Titanium dioxide (anatase) was purchased from Merck (Darmstadt, Germany). The coupling agents 3-(Trimethoxysilyl)-propylamine, N- [3-(Trimethoxysilyl)propyl]ethylenediamine and N1- [3-(Trimethoxysilyl)-propyl]diethylenetriamine were obtained from Merck (Darmstadt, Germany), Fluka (Buchs, Schwitzerland) and Aldrich (Steinheim, Germany). N-(3-Dimethylaminopropyl)-N'-ethylcarbodiimide Hydrochloride(EDC) was purchased from Sigma Chemical Company. N-Hydroxysuccinimide (NHS) was purchased from Merck. 2-(N-Morpholino)ethanesulfonic acid (MES) and Heparin Sodium salt were purchased from Fluka (Buchs, Schwitzerland).
Silanisation of titanium dioxide and cp-titanium
Modification of TiO2 was carried out under variation of a published preparation regime . 2.00 g TiO2 (25.0 mmol) and 2.00 ml 3-(Trimethoxysilyl)-propylamine (APMS, 11.3 mmol) were suspended under argon atmosphere in 50.0 ml anhydrous toluene following 2 h treatment with ultrasound (35 kHz, Sonorex RK 102, Fa. Bandelin, Berlin). Afterwards, the functionalised TiO2 was separated with a G4-frit from the solution, washed with toluene and then purified with ethyl acetate in a Soxleth-extractor for 2 h. Finally, the sample was dried in vacuum at room temperature. The reaction was alternatively performed with 2.46 ml N- [3-(Trimethoxysilyl) propyl]ethylenediamine (Diamino-APMS, 11.3 mmol) or 2.91 ml N1- [3-(Trimethoxysilyl)-propyl]diethylenetriamine (Triamino-APMS, 11.3 mmol).
cp-Titanium surfaces were modified in a similar manner. The substrates (d = 16 mm, h = 1 mm; Zapp, Düsseldorf) were firstly cleaned in 5% EXTRAN solution for 10 min using ultrasound following three washes with deionised water. Oxidation of the metal surface to produce a homogeneous oxide layer was either performed by using a solution of conc. H2SO4/30% H2O2 (1:1) for 3.5 h at room temperature or thermally by annealing the substrates at 750°C for 90 min in a furnace. These samples were then boiled in a solution of 7.00 ml APMS (39.7 mmol) in 70 ml toluene for 6 h.
Covalent attachment of heparin
In each case 500 mg of the functionalised TiO2 was suspended in 50.0 ml MES-buffer (50.0 mM, 40% (v/v) ethyl alcohol/water, pH = 5.5) and stirred for 15 min. Then, 95.0 mg N-(3-Dimethylaminopropyl)-N'-ethylcarbodiimide Hydrochloride (EDC, 0.50 mmol), 12.0 mg N-Hydroxysuccinimide (NHS, 0.10 mmol) and 100 mg heparin (~5.60 μmol) were added to the suspensions followed by stirring the reaction mixture at room temperature for 6 h. Finally, the solid was separated by centrifugation and the obtained modified powders were washed three times each with Na2HPO4-solution (0.10 M), NaCl-solution (2.00 M), deionised water and then dried in vacuum at 40°C over night. The modification of APMS modified titanium substrates with heparin was performed in a similar manner by stirring the metal samples in a solution of 19.0 mg EDC (0.10 mmol), 2.00 mg NHS (0.02 mmol) and 20.0 mg heparin in 50 ml MES-buffer for 6 h at room temperature.
Quantification of surface bound spacer and heparin
The quantification of the spacer was determined by the amount of terminal primary amino groups and could be measured photometrically by dint of the ninhydrin reaction after Moore and Stein [7-9]. Basically, this method is based upon the reaction of primary amino groups with ninhydrin to give the dye "Ruhemanns Purpur", which can be quantified photometrically at 570 nm. The amount of immobilised heparin was quantified by means of the toluidine-blue method, already described in literature . This method is based on complexation of the dye toluidine blue with heparin in aqueous medium. By the use of an excess of the dye, the residual "free" toluidine-blue can be determined photometrically at 631 nm and can be correlated with the amount of immobilised heparin.
Hydrolysis behaviour of immobilised heparin
Five probes with 250 mg of the TiO2 powders, respectively, modified with APMS, Di- and Triamino-APMS and heparin were given in glass flasks. Then, 5.00 ml PBS buffer was added and the suspensions were treated with ultrasound for 3 min. The flasks were densely closed with a plastic lid and then given inside an incubator (37°C) onto a shaker (70 shakes per minute). At scheduled times one probe of each heparinised samples was taken, the powder was separated from the liquid with a centrifuge and then dried in vacuum. The amount of residual immobilised heparin was quantified photometrically with the toluidine-blue method.
Determination of the biological potency of heparin
For the determination of the biological activity of the immobilised heparin, the chromogenic substrate Chromozym TH was used. The principle of this method is based upon heparin formation with antithrombin (AT-III), a heparin/AT-III-complex in an aqueous medium. In the presence of an excess of thrombin, formation of a heparin/AT-III/thrombin-complex follows. Excessive thrombin in solution catalyses the hydrolysis of the chromogenic substrate Chromozym TH into a dye, which can be measured photometrically at 405 nm . The higher the biological activity of the heparin remains after immoblisation, the more thrombin is complexed and the less dye will be hydrolysed.
Surface charge (Zeta-potential) of modified surfaces
Zeta potentials were measured with a Zeta-sizer 3000 (Malvern Instruments, Herrenberg, Germany) equipped with a HeNe-Laser (λ = 633 nm) and a standard electrophoresis cell at an electrical field force of ± 150 mV. The suspension medium was double distilled water. For data compilation the software PCS version 1.36 from Malvern Instruments, Herrenberg was used.
Fibrinogen adsorption to modified surfaces
The adsorption of fibrinogen onto the modified surfaces was measured in real-time using Quartz Crystal Microbalance with Dissipation Monitoring (QCM-D; Q-Sense, D300, Sweden). Modification of the quartz crystals was performed in three steps: firstly a thin TiO2 coating was applied using physical vapour deposition (PVD) technique  following chemical coupling of APMS, Diamino-APMS and Triamino-APMS and heparin similar to the methods described for cp-titanium samples. Fibrinogen adsorption was measured at room temperature using 5.00 ml of a solution of fibrinogen dissolved in phosphate buffered saline (PBS, 50.0 μg fibrinogen/ml PBS) for about 5 h. The adsorbed mass of fibrinogen on the differently heparinised surfaces was calculated according to the Sauerbrey equation, where the change in the resonance frequency Δf is proportional to the change in the adsorbed mass Δm:
with n being the number of the overtone (= 3) of the measured frequency and C = 17.7 ng/cm2Hz for the crystals used in this experiment .