When the working PH is far away from the PI, the protein will be highly charged. That's why I think large deviation from the PI will tend to minimize the folding, which is essential for its function. Globular proteins remain in highly folded state, so deviation from PI should have higher influence on them.
Incorrect. The "best" folding is not achieved at pI, that is for sure. By definition, which you should look up in a college level textbook, to get all the nuances, pI is the point when all native charged side groups are uncharged. This will remove all folding, except by covalent linkages. This will most likely ruin catalytic activity. Often, at pI, the protein associates best with itself, and precipitates out, as a solid, and may not even refold properly when the pH is changed back. Like wpenrose: said, you can't make predictions on optimal catalytic pH based on pI, or distance from it.
[sorry, I hit edit instead of quote - Yggdrasil]