Human mantle cell lymphoma Maver and Z138 cell lines were obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA). The cells were, respectively cultured in Iscove’s modified Dulbecco’s medium (IMDM) and RPMI-1640 (Gibco Invitrogen, Grand Island, NY, USA) medium containing 2 mM L-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin and 10% heat-inactivated fetal bovine serum (FBS) (Thermo Scientific HyClone, South Logan, UT, USA), at 37°C in a 5% CO₂ incubator.

The human B7-H3 (Gene ID: 80381) targeting small hairpin RNA (shRNA) sequence 5′-TCGTG TGCTGGAGAAAGATCAAACAGAGC-3′ and a negative non-targeted control sequence 5′-GCACTACCAGAGCTAA CTCAGATAGTACT-3′ were used to generate recombinant lentiviral particles (20). These recombinant lentivirus were prepared and titered to 5×10⁹ TU/ml (transfection unit), and the multiplicity of infection (MOI) was 10. Antibiotic-resistant clones were isolated and maintained in medium containing 200 μg/ml puromycin (Sigma-Aldrich, St. Louis, MO, USA). The B7-H3 knockdown was confirmed by RT-PCR, western blotting and fluorescence activating cell sorter (FACS). The infected cells comprised the B7-H3 shRNA (KD) and negative non-targeted control (NC) groups, and the non-infected cells were the control (CON) groups. These three groups of each cell line were used for the following experiments.

Total RNA was extracted and reverse transcribed for cDNA. The primers used for PCR were as follows: 5′-CTC TGCCTTCTCACCTCTTTG-3′ (forward) and 5′-CCTTGAG GGAGGAACTTTATC-3′ (reverse) for B7-H3 (134 bp) (23); 5′-TTGACGGTAAGGACGGACTC-3′ (forward) and 5′-ACTT GCAGTACTCCCCATCG-3′ (reverse) for matrix metalloproteinase-2 (MMP-2, 153 bp); 5′-TTGACAGCGACAAGAAG TGG-3′ (forward) and 5′-CCCTCAGTGAAGCGGTACAT-3′ (reverse) for matrix metalloproteinase-9 (MMP-9, 148 bp) (28); and 5′-TGACGTGGACATCCGCAAAG-3′ (forward) and 5′-CTGGAAGGTGGACAGCGAGG-3′ (reverse) for β-actin (205 bp). The PCR conditions were 94°C for 2 min, then 28–36 cycles (B7-H3/MMP-2/MMP-9: 36 cycles; β-actin: 28 cycles) at 94°C for 30 sec, 56–59°C (B7-H3: 56°C; MMP-2/MMP-9/β-actin: 59°C) for 30 sec, 72°C for 30 sec, and finally 72°C for 2 min.

The cells were lysed on ice, and the concentration of protein was determined using the BCA method. A total of 30 μg proteins were transferred onto nitrocellulose filter (NC) membranes after 10% SDS-PAGE. The membranes were blocked and incubated with primary antibodies, including rabbit anti-B7-H3 (1:500 dilution, 110 kDa) (clone EPNCIR122; Epitomics, Burlingame, CA, USA) or mouse anti-β-actin (1:1,000 dilution, 43 kDa) (clone C-2; Santa Cruz, CA, USA) monoclonal antibodies overnight at 4°C. After incubation with IRDye 800CW conjugated goat (polyclonal) anti-rabbit/antimouse IgG secondary antibody (1:10,000 dilution) (LI-COR, NE, USA) for 1 h, the fluorescent bands was visualized with an Odyssey infrared imaging system (LI-COR), and the gray values were analyzed using Odyssey V3.0 software.

Single cell suspensions were stained with antibodies on ice for 30 min. After three washes in PBS, the cells were analyzed using a flow cytometry system (FACSCalibur; BD Biosciences, San Jose, CA, USA). The monoclonal antibodies used to measure the expression of the cell surface markers by flow cytometry included B7-H3 (CD276)-APC (clone 7-517) from eBioscience (San Diego, CA, USA) and CD3-APC, CD19-APC, CD117-APC, CD20-FITC, CD10-PE, CD5-PE, CD38-FITC, CD56-PE, CD138-APC, CD34-PE, CD45-PerCP, and appropriate isotype controls from BD Biosciences.

The Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Tokyo, Japan) was used to study the effects of B7-H3 shRNA on cell proliferation. Resuspended cells were plated at 5×10⁴ cells/well in a 96-well plate for 24-, 48- or 72-h inoculation. A volume of 10 μl/well of CCK-8 solution was added into the plate. After incubation at 37°C for 5 h, the absorbance was measured at 450 nm using a microplate reader. The assay was performed in sextuplicate for each group.

A 5×10³ single-cell suspension was resuspended in 1 ml medium with 20% FBS and 0.9% methyl cellulose (Sigma, St. Louis, MO, USA). The samples were plated in 24-well plates and incubated for 14 days. A colony with >50 cells was counted as one positive colony. The colony-forming ability (CFA) = (colonies counts in experiment group/ control group) × 100%. Each experiment was repeated three times.

Female BALB/c nude mice, 5–6 weeks of age, 16–18 g of weight (Experimental Animal Center, Peking University Health Science Center, Beijing, China), were bred under specified pathogen-free (SPF) conditions. All of the care, experimental procedures and handling of animals were performed with approval of the Institutional Authority for Laboratory Animal Care of Peking University. Mice were pretreated by intraperitoneal (i.p.) injection, with 100 mg/kg cyclophosphamide (Jiangsu Hengrui Medicine Co. Ltd., China) once daily for two consecutive days. The next day, 2×10⁷ cells (the Maver and Z138 cells in CON, NC or KD group respectively, 6 mice per group) in 100 μl normal saline (NS) were injected subcutaneously (s.c.) into right axilla region of the nude mice.

The xenografted tumor volume (V) was measured every other day and calculated as: V (mm³) = length × width² × 0.5. The inhibition rate of tumor growth was calculated using the formula: (1 - average tumor weight of treated group/average tumor weight of control group) × 100%.

At the end of observation (20 days post-inoculation), the xenograft tumors of Maver and Z138 mice were excised and fixed in 10% neutrally buffered formalin for 24 h. After embedded in paraffin, deparaffinized and rehydrated, the histologic sections were subjected to heat-induced epitope retrieval using microwave in 1 mmol/l EDTA buffer (pH 8.0) and quenched for endogenous peroxidase activity with 3% hydrogen peroxide. Then the sections were stained with an rabbit polyclonal antibody to the cell proliferation-associated antigen Ki-67 (1:500 dilution) or a mouse monoclonal antibody to proliferating cell nuclear antigen (PCNA) (1:300 dilution) (Dako Corp., Copenhagen, Denmark), respectively, followed by incubation with the secondary antibody (Dako Corp.) according to the avidin-biotin-peroxidase method. Phosphate-buffered saline (PBS) were used as negative controls. After incubation with 3,3′-diaminobenzidine (DAB) and counterstained with hematoxylin, the sections were analyzed in 5 randomly selected microscopic fields (400x). The Ki-67 labeling index or the positive expression rate of PCNA = (positively stained cells counts / the total number of nucleated cells) × 100%.

The Maver and Z138 cells in each group were collected, washed with PBS and fixed in 70% ethanol overnight at −20°C. The cellular DNA was stained with propidium iodide (PI) (500 μg/ml) (Biosea, China) (10 μl/10⁵ cells) for 10 min at room temperature. The DNA content and cell number were determined by FACS analysis, and the cell cycle profiles were analyzed using the ModFit program (Verify Software House, Inc.). The proliferation index (PI) was calculated using the following equation: PI = [(S+G2M) / (G0/G1+S+G2M)] × 100%. Each experiment was repeated three times.

For the in vitro migration and invasion assays, 4×10⁵ or 2×10⁵ cells were resuspended in serum-free medium and placed on the top of an 8-μm pore size Transwell chamber (8.0 μm PC, Corning-Costar, Corning, NY, USA) or Matrigel (1:5 dilution) (BD Biosciences, San Jose, CA, USA) invasion chambers. The lower chambers contained medium with 10% FBS. After 24 h of incubation, the migrating cells in the lower chamber or invading cells on the bottom of each well were stained with 4′,6-diamidino-2-phenylindole (DAPI) (1 mg/ml, Solarbio, China) or 0.1% crystal violet following by fixation in methyl alcohol for 30 min, respectively. Then, the number of cells in 6 randomly selected microscopic fields (200x) was counted with a BX51 fluorescence microscope (Olympus, Japan) or a DMIL inverted phase microscope (Leica, German). The migration rate = (cells counts in the lower chamber / total number on the top of Transwell chamber) × 100%. MMP-2 and MMP-9 were detected by RT-PCR to further determine the abilities of tumor cells to penetrate the cell matrix. Each experiment was repeated three times.

Rituximab (R, 500, 1,000, 1,500, 2,000, 2,500, 3,000, 3,500 and 4,000 μg/ml) [MabThera, Roche Pharma Ltd., Germany] or bendamustine (Ben, 1, 2, 4, 8, 10, 16, 20 and 40 μg/ml) (Ribomustin, Ribosepharm GmbH, Germany) was added to three groups of Maver and Z138 cells for 24, 48 or 72 h respectively, and then the absorbance was detected using the CCK-8 reagent to evaluate the effects of cell proliferation inhibition by the chemotherapy drugs.

A total of 4×10⁵ Maver cells in each group were grown in triplicate in 6-well plates with 3,500 μg/ml R and/or 4 μg/ml Ben for 12 and 24 h, while the same counts of Z138 cells with 2,500 μg/ml R and/or 4 μg/ml Ben. A volume of 10 μl Annexin V-FITC (20 μg/ml) (Biosea, China) was added to the collected cells. After incubation for 15 min at room temperature, 300 μl binding buffer was added. Then, we added 10 μl PI (50 μg/ml) (Biosea, China) to the mixtures. The cells were examined by flow cytometry within 1 h to determine the cell apoptosis rates induced by the chemotherapy drugs. CellQuest software (Becton-Dickinson) was used for data acquisition and analysis.

The Maver cells (8×10⁶ cells/dish) were treated with 3,500 μg/ml R and/or 4 μg/ml Ben for 0 and 24 h, while the same counts of Z138 cells were treated with 2,500 μg/ml R and/or 4 μg/ml Ben. The cells were then collected and lysed on ice for 1 h. The concentration of protein was detected using the BCA method. A total of 150 μg protein was detected using the caspase-3 Colorimetric Assay kit (Keygen, China). The activity of caspase-3 = OD₄₀₅ in the experiment group/OD₄₀₅ in the control group.

Mice bearing non-infected Maver and Z138 cell xenografts were allocated randomly into 10 groups respectively (6 mice per group), including: a) an equal volume of NS, b) R (50 mg/kg), c) negative non-targeted control plasmid (pNC) + R, d) short hairpin RNAs targeting B7-H3 plasmid (pKD) + R, e) Ben (25 mg/kg), f) pNC + Ben, g) pKD + Ben, h) R+ Ben, i) pNC + R+ Ben, j) pKD + R + Ben. The treatments began at day 8 and day 10 after inoculation, respectively, for Maver and Z138 mice, when tumors reached an average volume of 100 mm³. The chemotherapy drugs and NS were i.p. administered, while the mixture of plasmid DNA (10 μg) and Lipofectamine 2000 (30 μl; Invitrogen) in 50 μl NS were intratumorally injected. All of the treatments were performed every other day, for a total of three times.

The data are shown as the mean ± standard deviation (SD) of triplicate values for each experiment. Statistical comparisons were performed using Student’s t-test. A value of p<0.05 was considered statistically significant. The statistical analysis was performed using SPSS 18.0 software (Chicago, IL, USA).

B7-H3 knockdown in Maver and Z138 cells was performed using lentivirus transduction to stably express shRNA targeting B7-H3. There was no significant difference of B7-H3 expression between the parental non-infection (CON) cells and the transfection negative non-targeted control cells (NC) in each cell line by RT-PCR, western blotting and FACS (p>0.05). The B7-H3 expression in the shB7-H3/Maver and shB7-H3/Z138 cells (KD groups) was decreased compared to the relative NC groups (p<0.05). The inhibition rates of mRNA expression in Maver and Z138 cells were 84.5 and 81.2%, whereas the nuclear and cytoplasmic proteins were reduced by 80.3 and 74.5%, and the membrane proteins were the most significantly inhibited by 86.9 and 82.4%, respectively (Fig. 1).

To evaluate the effects of B7-H3 knockdown on tumor cell proliferation in vitro, we used CCK-8 and colony formation assays. The stable knock-down of B7-H3 in Maver and Z138 cells significantly reduced cell growth. Compared to the relative NC groups, the growth of shB7-H3/Maver and shB7-H3/Z138 cells was decreased by 41.7 and 37.5% after 72 h of incubation, respectively (p=0.015 and 0.028) (Fig. 2A). The colony formation assay further confirmed that B7-H3 silencing inhibits Maver and Z138 cell proliferation. At day 14 of incubation, the colony-forming ability (CFA) of shB7-H3/Maver and shB7-H3/Z138 cells was significantly decreased by 71.2 and 77.2%, respectively compared with the NC groups (Fig. 2B).

To detect the in vivo effects of B7-H3 knockdown of tumorigenicity, we established xenograft models by subcutaneously injecting Maver and Z138 cells into right axilla region of BALB/c nude mice. The increases of tumor volumes in B7-H3 silencing Maver and Z138 mice were slowed down comparing to their NC groups injected with non-targeted sequence transfected cells, while no significant differences were observed between their NC groups and the non-infection groups (p>0.05) (Fig. 2C). At day 20 after inoculation, the tumors were excised from the mice and weighed, and the inhibition rates of tumor growth with B7-H3 knockdown were 59.1 and 65.0% in Maver and Z138 xenograft models (p=0.010 and 0.003) (Fig. 2D). Furthermore, we assessed two markers reflecting tumor cell proliferation activity, Ki-67 and PCNA, through immunohistochemical staining. There were apparently fewer positively Ki-67 and PCNA stained cells in the B7-H3 knockdown groups than in the relative NC groups of Maver and Z138 xenograft models. The Ki-67 labeling indexes of the excised xenografts in B7-H3 knockdown groups were 34.1±5.2 and 42.3±4.1%, respectively, significantly lower than in the relative NC groups in Maver and Z138 mice (p=0.001 and 0.001). The B7-H3 knockdown also decreased the positive expression rates of PCNA compared with the NC groups in the two types of MCL xenograft models (38.9±3.9 vs. 76.4±5.9 and 39.9±2.4 vs. 80.3±5.5%) (Fig. 2E).

We used flow cytometry comparing the G0/G1, S and G2/M phases to determine whether B7-H3 expression affects cell cycle progression. Fig. 3A and B, respectively, show that the Maver and Z138 cell cycle progression was inhibited after B7-H3 silencing. The proliferation index (PI) was decreased by 18.46% in the KD group comparing to the NC group of Maver cells (p=0.002); while it was reduced by 13.49% in B7-H3 knockdown Z138 cells (p=0.034). This suggests that the knockdown of B7-H3 arrests MCL cell cycle at the G0/G1 phase to inhibit cell proliferation.

We used Transwell migration and invasion assays to compare the cell migration rate and invasive capacity in each group to determine whether B7-H3 acts as a tumor migration and invasion regulator. Fig. 4A shows that the migration rates of the shB7-H3/Maver and shB7-H3/Z138 cells to the lower chamber were significantly reduced after 24-h incubation compared with the NC groups, with a 6.9 vs. 34.4% and a 7.4 vs. 63.8% reduction, respectively. In addition, the invasive capacity of shB7-H3/Maver and shB7-H3/Z138 cells was reduced by 85.5 and 80.1% compared to the NC groups (Fig. 4B). Both the cell migratory and invasive potential in the CON and NC groups of Maver and Z138 cells were similar (p>0.05). Furthermore, we measured the invasion-related proteins by RT-PCR, and found that MMP-2 and MMP-9 were lower in shB7-H3/Maver and shB7-H3/Z138 cells than in the NC groups (Fig. 4C). These results indicate that silencing B7-H3 can impede cell migration and inhibit cell invasion via downregulating the expression of MMP-2 and MMP-9.

To determine whether B7-H3 knockdown affects drug-induced cytotoxicity and apoptosis, we selected R and Ben, which is increasingly being used for the frontline treatment of mantle cell lymphoma (29). Following treatment with various concentrations of R or Ben for 24, 48 or 72 h, a dose-dependent and time-dependent inhibition of cell growth was observed in each group of Maver and Z138 cells using the CCK-8 assay (data not shown). Concentrations of 3,500 μg/ml R and/or 4 μg/ml Ben were treated in Maver cell groups, while 2,500 μg/ml R and/or 4 μg/ml Ben were selected for the Z138 cell groups, and then the absorbance was compared (Fig. 5). The cell survival rates in the KD groups of Maver and Z138 cells were significantly decreased compared with their NC groups (p<0.05), and these two drug combinations synergistically inhibited MCL cell proliferation.

Exposure to 3,500 μg/ml R and/or 4 μg/ml Ben for 12 and 24 h in Maver cells or 2,500 μg/ml R and/or 4 μg/ml Ben in Z138 cells suggested that B7-H3 silencing promoted apoptosis in a time-dependent manner. The apoptosis rates of shB7-H3/Maver cells in the drug combination groups were 63.84±7.07% for 12 h and 82.43±4.68% for 24 h, which were the most significantly increased (p=0.008 and 0.001). The same conclusion can be drawn for Z138 cells (68.06±7.01% for 12 h and 95.99±3.84% for 24 h) (p=0.006 and 0.002) (Fig. 6A–D). After exposure to R and/or Ben for 24 h, we measured the activity of the apoptosis-related protein caspase-3. The results demonstrated that the activity of caspase-3 in the KD group of Maver cells compared to the NC group was significantly increased treated with R, Ben, and R+Ben (3.32±0.37, 3.36±0.41, and 4.57±0.50 vs. 2.01±0.35, 2.33±0.26, and 2.93±0.61, respectively) (p=0.011, 0.022, and 0.023) (Fig. 6E). Also, the same conclusion can be drawn for Z138 cells (3.45±0.43, 3.79±0.31, and 4.65±0.31 vs. 2.42±0.34, 2.68±0.52 and 3.40±0.35, respectively) (p=0.032, 0.034, and 0.010) (Fig. 6F). These results indicate that silencing B7-H3 increases drug-induced cytotoxicity and promotes drug-mediated apoptosis by increasing the caspase-3 activity in vitro.

In order to explore the impact of B7-H3 knockdown on the antitumor activity of chemotherapy drugs in vivo, we constructed tumor-bearing mouse models injecting into non-infected Maver and Z138 cells, respectively. The treatments began on day 8 or day 10 in Maver or Z138 xenograft model, when the average tumor volumes reached 100 mm³. As shown in Fig. 6A and E, the plasmids of B7-H3 shRNA (pKD) combined with R and/or Ben were more effective in reducing the established Maver and Z138 tumor growth comparing to the groups of non-targeted control plasmid (pNC) combining chemotherapy, while there were no significant differences in the tumor volumes between the pNC combined with chemotherapy groups and the chemotherapy groups alone. At the end of observation, the inhibition rates of tumor growth in pKD combined with R, Ben, and R+Ben groups in Maver xenograft model were 83.3, 80.9 and 92.3% respectively, which were higher than in the pNC combined with chemotherapy groups with 62.2, 69.5 and 85.4% (p=0.019, 0.049 and 0.043) (Fig. 7B and C). The same conclusion can be drawn for Z138 xenograft model, and the inhibition rates of tumor growth were 83.7, 84.8 and 92.9% vs. 66.8, 64.7 and 82.1% (p=0.030, 0.009 and 0.027) (Fig. 7F and G). Both the Maver and Z138 groups of B7-H3 shRNA combined with two drugs received the best antitumor activity. Besides, all of the mice treated with B7-H3 shRNA exhibited no body weight loss at the end of our experiment (Fig. 7D and H). These results indicate that the B7-H3 silencing can apparently enhance chemosensitivity to mantle cell lymphoma in the xenograft model.