[EDIT: On rereading the question, it seems I have taken an overly pedagogic approach, underestimating the level of the asker. Sorry about that. Consider it a service to any hobby readers of the future.]
Protons can be quite precisely described as three quarks held together by the strong force as mediated by gluons. As you say a gluon can emit a quark-antiquark pair, which can annihilate into a gluon - but this hardly means that the protons "contains" these pairs. The key concept to understand is when something is "energetically favorable". A ball on a hill can win energy by rolling down to the bottom - this is an energetically favorable process.
As far as we know, protons are in an energetically favorable position - they have found a minimum in energy, and will not decay into something else unless we put energy into it. That means that you will not measure a proton suddenly emitting a meson (a quark - antiquart particle) without you investing considerable energy into it. If you make that investment, however, you can get all sorts of fun particles. This is what they are doing at LHC - knocking together protons to force the quarks out of their comfy energy minimum and into creating a bunch of exotic particles.
So no, protons do not contain anything else than three quarks in the way that a stable state of those quarks are our best model of the proton, and no are suddenly emitted from the proton. What I suspect you might be thinking of is "virtual particles". These are a little complicated. Consider a gluon travelling between two quarks, mediating their strong attraction (or confinement if you will). On the way, that gluon might decide to split into a quark - antiquark pair, which then almost immediately annihilate into a gluon which continues as if nothing happened. This is a lot less energetically favorable than the gluon just staying a gluon, but it still appears to happen every now and then. We need to include them in the model to get the correct results, and they are allowed due to quantum uncertainty. But! These particles are virtual, not real. There is no measurement which can ever observe them. In some sense, they exist more as math than as physics.
An analogy in easier to understand terms could be a light travelling through empty space as a photon. The photon has the virtual process of emitting an electron and a positron which then annihilate back into a photon which goes on as if nothing happened. This does mean that empty space is a bit more complicated than first thought, but it can hardly be interpreted as empty space being full of electrons and positrons. After all, empty space is empty.
I can recommend looking into Feynman diagrams to get a better feel for the situation. The relevant concepts are that virtual processes correspond to lines which start and stop inside the diagram and thus never escape the diagram to be measured, and the fact that the more complicated a diagram looks, the more energy it costs and thus the more rare it is.
Sorry for the long answer. I hope it helps!