When using screws in MDF or plywood, it's always a good idea to drill a pilot hole to ensure that the screw is doing its work to hold the panels tight. Use coarse threaded screws, and don't over torque them when you put it together. If you aren't using a lot of clamps, this would work well also when using Gorilla glue to keep the joints tight while the glue cures. Titebond III works very well, and is the better route to take if you are brad nailing the panels.
Gorilla glue (and urethane glues in general) have their place. Assembling boxes is not a place where the particular properties of urethane glues is ideal (waterproofness being the primary). Titebond I or III (or the Elmer's equivalent) are far better and will make for much stronger joints.
For securing T-nuts, urethane glue will work better than a PVA glue.
Gorilla brand doesn't even perform as well as the Elmer's urethane glue. It's all marketing.
I think Elmers has a different formula now. It has a different bottle and everything so it may be fine. The last time I tried it was about 7 years ago. It seemed very weak. It would expand very quickly and dry very hard and brittle. To test I glued 2 pieces of mdf together making a butt joint, clamping until it was dried. There was about a 12" long joint glued together. When dried i could easily break it apart just by pulling on the 2 pieces by hand. With the gorilla glue the mdf would peel off on one edge and the glue itself wouldn't come apart.
As far as problems with expansion go, just remember you don't need a lot of it. Since it does expand, you can use a lot less.
John E. Janowitz Acoustic Elegance, LLC "Learn from the mistakes of others... you can't possibly live long enough to make them all yourself"
Urethane glues in general don't tend to form a good joint. It's not expansion that's the problem, it's joint strength. Is it strong enough? Of course - for speaker boxes we're not talking about tons of pressure. So it's not a "don't use this" type "not best for the application". Using gorilla glue will not be a problem. It's just expensive and you gain nothing.
Fine Woodworking published a set of tests this last year on glues (I believe it was Case Western that actually did the lab work).
Polyurethane joint strength is stronger then most materials glued using it, if glued properly.
I built a NHT 1259 based tower sub years ago using polyurethane. It had double wall construction with solid 2x2" 20 year aged red oak corners. I misrouted a side and forcibly bent one of the corners pieces to fit, in an arch, using a dozen webclamps. short story is the joint never let go after only twenty minutes of clamp time.