occurring at the membrane contact sites, ORP5 and ORP8 were shown to transfer PS from the ER to the plasma membrane and PI(4)P and PI(4,5)P 2 from the plasma membrane to the ER (15,16). Interestingly, ORP5 and ORP8 were also reported to transport PS from the ER to mitochondria and maintain proper mitochondrial function (17).
Despite seemingly working together to transport lipids at the membrane contact sites, ORP5 and ORP8 appear to have different functions in the regulation of cellular homeostasis (18). Notably, ORP8 expression reportedly inhibits cell proliferation and tumor growth while enhancing apoptosis (19). In contrast, ORP5 expression has been linked to increased cancer cell invasion and metastasis. For example, a previous report showed that the invasion rate of both hamster and human pancreatic cancer cells is enhanced by ORP5 overexpression and reduced by ORP5 depletion (20). Importantly, analysis of clinical samples suggested that poor prognosis in human pancreatic cancer is associated with high expression levels of ORP5 (20). In a recent study, tissue microarray analysis revealed that ORP5 is highly expressed in lung tumor tissues, especially in the lung tissues of metastasispositive cases (21).
Why overexpression of ORP5 is positively correlated with cancer cell invasion and tumor progression is unknown. Specifically, it is unclear whether this correlation involves alterations of cell signaling pathways. In the current study, we provide evidence that mTORC1 signaling is positively regulated by ORP5. Our data suggest that ORP5 facilitates cancer cell proliferation and motility at least in part through mTORC1 activation.

ORP5 regulates mTORC1
Overexpression of ORP5 promotes cell proliferation and migration Elevated expression of ORP5 has been associated with pancreatic and lung cancers (20,21). One of the aims of this study is to verify whether increased expression of ORP5 can drive cell proliferation. For this purpose, we chose HeLa cells because the level of endogenous ORP5 in HeLa cells is relatively low. This system not only allowed us to test wild type ORP5, but also mutant forms of ORP5 in cell proliferation. To study the effect of ORP5 overexpression, we used a retroviral vector system (Clontech pQCXIN) and generated a stable line of HeLa cells overexpressing ORP5. Using a goat polyclonal ORP5 antibody, we detected a robust band of ~110 kDa corresponding to ORP5 in the stable cell line (HeLa/ORP5) ( Figure 1A, lane 2). It has been shown that overexpression of ORP5 induces pancreatic cancer invasion and lung tumor formation (20,21). We predicted that overexpression of ORP5 in HeLa cells may facilitate cell proliferation and motility. To test this hypothesis, we carried out cell proliferation and migration assays in HeLa/ORP5 and Mock cells.
Indeed, compared to the control, HeLa/ORP5 cells had a significantly higher proliferation rate (increased by ~50%) ( Figure  1B). Cell migration rate was also significantly enhanced in HeLa/ORP5 cells by ~60% compared with the Mock control ( Figure 1C). If ORP5 overexpression facilitates cell proliferation and migration, depletion of ORP5 should blunt this effect. We used a set of three specific siRNAs against ORP5 to efficiently reduce ORP5 expression in HeLa/ORP5 cells ( Figure 1A, lane 6-8). As expected, depletion of ORP5 inhibited both cell proliferation and migration to an extent comparable to that of the mock control cells ( Figure 1D and 1E).

A functional ORP5-ORD is critical for facilitating cell growth and motility
ORP5 is a lipid transfer protein and its ORD domain is responsible for extracting, binding and transporting lipid cargos between the ER and other organelle membranes (12,16,17). To investigate whether a functional ORD of ORP5 is required for facilitating cell proliferation and migration, we mutated some critical residues within ORP5-ORD conserved for binding PS (L389D) (14), PI(4)P (K446A, H478A/H479A, K670A) or possibly cholesterol (H538A/K540A) (16,22) ( Figure  2A). Transient transfection in HeLa cells with GFPfused wildtype ORP5 and these mutant cDNAs displayed comparably high transfection efficiency under fluorescent microscope (70-80%, Figure S1). The transiently transfected HeLa cells were used for testing cell proliferation and migration. Compared with empty vector control, wild-type ORP5 transfection significantly enhanced cell proliferation and migration ( Figure 2B, C and D), which is consistent with the observations in HeLa/ORP5 vs HeLa control cells ( Figure 1B and C). A significant reduction of cell proliferation and migration rates was seen in the cells transfected with lipid binding mutants compared with wild-type ORP5 ( Figure 2B, C and D). Importantly, the PS binding mutation (L389D) and the PI(4)P binding mutations (H478A/H479A) also resulted in a significant reduction of cell invasiveness based on the assay using the Matrigel invasion chamber ( Figure 2E), highlighting the critical importance of a functional ORP5-ORD in facilitating cell motility. Together, these data suggest that lipid binding/transfer function of ORP5 is essential for its role in promoting cell growth.

ORP5 interacts with mTOR
We hypothesized that the role of ORP5 in promoting cell proliferation may involve other ORP5-interacting proteins. To this end, we performed liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis to identify potential ORP5 interacting proteins.

ORP5 regulates mTORC1
The goat polyclonal ORP5 antibody was incubated with the cell lysates prepared from HeLa/ORP5 and immunoprecipitation performed with normal goat IgG as a control. After immunoprecipitation, the efficiency of ORP5 pull down was verified by Coomassie blue staining and anti-ORP5 Western blot analysis ( Figure S2A, B). LC-MS/MS analysis of ORP5 immuoprecipitates identified a list of proteins ( Figure S2C), among which we selected mTOR (MASCOT ID score: 32.5 -49.6; peptide hits = 26) for further investigation since this serine/threonine-protein kinase is a master regulator of cell proliferation and survival (1).
We performed immunoprecipitation experiments to confirm ORP5-mTOR interaction.
Firstly, we analyzed samples pulled-down by the control or ORP5 antibody from the HeLa/ORP5 cell lysates. Western blot analysis using antibodies to ORP5, mTOR, Calnexin (an ER integral-membrane protein), Akt (an upstream effector of mTOR) and p70-S6K (a downstream of mTOR signaling) showed that only mTOR was detected along with ORP5 ( Figure 3A, lane 4). Secondly, we transiently transfected HeLa cells with cDNAs for mCherry empty vector (EV), mCherry-fused ORP5 or ORP8, and used anti-RFP antibody to carry out immunoprecipitation. Anti-mTOR Western blot analysis demonstrated that mCherry-ORP5, and not mCherry-ORP8, interacted with mTOR ( Figure 3B, lane 5). We next used the in situ proximity ligation assay (PLA) to further verify ORP5-mTOR interaction. PLA allows direct visualization and quantitation of protein-protein interaction. When HeLa/ORP5 cells were incubated with ORP5 primary antibody alone, only negligible PLA signal was observed and hardly detectable. However, when the cells were treated with both ORP5 and mTOR primary antibodies, there was a robust, significantly amplified PLA signal that could be clearly detected in every single cell ( Figure 3C, D). The PLA signal observed in these experiments was specific, as the signal was significantly weakened in cells depleted with ORP5 by siRNAs ( Figure 3E, F).
Next, we wanted to examine which domain of ORP5 is required for its interaction with mTOR. We constructed GFP-fused ORP5 deletion mutants that lack the N-terminal PH domain (GFP-ORP5ΔPH), ORD (GFP-ORP5ΔORD), as well as both the PH domain and ORD (GFP-ORP5ΔPHΔORD) ( Figure 4A). GFP empty vector (EV), GFP-fused wild-type ORP5, or the deletion mutants were cotransfected into HEK-293 cells together with Myc-mTOR cDNA.
Anti-GFP immunoprecipitation analysis revealed that the PH domain of ORP5 is dispensable for its interaction with mTOR as GFP-ORP5ΔPH similarly associated with myc-mTOR when compared to wild-type GFP-ORP5 ( Figure 4B, lane 8). In stark contrast, deletion of ORD severely diminished the physical association between ORP5 and mTOR ( Figure 4B, lane 9, lane 10), suggesting that ORP5-ORD is required for ORP5-mTOR interaction. Consistent with this observation, the reverse anti-Myc immunoprecipitation also demonstrated that mTOR strongly associates with wild-type GFP-ORP5 but not GFP-ORP5ΔORD ( Figure 4C, lane 5 vs lane 6). The dependency on ORP5-ORD appears to be separable from its lipid transfer activity, as the point mutations within ORP5-ORD (L389D and H478A/H479A) abolishing PS or PI(4)P binding did not affect ORP5-mTOR interaction ( Figure 4D, 4E). This may be due to the observation that mutations abolishing lipid transfer capability do not alter the overall conformation of ORP5-ORD (14,22).
Taken together, these results demonstrate that ORP5 interacts with mTOR and that ORP5-ORD is required for this interaction.

ORP5 is involved in mTORC1 signaling
by guest on July 24, 2018 http://www.jbc.org/ In mammalian cells, mTOR forms two functionally distinct complexes, namely mTORC1 and mTORC2. mTORC1 contributes to cell growth and proliferation by directly phosphorylating S6 kinase (S6K), which activates ribosomal S6 protein, ultimately promoting protein synthesis (3). The role of ORP5 in facilitating cell proliferation and the association between ORP5 and mTOR prompted us to ask whether ORP5 plays any role in mTORC1 signaling.
We treated HeLa/Mock and HeLa/ORP5 cells with control or ORP5 siRNAs and analyzed the phosphorylation of S6K and S6. When treated with control siRNA, a significant increase of S6K and S6 phosphorylation was seen in HeLa/ORP5 cells ( Figure 5A, lane 3 vs lane 1, 5B, 5C), indicating that overexpression of ORP5 in HeLa cells induces mTORC1 activation. ORP5 depletion resulted in downregulation of the phosphorylation of S6K and S6, and this effect was more drastic in HeLa/ORP5 cells ( Figure 5A, lane 4, B, C). Treatment of HeLa/ORP5 cells with two different ORP5 siRNAs showed similar results in terms of S6 phosphorylation ( Figure 5D, E). The downregulation of S6 phosphorylation by ORP5 knockdown appears to be mTORC1-dependent, since in our system a complete inhibition of S6K activation by rapamycin treatment also shut down S6 phosphorylation ( Figure S3). Nascent protein synthesis assay in HeLa/ORP5 cells treated with control or ORP5 siRNAs indicated that ORP5 knockdown significantly decreased protein synthesis ( Figure 5F, G), a phenotype that reflected the downregulation of S6K and S6 phosphorylation downstream of mTORC1. These data indicate that, at least when overexpressed, ORP5 is positively associated with mTORC1 activation.

Knockdown of endogenous ORP5 impairs S6K activation and cell proliferation
Next, we investigated whether endogenous ORP5 regulates mTORC1 signaling. The expression of ORP5 has been linked to the induction of pancreatic cancer cell invasion (20). This prompted us to examine the effect of ORP5 depletion on mTORC1 signaling and cell proliferation in PANC-1 cells, a pancreas ductal adenocarcinoma cell line. The two specific ORP5 siRNAs efficiently reduced endogenous ORP5 protein levels in PANC-1 cells ( Figure 5H). Anti-phosphorylated S6K immunoblotting analysis clearly revealed that S6K phosphorylation was downregulated upon the depletion of ORP5 ( Figure 5H). Accordingly, ORP5 depletion severely impaired normal cell growth and cell number expansion ( Figure 5I). This observation was also in accordance with the results from cell proliferation assay, which indicated a significant impairment of cell proliferation when ORP5 was depleted in PANC-1 cells ( Figure 5J). We observed similar results in another pancreatic cancer cell line, Capan-1 cells. In Capan-1 cells, there was a clear decrease of mTORC1 activation and significant reduction of cell growth when ORP5 was depleted ( Figure S4). Importantly, the growth defect in ORP5depleted Capan-1 cells was rescued when TSC1, an inhibitory factor of mTORC1 activation (23), was silenced ( Figure S4), highlighting the link between ORP5 and mTORC1.

ORP5 depletion impairs the localization of mTOR to lysosomes
Targeting of mTORC1 to the lysosomal surface is essential for its activation (24). We hypothesized that depletion of ORP5 may interfere with the translocation of mTORC1 to lysosomes, thereby impairing mTORC1 signaling. To test this hypothesis, we examined the co-localization of endogenous mTOR with LAMP-1, a lysosomal membrane marker, by immunofluorescence analysis in HeLa/ORP5 cells. In control siRNA treated cells and in agreement with the previous study (23), the majority of endogenous mTOR localized to LAMP-1-positive lysosomal ORP5 regulates mTORC1 structures ( Figure  6A, top panel). Quantification of mTOR/LAMP-1 colocalization revealed a significant decrease of the association of mTOR with lysosomes ( Figure 6A, 6B) in ORP5 siRNAs-treated cells. Similar results were obtained from the immunofluorescence experiments performed in PANC-1 cells. Knocking down of endogenous ORP5 in PANC-1 cells also significantly decreased mTOR/LAMP-1 co-localization ( Figure 6C, 6D), without affecting overall mTOR protein levels in the cells examined ( Figure 6E).

DISCUSSION
OSBP/ORPs are unified by their characteristic ORDs that render them capable of sensing, binding, and transporting lipids between intracellular membranes. Previous studies have implicated some members of the OSBP/ORPs family in the regulation of cell signaling pathways, such as OSBP in ERK1/2 activation (25) and ORP4L in Ca 2+ signaling (26). Here we uncover another connection between OSBP/ORPs and cellular signaling events. We show that overexpression of ORP5 promotes cell proliferation and motility, and that OPR5 expression is positively correlated with mTORC1 signaling. In this context, we present evidence that ORP5 interacts with mTOR and contributes to the targeting of mTORC1 to lysosomal surface for activation.
As a lipid transfer protein, ORP5 has been shown to play a role in intracellular lipid transport at membrane contact sites associated with the ER (12,(15)(16)(17).
Membrane lipid homeostasis regulated by ORP5 and other lipid transfer proteins has an impact on mTORC1, which itself is a key regulator of lipid metabolism, as well as protein synthesis and cell proliferation (4). Notably, ORP5 appears to be a phosphatidylserine transporter (14) and has been shown to deliver phosphatidylserine from the ER to the plasma membrane and mitochondria (15)(16)(17). Phosphatidylserine is critical for the organized distribution of cholesterol in the cytosolic leaflet of the plasma membrane and possibly other organelles (27).
Hence, phosphatidylserine transport mediated by ORP5 may be coupled with cholesterol movements between intracellular membranes. Since proper intracellular cholesterol trafficking is essential for mTOR signaling (28), it is possible that phosphatidylserine transport by ORP5 is favored for mTORC1 activation in certain cancer cells, contributing to increased cell proliferation. It is worth noting that membrane phosphatidylserine binds to specific residues in the PH and regulatory domain of Akt, an upstream effector of mTORC1 (29). The phosphatidylserine-Akt interaction is required for Akt activation and downstream signaling (29). This may represent another possible explanation of why ORP5 expression is positively correlated with mTORC1 signaling. In agreement with this possibility, we found that in HepG2 cells, ORP5 depletion impaired insulin-induced Akt activation and mTORC1 signaling (Du and Yang, unpublished data). Therefore, the ability of ORP5 to transport phosphatidylserine seems to be critical for the regulation of mTORC1. In line with this, mutation of a key residue in ORP5-ORD (L389D) responsible for phosphatidylserine binding and transfer failed to promote cell proliferation, migration, and invasion. Similar effects obtained from ORP5-ORD mutants incapable of transporting phosphoinositides or other lipids could be secondary to the impairment of phosphatidylserine transport. However, we could not rule out the possibility that these mutants have a direct impact on the homeostasis of membrane phosphoinositides, resulting in the downregulation of the Akt/mTORC1 pathway (30). It should also be noted that while ORP5 overexpression in HeLa cells was needed for assessing the gain-offunction effects of ORP5 and its mutants, it ORP5 regulates mTORC1 remains possible that the use of ORP5 overexpression in HeLa cells yields nonphysiological results.
An important finding of the current study is that proteomics analysis identified mTOR as one of possible ORP5 interacting proteins.
The results from the coimmunoprecipitation and proximity ligation assays confirmed that ORP5 is closely associated with mTOR, suggesting that ORP5 might be able to directly regulate mTORC1. Interestingly, the association between ORP5 and mTOR depends on ORP5-ORD, and this dependency does not seem to rely on the lipid transfer activity of ORP5. Future studies are needed to pinpoint critical regions or residues responsible for ORP5-mTOR interaction. It is also possible that deletion of ORD influences the localization of ORP5 and hence the interaction between ORP5 and mTOR. For example, loss of ORD has been shown to dissociate ORP5 from the mitochondrial surface (17). Given the observation that mTORC1 can be found on mitochondria and regulates mitochondrial functions (31), the dissociation of ORP5 from mitochondria due to ORD deletion could justify the loss of the interaction between ORP5 and mTOR. Future work is necessary to investigate whether a possible ORP5/mTORC1 cooperation at the mitochondrial membrane controls the proper function of mitochondria. Finally, it should be noted that the interaction between ORP5 and mTOR was detected only when ORP5 was overexpressed. Thus, while there is a strong functional link between ORP5 and mTORC1 pathway, the physical interaction between endogenous proteins remains to be demonstrated and warrants further analyses in the future.
Our data indicate that overexpression of ORP5 up-regulates, and depletion of ORP5 down-regulates mTORC1 activation. How are these effects linked to ORP5-mTOR interaction? In response to amino acids or other stimuli, mTORC1 is translocated to the lysosomal surface for activation (24). We found significant impairment of mTOR localization to lysosomes in cells depleted with ORP5. At least in part, this could account for the downregulation of mTORC1. How the impairment of mTORC1 translocation to lysosomes is caused by ORP5 depletion remains to be elucidated. ORP5 interacts with lysosomal membrane protein Niemann Pick C1, and ORP5 without the ER anchoring transmembrane domain has been shown to be enriched in late endosome and lysosomes (12), suggesting an association between ORP5 and lysosomes. In this regard, ORP5 may recruit mTORC1 to lysosomes through the interaction with mTOR. This would explain why mTORC1 translocation to the lysosomal surface is impaired when ORP5 is deficient. However, it is also possible that the regulation of mTORC1 activation by ORP5 could be independent of the interaction between ORP5 and mTOR. As mentioned earlier, lipid transfer or targeting activity by ORP5 may be the key to regulate mTORC1 activation. For instance, ORP5 may regulate the homeostasis of phosphatidylinositol 3,5-bisphosphate (PI(3,5)P 2 ), which is concentrated in endolysosomal membrane (32).
Of note, PI(3,5)P 2 has been shown to bind to the mTORC1component raptor and could contribute to lysosomal targeting of mTORC1 (33). Future studies involving co-localization between PI(3,5)P 2 and mTORC1, as well as in vitro lipid transfer assay and lipidomics analysis, are required to test this hypothesis.
In summary, we show that the lipid transfer function of ORP5 is critical for ORP5 to promote cell proliferation. Importantly, we identify a novel link between ORP5 and mTORC1 signaling which provides another explanation for the involvement of ORP5 in cancer cell proliferation. Our work suggests that ORP5 regulates mTORC1 ORP5 may be a therapeutic target, alongside mTORC1 inhibition, to treat certain cancers.

Cell Culture and Transfection
HeLa, PANC-1 and Capan-1 cells were obtained from ATCC. Monolayers of cells were maintained in DMEM supplemented with 10% FBS (for Capan-1 cells 20% FBS was used), 100 units/ml penicillin, and 100 µg/ml streptomycin sulphate in 5% CO 2 at 37°C. DNA transfection was performed using Lipofectamine™ LTX and Plus Reagent (Life Technologies) according to manufacturer's instruction. siRNA transfection was carried out in cells grown in full serum medium according to standard methods using Lipofectamine™ RNAiMAX transfection reagent (Life Technologies).

Generation of HeLa/ORP5 stable cell line
Retrovirus particles were produced in the packaging cell line 293T Phoenix-AMPHO expressing pQCXIN-ORP5. HeLa cells were transduced with the retrovirus encoding ORP5 or mock control for 48 h. Transduced cells were selected with growth medium containing 500 µg/ml G418 (Sigma Aldich) for 10 d. During the selection, cells were passaged every 2 d until a stable line of cells expressing ORP5 was established.

Antibodies
Antibodies used were rabbit polyclonal to mTOR

Immunoblot Analysis
Samples were mixed with 2 × laemmli buffer, boiled for 5 min at 95°C or incubated for 10 min at 70°C, and then subjected to 7.5% or 10% SDS-PAGE. After electrophoresis, the proteins were transferred to Hybond-C nitrocellulose filters (GE Healthcare). Incubations with primary antibodies were performed at 4°C overnight. Secondary antibodies were peroxidase-conjugated AffiniPure donkey anti-rabbit or donkey antimouse IgG (H+L; Jackson ImmunoResearch Laboratories) used at a 1:5000 dilution. The bound antibodies were detected by ECL Western blotting detection reagent (GE Healthcare or Merck Millipore) and visualised with Molecular Imager® ChemiDocTM XRS+ (Bio-Rad Laboratories).

Immunoprecipitation Assay
Transfected cells grown in 100-mm dishes were harvested, washed with cold PBS, resuspended in 1 ml of cell lysis buffer (50 mM Tris-HCl, pH 7.8, 100 mM NaCl, 1% Triton X-100) containing Protease Inhibitor Cocktail (100X, Sigma-Aldrich) and Phosphatase Inhibitor Cocktail (100X, CST). Cell lysates were passed through a 22-gauge needle twenty times, incubated on ice for 30 min, and clarified by centrifugation at 18,000 × g for 15 min at 4°C. Immunoprecipitation of the lysates with a polycolonal antibody against NPC1 was performed using the Dynabeads® Protein G (Life Technologies) according to manufacturer's instructions. The immunoprecipitated pellets were resuspended in 60 µl 2 × laemmli buffer (Sigma-Aldrich) and then incubated for 10min at 70ºC. The resultant samples were subjected to SDS/PAGE and immunoblotting.

MTS Assay
Cells were seeded in 96-well plates at 2.5×10 3 cells/well and transfected with siRNA (20 nM) for 72 h. For non-transfected or plasmid cDNA transfected cells (24 h), cells were seeded at 5×10 3 cells/well in 96-well plates. Six wells were seeded for each transfection. Cell proliferation was measured with a MTS assay kit (CellTiter 96® AQueous One Solution Cell Proliferation Assay, Promega). For each reading, 20 µl of MTS solution was added, plates were incubated for 1-2 h, and absorbance read on a Spectra MAX340 Microplate Reader at 490 nm.

Crystal Violet staining
Cells were fixed with 4% paraformaldehyde for 5 min at room temperature. The fixed cells were stained with 0.05% (w/v) crystal violet diluted in distilled water for 30 min at room temperature. Cells

ORP5 regulates mTORC1
were washed twice with water and drained to dry. The dishes were then scanned and cell numbers were counted using Fiji software (35)

Cell Migration Assay
Cells were seeded in 24-well plates containing Culture-Insert 24 (Ibidi ® cells in focus, Planegg, Germany). For overexpression or knockdown experiments, cells were transfected with plasmid cDNAs for 24 h or siRNAs for 48 h, respectively, prior to seeding. After Culture-Insert was removed, images were taken at 0 and 24 h using an Olympus CKX41 microscope (Olympus, Japan) fitted with an Infinity 1-2CB camera (Lumenera Corporation, Ontario, Cannda) connected to a computer. Percentage of the cell migration was quantitated using the Fiji Software (35). Data from 24 h were normalised to those from 0 h.

Proximity ligation assay
HeLa/ORP5 cells were seeded in 6well plate containing coverslips at 4×10 5 cells/well and grown for 48 h. For knockdown experiment, cells were seeded at 2×10 5 cells/well and treated with siRNAs for 72 h. Cells were fixed with 4% paraformaldehyde (Electron Microscopy Sciences) for 15 min and permeabilized with 0.2% Triton X-100 (Sigma-Aldrich) for 10 min at room temperature. The slides were then blocked with 3% BSA in PBS for 1 h and incubated with appropriate combinations of goat anti-ORP5 (Abcam Cat.# Ab59016, 1:50) and rabbit anti-mTOR (CST #2983, 1:400) antibodies diluted in blocking solution for 1 h at room temperature. After washing, the coverslips were incubated with Duolink PLA Rabbit MINUS and PLA Mouse PLUS proximity probes (Olink Bioscience, Uppsala, Sweden) and proximity ligation was performed using the Duolink detection reagent kit (Olink Bioscience) according to the manufacturer's protocol. Fluorescence images were acquired using an Olympus Fluoview FV1200 confocal microscope with a 60x/1.35 UPlanSApo objective. Images were prepared and analyzed with Fiji software (35). All data were analyzed using GraphPad PRISM software.

Cell invasion assay
Invasion assay was performed using the Corning ® BioCoat™ Matrigel ® Invasion Chambers (Corning Life Sciences). The chambers of the 24-well cell culture inserts were rehydrated with serum-free medium at 37°C, 5% CO 2 atmosphere for 2 h, then transfer to a 24-well plate containing normal growth medium. To the upper chamber, 0.5 ml of cell suspension (2.5×10 4 cells) were added. The invasion chambers were incubated for 24 h at 37°C in the cell culture incubator. Non-invasive cells on the upper insert membranes were removed using cotton tipped swabs by gentle scrubbing. Invasive cells on the lower insert membranes were fixed with 100% methanol and stained with 1% Toluidine Blue (Sigma-Aldrich). The membranes from the insert housing were removed by cutting and mounted on microscope slides. The invading cells were counted under the microscope at 100X magnification. Data is expressed relative to cell numbers migrating through control membranes.

Protein synthesis assay
HeLa/ORP5 cells were seeded in 6well plate containing coverslips at 2×10 5 cells/well treated control siRNA or ORP5 siRNAs for 72 h. The detection of nascent protein synthesis was carried out using the Click-iT ® HPG Alexa Fluor ® 488 Protein Synthesis Assays Kit (Life Technologies, Cat.#C10428) according to manufacturer's instructions. Fluorescence images were acquired using an Olympus FluoView FV1200 confocal microscope equipped with DAPI filter and FITC filter for Alexa Fluor ® 488. Nascent protein synthesis was assessed by determining the signal intensity in the fluorescent channel in the ring around the nucleus as defined by

ORP5 regulates mTORC1
NuclearMask™ Blue Stain. Images were prepared and analyzed with Fiji software (35). All data were analyzed using GraphPad PRISM software.

Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis
To identify potential ORP5 interacting proteins, HeLa/ORP5 cell lysates were immunoprecipitated with control goat IgG or goat anti-ORP5 polyclonal antibody. The immunoprecipitates were subject to gel electrophoresis and the gel was stained with Coomassie blue. Stained protein bands were excised and in-gel trypsin digestion was performed prior to LC-MS/MS analysis, which was carried out as previously described (36). All MS/MS spectra data from human proteins were searched using Mascot search engine and the results from control and ORP5 immuoprecipitates were combined and filtered using Scaffold software (Version: Scaffold_4.0.7). Positive hits from ORP5 immunopreciptates were identified with the following criteria: peptide thresholds ≥ 90.0% ; protein thresholds ≥ 99.0% and ≥ 2 peptides.

Statistical Analyses
Statistical analysis between groups was performed using GraphPad PRISM software (Version 6.03) with Student's unpaired t tests or One-way ANOVA. Data are expressed as mean + SD unless otherwise stated. Significant differences are indicated in the figures. (C) Cell migration assay in HeLa/Mock and HeLa/ORP5 cells. Relative cell migration was analyzed using ImageJ (mean + SD; **** p < 0.0001; n = 6).
All data are representative from three to four independent experiments with similar results. mutants. Relative cell migration was analyzed using ImageJ (mean + SD; **** p < 0.0001; n = 6).
All data are representative from three to five independent experiments with similar results. (C) Proximity ligation assay (PLA) using goat ORP5 antibody together with or without rabbit mTOR antibody in HeLa/ORP5 cells.
(E) Proximity ligation assay in HeLa/ORP5 cells treated with control siRNA or ORP5 siRNA.
(F) Average PLA intensity in (E) was quantitated using ImageJ (mean +SD; *** p < 0.001; n > 30 cells).  (D) HeLa/ORP5 cells were treated with control and two different ORP5 siRNAs. Cells were starved in serum free medium overnight prior to harvest for immunoblotting analysis.
(H) PANC-1 cells were treated with control or ORP5 siRNAs for 72 h followed by harvest for immunoblotting analysis.
(I) PANC-1 cells were treated with control or ORP5 siRNAs for 72 h. Cells were imaged under a widefield microscope.
(C) PANC-1 cells were treated with control siRNAs or siORP5 for 72 h followed by immunofluorescence of LAMP-1 and mTOR.
(E) PANC-1 cells were treated with control or ORP5 siRNAs for 72 h followed by harvest for immunoblotting analysis.     -100