Thirty milliliters of peripheral blood were drawn from normal, healthy volunteers into 3.2% sodium citrate vacutainers (Becton Dickinson VACUTAINER™ Systems, Rutherford, New Jersey). Whole blood diluted by adding 20 μl of blood into 180 μl 1% crystal violet stain in 0.5% acetic acid. A leukocyte count was performed in an improved Neubauer counting chamber.

Venepuncture and cell isolation for this study was performed in accordance with the University of Tennessee Institutional Review Board approval (IRB# 6788B).

sFn is made in the required amount for each experiment. To produce sFn monomer, an excess of GPRP-NH2 is added (4 mM final). To make 1 ml of sFn solution, 50 μl fibrinogen (10 mg/ml; plasminogen-, fibronectin-, and factor XIII free; American Diagnostica Inc., Greenwich, Connecticut), 84 μl of the fibrin polymerization inhibitor Gly-Pro-Arg-Pro-amide (24 mM; GPRP-NH2; Sigma Chemical Company), followed by 1.25 μl thrombin (100 U/ml; Sigma Chemical Company) were added to 865 μl of RPMI 1640.

For experiments investigating peptide inhibition of sFn binding to monocytes and tumor cells, FITC-labeled fibrinogen (Molecular Probes, Eugene, OR) was used in place of unlabeled fibrinogen above.

The A375 human amelanotic malignant melanoma cell line was maintained in continuous cell culture. Cells were detached from plastic using trypsin/EDTA (0.25%: Hyclone, Logan, Utah) and washed in RPMI 1640 tissue culture medium containing 10% fetal bovine serum (10% FBS) centrifuged at 200 × g, resuspended in 5 ml of10% FBS, and counted. For static adherence assays, 4 × 10⁴ cells were added to the wells of a 96 well flat-bottomed microtiter plate and incubated for 24–48 or until confluent. For cytotoxicity experiments, cells were incubated with 5 μl of Calcein-AM (Molecular Probes 27) for 30 min, washed, counted and used at appropriate concentrations in a 96 well round-bottomed plate, to which effector cells were added. For flow experiments, the cell suspension was added to 40 mm glass coverslips together with 10% FBS in petri dishes and incubated until nearly confluent (24–48 h).

Six tubes of citrated blood were diluted 1:1 with RPMI, and 15 ml layered over 8 ml Lymphoprep™ in plastic universal containers. These discontinuous density gradients were centrifuged for 25 min at 450 × g and the mononuclear cell interface removed using a sterile plastic pasteur pipette. The cells were washed three times by resuspension and recentrifugation at 200 × g for 10 min in RPMI/FBS so as to remove platelets. The final cell pellets were resuspended in 10 ml RPMI. A cell count was performed by adding 20 μl of blood into 180 μl 0.5% trypan blue vital stain, and cells were counted in an improved Neubauer counting chamber.

Mononuclear cells obtained from the Lymphoprep™ density gradient were adjusted to 2 × 10⁶ cells/ml, and aliquots of 5 × 10⁷ cells added to 75 cm² serum coated tissue culture flasks. The flasks were incubated at 37°C, in 5% CO₂, for 1.5 h in order to allow monocyte adherence to the plastic. The non-adherent cells (lymphocytes) were decanted. A cell count was performed on the final cell suspension an aliquot were dried onto microscope slides for analysis of cell purity. The flasks containing the remaining adherent cells (monocytes) were rinsed twice with RPMI and 10 ml of 3 mmol/L EDTA in RPMI/FBS added. The flasks were then incubated at 37°C for 12 min, or until the cells have detached from the plastic (determined by observation under an inverted microscope). The cells were removed from the flasks using a 5 ml pipette, transferred into universal containers, and washed three times in 10% FBS to remove any residual EDTA. Cell purity was determined by differential counting of lymphocytes and monocytes using May-Grunwald/Giemsa stain (Sigma). It was important that the monocytes were used in experiments immediately to avoid re-adherence to the plastic, which would result in a decreased cell yield. In 19 experiments, the mean monocyte purity was 92 ± 5%.

To produce 1 mL of sFn monomer solution, 50 μl fibrinogen (10 mg/ml; plasminogen-, fibronectin-, and factor XIII free; American Diagnostica Inc., Greenwich, Connecticut), 84 μl of the fibrin polymerization inhibitor Gly-Pro-Arg-Pro-amide (24 mM; GPRP-NH2; Sigma Chemical Company), followed by 1.25 μl thrombin (100 U/ml; Sigma Chemical Company) was added to 865 μl of RPMI 1640.

Tumor cells were detached from plastic using trypsin (0.25%)/EDTA (0.1 μM) and washed in cell culture medium. Forty thousand cells (in 200 μl 10% FBS) were added to each well of a 96 well, flat bottomed tissue culture plate and incubated at 37°C until confluent. Six wells were trypsinized and the cells mixed 1:1 with Trypan blue stain (0.5% in PBS), and counted in a hemocytometer. The mean count was recorded. Adherent cells were left untreated (controls) or incubated with sFn, fibrinogen (0.5 mg/ml), GPRP-NH2 (4 mM) or thrombin (0.125 U/ml).

After washing with RPMI, 5 μl of Calcein AM (Molecular Probes, Eugene, OR) stock solution in DMSO were added to 5 ml of effector cell preparation (Leukocytes; 2 × 10⁶ cells/ml) and incubated for 30 min at 37°C. Effector cells were left untreated (controls) or incubated with sFn or fibrinogen (0.5 mg/ml), GPRP-NH2 (4 mM) or thrombin (0.125 U/ml). Total fluorescence was determined by addition of 200 μl of fluorescently labeled effector cells to three wells and fluorescence measured. Minimal fluorescence (blank) was measured in three wells containing adherent cells to which only RPMI is added. After a further wash, cells were made to 1 × 10⁶/ml and 200 μl added to appropriate wells, and incubated at 37°C for 1 h. The plates were washed three times with RPMI followed by addition of 200 μl of 0.5 M NaOH to lyse the cells. The supernatants were removed into black 96-well microtiter plates and fluorescence was measured on a Perkin-Elmer Victor 3 plate reader. Specific adherence was determined by:

(Test - Blank/Total - blank)*100%.

Lyophilized murine anti-human monoclonal blocking antibodies directed against α_Lβ2 (clone 25.3) and CD54 (clone 84H10) were obtained from Beckman-Coulter (Miami, FL), as were purified immunoglobulin matched isotypic controls (IgG1 (mouse) Clone 679.1Mc7). Blocking monoclonal anti-α_Mβ2 (clone ICRF44) was obtained from PharMingen (BD Biosciences, San Jose, CA). All antibody inhibition experiments were performed using pre-incubation of target cells with 5 μg antibody/10⁶ cells. Experiments were performed in triplicate wells and the results were expressed as the mean ± SD for three separate experiments).

Tumor cells were detached from plastic using trypsin (0.25%)/EDTA (0.1 μM), washed and resuspended in 2 ml of culture medium and counted. Cell aliquots were untreated or pretreated as described in the previous section, depending on the experiment to be performed. The cell concentration was adjusted to 1 × 10⁵ cells/ml and 100 μl added to the wells of a round-bottomed 96 well microtiter plate. One hundred microliters of untreated or appropriately pre-incubated monocytes were added to the tumor cells in appropriate wells at an effector target cell ratios of 20:1, and the plates were centrifuged at 250 × g for 10 min and 100 μl of supernatant carefully removed into the appropriate wells of an optically clear 96 well flat bottomed plate. LDH activity was measured using a standard kit. (Roche Molecular Biochemicals, Indianapolis, Indiana). One hundred microliters of reaction mixture were added to the wells and incubated at room temperature for 30 min in the dark. The absorbance of each well was measured using a Perkin-Elmer Victor 3 plate reader equipped to measure absorbance. Specific cytotoxicity was determined by (test release- blank/total release – blank)*100%. Initial experiments were performed to determine the optimum time course and effector:target (E:T) cell ratio for monocyte cytotoxicity (data not shown). Based on these results, sFn inhibition studies were performed using and E:T ratio of 20:1 in an 18 h assay, which was consistent with other reports 2829.

To simulate the fluid shear stresses present in vivo, coverslips containing Calcein AM labeled, adherent A375 were loaded into the FCS2 stage incubator (Bioptechs Inc., Butler, PA.) and mounted on the stage of a Leica DMIRB inverted fluorescence microscope equipped with a Hamamatsu Color Chilled 3CCD camera (C5810 model) attached to a computer for data acquisition. The FCS2 incubator is a closed near-laminar flow, temperature-controlled perfusion chamber which facilitates direct observation and scanning of the cells on the coverslip during perfusion with solutions or cell suspensions. It was designed to maintain accurate thermal control and allow high-volume near-laminar flow perfusion. A fluid pathway was formed (Dimensions 0.5 mm × 14 mm × 25 mm) by separating the microaqueduct slide from the coverslip containing cells with a single silicone gasket with a rectangular bore, generating near-laminar flow conditions during perfusion. The stage incubator was initially connected to a peristaltic pump on the afferent side using 0.062 inch bore S/P medical grade silicone tubing (Fisher Scientific, Suwanee, GA), with a three way in-line tap closed to the incubator inlet. The efferent side was connected to waste. After connection of the electronic temperature control, the coverslips were perfused with RPMI to briefly buffer the cells at 37°C for fifteen minutes (flow rate of 0.5 ml/min). The inlet was then connected to the tube containing the appropriately treated monocyte suspension (3 mL). The efferent tubing was also connected to the same tube, and Di-I 30 labelled monocytes were recirculated across the tumor cells for 1 h, after which, the tubing was set up to the initial configuration and the cells were again washed with RPMI for 10 min to remove unbound monocytes. Individual still images in five randomly chosen fields of view were captured and stored on the computer. The number of monocytes (red) and tumor cells (green) were counted in each field of view and the mean monocyte/tumor cell ratio was calculated.

The wall shear stress was calculated using the momentum balance for a Newtonian fluid. The viscosity of water at 37°C was used as an approximation of the viscosity of the buffer used (RPMI 1640; 0.007 poise). As described in the following equation, the wall shear stress experienced by the monolayer is equal to: T = 3μQ/2a²b, where T=wall shear stress, μ= coefficient of viscosity (0.007 Poise), Q= volumetric flow rate (0.0083 cm³/s), a= half channel height (0.025 cm), and b=channel width (1.5 cm). The wall shear rate is given by T/μ. Experiments were conducted to maximize adherence in the model, within physiologically relevant shear rates (In post-capillary venules, shear rate has been reported to range between 35 and 560 s^-1 3132. This range is believed to be characteristic of the stresses that a leukocyte must resist to form a stable adhesion with a vessel wall).

Based on the reported activities of the various fibrin(ogen) blocking peptides, four were selected which would possibly also inhibit fibrin mediated immunosuppression. Table 1 shows our laboratory designation, the amino acid sequence, its molecule of origin, and which molecule it binds to.

Experiments were performed to determine if the specific fibrin(ogen) adherence inhibitors also affected 1. fibrinogen clotting in the presence of thrombin, or 2. normal coagulation in re-calcified plasma.

1. Fibrinogen (0.5 mg/ml, 0.05 mmol/L), and thrombin (0.125 U/ml) were used at the concentration demonstrated to induce maximal inhibition of adherence and cytotoxicity in microplate assays. Thrombin (0.625 μl (stock 100 U/ml) was added to RPMI 1640 medium containing 25 μl (stock 10 mg/ml) thrombin alone or in the presence of 4 mmol/L specific peptide or peptide combination in a total volume of 1 ml. The tubes were left at room temperature and clotting was determined by observed gelation of the solution every 30 s, and the approximate clotting time was recorded. An additional control was also tested in which the fibrin polymerization inhibitor GPRP-NH2 was added to fibrinogen in the presence of P1-P4 prior to thrombin addition. The tube was left at room temperature to determine if clotting was observed.

2. The effect of the specific peptides on the clotting of re-calcified normal human plasma was also examined to determine if the normal coagulation cascade was inhibited in their presence. Plasma alone or in the presence of single or combined peptides P1-P4 (4 mM final concentration: 20 μl of 100 mM stock) was re-calcified with CaCl₂ (50 μL; 25 mM stock) in a final volume of 500 μL. Tubes were left at room temperature and clotting time was then recorded.

After fluorescence labeling of each cell type, stock peptides were added to both confluent coverslips and/or 1 milliliter suspensions of monocytes (P1 and P2) and/or sFn solution (P3 and P4) for a final treatment concentration of 4 mM (40uL added to 1 mL suspension). Cells were treated with peptides for 20 minutes at room temperature, followed directly by washing of cells (by centrifugation and resuspension – monocytes, or perfusion with RPMI 1640 for 10 min – tumor cells). For sFn peptide treatment, peptides remained in the sFn solution during incubation. Monocyte adherence to tumor cells under flow conditions was then performed as described above.

Oregon Green labeled fibrinogen was obtained from Molecular Probes (Oregon Green 488 human fibrinogen conjugate (F-7496)). Lyophilized stock fibrinogen (5 mg) was made up in RPMI 1640 to a total volume of 5 ml, aliquotted (25 μl) and stored at -20°C until time of experiment. For experiments, labeled sFn was prepared as described for sFn above, and incubated for 20 minutes at 37°C to allow sFn formation.

Stock peptides (P1 and P2) were added to both confluent coverslips containing unlabeled A375 cells and/or 1 milliliter suspensions of unlabeled monocytes and/or labeled sFn (P3 and P4) at a final concentration of 4 mM, for 20 minutes at room temperature, followed by washing in RPMI 1640. Peptides were retained in the sFn solution during adherence. Untreated or peptide pre-treated monocytes and/or tumor cells were incubated for 20 min with labeled-sFn (+/- peptide treatment as appropriate), and washed in RPMI 1640. The monocyte pellet was resuspended in 1 ml of phenol red free RPMI 1640 and 10 μl added to a microscope slide, coverslipped and sealed for microscopy. Coverslips containing tumor cells were inverted onto a microscope slide together with 50 μl of Pro-long Gold™ (Molecular Probes) mounting medium, and allowed to harden for microscopy. An Olympus BX61 was used (100× oil immersion objective) to observe samples and capture images. Using the same microscope settings to observe differences in fluorescence between samples, five representative fields were captured of each slide for each pre-treatment. Images were processed using Image Pro Plus™ (Media Cybernetics, Silver Spring, MD) and deconvolved using AutoDeblur™ deconvolution software (Media Cybernetics, Silver Spring, MD).

In static microplate adherence and cytotoxicity assays (Figures 2, 3 and 6) each data point was obtained as the mean +/- SD for three replicates. Each experiment was performed at least three times. Student's t-test for independent (unpaired) variables was used to determine significant differences between groups.

For experiments investigating adherence under flow conditions (Figures 7 and 8), each experiment used monocytes from a single individual and the experiment was internally controlled on each occasion by inclusion of untreated and sFn/sFn treated cells. Each experiment was performed at least five times. Inhibition of adherence compared to the untreated control, as well as peptide blocking of sFn inhibition was determined by direct comparison with its own internal control in each experimental protocol. Thus, the number of untreated and sFn treated controls was greater than the number of tests for each single or combined peptide. Since there was also variation in monocyte adherence from individual donors, each peptide effect was compared to the whole population of either untreated or sFn treated control, using Student's t-test for independent (unpaired) variables.

Experiments were performed to determine the effect of sFn pre-incubation of either monocytes, tumor cells, or both on monocyte adherence to A375 melanoma cells in static microtiter plate assays (Figure 2). In the absence of sFn, monocytes adherence was 32.9 ± 1.3%. Pre-treatment of tumor cells with sFn significantly (P < 0.01 compared to untreated control) increased adherence (68.5 ± 0.7%). Addition of sFn treated monocytes to untreated tumor cells also increased adherence (38.75 ± 1%; P < 0.05 compared to untreated control), but to a significantly lower degree than tumor cell pretreatment with sFn (P < 0.01 compared to sFn treated tumor cells). However, when both monocytes and tumor cells were both pre-treated with sFn a pronounced inhibition of adherence was observed (15.95 + 1%; P < 0.01 compared to untreated control).

Monocyte cytotoxicity against A375 melanoma cells was measured in a static microplate assay (Figure 3). At an effector:target cell ratio of 20:1 monocyte cytotoxicity was 28.6 + 0.7%. Pre-treatment of A375 cells with sFn did not significantly (P >0.05) affect cytotoxicity (27.1 + 0.7%) compared to the untreated control, but a significant reduction (P <0.05) was observed when monocytes were sFn treated (20 + 0.7%). Maximal inhibition of cytotoxicity compared to untreated cells was observed when both effector and target cells were sFn treated (17.3 + 0.3%; P < 0.01 compared to Untreated control, sFn treated monocytes and to sFn treated A375 cells; n = 3).

Initial experiments were conducted to maximize adherence in the model, within physiologically relevant shear rates. Figure 4 shows the percentage adherence (number of monocytes/field/number of tumor cells/field) at flow rates of 0.5 – 2.5 ml/min. Maximal monocyte adherence was observed at a flow rate of 0.5 ml/min corresponding to a shear stress of 0.93 dynes/cm², with a corresponding shear rate of 132.9 s^-1. This shear rate was within the range of the stresses (35 and 560 s^-1) 3132 which a leukocyte must resist to form a stable adhesion with a postcapillary vessel wall. Experiments investigating sFn inhibition of adherence were performed in further experiments at a flow rate of 0.5 ml/min. Figure 5 shows two representative images of monocytes (red) adherent to tumor cells (green). In Figure 5A untreated monocytes were adhered to untreated tumor cells. Less adherence was observed when both cells were pre-treated with sFn (B).

Pre-treatment of monocytes with monoclonal anti-α_Lβ2 antibody (5 μg/10⁶cells) inhibited their adherence to untreated A375 cells by 80 ± 7%. (Figure 6), whereas only 23 ± 0.3% inhibition was observed when A375 cells were also pre-treated with sFn (P < 0.05 compared to untreated A375 cells). Conversely, monocyte pre-treatment with anti-α_Mβ2 monoclonal antibody only inhibited adherence by 22.5 ± 2.6% to untreated A375 cells, and by 80 ± 0.5% to sFn pre-treated tumor cells (P < 0.05 compared to untreated A375 cells). Incubation of A375 cells with anti-CD54 inhibited monocyte adherence of untreated monocytes by 53 ± 0.3%, and sFn pre-treated monocytes by 86 ± 0.2%, which were significantly greater (P < 0.05) compared to untreated monocytes). Pre-incubation of either cell type with appropriate isotypic control or an irrelevant antibody (CD4) did not significantly (P > 0.05) inhibit adherence.

Experiments were performed to determine if fibrin(ogen) blocking peptides affected the ability of fibrinogen to clot. Using purified fibrinogen in the absence or presence of P1 + P2, P3 + P4, or all four peptides together, clotting was observed within approximately 5 minutes. Addition of GPRP-NH2 to purified fibrinogen prior to thrombin addition failed to clot within 15 minutes, indicating the production of soluble, rather than polymerized fibrin. Using no treatment, or the same combinations of peptides, clotting of recalcified plasma was observed under all conditions in approximately 4 minutes.

Figure 7 shows the effect of two specific peptides, designated P1 and P2 on sFn inhibition of monocyte adherence to A375 melanoma cells in a flowing microscope stage incubator. P1 represents the sFn major binding site for CD54 and P2 represents the sFn major binding site for α_Mβ2. From the left, the first bar (+/+) shows the mean (+SD) inhibition when both monocytes and tumor cells were pre-incubated for 20 min with sFn. SFn considerably reduced monocyte adherence (P < 0.05 compared to untreated cells). When both monocytes (which express α_Mβ2 and some CD54) and tumor cells were pre-treated with both P1 and P2 followed by sFn prior to assay (Bar2 +P1P2/+P1/P2) sFn inhibition was considerably reduced to a level which was not significantly different to the untreated control, as was also observed when cells were treated with peptides in the absence of sFn (Bar3; P1P2/P1/P2). These results suggest that by blocking the major receptor sites for sFn on α Mβ2 and CD54 binding of sFn is blocked, allowing α Lβ2 (LFA1 which does not bind to sFn) and α_Mβ2 to bind to CD54 restoring adherence. If A375 cells are treated with P1 to block CD54 binding to fibrin (even when sFn was added subsequently) and monocytes are treated with sFn, inhibition was still observed (Bar4; +P1/+) since CD54 was blocked by P1. Similarly, inhibition was still observed when monocytes were treated with P2 to block sFn binding to α Mβ2, and tumor cells were treated with sFn (Bar 5; +/+P2). As controls, tumor cells were incubated with P1 (but not sFn) and monocytes were treated with sFn. Inhibition was still observed because P1 blocked sFn coated α Mβ2 binding (Bar 6; P1/+). Similarly, inhibition was observed when P2 blocked sFn binding to monocytes and tumor cells were treated with sFn (Bar 7; +/P2). The final 3 bars (Fg/Fg, thr/thr, GPRP-NH2/GPRP-NH2) showed that no significant (P > 0.05 compared to untreated cell adherence) inhibition was observed when cells were treated with soluble fibrinogen, thrombin, or GPRP-NH2). These results demonstrate the peptides P1 and P2 are effective in inhibiting sFn inhibition of monocyte adherence by a mechanism involving its blocking of CD54 to α_Mβ2.

Figure 8 shows the effect of two specific peptides, designated P3 and P4 on sFn inhibition of monocyte adherence to A375 melanoma cells in a flowing microscope stage incubator. P3 represents the CD54 major binding site for sFn and P4 represents the α Mβ2 major binding site for sFn. From the left, the first bar (+/+) shows the mean (+SD) when both monocytes and tumor cells were pre-incubated for 20 min with sFn. SFn considerably reduced monocyte adherence (P < 0.05 compared to untreated cells). Pre-treatment with sFn with P3 and P4 prior to its incubation with tumor cells and monocytes (which should block its binding) resulted in a marked increase in cell adherence, which was not significantly different than the untreated control (Bar 2; +P3P4/+P3P4;P < 0.05), thus reversing sFn inhibition of adherence. As expected, addition of peptides P3 and P4 to cells did not significantly affect adherence (Bar 3; -P3P4/-P3P4; P > 0.05 compared to untreated). When sFn was treated with P3 and incubated with tumor cells, and monocytes were untreated, little or no inhibition was observed, because α Mβ2 could still bind to CD54 directly (Bar4; +P3/-). Similarly, when sFn was pre-treated with P4 and incubated with monocytes, and tumor cells were untreated, no inhibition was observed, because tumor cells would not bind P3-sFn, and the sFn on the monocytes could bind to CD54 on the tumor cells (Bar 6; +P3/+). Similarly, when sFn was pre-treated with P4 and incubated with monocytes, and tumor cells were incubated with sFn, no inhibition was observed, because the free α Mβ2 could still bind to sFn on the tumor cells (Bar 7; +/+P4).

These results show that blocking of sFn with peptides representing its major CD54 and α Mβ2 receptor sites effectively inhibited its ability to block adherence of monocytes to tumor cells.

Having demonstrated that specific sFn blocking peptides reversed sFn inhibition of monocyte/tumor cells adherence, experiments using fluorescently labeled fibrinogen to prepare sFn were performed to observe whether sFn binding to cells was also decreased. After sFn binding (+/-) peptide pre-treatment, slides were observed on an Olympus BX61 fluorescence microscope equipped with an Olympus (COOL-1300QS) digital camera for image acquisition. Oregon Green fluorescence was detected using a 535 nm long-pass dichroic filter. After setting up the microscope to detect tumor cells or monocytes, all settings were kept identical for subsequent slides. Figure 9 shows binding of FITC-sFn to A375 cells (A). Fluorescent sFn bound strongly to both the cell types. However, when tumor cells were pre-treated with P1 + P3 (B), or sFn was pre-treated with P3 + P4 (C), almost no binding was observed. Similarly, sFn bound strongly to monocytes (D). Monocyte Pre-incubation with P1 + P2 considerably reduced sFn fluorescence, as did sFn pre-incubation with P3 + P4, demonstrating that the blocking peptides cause reduced sFn binding to cells.