This equation is incredibly generic and describes many phenomena outside of nuclear phenomena.
For example: place the the following setup in a box (a spring and some bars) and weigh them:
.
Now, loosen the spring and repeat. You should measure a smaller mass because you've removed some of the energy.
In reality you couldn't possibly measure the difference in this experiment. The energy content of a typical spring is in the millijoules, so the mass difference is $E/c^2 = 10^{-20}\;\textrm{kg}$. That's the mass of a virus. However, in atomic system, such a difference is (barely) measurable.
Mass of vacuum
Now, for a much weirder example: the vacuum state. It's well known from quantum mechanics that the energy of a standing wave, such as a spring or photon in a box, has a minimum zero-point energy. An empty box is filled with many possible photon modes, even if it has no photons in it. Each of these modes has some energy. If that doesn't make any sense, that's okay, here's my point: quantum mechanics predicts that an empty box has energy.
That means an empty box has mass. Is it a lot? Well, let's consider only photons. I want to calculate the total energy of all the modes in a box, so I integrate the number of standing waves per energy times the energy.
$$ E_\textrm{zero} = \int \rho(E) E \;dE = \frac{8\pi V}{(hc)^3} \int E^3\;dE $$
Now, this is the kind of integral you never want to see because it's badly divergent. Physicists usually try to bluff their way out of these problems by saying: "I know how physics works at low energy, but I have no idea what happens at very high energy. So, I'll just integrate until I get to the part where something else happens." So, we integrate up to $E_\textrm{cutoff}$ and find
$E_\textrm{zero} = (2\pi V E_\textrm{cutoff}^4)/(hc)^3$. What do we use as a cutoff? An optimist says that we understand everything up to the Planck energy of $10^{28}\;\textrm{eV}$. A pessimist says we've got everything up to the LHC, which is $10^{14}\;\textrm{eV}$.
$$ E_\textrm{zero} = \left\{ \begin{array}{ll} 10^{111}\;\textrm{J/m}^3 & \textrm{optimist} \\ 10^{55}\;\textrm{J/m}^3 & \textrm{pessimist} \end{array}\right.$$
These are some very large numbers. Note that the optimist and pessimist disagree by 56 orders of magnitude! Do they have meaning? Well, yes and no.
- Yes, when we look at differences. If there is an energy density, as shown here, there is also a force. You would expect that a box would feel a force inward. This has been observe and is called the Casimir Force. It turns out that the cutoff energy we used does not matter.
- No, when we consider the actual value. If that's the energy density of the box, then the mass density is $10^{38} - 10^{94} \;\textrm{kg/m}^3$. (We're finally back to your question!) We know from astronomical observations of the universe's expansion that the actual mass density is something around $10^{-26} \;\textrm{kg/m}^3$. So, we're off by something around 64 to 120 orders of magnitude. This is an unbelievably large error (and error bar!).
So, instead of telling you why $E=mc^2$ is useful, I've shown you an example where it's 120 orders of magnitude off, which happens when you compare a prediction from quantum mechanics with observations from cosmology. I hope that's more interesting.