Soluble fibrin inhibits monocyte adherence and cytotoxicity against tumor cells: implications for cancer metastasis

Abstract

Background: Soluble fibrin (sFn) is a marker for disseminated intravascular coagulation and may have prognostic significance, especially in metastasis. However, a role for sFn in the etiology of metastatic cancer growth has not been extensively studied. We have reported that sFn cross-linked platelet binding to tumor cells via the major platelet fibrin receptor alphaIIb beta3, and tumor cell CD54 (ICAM-1), which is the receptor for two of the leukocyte beta2 integrins (alphaL beta2 and aM beta2). We hypothesized that sFn may also affect leukocyte adherence, recognition, and killing of tumor cells. Furthermore, in a rat experimental metastasis model sFn pre-treatment of tumor cells enhanced metastasis by over 60% compared to untreated cells. Other studies have shown that fibrin(ogen) binds to the monocyte integrin alphaM beta2. This study therefore sought to investigate the effect of sFn on beta2 integrin mediated monocyte adherence and killing of tumor cells.

Methods: The role of sFn in monocyte adherence and cytotoxicity against tumor cells was initially studied using static microplate adherence and cytotoxicity assays, and under physiologically relevant flow conditions in a microscope perfusion incubator system. Blocking studies were performed using monoclonal antibodies specific for beta2 integrins and CD54, and specific peptides which inhibit sFn binding to these receptors.

Results: Enhancement of monocyte/tumor cell adherence was observed when only one cell type was bound to sFn, but profound inhibition was observed when sFn was bound to both monocytes and tumor cells. This effect was also reflected in the pattern of monocyte cytotoxicity. Studies using monoclonal blocking antibodies and specific blocking peptides (which did not affect normal coagulation) showed that the predominant mechanism of fibrin inhibition is via its binding to alphaM beta2 on monocytes, and to CD54 on both leukocytes and tumor cells.

Conclusion: sFn inhibits monocyte adherence and cytotoxicity of tumor cells by blocking alphaL beta2 and alphaM beta2 binding to tumor cell CD54. These results demonstrate that sFn is immunosuppressive and may be directly involved in the etiology of metastasis. Use of specific peptides also inhibited this effect without affecting coagulation, suggesting their possible use as novel therapeutic agents in cancer metastasis.

Figure 1

Schematic diagram showing the amino acid sequences, sites of origin and effector molecules for four peptides (designated P1 – P4) reported to inhibit fibrin(ogen) binding to α_Mβ2 (orange) and CD54 (blue).

Figure 2

Effect of soluble fibrin on monocyte adherence to tumor cells. Calcein AM labeled PBM incubated with A375 cells after pre-treatment of A375 and/or PBM with RPMI or sFn prior to assay. sFn pre-treatment of tumor cells significantly increased adherence to untreated A375 cells compared to the untreated control (P < 0.01: n = 3). Preincubation of monocytes also marginally increased adherence to untreated A375 cells (P < 0.05: n = 3) compared to the untreated control, but to a significantly lower degree than with sFn treated A375 cells (P < 0.01 compared to monocyte sFn). sFn pre-treatment of both effector and target cells resulted in a significant inhibition of adherence (P < 0.05: n = 3) compared to the untreated control.

Figure 3

Effect of sFn pre-treatment on monocyte cytotoxicity against tumor cells. Calcein AM labeled PBM incubated with A375 cells after pre-treatment of A375 with or without sFn and pre-treatment of PBM with or without sFn prior to assay. sFn pre-treatment of monocytes was slightly inhibitory (P <0.05 compared to untreated control; n = 3). Significantly greater inhibition was observed when A375 cells were sFn pre-treated (P <0.01 compared to untreated and to monocyte treated cells; n = 3). Maximal inhibition of PBM cytotoxic activity occurred when both cell types were treated with sFn (P < 0.01 compared to untreated, monocyte treated or A375 treated cells: n = 3).

Figure 6

Effect of monoclonal anti-α_Lβ2, -α_Mβ2 and CD54 on monocyte adherence to tumor cells under flow conditions in the presence or absence of sFn. Anti-α_Lβ2 inhibited monocyte adherence to untreated cells (P <0.05; n = 3), but was significantly (P < 0.01 compared to non-sFn treated cells; n = 3) less effective in blocking monocyte binding to sFn pre-treated tumor cells. Conversely, anti-α_Mβ2 inhibited monocyte adherence to sFn pre-treated tumor cells to a significantly (P < 0.01) greater extent than to untreated tumor cells. Anti-CD54 inhibited monocyte adherence to untreated tumor cells by over 50%, and by over 80% when tumor cells were pre-incubated with sFn. Isotypic control IgGs or an irrelevant monoclonal antibody (CD4) did not affect monocyte adherence.

Figure 7

Effect of specific blocking peptides designated P1 (binds to CD54) and P2 (binds to α_Mβ2) on sFn inhibition of monocyte/tumor cell adherence under flow conditions. (from left to right): sFn pre-treatment of monocytes and A375 cells (n = 25) significantly (P < 0.01 compared to untreated control; n = 30) inhibited monocyte adherence. Pretreatment of cells with P1 and P2 restored cell adherence to levels not significantly different to that of the untreated control (P < 0.05 to fibrin (n = 10); P > 0.05 to control). Pre-treatment of tumor cells with P1 and monocytes with sFn or tumor cells with sFn and monocytes with P2 inhibited adherence to a similar level to that of sFn treatment of both cells (P > 0.05 to fibrin; n = 5 in each case). Pretreatment of effector and tumor cells with fibrinogen (Fg), thrombin or GPRP did not significantly inhibit adherence (P > 0.05 compared to untreated cells; n = 5)

Figure 8

Effect of specific blocking peptides designated P3 (binds CD54 binding site on sFn) and P4 (binds to α_Mβ2 binding site on sFn) on sFn inhibition of monocyte/tumor cell adherence under flow conditions. (from left to right): sFn treatment of monocytes and A375 cells (n = 25) significantly (P < 0.01 compared to untreated control; n = 10) inhibited cell adherence. Pretreatment of sFn with P3 and P4 restored cell adherence to levels not significantly different to the untreated control (P < 0.05 to fibrin; n = 5; P > 0.05 to control). Pre-treatment of tumor cells and/or monocytes with either P3+P4, P3 alone or P4 alone were not inhibitory (P > 0.05; n = 5 in each case, compared to untreated control).

Figure 4

Effect of perfusion flow rate on monocyte adherence to tumor cells. Monocytes (1 × 10⁶/ml) were perfused across a monolayer of A375 cells attached to a coverslip in a perfusion stage incubator for 1 h at 37°C, and non-adherent cells were washed off by perfusion to waste for 10 min. Monocyte adherence was maximal at a flow rate of 0.5 ml/min, and linearly decreased as the flow rate was increased to 1, 1.5, and 2 ml/min.

Figure 5

Tumor cells were grown on 40 mm coverslips to confluence and labeled with Calcein AM (green). Monocytes were labeled with DiI (C18; red), made to 1 × 10⁶ cells/ml and continuously perfused across the tumor cells at 0.5 ml/min for 1 h. (Left) monocyte adherence to untreated A375 cells. (Right) monocyte (pre-treated with sFn) adherence to sFn pre-treated A375 cells. sFn pre-treatment of both effector and target cells inhibited monocyte adherence.

Figure 9

Oregon Green labeled fibrinogen (0.5 mg/ml; Molecular Probes, Eugene, OR) was treated with thrombin (1.25 U) in the presence of 4 mM GPRP-NH2 to produce fluorescently labeled sFn. A375 cells were incubated with labeled sFn for 20 min in a Bioptechs FCS2 enclosed stage incubator. The residual sFn was washed away by perfusion and the cells were imaged on an Olympus BX61 fluorescence microscope equipped with a long pass 535 nm dichroic filter. Considerable binding of sFn was observed. A is a representative image showing tumor cell sFn binding. In contrast, little or no binding was observed when cells were pre-incubated with peptides P1 + P2 (B), or sFn with P3 + P4 (C). Similarly, sFn bound readily to monocytes (D), but was inhibited when cells were pre-incubated with P1 + P2 (E), or sFn was pre-treated with P3 + P4 (F).

Figure 10

Schematic diagram summarizing the proposed mechanism of sFn mediated inhibition of monocyte adherence (and consequently cytotoxicity) to tumor cells. A. α_Lβ2 (orange) binds preferentially compared with α_Mβ2 (blue) to tumor cell CD54. B. Pre-treatment of tumor cells with sFn inhibits α_Lβ2 (α_Lβ2 does not bind fibrin(ogen) binding, but enhances α_Mβ2 mediated adherence). C. Pre-treatment of monocytes with sFn (sFn binds to α_Mβ2) allows adherence by both α_Lβ2 and sFn bound α_Mβ2 to tumor cell CD54. D. Pre-incubation of monocytes and tumor cells with sFn inhibits both α_Lβ2 and α_Mβ2 binding to sFn coated CD54.