In this work, a purified acid urease preparation was covalently immobilised onto porous chitosan beads of different size. The covalent binding method was found to be more efficient than the adsorption cross-linkage one whatever the glutaraldehyde-to-chitosan bead ratio (Y GA/CHI ) used. At the optimal Y GA/CHI ratio of 0.625gg −1 , the specific activity (A Bi ) of the biocatalysts decreased from circa 300 to 70IUg −1 wet support, as the bead average diameter (d P ) increased from 0.14 to 2.2mm. Generally, A Bi reduced less than 5% after preservation in the wet form at 4°C for 150–170 days. Only the biocatalyst prepared using the Chitopearl BCW-3001 lost about 40% of its initial activity.The kinetics of urea degradation in a model wine solution using these biocatalysts was of the pseudo-first order with respect to the urea concentration in the liquid bulk, the apparent pseudo-first order kinetic rate constant (k Ii ) ranging from about two thirds to one fifth of that (k If ) pertaining to free acid urease. In the operating conditions tested, the reaction kinetics was estimated as unaffected by the contribution of the external film and intraparticle diffusion mass-transfer resistances. When the model wine solution was enriched with the high-inhibitory tannins extracted from grape seeds, at the maximum level tested (374±2gGAEm −3 ) k Ii reduced to no more than (58±9)% of k If , this proving quite a higher protective action against such compounds for the chitosan-based biocatalysts towards free or Eupergit ® C 250 L-immobilised acid urease.