Chemical coupling of silane spacers and heparin to TiO2 was followed by measuring the zeta-potentials from the unmodified, with spacer functionalised or rather heparinised powders (Figure 2). Unmodified TiO2 had a negative zeta potential of -26.1 ± 10.5 mV due to its deprotonated hydroxyl groups at pH 7 in water. After modification with the different spacer molecules, the zeta-potentials turned positive to + 44.1 ± 4.4 mV (APMS), + 40.6 ± 6.2 mV (Diamino-APMS) and 45.3 ± 5.5 mV (Triamino-APMS), due to protonated primary amino groups (RNH3 +) of the spacer molecules in water. Finally, the covalent attachment of heparin changes the zeta-potentials to negative values of -37.2 ± 0.9 mV (APMS), -39.3 ± 1.0 mV (Diamino-APMS) and 39.0 ± 0.2 mV (Triamino-APMS). The strong decrease of the zeta-potentials is caused by negative sulphate and carboxyl groups of the drug in the aqueous medium at pH 7.
The amount of covalently attached primary amino groups on the titania surface (powder and metal sheets) was determined photometrically by the ninhydrin reaction and is shown in Table 1. The amount of immobilised terminal amino groups on TiO2 powder was the highest for APMS and decreased for Di- and Triamino-APMS.
Quantification of immobilised heparin was done by the toluidine-blue method and was only applicable for the drug modified powders (Table 2
). On the Ti discs, no heparin was detectable by this method, probably because the amount of bound drug was less than the detection limit of 10 μ
g by this method. On the APMS spacer, the greatest amount of 53.3 ± 3.6 ng/cm2 heparin was attached compared to Di- and Triamino-APMS (Table 2
The potency of the immobilised heparin was measured photometrically by dint of the chromogenic substrate Chromozym TH at λ = 405 nm. Comparing the potencies of the covalently attached heparin on Di- and Triamino-APMS with APMS, it becomes obviously that longer spacer molecules tend to result in higher biological activity of heparin (Figure 3). The influence of the secondary amino groups on the immobilisation of heparin was investigated by measuring the hydrolysis rate of the drug under in vitro simulated physiological conditions (37°C, PBS, 70 shakes per minute). Figure 4 shows that spacers like Triamino-APMS with many secondary amino groups has the highest hydrolysis rate of the drug during the first 100 h, compared to Diamino-APMS and APMS.
Figure 5 shows changes of the frequency Δf and the dissipation ΔD for the various surfaces during QCM-D measurement. The addition of fibrinogen dissolved in PBS buffer occurred after 5 min and resulted in a decrease of Δf and an increase of ΔD. Since Δf correlates with a mass increase of the crystal due the adsorbed protein layer and ΔD depends on the viscoelastic properties of the layer, the results demonstrate that the amount of adsorbed fibrinogen varied in the range APMS > Diamino-APMS > Triamino-APMS for the differently heparinised surfaces. The amount of adsorbed protein was calculated to be 539 ng/cm2 (APMS), 476 ng/cm2 (Diamino-APMS) and 382 ng/cm2 (Triamino-APMS) after 5 h adsorption time using the Sauerbrey equation. The adsorbed protein layers showed different viscoelastic properties as seen by the dissipation curves, the layer viscosity increased in the range APMS