This depends on the protein. Some proteins could lose function by just mutating a single residue, whereas, others can tolerate very many mutations, and can still carry out their function. So there is no general rule of thumb. It depends on individual proteins.

Hope this helps!

All the best,

Manjunatha

In order to answer the question following idea would help.

Protein structural changes caused by several reason can be classified and statistical analyses can help us to identify how much each of that changes can influence the function of proteins. It is obvious that we should started to collect a data of mutant proteins.

As mr Kogenaru said (above) it depends on the protein. Also on whic residue is replaced and how similar is the new residue, Changing valine to isoleucine will not bring as many changes as changing valine to proline for example.

Thank you for your answers.

The thing is that I was not focusing on mutations. I was thinking about having the same amino acids in the active site with only marginal changes to their position. The change of their place could be affected by either one or hundred mutations in the other part of the protein - it doesn't matter. What matters is how much can the same amino acids of active site move before the active site loses function? E.g. a half of angstrom, one angstrom, two, three...