Tumour-associated macrophages mediate the invasion and metastasis of bladder cancer cells through CXCL8

Patient selection and data collection

The Second Hospital of Lanzhou University approved this research, and all participants provided informed consent for this study. Immunohistochemical staining was performed on 55 pathologically diagnosed bladder cancer patients who underwent total or partial cystectomy between September 2017 and December 2018. All hematoxylin and eosin (H&E)-stained specimens obtained by initial cystectomy were reassessed independently by experienced uropathologists.

Immunohistochemical staining and quantification

Paraffin-embedded sections for immunohistochemical detection were first deparaffinized in xylene and rehydrated in ethanol, and added to a pressure cooker with 0.01 M citrate (pH = 6.0) buffer and antigen recovery. After pressure was reached, the sections were incubated in the pressure cooker for 2 min. Endogenous peroxidase activity was blocked using 1% hydrogen peroxide in methanol according to the Streptavidin-Peroxidase Detection Kit (ZSGB-BIO, China) after blocking in 1% bovine serum albumin (BSA) for 10 min. Slides were incubated overnight at 4 °C with anti-human CXCL8 antibody (Abcam, CA, goat polyclonal, diluted 1/1,000 in 1% BSA), anti-CD163 antibody, anti-VEGF antibody, anti-MMP-9 antibody, and anti-E-cadherin antibody (Santa Cruz, CA, goat polyclonal, diluted 1/250 in 1% BSA), respectively. Slides were then incubated with biotinylated anti-goat IgG secondary antibody for 10 min at room temperature. Sections were subsequently stained with 3, 3′-diaminobenzidine (DAB; Vector Laboratories), counterstained with hematoxylin (Solarbio, Peking, China), dehydrated, and coverslipped. The results were observed under a microscope (OLYMPUS BH-2 Microscopes, USA), and the positive staining rate was counted.

The semi-quantitative methods used for immunohistochemistry results were as follows: CD163-positive round cells were counted in cancerous areas using at least three independent high-power microscopic fields (HPF; 400×, 0.0625 µm²). The mean number of CD163-positive cells was considered the number of tumour-infiltrating TAMs. To quantify the expression levels of CXCL8, MMP9, VEGF, and E-cadherin in bladder cancer cells, tumour cells were counted in at least three independent fields (HPF; 400×, 0.0625 µm²), and that number was divided by the total cancer cells to calculate the percentage of positive cells (1–100%). The evaluation was carried out blindly by two investigators, without knowledge of the patients’ outcome or other clinic pathological characteristics.

Cell lines and reagents

Three bladder cancer cell lines, 5637, UM-UC-3, and T24, as well as acute monocytic leukemia cell line THP-1 and EA⋅hy926 (Cell Bank of the Chinese Academy of Sciences, Peking, China) were used in the present study. Cell lines were maintained in RPMI-1640 medium (Gibco, Thermo Fisher Scientific, USA) supplemented with 10% fetal bovine serum (Pan Biotech, Germany), 100 units/mL penicillin, and 100µg/mL streptomycin in a standard humidified incubator at 37 °C in 5% CO₂. Phorbol-12-myristate-13-acetate (PMA) was purchased from Multis Sciences (LIANKE) Biotech (Hangzhou, China). Recombinant protein IL-4 (PeproTech, USA) was also used in the described experiments.

Generation of TAMs from THP-1

To generate TAMs from monocytic THP-1, cells were seeded 1 × 10⁶ cells in a dish and treated with PMA (200nM) for 24 h for differentiation into resting macrophages (M0 cells). For M2 polarization, cells were treated with IL-4 (20 ng/mL) for an additional 48 h (Genin et al., 2015; Mantovani et al., 2004). qRT-PCR analysis for CD163 was performed to confirm TAM generation.

Enzyme-linked immunosorbent assay (ELISA)

A human CXCL8 kit and MMP-9 kit were purchased from Neobioscience Technology Company, China, and ELISA was performed according to the manufacturer’s instructions. Both standard and tissue samples were added to a 96-well plate with 0.1 mL of biotinylated antibody and incubated at 37 °C for 60 min. After gentle washing, enzyme conjugate working solution was added to each well and the samples incubated for 30 min. Chromogenic substrate (TMB) was then added and the samples incubated for 15 min in the dark. After stopping with stop solution, the optical density at 450 nm was quantified using a microplate reader (Bio Tek Instruments, USA) for further analysis.

Quantitative real-time polymerase chain reaction (qRT-PCR)

RNA was extracted from cells using TRIzol (Takara, Japan) as per the manufacturer’s instructions. Conversion to cDNA was achieved using cDNA PrimeScript™ RT Master Mix (Perfect Real Time) (Takara, Japan) in a reverse transcription PCR instrument (Bio-Rad Laboratories, USA). Quantitative qRT–PCR was carried out using the CFX Real-Time PCR System (Bio-Rad Laboratories, USA) in a 15-µl reaction volume containing first-strand cDNA, TB Green™ Premix Ex Taq™ II (Tli RNaseH Pluse). Primer sets can be found in Table 1. Relative fold changes in mRNA levels were calculated after normalization to GADPH using the comparative Ct method.

Western blot analysis

After the treated cells were collected and washed with PBS and RIPA buffer lysis, they were placed on ice for 30 min. The concentration of supernatant protein was determined using the BCA method (Solarbio, Peking, China) after centrifugation at a speed of 14,000 r.p.m. for 15 min. The same amount of total protein was resolved using SDS-PAGE and transferred to the PVDF membrane. Antibodies to the target protein were tested overnight at 4 °C. The primary antibodies used in this study were anti-E-cadherin (dilution,1:100), anti-VEGF (dilution,1:200), and anti-actin antibody (dilution, 1:1000; ZSGB-BIO, China) as an internal loading control. After incubation with IRDye® near-infrared fluorescence secondary antibody, the protein was detected using Odyssey Near-Infrared Fluorescence Imaging Systems (LI-COR Biosciences, USA).

Cell invasion assay using conditioned media obtained from TAMs

THP-1-derived TAMs were seeded in serum-containing media onto cell culture dishes at a density of 1,000,000 cells/dish. Twenty-four hours later, the medium was replaced with serum-free RPMI1640 medium, and cells were incubated for an additional 48 h. The CM was collected from manipulated TAMs, followed by centrifugation to remove cells or cell debris. The invasion assay was performed using Costar® Transwell® inserts (Corning Incorporated, USA) according to the manufacturer’s directions, and 24-well plates with 8-µm pore membranes were lightly coated with 2 mg/mL Matrigel (Solarbio, Peking, China). T24, 5637, and UM-UC-3 cells were added to each insert at a density of 30,000 cells/well in the CM described above. The lower chamber contained RPMI1640 medium with 10% FBS as a chemoattractant. The cells were maintained in a humidified incubator with 5% CO₂ at 37 °C for 48 h. Invading cells on the membrane were then fixed using paraformaldehyde, stained with 0.1% crystal violet (Solarbio, Peking, China), and positively stained cells were counted. The cell numbers counted from five random fields were averaged.

Wound healing assay

Cells were seeded in six-well plates and incubated overnight. A vertical bar was drawn using a mark pen on the back of the six-well plate, and a 1mL pipette tip was used to transversely scratch the cell monolayer. After washing the scraped cells from the wells with PBS, serum-free medium, CM, and CM with added anti-CXCL8 neutralizing antibody were added to the wells. Tumour cell migration was observed at 0, 6, 12 and 24 h. Wound healing percentage was calculated using Image J software.

Vascular endothelial cell tube formation experiment

Matrigel (8.7 mg/mL) was placed in a refrigerator at 4 °C to slowly melt overnight, and 50µL was spread in each well of a 96-well plate and placed in an incubator at 37 °C for 1 h. The EA⋅hy926 cells were prepared into a cell suspension of 2 × 10⁵ cells/mL, which was mixed well and added to a 96-well plate at a volume of 100 µL per well. The CM prepared in advance was then added at a volume of 50 µL per well, incubated in an incubator at 37 °C and 5% CO₂, and observed after 8 h (Arnaoutova & Kleinman, 2010; Lee & Kang, 2018; Ma et al., 2017). We observed the tube formation of endothelial cells and counted the number of vessels and visible nodes.

Statistical analysis

All experimental results were expressed as mean ± SD and analyzed for statistical significance by a two-tailed Student’s t-test. The interrelationship was examined using Spearman’s analysis. Data were statistically analyzed and plotted using IBM SPSS version 23 (SPSS Inc., Chicago, IL, USA) and Prism software version 5 (San Diego, CA, USA), respectively. A p-value <0.05 was considered statistically significant.

The association between the expression of CXCL8 in bladder cancer and clinicopathological features

The clinicopathological characteristics of 55 bladder cancer cases are outlined in Table 2. Representative pictures of immunohistochemical staining for CXCL8, CD163, MMP9, VEGF, and E-cadherin are shown in Figs. 1A–1J. Since CXCL8 is mainly expressed in the cytoplasm of tumour cells, the expression levels were significantly higher in bladder cancer tissues than in normal tissues (Fig. 1K). Strong CXCL8 expression was observed in the tumor tissue of high-grade tumors, while low-grade tumors had relatively lower expression levels. The expression of CD163 (a marker of TAMs), MMP9, VEGF, and E-cadherin was quantitatively assessed and compared with the CXCL8 expression score. The results showed that a high expression of CXCL8 in tumour tissues was associated with a high infiltration of TAMs, high expression of VEGF and MMP9, and a low expression of E-cadherin (Figs. 1L–1O). When compared to clinicopathological variables, a higher CXCL8 expression score was associated with higher T category and tumor grade (Table 2). These findings suggest that the infiltration of TAMs and the expression of CXCL8 in bladder cancer exhibit the aggressiveness of the remodeled stroma.

TAM-like PBM-derived macrophages can secrete CXCL8

To demonstrate how macrophages in the microenvironment can secrete CXCL8, we used TAM-like PBM-derived macrophages with IL-4-induced M2 phenotype (Figs. 2A–2C). The TAM-like PBM-derived macrophages were cultured in a serum-free medium, which was collected daily. The amount of CXCL8 secreted in the collected culture medium was then measured using an ELISA experimental method. It can be concluded that different concentrations of IL-4 in vitro induce different amounts of CXCL8 secreted by TAM-like PBM-derived macrophages (Fig. 2D). The highest amount of CXCL8 was secreted when the concentration of IL-4 was 20 ng/mL (Fig. 2E). After analyzing the expression of CXCL8 and CD163 using qRT-PCR, we concluded that CXCL8 expression significantly increased (Fig. 2F) and TAM-like PBM-derived macrophages induced by IL-4 significantly expressed the specific M2 macrophage marker CD163 (Fig. 2G).

CXCL8 derived from TAM-like PBM-derived macrophages promotes the invasion of bladder cancer

To examine the potential consequences of recruiting macrophages into bladder cancer tumour microenvironments, we investigated the potential effect of TAM-derived CXCL8 on bladder cancer cell invasion. We tested invasion ability after culturing bladder cancer cells 5637, T24, and UM-UC-3 with/without CM, using 8 µm membrane inserts as described in the Materials and Methods. The results show that CM increases bladder cancer cells’ invasion capacity, and this effect can be suppressed using an anti-CXCL8 neutralizing antibody (Figs. 3A–3J). RNA subsequently extracted from bladder cancer cells after the CM treatment described above showed that the expression of MMP-9 had significantly increased in the treated cells (Fig. 3K). This was also demonstrated by ELISA at the protein level (Fig. 3L), indicating that CXCL8 can promote bladder cancer invasion by stimulating bladder cancer cells to secrete MMP-9.

CXCL8 derived from TAM-like PBM-derived macrophages increases the migration of bladder cancer cells

We conducted a wound healing experiment to investigate the effect of CXCL8 on bladder cancer cell migration. We planted 5637, T24, and UM-UC-3 bladder cancer cells on a six-well plate, attached them to the wall for 24 h, and then added CM (Figs. 4A–4R). We found that the cell migration increased significantly after the addition of CM. After neutralizing CXCL8 in CM with anti-CXCL8 neutralizing antibody, the accelerated migration rate was observed to be significantly reduced. This suggests that CXCL8 can promote the migration of cancer cells (Fig. 4S).

CXCL8 derived from TAM-like PBM-derived macrophages promotes the formation of blood vessels in tumours

Angiogenesis is associated with the benign-malignant transition of tumours, with resulting blood vessels providing tumour cell nutrients and oxygen for their propagation, invasion, and metastasis. The blood vessel formation process is complex, and TAMs are one of the major contributors (Lin & Pollard, 2007). To investigate the proangiogenic ability of CXCL8 derived from TAM-like PBM-derived macrophages, we used the serum-free RPMI-1640 and CM described above to culture T24 cells for 24 h, and extracted the culture medium, which was centrifuged to become new CM (called CM-C and CM-T24, respectively). Using the endothelial cell tube formation assay described above, we added CM-C (as control) and CM-T24 as EA⋅hy926 cells were added to the 96-well plate. After 8 h of culture, we observed significantly more endothelial cell tube formation in wells with CM-T24 added than in those with CM-C. We used CM to culture T24 cells with an additional anti-CXCL8 neutralizing antibody, and then cultured EA⋅hy926 cells in the extraction medium (CM-T24-NA) described above. We found that the tube formation of EA⋅hy926 cells was significantly inhibited (Figs. 5A–5D). Total RNA and total protein were then extracted from the CM-treated bladder cancer cells. QRT-PCR and WB analysis showed that the expression level of VEGF in CM-treated bladder cancer cells increased, indicating that CM can induce angiogenesis by promoting the secretion of VEGF in bladder cancer cells. The inhibition of CXCL8 with the CXCL8-neutralizing antibody reduced the ability of CXCL8 to promote VEGF secretion in bladder cancer cells, suggesting that CXCL8 in CM plays a role in this process (Figs. 5E–5G).