Tumour cell lines HT-29 and FaDu produce proinflammatory cytokines and activate neutrophils in vitro: possible applications for neutrophil-based antitumour treatment

Abstract

There is evidence that polymorphonuclear neutrophils (PMNs) can exert severe antineoplastic effects. Cross-talk between tumour cells and endothelial cells (ECs) is necessary for the accumulation of PMN around a tumour. This work reports the ability of two PMN-sensitive, human, permanent cell lines-colorectal adenocarcinoma (HT-29) and pharyngeal squamous-cell carcinoma (FaDu) cells-to act as inflammatory foci. PMNs were cytotoxic to both lines, the adhesion of the PMNs to the tumour cells being important in this effect. The tumour cells released appreciable amounts of IL-8 and GROalpha, and induced the transmigration of PMN through human microvascular-EC monolayers. Conditioning media associated with both lines induced the adhesion of PMN and the surface expression of ICAM-1 in microvascular-EC. In addition, FaDu-conditioning-medium strongly induced the production of proinflammatory cytokines by microvascular-EC. These results support the idea that tumour cells might normally induce a potent acute inflammatory response, leading to their own destruction.

Figure 1

Representative microscopic photographs (×100) of tumour cells exposed overnight to PMN. Tumour cells were cultured in 12-well culture plates containing 12 mm diameter cover glasses. 5 × 10⁶ PMN were placed on culture inserts that were located over the wells containing the tumour cells. Control, medium without PMN in the inserts; TW 3 μm, PMN placed in 3 μm pore size inserts; TW 0.4 μm fMLP, PMN activated with 0.1 μmoles/L fMLP placed in 0.4 μm pore size inserts; and TW 3 μm Ab:CD18, CD162, CD44, PMN placed in 3 μm pore size inserts with 10 μg/mL of mouse anti-human CD44 monoclonal antibody, 10 μg/mL of mouse anti-human CD162 monoclonal antibody and 10 μg/mL of mouse anti-human CD18 monoclonal antibody. Covers were then stained with AO-EB. AO permeates throughout the cells and renders the nuclei green. EB is taken up by the cells only when cytoplasmic membrane integrity is lost, and stains the nuclei red. N = 3 with similar results.

Figure 2

Adhesion of PMN to tumour cells. 2 × 10⁶ DiI-labeled PMN were added to culture wells containing confluent dermal fibroblasts (fibroblasts), HMVEC (HMVEC), HT-29 (HT-29), HT-29 with 10 μg/mL of mouse anti-human CD44 monoclonal antibody, 10 μg/mL of mouse anti-human CD162 monoclonal antibody and 10 μg/mL of mouse anti-human CD18 monoclonal antibody in the medium (HT-29 + AbCocktail), FaDu (FaDu), FaDu with the Ab Cocktail (FaDu + AbCocktail), and HMVEC previously treated with 10 U/mL of human recombinant IL-1β for 5 hours (HMVEC + IL-1B) and incubated at 37°C for 30 minutes. After washing, adhered PMN were evaluated as described in the Methods Section. Bars represent the mean ± SEM of 7 independent experiments performed by sixtoplicate. Statistical significance was assessed using ANOVA test; *P < .05, **P < .01, and ***P < .001 when compared with HMVEC group.

Figure 3

Tumour cell-induced PMN transmigration through human microvascular endothelium. DiI-labeled PMN were added to 3 μm pore size cell culture inserts previously coated with HMVEC cultured at confluence. Inserts were placed upon culture wells containing confluent human dermal fibroblast (fibroblasts), HT-29 (HT-29), FaDu (FaDu) and HMVEC previously treated with 10 U/mL of human recombinant IL-1β for 5 hours (HMVEC + IL-1B). See Methods. After 3 hours at 37°C, PMN in the lower side of the system were evaluated as described in the Methods section. Bars represent the mean±SEM of 7 independent experiments performed by sixtoplicate. Statistical significance was assessed using ANOVA test; *P < .05, and ***P < .001 when compared with fibroblasts group.

Figure 4

Protein array analysis of culture medium from HT-29, FaDu and human dermal fibroblasts. Culture medium (DMEM plus 10% FBS) was replaced and collected after 48 hours. Controls were performed with DMEM plus 10% FBS. The proteins that were analyzed are indicated in the bottom panel; pos, positive controls, neg, negative controls. Bars are the densitometric evaluation of samples (right panel); AU, arbitrary units. Values were calculated as the difference between the density value of samples and controls. Bars represent the mean of two independent experiments. Doted lines indicate the value which comprise 75% of the proteins.

Figure 5

Quantitative evaluation of the IL-8, growth related oncogene-α (GROα), and vascular endothelial growth factor (VEGF) released by HT-29 and FaDu in 48 hours. Culture medium (DMEM plus 10% FBS) was replaced and collected after 48 hours. Protein levels were determined by ELISA. Bars represent the mean ± SEM, n = 4. Statistical significance was assessed using Students t-test;***P < .001 when compared with HT-29.

Figure 6

Protein array analysis of culture medium from HMVEC exposed to IL-1β. Culture medium was replaced by DMEM plus 10% FBS and 10 U/mL of human recombinant IL-1β, and collected after 48 hours. Controls were performed with DMEM plus 10% FBS and 10 U/mL of human recombinant IL-1β. The proteins that were analyzed are indicated in the bottom panel of Figure 4. Densitometric evaluation of samples is also shown (right panel). Values were calculated as the difference between the density value of samples and controls. Columns represent the mean of 2 independent experiments. The density value which comprise 75% of the proteins is indicated by the doted line. AU, arbitrary units.

Figure 7

Production of cytokines by HMVEC stimulated with tumour cell conditioning medium. HMVEC cultures were exposed to 48 hours-conditioning medium of HT-29 and FaDu for the indicated periods of time. Production of cytokines by HMVEC was calculated as the difference between the value of samples and the initial amount in the conditioning mediums from tumour cells. Bars represent the mean ± SEM of 5 independent experiments performed by triplicate. Statistical significance was assessed using multivariate Wilks statistic and one sided Dunnet test to compare every treatment groups with the control group; *P < .05, **P < .01 and ***P < .001.

Figure 8

(a) Adhesion of PMN to HMVEC exposed to tumour cell conditioning medium. 2 × 10⁶ DiI-labeled PMN were added to the culture wells containing confluent HMVEC previously treated overnight with none (control), HT-29 (HT-29) and FaDu (FaDu) 48 hours conditioning medium, and 10 U/mL of human recombinant IL-1β (IL-1B) and then incubated at 37°C for 30 minutes. After washing, adhered PMN were evaluated as described in the Methods section. Bars represent the mean±SEM of 7 independent experiments performed in sixtoplicate. (b) Surface expression of ICAM-1 in HMVEC exposed to tumour cell conditioning media. HMVEC previously treated overnight with none (control), HT-29 (HT-29) and FaDu (FaDu) 48 hours conditioning medium, and 10 U/mL of human recombinant IL-1β (IL-1B) were detached and analyzed for ICAM-1 expression by flow cytometry (see Methods). Upper panels represent histograms (out of 4) showing the effect of tumour cell conditioning medium on ICAM-1 expression. Bottom panels show quantitative data. Bars represent the mean ± SEM of 4 independent experiments. Statistical significance was assessed using ANOVA test; *P < .05, **P < .01, and ***P < .001 when compared to control group.