You have different approaches available : protein precipitation with strong acids (TCA or PCA), solvent extraction (provided proteins and molécules of interest behave differently toward the organic solvent), filtration, desalting chromatography. For the last 2 the mass difference between protein and molecules of interest has to be significative as it is the basis of separation.
yes. The techniques I mentioned are very general. In case of proteases, what you could do first is to heat inactivate the enzyme, spin the sample, recover the supernatant and then consider an other deproteinization step. If the protase is the only protein in the mixture maybe heat treatment is enough to precipitate it...