It's not so easy to answer that question, and the last I knew it's only been done unambiguously via NMR for a few enzyme systems. Most crystal structures are fits to e- density and refinement with a program like X-PLOR, and the His sidechain rotation can be subject to both the forcefield and uncertainities in the e- density. Still, it's not unreasonable to assume the crystal structure is correct, and then determine the protonation states by inspection. If either NE or ND of the His ring are suitably positioned to make a good hydrogen bond if protonated, then place the proton on that N atom.