I would put all mid/side tracks into Pro Tools as stereo tracks, with mid on the "left" channel and side on the "right". That way all editing will be sure to apply to both the mid and side signals identically. Then you can put in a decoder plug-in or send them to a bus for "manual" decoding. Or you could break them out again after editing.
Another approach would be to create an edit group for each mid/side pair of tracks and make sure you have the parameters set correctly that you want to be grouped when edited. This would seem to be the most professional way to do it, IMHO, but not necessarily the easiest.
Finally, you could decode and render/print the mid/side raw audio to stereo tracks before editing and processing them, but then you lose the mixing ability of mid versus side balance. This last option would be the best one if you only recorded mid/side for the purposes of recording based on the source, environment, and/or mics available, and you weren't hoping to have the mid/side mixing advantages or making mid-only or mono-compatible stems.
Regarding plug-ins, you might want to have a decoder inserted first or "manually" decode the m/s before stereo plugins where you want to take advantage of the stereo action of the plug-in (e.g., reverb and delay). For dynamics processing you can often get away with treating it like a stereo pair. You can also get some interesting effects by separately processing the mid and side tracks when it comes to EQ and dynamics. Really, it depends on what you're trying to do.