Fibroblast activation protein protects bortezomib-induced apoptosis in multiple myeloma cells through β-catenin signaling pathway

Abstract

Multiple myeloma (MM) is a malignant plasma cells proliferative disease. The intricate cross-talk of myeloma cells with bone marrow microenvironment plays an important role in facilitating growth and survival of myeloma cells. Bone marrow mesenchymal stem cells (BMMSCs) are important cells in MM microenvironment. In solid tumors, BMMSCs can be educated by tumor cells to become cancer-associated fibroblasts (CAFs) with high expression of fibroblast activation protein (FAP). FAP was reported to be involved in drug resistance, tumorigenesis, neoplastic progression, angiogenesis, invasion, and metastasis of tumor cells. However, the expression and the role of FAP in MM bone marrow microenvironment are still less known. The present study is aimed to investigate the expression of FAP, the role of FAP, and its relevant signaling pathway in regulating apoptosis induced by bortezomib in MM cells. In this study, our data illustrated that the expression levels of FAP were not different between the cultured BMMSCs isolated from MM patients and normal donors. The expression levels of FAP can be increased by tumor cells conditioned medium (TCCM) stimulation or coculture with RPMI8226 cells. FAP has important role in BMMSCs mediated protecting MM cell lines from apoptosis induced by bortezomib. Further study showed that this process may likely through β-catenin signaling pathway in vitro. The activation of β-catenin in MM cell lines was dependent on direct contact with BMMSCs other than separated by transwell or additional condition medium from BMMSCs and cytokines.

Introduction

Multiple myeloma (MM), characterized by increased blood calcium level, renal insufficiency, anemia, and bone lesions (CRAB), is a malignant plasma cell proliferative disorder with about 22 000 new cases annually and about 20% of deaths from all hematological malignancies in the US.¹ With the advent of new drugs, such as bortezomib, lenalidomide, pomalidomide, perifosine, etc., 5-y survival rate of MM increased to more than 40%.¹ However, MM is still regarded as an incurable disease in spite of new treatment regimens.² Interaction of MM cells and cells in the bone marrow (such as mesenchymal stem cells, osteoblasts, osteoclasts, endothelial cells, immunocytes, and adipocytes) constitutes a complex tumor microenvironment that support proliferation, growth, chemoresistance, invasion, and metastasis of MM cells.³ Bone marrow mesenchymal stem cells (BMMSCs) are important cells in MM microenvironment. It is now becoming apparent that BMMSCs mediate chemoresistance, angiogenesis, and progression of MM cells though either direct contact or secreted cytokines, chemokines and exosomes.^4-6 In solid tumors, previous studies have realized that tumor microenvironment has become a coconspirator in process of carcinogenesis.⁷ Tumor-stroma interaction modifies malignant cells and “nonmalignant” cells leading them to reciprocal activation and coevolution and turned tumor microenvironment into a fertile soil to support the growth and chemoresistance of tumor cells.^8,9 Cancer-associated fibroblasts (CAFs) have been considered as important component in tumor microenvironment and play vital roles in tumor behaviors.^8,10,11 However, the precise origin of CAFs is still less known.¹¹ As for the current knowledge, BMMSCs was recognized as one of its origins.^12-14 Studies have shown that tumor cells recruit BMMSCs through secreted cytokines or chemokines and educate BMMSCs to become CAFs to support the growth, proliferation, angiogenesis, invasion, and metastasis of malignant cells.^12-14 Current knowledge has revealed that CAFs express surface markers such as fibroblast activation protein (FAP), fibroblast surface protein (FSP), α-smooth muscle actin (α-SMA), platelet derived growth factor receptor-α/β (PDGFRα/β), tenascin-C (TN-C), vimentin, and fibronectin.^11,14,15 FAP, a member of serine protease family possessing of gelatinase activity and type I collagenase activity, is a vital type II trans membrane protein which is expressed in more than 90% epithelial tumor stromal cells but not normal cells.¹⁶ Researches have shown that FAP plays an important roles in tumor proliferation,¹⁷ angiogenesis,^18,19 immune escape,²⁰ invasion and metastasis,^21,22 and tissue remolding.^23,24 Considering its important roles in tumor progression and rare expression in normal tissue, FAP is becoming an important therapy target in tumors.²⁵ It is still less known whether or not BMMSCs from MM patients or normal donors had a different expression of FAP and whether or not FAP had important role in mediating MM cells behavior. In this study, we mainly focus on the expression of FAP on BMMSCs and the role of FAP in BMMSCs mediated protection on MM cell lines.

Results and Discussion

Identification of BMMSCs

BMMSCs were first observed by the German pathologist Cohnheimin 140 y ago.²⁶ Dramatic interest in MSCs has risen in last two decades. In terms of the current knowledge, there is now a general consensus that three criteria are applied to define MSCs.^27-30 First, cultured MSCs are fibroblast-like cells and adherence to plastic. Second, cultured MSCs express CD105 (SH2), CD73 (SH3/4), and CD90 (Thy-1), but are negative for the expression of hematopoietic markers, such as CD34, CD45, CD14 or CD11b, CD19 or CD79α, and HLA-II. Third, MSCs are capable of differentiating into osteoblasts, adipocytes, chondroblasts in vitro. MSCs in bone marrow are very rare with only 1 MSC every 10 000 nucleated cells.^26,30 In our experiments, cultured BMMSCs were adherent to the flasks and fibroblast-like in morphology (Fig. 1A). BMMSCs of p2 had a low expression of CD14 (1.31 ± 0.37%), CD34 (0.23 ± 0.05%), CD45 (1.20 ± 0.73%), and high expression of CD90 (94.65 ± 1.53%) and CD105 (97.82 ± 0.76%) in three independent experiments. FACS analysis of these CD molecular on BMMSCs is shown in Figure 1B. Taken together, these results met the minimal criteria for defining MSCs.²⁹

Figure 1. The phenotype of BMMSCs. (A) BMMSCs had a fibroblast-like morphology and adhered to the flask. (B) BMMSCs were positive for CD90 and CD105 but negative for CD14, CD34, and CD45.

Expression of FAP in BMMSCs

FAP was first discovered in the late 1980s with monoclonal antibody (mAb) F19.^31,32 In 1990, a dimeric 170-kDa membrane-bound gelatinase was identified in aggressive malignant human melanoma cell line, LOX and named “seprase” in 1994.^33,34 Subsequently, FAP and seprase were proved to be the same transmembrane protease with cloning and sequence analysis.^35,36 FAP was expressed in 90% tumor stroma of epithelia cancer,¹⁶ malignant melanoma cells, invasive ductal carcinoma cells of breast cancer, cancer cells of colorectum, stomach, uterine cervix, and glioma cells,^33,37-41 fibroblast foci, and fibrotic interstitium of idiopathic pulmonary fibrosis, fibroblast-like synoviocytes of rheumatoid arthritis, and stellate cells of cirrhosis,^42-44 BMMSCs, and cultured fibroblasts.^38,45 FAP had important role in tumor progression,¹⁷ invasion,²² metastasis,²¹ and immune escape.²⁰ HEK293 cells and human mammary adenocarcinoma cells MDA-MB-231 cotransfect with FAP cDNA could more easily to form subcutaneous tumors.^17,19 Researches have shown that BMMSCs can be educated by TCCM to become CAFs^12,46,47 and to promote tumor progression.¹³ Studies have revealed that BMMSCs from MM patients display distinct profile as compared with those from normal donors.^5,48-50 So, we presumed that MM microenvironment may also educate BMMSCs to become CAFs and have an elevated expression of FAP than normal BMMSCs.

First, we checked the expression of FAP on BMMSCs between MM patients and normal donors. As shown in Figure 2, there were no differences on the expression of FAP between MM patients and normal donors detected by qRT-PCR (P > 0.05), FCM (P > 0.05), and immunofluorescence. We speculated that two reasons maybe contributed to this result. First, cultured BMMSCs, like MSCs involving in healing wounds,^16,37,51 had already activated in the process of culture to express FAP. Second, BMMSCs from MM patients may lack of stimulation by MM cells and/or secreted factors because of long-term in vitro culture. So, next, we added TCCM from myeloma cells or cocultre with RPMI8226 cells to stimulate BMMSCs for 7 d and detected the expression of FAP by qRT-PCR. As shown in Figure 3, expression of FAP in BMMSCs had a moderate increase in TCCM stimulated group (n = 8) and cocultured group (n = 4) than non-stimulated group and non-cocultured group (P = 0.0256 and P = 0.035, respectively).

Figure 2. The expression of FAP on BMMSCs. (A) BMMSCs cell lines and (B) BMMSCs from a normal donor. (C and D) BMMSCs from MM patients at diagnosis. (E) The result of FAP expression on BMMSCs detected by qRT-PCR and (F) detected by FCM.

Figure 3. The expression of FAP after BMMSCs being stimulated. (A) BMMSCs were stimulated by TCCM (P = 0.0256, n = 8). (B) BMMSCs were cocultured with RPMI8226 cells (P = 0.035, n = 4). Both were detected by qRT-PCR.

FAP protect RPMI8226 and CAG cells from apoptosis induced by bortezomib

It is well known that the crosstalk of MM cells with BMMSCs through direct contact or secreted cytokines can facilitate MM progression,^4,52 survival and growth,^53,54 chemotherapy resistance,^55-57 and angiogenesis.^58,59 But the function of FAP, expressed on activated BMMSCs, in MM cells behavior is still less known and need to be further discussed. As shown in Figure 4, BMMSCs were able to protect bortezomib (20 nM) induced RPMI8226 (n = 3) and CAG (n = 2) cells apoptosis in cocultured system detected by FCM (Fig. 4B). This result was consistent with previous study.⁵³ Knockdown of FAP with siRNA weakened this protective effect. The apoptotic cells in FAP-siRNA group increased from about 6% to 16% (FAP-siRNA vs NC-siRNA, 50.94 ± 13.13% vs 42.76 ± 12.34%, P = 0.0014, n = 5) (Fig. 4B). A representative apoptosis analysis of CAG and a representative knockdown effect analysis of FAP on BMMSCs detected by FCM were shown in Figure 4A and C respectively. Next, we detected the apoptosis related protein with western blot in RPMI8226 and CAG cells in coculture system with either FAP knockdown or not in the presence of bortezomib. Our result demonstrated that in cocultured system knockdown of FAP in BMMSCs with siRNA activated caspase 3 and PARP-1 of RPMI8226 and CAG cells. Pro-apoptotic protein Bad and Bak (Fig. 5) had a moderate elevate and anti-apoptotic protein survivin (Fig. 6A) had a slight repression. These results further supported the result of FCM.

Figure 4. The protective effect of FAP on MM cell lines. (A) A representative apoptosis analysis of CAG in coculture system with FAP knockdown or not in the presence of 20 nM bortezomib detected by FCM. Cells of CD138-positive and annexin V-positive were considered as apoptotic cells. (B) FAP could protect RPMI8226 (n = 3) and CAG (n = 2) cells from bortezomib induced apoptosis (FAP-siRNA vs NC siRNA, P = 0.0014, n = 5). (C) A representative knockdown effect of FAP detected by FCM.

Figure 5. The probable role of FAP in BMMSCs mediated protection of RPMI8226 and CAG cells. BMMSCs could protect MM cell lines from apoptosis induced by 20 nM bortezomib through reducing the activation of caspase 3 and PARP-1 and the expression of pro-apoptotic proteins, such as Bad and Bak. Knockdown FAP could weaken this protection effect.

Figure 6. The probable role of FAP in BMMSCs mediated protection of MM cell lines. (A) β-catenin of MM cell lines was significantly activated when cocultured with BMMSCs. Knockdown of FAP in BMMSCs could decrease the expression of β-catenin and its downstream proteins such as c-myc, cyclin D1, and survivin and increase the expression of pp53 in RPMI8226 and CAG cells. (B) RPMI8226 cells pretreated with 40 μM XAV939 for 24 h, and then cocultured with BMMSCs (knockdown FAP or not). As a result, XAV939 could enhance bortezomib induced RPMI8226 cells apoptosis. (P = 0.017 and 0.031 in BMMSCs-NCsiRNA and BMMSCs-FAPsiRNA groups, respectively).

The probable mechanism for FAP to protect MM cells

Researches have demonstrated that BMMSCs secreted cytokines (IL-6, TNF-α, IGF-I, VEGF, CD44, ICAM-1, etc.) or adhesion of MM cell lines with BMMSCs could activate multitude of signaling pathways such as PI3K/Akt/mTOR, NFκB, JAK-STAT, MEK/ERK, and Notch pathways to support survival, proliferation, and migration as well as angiogenesis of MM cells.^60,61 However, BMMSCs-mediated Wnt/β-catenin signaling pathway activation is still less known. Our group demonstrated that Wnt/β-catenin signaling pathway is involved in the protection of BMMSCs in acute lymphoblastic leukemia (ALL) cells in previous study.⁶² Reports have shown that the activation of canonical Wnt/β-catenin signaling pathway in MM is ubiquitous.^63,64 Wnt/β-catenin signaling pathway has an important role in MM proliferation,⁶³ growth,⁶⁵ chemoresistance,⁶⁶ MM-related bone disease,⁶⁵ migration, and invasion.⁶⁷ Upregulated β-catenin can cause the activation of cyclin D1, c-myc, and survivin and repress the activation of p53.^63,68 A previous paper has hypothesized that bone marrow stromal cells-derived Wnt is essential for the activation of Wnt/β-catenin signaling in myeloma cells according to three observations.⁶⁹ The relationship of FAP and Wnt/β-catenin for MM cells behavior is still unknown. Here, we demonstrated that the expression of β-catenin in RPMI8226 and CAG cells were significantly increased when cocultured with BMMSCs for 48 h even in the presence of bortezomib. This result supported that β-catenin likely plays an important role in BMMSCs mediated chemoresistance including bortezomib and other novel chemotherapeutic drugs. Knockdown of FAP with siRNA could obviously suppress the expression of β-catenin (Fig. 6A). Subsequently, we detected the major downstream protein of β-catenin, such as survivin, cyclin D1, and c-myc and regulatory protein of β-catenin, such as pp53 by western blot. As shown in Figure 6, knockdown of FAP suppressed the downstream proteins of c-myc, survivin, and cyclin D1 and upregulate the expression of pp53. These results showed that FAP may play an important role in participating BMMSCs mediated protection of bortezomib-induced MM cell lines apoptosis and this protection effect may likely through β-catenin parthway.

To verify β-catenin signaling pathway participating BMMSCs- and FAP-mediated protection effect. We inhibited β-catenin signaling pathway with XAV939.⁷⁰ As shown in Figure 6B, 40 μM XAV939 could enhance cocultured RPMI8226 cells apoptosis. In BMMSCs group without FAP knockdown, the apoptotic cells of RPMI8226 was increased about 9% after treated with bortezomib and XAV939 compared with bortezomib single for 48 h (45.01% ± 6.65% vs 54.08% ± 4.67%, P = 0.017). In BMMSCs group with FAP knockdown, the apoptotic cells of RPMI8226 was increased about 6% after treated with bortezomib and XAV939 compared with bortezomib single for 48 h (53.52% ± 4.90% vs 59.44% ± 3.71%, P = 0.031).

Researches have demonstrated that cytokines,⁷¹ such as IGF-I,⁷² MCP-1,⁷³ IFN-γ,^74,75 TNF-α,⁷⁶ TGF-β,⁷⁷ etc. could active the expression of β-catenin. In order to verify whether the activation of β-catenin was dependent on direct contact with BMMSCs, we added cytokines mentioned above, conditioned medium (CM) derived from BMMSCs and cocultured RPMI8226 or CAG cells with BMMSCs using transwell chamber. As shown in Figure 7, β-catenin in RPMI8226 and CAG cells were not activated when separated by transwell chamber or by cytokines and CM derived from BMMSCs. This result demonstrated that the activation of β-catenin was dependent on direct contact with BMMSCs.

Figure 7. MM cell lines treated with additional cytokines, BMMSCs-derived conditioned medium or MM-BMMSCs cocultured system separated with transwell chamber could not activate the expression of β-catenin in MM cell lines demonstrating that the activation of β-catenin dependent on direct contact. Coaktail means the mixture of cytokines.

Taken together, our data indicated that there is no difference on the expression of FAP in the BMMSCs isolated from MM patients and normal donors. The expression of FAP can be increased under TCCM stimulation or coculture with MM cells lines. Further study demonstrated that BMMSCs and FAP can protect MM cells from apoptosis induced by bortezomib, which is likely through β-catenin signaling pathway.

Materials and Methods

Cell cultures

Human multiple myeloma cell lines RPMI8226, MM.1R, CAG, and the cell line of human bone marrow mesenchymal stem cells (BMMSCs) were kindly provided by Dr Qing Yi (Department of Cancer Biology, Lerner Research Institute, Cleveland Clinic). Cells were cultured in RPMI1640 (Thermo Scientific) medium containing 10% heat-inactivated fetal bovine serum (FBS) (Life Technologies), 1% l-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin at 37 °C in 5% CO₂. Culture of BMMSCs were according this paper.⁷⁸ Human bone marrow mononuclear cells (BMMNCs) from normal donors, obtained after informed consent in keeping with ethical guidelines declared by the First Affiliated Hospital of Zhejiang University and the Declaration of Helsinki, were isolated by lymphocyte separation medium and cultured in flasks, half of the medium was replaced every 3–7 d. After being cultured for 2 to 4 wk, cells became adherent and fibroblast-like cells, reaching to about >90% confluency. Cells were digested with trypsin-EDTA and subcultured. Passage 2 to 6 MSCs were used for the experiments.

Identification of BMMSCs

Accutase^TM solution (Merck Millipore) was used to detach the cells from culture flasks. Cells were washed twice with phosphate-buffered saline (PBS). Then cells were incubated with anti-CD14-FITC, anti-CD34-FITC, anti-CD90-APC, anti-CD105-PE antibodies (BD PharMingen), and anti-CD45-PE-Cy7 antibody (Biolegend) for 15 min according to the manufacturer’s instructions. Mouse IgG1, κ (Biolegend) were used as isotype control. Fluorescence staining was performed using a FACScan (Becton Dickinson).

Flow cytometry analysis for detection of FAP

Cells were detached from culture flasks with Accutase^TM solution (Merck Millipore). Cells were washed with PBS and non-specific antigens were blocked with 5% goat serum (Beijing Biosynthesis Biotechnology Co., LTD) for 1 h. Then cells were incubated with mouse anti-FAP (Santa Cruz Biotechnology) for 2 h. After that, cells were washed one more time and incubated with Alexa Fluor 488-labeled goat anti-mouse IgG (Invitrogen) for 30 min as secondary antibody. Analyses of ñuorescence staining were performed using a FACScan (Becton Dickinson).

Immunofluorescence

BMMSCs were seeded on the cover glass in 6-well plates for 24 h before being used for experiments. Next day, cells were washed three times with PBS for 5 min every time, and fixed in 4% paraformaldehyde for 30 min in room temperature. Cells were then washed three times with PBS and permeabilized within 0.3% Triton X-100 for 10 min. After being washed three times with PBS, non-specific antigens were blocked with 5% goat serum (Beijing Biosynthesis Biotechnology Co., LTD) for 1 h. After being washed three times with PBS, cells were incubated with anti-fibroblast activation protein (FAP) (Santa Cruz Biotechnology) at 4 °C overnight. After washing, cells were incubated avoiding light with Alexa Fluor 488-labeled goat anti-mouse IgG (Invitrogen) for 30 min. After washing, cells were incubated avoiding light with DAPI (4’,6-diamidino-2-phenylindole, 1:5000; Wako) for 5 min. After that, cells were visualized using fluorescence microscopy (Olympus) and photos were taken.

Quantitative real-time RT-PCR

Total RNA was extracted from BMMSCs using RNAiso™ Plus (TaKaRa) according to the manufacturer’s instructions. cDNA was generated from 1 μg of total RNA with PrimeScript^TM T reagent Kit with DNA Eraser (Perfect Real Time, Takara) according to the manufacturer’s instructions. Semiquantitative real-time PCR (qRT-PCR) was performed with SYBR^® Premix Ex Taq^TM II (Tli RNaseH Plus) (Takara) following the manufacturers’ instructions using iQ5 Multicolor Real-Time PCR Detection System (Bio-Rad Inc.). Semiquantitative real-time PCR reactions showed no non-specific amplifications. Sequences of primers for FAP (NM_004460) amplification were: forward, 5′-GTATTTGGAG TTGCCACCTC TG; reverse, 5′- GAAGGGCGTA AGACAATGCA C; GAPDH was used as an endogenous control. The sequences for GAPDH amplification were: forward, 5′-AAGGTGAAGG TCGGAGTCAA C; reverse, 5′-GGGGTCATTG ATGGCAACAA TA. The qRT-PCR program of FAP and GAPDH were: initial step at 95 °C for 30 s followed by 40 cycles of 95 °C for 5 s and 60 °C for 30 s. The specificity of primers was verified by melt curve. The 2^−ΔΔCT method was used to calculate relative quantification of FAP in BMMSCs.

Preparation of tumor cells conditioned medium (TCCM) and stimulation of BMMSCs by TCCM or coculture with RPMI8226 cells

RPMI8226 and MM1.R cells were cultured at a density of 1 × 10⁶/mL in RPMI-1640 containing 10% FBS. Sixteen hours later, culture medium were collected, and stored at 4 °C for up to 1 wk before use.⁷⁹ To examine the effects of myeloma-derived factors on the expression of FAP in BMMSCs, we added TCCM to the cell cultures of BMMSCs (50% TCCM to 50% complete medium) on day 1 and on day 4 when medium was replaced.⁷⁹ The division of RPMI8226 cells was blocked by addition of mitomycin C (10 μg/mL) for 2 h. Then BMMSCs were cocultured with pretreated RPMI8226 cells and RPMI8226 cells was changed every 2 d. After 7 d stimulation, BMMSCs were collected, washed with PBS, and cryopreserved with RNAiso™ Plus (TaKaRa) in –80 °C.

Transfection and cell apoptosis analysis by flow cytometry

The day before transfection, BMMSCs were planted in 12-orifice plates. When reaching about 90% confluency, cells were transfected with FAP siRNA (Genepharma Co. Ltd), forward (5′ → 3′): CGCCCUUCAA GAGUUCAUAT T, reverse (5′ → 3′): UAUGAACUCU UGAAGGGCGT T, using lipofectamine™ 2000 (Invitrogen). After 6 h, opti-medium was removed (Invitrogen), RPMI8226 or CAG cells were added to coculture with transfected BMMSCs (MM cells:BMMSCs = 1–2:1). Meanwhile, 20 nM bortezomib (Millennium Pharmaceuticals, Inc.) was added to the well to induce apoptosis in MM cells. After 48 h, cocultured RPMI8226 or CAG cells were separated from the BMMSCs monolayer by gently and careful pipetting with medium and washed twice with PBS. Then, cells were stained with fluorescein isothiocyanate (FITC)-conjugated annexin V (Biouniquer) and APC-conjugated anti-CD138 (Biolegend) following the manufacturers’ instructions. Annexin V-FITC positive cells were regarded as apoptotic cells. The interfering effect was detected by qRT-PCR and flow cytometry.

Western blot analysis

To examine intracellular signaling of cocultured RPMI8226 and CAG cells after FAP knockdown and bortezomib-induced apoptosis, we detect the changes of apoptosis-related proteins and β-catanin-related proteins. BMMSCs (2 × 10⁵) and RPMI8226 or CAG cells (5 × 10⁵) were cocultured in 6-well plates after BMMSCs were transfected with FAP-siRNA. RPMI8226 and CAG cells were also treated with conditioned medium derived from BMMSCs (2 × 10⁵) or 100 ng/mL cytokines (IFN-γ, IGF-I, MCP-1, TGF-β, TNF-α, and mixture of cytokines. All cytokines were from PeproTech), or were cocultured with BMMSCs separated by 0.4 μm transwell chambers (Corning Costar). After 48 h, RPMI8226 and CAG cells were collected and washed twice with PBS. Totol cell lysates were performed using lysis buffer supplemented with 1 mM PMSF (Beyotime). The protein concentration was detected using bicinchoninic acid (BCA, Sangon Biotech) assay according to the protocol. An equal amount of 20–40 µg of protein was separated by 10–12% sodium dodecyl sulfate/PAGE (SDS-PAGE) and transferred to a polyvinylidene difluoride membrane (PVDF) (Merck Millipore) using an electroblot apparatus (Bio-Rad Inc.). After being blocked with 5% non-fat milk for 2 h. PVDF membranes were incubated with the corresponding primary antibodies overnight at 4 °C. Anti-caspase 3, β-catenin, c-Myc, and pp53 antibody were both from Cell Signaling Technology. Anti-PARP-1, survivin, cyclin D1, bak, bad antibodies were from Abcam. Anti-β-actin antibody from Sigma-Aldrich. The membranes were then washed and incubated with the horseradish peroxidase (HRP)-conjugated antibody (Jackson ImmunoResearch Laboratories, Inc.; 1:5000) in TBST at room temperature for 2 h. After being washed, the bands on the membrane were detected on X-ray films using an enhanced chemiluminescence (ECL) detection kit for HRP (Biological Industries, Beit Haemek LTD).

Cell apoptosis analysis by flow cytometry after inhibition of β-catenin signaling pathway

To verify whether β-catenin signaling pathway participated in BMMSCs- and FAP-mediated protection effect of MM cells or not. We inhibited β-catenin signaling pathway with XAV939 (Sigma-Aldrich). RPMI8226 cells were pretreated with XAV939 (40 μM) for 24 h before cocultured with BMMSCs (knockdown FAP or not) and exposed to 20 nM bortezomib. After 48 h, cocultured RPMI8226 cells were separated from the BMMSCs and washed twice with PBS. Then, cells were stained with APC-conjugated anti-CD138 and FITC-conjugated annexin V. Annexin V-FITC positive cells were regarded as apoptotic cells.

Statistical analysis

Statistical analysis was performed using the Student t test to compare various experimental groups and all tests were considered statistically significant if P < 0.05.

10.4161/cbt.29924