Reticulocalbin 1 is required for proliferation and migration of non‐small cell lung cancer cells regulated by osteoblast‐conditioned medium

Abstract Reticulocalbin1 (RCN1) is implicated in tumorigenesis and tumour progression. However, whether RCN1‐mediated bone metastasis of non‐small cell lung cancer (NSCLC) cells was elusive. Here, we assessed the effect of osteoblast‐conditioned medium (CM) on proliferation and migration of NSCLC cell line, NCI‐H1299 and NCI‐H460 cells, and identified the soluble mediators in CMs from osteoblasts and NSCLC cells using MTT, Clonogenicity, Transwell, wound healing, RT‐PCR, and Western blotting assays, and LC‐MS/MS analysis, respectively. Furthermore, the role of RCN1 was investigated in NSCLC cells cultured with or without osteoblast‐CM. Tumour growth and bone resorption were measured in a nude mouse model bearing NCI‐H1299 cells transduced with shRNA/RCN1 vector using in vivo imaging technique and micro‐CT. The results showed that RCN1 with a higher abundance in osteoblast‐CM, which was present in extracellular vesicles (EVs), enhanced RCN1 expression in NSCLC cells. Osteoblast‐CM partially offset the inhibitory effect of RCN1 depletion on proliferation and migration of NSCLC cells. RCN1 depletion‐induced endoplasmic reticulum (ER) stress caused by increasing GRP78, CHOP, IRE1α, p‐IRE1α, p‐PERK and p‐JNK, which was positively regulated by self‐induced autophagy, contributed to suppression of proliferation and migration in NCI‐H1299 cells. Therefore, osteoblasts produced RCN1 to transfer into NSCLC cells partially through EVs, facilitating proliferation and migration of NSCLC cells via blocking ER stress. RCN1 could be required for proliferation and migration of NSCLC cells regulated by osteoblast‐CM.


| INTRODUC TI ON
Osteoblast is one of main cell types in bone microenvironment and has been reported to play key roles in cancer dissemination to bone and subsequent metastases. 1,2 Accumulating evidence shows that osteoblasts could produce factors into bone microenvironment, thereby affecting proliferation and metastases of cancer cells. 3 Osteoblast-derived sphingosine 1-phosphate induces proliferation and confers resistance to therapeutics to bone metastasis-derived prostate cancer cells. 4 HIF signalling in osteoblast-lineage cells promotes systemic breast cancer growth and metastasis in mice. 5 Lung tumour-associated osteoblast-derived bone morphogenetic protein-2 increased epithelial-to-mesenchymal transition of cancer by Runx2/Snail signalling pathway. 6 Hence, advances in understanding the function and regulation mechanism of osteoblast-derived factors in tumour growth and bone metastasis will aid in controlling and ultimately preventing cancer cell metastasis to bone. Reticulocalbin1 (RCN1), a member of the CREC family, is an endoplasmic reticulum (ER)-resident Ca2+-binding protein. 7 In addition to localizing in the ER lumen, RCN1 is also expressed on the cell surface, such as bone endothelia cell and prostate cancer cell, and its expression is heterogeneous in each constituent cell of some organs. 8 RCN1 has been drawn intensive attention due to its role in tumorigenesis and tumour progression. [8][9][10][11][12][13] For example, RCN1 is expressed in the highly invasive breast cancer cell lines, not in poorly invasive ones. 9 RCN1 appears as a potential renal cell carcinoma tumour marker in clinical proteomic analysis. 10 RCN1 plays an important role in the development of doxorubicin-resistant uterine cancer. 11 Specially, in non-small cell lung cancer (NSCLC), overexpression of RCN1 correlates with poor prognosis and progression 12 and the efficacy of platinum-based adjuvant chemotherapy relates with co-expression of RCN1 and GSTπ. 13 The involvement of RCN1 is thus required for lung cancer progression, although its regulatory mechanism is not yet clear.
Recent evidence displays that mRNA of RCN1 has been detected in extracellular vesicles (EVs) from porcine adipose tissuederived mesenchymal stem cells 14 and osteoblasts. 15 Compared to non-sclerotic osteoblast secretome in osteoarthritis subchondral bone, RCN1 is one of 13 proteins significantly less secreted by sclerotic osteoblasts. 15 Hence, there is a possibility that RCN1 is one of osteoblast-derived factors to carry out its function in bone microenvironment. However, whether and how osteoblasts come in touch with proliferation and migration of NSCLC cells through producing RCN1 into bone microenvironment is not elucidated.
Here, we assessed the effect of osteoblast-conditioned medium

| Osteoblast-CM preparation
Human and rat primary osteoblasts were seeded at a density of

| Cell proliferation
NSCLC cells were seeded in 96-well plates and cultured in different types of medium for 48 h. Cell viability was measured by 3-(4, 5-Di methylthiazol-2-y)-2,5-diphenyl-tetrazolium bromide (MTT) assay at the end of the assay period. 16,17 Clonogenicity was also used to assess cell proliferation. Cells cultured in 6-well plates for 48 h were fixed with 100% methanol and stained with 0.1% crystal violet, followed with observation of colonies under an Olympus BX41 microscope equipped with a digital camera (Olympus, Tokyo, Japan) at 4x magnification. 16

| Cell migration
Cell migration was assessed by a wound-healing assay. Briefly, NSCLS cells were allowed to reach full confluence in 6-well plates. A uniform scratch was made down the centre of the well using a 10 μl micropipette tip. After washed once with PBS, cells were cultured in different types of medium according to experimental designs. Images at 4× magnification were acquired at 48 h after scratching. 4 measurements were taken in each well, and the wound area was calculated using Image-Pro Pus 6.0 system. 17 Transwell migration assay was also used to assess cell migration. According to the previous studied, 16,17 cells in serum-free RPMI-1640 were placed into the top chambers of Transwell inserts set with 8 μm pore filters and different types of media were added to the bottom chamber, respectively. Cells on the top of the membrane were removed with a cotton swab. The chamber of Transwell was fixed with 4% paraformaldehyde and stained with 0.1% Giemsa stain. The migrated cells were observed under an Olympus BX41 microscope equipped with a digital camera (Olympus, Tokyo, Japan) at 4 × magnification.

| LC-MS/MS analysis
LC-MS/MS analysis was carried out as previously described. 18,19 Briefly, combined with a nanoscale EASY-nLC 1200UHPLC system

| MS data analysis
Proteome Discoverer (PD, version 2.2) was carried out to process raw data, and the reviewed SwissProt human proteome database (20259 entries) was used to data search. All searches were carried out with precursor mass tolerance of 20 ppm, fragment mass tolerance of 0.02 Da and three trypsin missed cleavages allowed. Only peptides with at least six amino acids in length were considered. The peptide and protein identifications were filtered by PD to control the false discovery rate (FDR) <1%. At least one unique peptide was required for protein identification.

| Extraction of osteoblast-derived EVs
As described previously, 22 reaching 80% confluency, osteoblasts were rinsed with PBS and cultured un serum-free medium for 48 h.
Osteoblast-CM was collected and centrifuged at 300 × g for 15 min, 2000 × g for 15 min, and 10,000 × g for 30 min at 4°C to remove dead cells and cell debris. The supernatants were ultra-centrifuged at 120,000 x g for 3 h to separate EVs from CM, followed with the identification of EVs performed by western blotting assay.

| Western blotting assay
Protein extracts were subjected to SDS-PAGE (8-15%) and transferred to a PVDF membrane (GE Healthcare, Hertfordshire, UK) as described previously. 23,24 The membrane was incubated with various primary antibodies as required at 4°C overnight, followed by the addition of the corresponding secondary antibodies at room temperature for 1 to 2 h (Table 1). An enhanced chemiluminescence (ECL) detection kit was used to detect antibody reactivity (Pierce, Rockford, IL, USA).

| Plasmid construction and transfection
Short hairpin RNA (shRNA) targeting RCN1 (shRCN1) was pur- to synthesize cDNA. Real-time PCR was then performed using a Light Cycler (Roche) with a SYBR Premix Ex Taq II Kit (Takara, Dalian, China). Results were analysed as described in previous studies. 25,26 The primers used for quantitative PCR are described in Table 2.

| Micro-computed tomography
Mouse legs were fixed with 4% paraformaldehyde for 48 h prior to micro-CT analysis as described previously. [29][30][31][32] It was examined that the qualitative and semi-quantitative analysis of bone destruction in the two groups was measured using high-resolution micro-

| Statistical analysis
Differences between the groups were examined for statistical significance using t test (between two groups) and one-way ANOVA (among three groups or more) with GraphPad Prism 6 software (GraphPad Software). A value of p < 0.05 was considered significant.

| Osteoblast-CM facilitated proliferation and migration in NSCLC cells
To determine the effect of osteoblast-CM on proliferation and migration of NSCLC cells, NCI-H1299 and NCI-H460 cells were cultured with different types of media, respectively. As shown in Figure 1A,

| RCN1 with a higher abundance in osteoblast-CM enhanced RCN1 expression in NCI-H1299 and NCI-H460 cells
To confirm the crucial elements in osteoblast-CM regulating cell proliferation and migration of NSCLC cells, the soluble mediators in human or rat osteoblast-CM and NCI-H1299 cell-CM were identified and analysed using LC-MS/MS. As described in Figure 3A, between human and rat osteoblast-CM (H-OB and R-OB), from the top 40% high-abundance proteins, 122 types of proteins were shared.
Furthermore, between human osteoblast-CM and NCI-H1299 cell-CM (H-OB and HCI-H1299), there were 84 types of shared proteins in the top 10% differentially expressed proteins that were F I G U R E 2 Conditioned medium (CM) from human or rat primary osteoblasts promoted migration in human lung cancer NCI-H1299 and NCI-H460 cells. A&B, NCI-H1299 and NCI-H460 cells were cultured in αMEM, 1640,1640+αMEM (1:1), 1640+CM (1:1) from rat osteoblasts for 48 h, respectively. As described in the Material and methods section, the cell migration was assessed using a wound healing assay (A). Relative mRNA level of MMP2 was measured via RT-PCR as described in the Material and methods section (B). C, NCI-H1299 cells were cultured in F12, 1640,1640+F12(1:1), 1640+CM (1:1) from human osteoblasts for 48 h, the cell migration was assessed using a wound healing assay as described in the Material and methods section. D, NCI-H1299 and NCI-H460 cells were culture in F12, 1640,1640+F12(1:1), 1640+CM (1:1) from human osteoblasts for 48 h, relative mRNA level of MMP2 was measured via RT-PCR as described in the Material and methods section. The data are representative of three independent experiments (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001)

| Osteoblast-CM partially offset the inhibitory effect of RCN1 depletion on proliferation and migration in NCI-H1299 cells
As shown in Figure

| RCN1 depletion-induced ER stress contributed to suppression of proliferation and migration in NCI-H1299 cells
Tumour microenvironment stresses could produce ER stress, which has been highlighted to the involvement in the development and progression of lung cancer. 33,34 To confirm whether the regulation of cell proliferation and migration by RCN1 was associated with ER stress in NSCLC cells, we investigated the protein expression levels of some ER stress-related biomarkers. The results in Figure 5A showed that RCN1 depletion by shRCN1-2 vector enhanced GRP78, p-PERK(Thr980), CHOP, IRE1α, p-IRE1α(Ser724) and p-JNK(Thr183/ Thr183/Thr221) levels without the alteration of total PERK and JNK levels, compared with shcon group. Nevertheless, the effect of RCN1 depletion by shRCN1-2 vector on ER stress was neutralized by human osteoblast-CM, even offset ( Figure 5A). Given that ER stress can induce apoptosis and autophagy, 33,35 the expression levels of apoptosis-and autophagy-related biomarkers were simultaneously detected in NCI-H1299 cells transduced with shRCN1-2 vector with or without ER stress inhibitor 4-PBA. The results shown in Figure 5B exhibited that PCNA and Bcl-2 levels were reduced by shRCN1-2 vector, accompanied with an increase in CHOP level. The addition of ER stress inhibitor 4-PBA led to decreased PCNA and Bcl-2 levels compared with shRCN1-2 group ( Figure 5B). On the contrary, the depletion of RCN1 by shRCN1-2 vector result in increased Beclin1 and ATG5 levels, accompanied with decreased p-mTOR and p-ULK1 levels in NCI-H1299 cells (Upper panel in Figure 5C). However, compared with shRCN1-2 group, the addition of ER stress inhibitor 4-PBA only led to a slight increase in the ATG5 and Beclin1 levels (Lower panel, F I G U R E 3 RCN1 with a higher abundance in osteoblast-CM enhanced RCN1 expression in NSCLC cells. A-D, As described in the Material and methods section, the soluble mediators in human or rat osteoblast-CM and NCI-H1299 cell-CM were identified and analysed using LC-MS/MS, prior to analysis by Metascape online service and calculation and visualization by Draw Venn by Diagrams serve and InteractiVenn tools. E, Using a gradient centrifugation as described in the Material and methods section, osteoblast-derived exosomes were extracted and the CD9, CD63, CD81 and RCN1 protein levels were detected via Western blotting as described in the Material and methods section. F, NCI-H1299 and NCI-H460 cells were cultured in αMEM, 1640,1640+αMEM(1:1), 1640+CM(1:1) from rat osteoblasts for 48 h, the RCN1 and β-actin protein levels and mRNA level of RCN1 were measured via Western blotting and RT-PCR, respectively, as described in the Material and methods section. G, NCI-H1299 cells were culture in F12, 1640,1640+F12(1:1), 1640+CM (1:1) from human osteoblasts for 48 h, the RCN1 and β-actin protein levels and mRNA level of RCN1 were measured via Western blotting and RT-PCR, respectively, as described in the Material and methods section. The data are representative of three independent experiments (**p<0.01, ***p<0.001, ****p<0.0001) Figure 5C)

| DISCUSS ION
Previous studies have demonstrated that osteoblast-CM was a potent chemoattractant for bone metastasis of cancer cells. 4,6,36,37 For example, factors derived from osteoblasts induce an expression signature that identifies prostate cancer metastasis. 36 Bone metastatic breast cancer cells utilize osteoblast-derived cytokines to facilitate breast cancer cell colonization and survival upon arrival in the bone microenvironment. 37 Herein, we found that RCN1, which correlates with poor prognosis and progression of NSCLC, was identified to express with a higher abundance in human or rat primary osteoblast-CM as well as in EVs. Subsequently, os- Osteoblast-CM offset partly the inhibitory effect of RCN1 depletion on cell proliferation and migration in NCI-H1299 cells. A&B, NCI-H1299 cells were stably transduced with shRNA/RCN1-1/2/3 or transiently transfected with GFP-RCN1 vectors. The cell viabilities were detected via MTT assay and the PCNA, RCN1, Bcl-2 and β-actin protein levels were measured via Western blotting as described in the Material and methods section (A). Relative mRNA level of MMP2 was measured via RT-PCR as described in the Material and methods section (B).C&D, NCI-H1299 cells were stably transduced with shRNA/RCN1-2 vector and cultured in rat osteoblast-CM. The cell viabilities were detected via MTT assay(C) and the MMP2, Bcl-2, and β-actin protein levels (D) were measured via Western blotting as described in the Material and methods section. E, NCI-H1299 cells were cultured in CM from rat osteoblasts transduced with shRNA/ RCN1-2 vector. The RCN1, Bcl-2, MMP2 and β-actin protein levels were measured via Western blotting. F&G, NCI-H1299 cells transduced with shRNA/RCN1-2 vector were injected in the proximal metaphysic of the tibia in nude mice for 8 weeks. The radiance (p/sec/cm2/sr) of tumour mass was measured using the IVIS Lumin II in vivo imaging (F). The BV/TV in cortical bone in the tibia was measured using high resolution micro-CT scanning(G). The data are representative of three independent experiments (*p<0.05, **p<0.01, ****p<0.0001) vector, it is speculated that there might exist a complicated mechanism to regulate apoptosis, which was not completely dependent on RCN1 depletion-induced ER stress or autophagy.
In conclusion, our findings determined that osteoblasts produced RCN1 to transfer into NSCLC cells partially through encapsulating in EVs. Furthermore, osteoblastic RCN1 contributed to proliferation and migration of NSCLC cells through suppressing ER stress that was associated with decreased GRP78, CHOP, IRE1α, p-IRE1α, p-PERK and p-JNK, which could be positively regulated by self-induced autophagy ( Figure 5E). Therefore, RCN1 could be required for proliferation and migration of NSCLC cells regulated by osteoblast-CM.

ACK N OWLED G EM ENTS
This study was supported by the National Natural Science Foundation of China (No. 81572189 and 81972091). We would like to thank MS Jiqian Xia for statistical data analytics support.

CO N FLI C T S O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data support the findings of this study are available in the supplementary material of this article.