The short answer to your question is that the energy release in nuclear fission or fusion is a larger fraction of the mass of the fuel than in chemical processes because the binding forces (and hence the binding energies) between the protons and neutrons in nuclei are much larger than those for atoms in a molecule or solid.

To understand this first let's look at a coal-burning power plant (that is for combustion or for any other energy-releasing chemical reaction). The usual rule taught in chemistry is that mass is conserved. The precise version of this statement is that the sum of the masses of the atoms is the same before and after any chemical process, since atoms are not created or destroyed in chemical processes. But every stable molecule has a mass that is a tiny bit less than the sum of the masses of the atoms it contains. It is less by an amount (binding energy)/c-squared, where c is the speed of light (m=E/c2 is just another way of writing E=mc2).

The energy released in any chemical combustion process is just the difference in binding energy between the molecules present before burning and those present after the burning. The typical binding energy of a molecule is a few parts in a billion of the mass-energy (mass times c-squared) of the molecule. (That's why we never see a measurable mass change in chemical reactions, our chemistry lab experiments never have a balance that's accurate to that level.)

Respiration results in chemical combustion process- if energy is released in a chemical reaction, then the mass of the product MUST be less than that of the sum of the original components.

A twisted example I call "Schroedinger's Cat's Revenge": Suppose you had a human in a box with a precisely measured amount of oxygen and the only thing allowed to leave the box was the infrared radiation from the human's body heat- NOTHING is allowed to enter the box. Humans radiate ~100 Watts of heat. When the unfortunate human dies from asphixiation, will the box weigh more or less than it did when the Human was placed in it? Of course the box will contain less mass, since energy was lost... If SDAI doubts this, I would be happy to arrange a demonstration if he is the volunteer for the box. ;-)

Chemical reactions change mass... But every stable molecule has a mass that is a tiny bit less than the sum of the masses of the atoms it contains.

The same holds true for nuclear reactions...

The atomic mass of an isotope is the relative mass of the isotope, scaled with carbon-12 as exactly 12. No other isotopes have whole number masses due to the different mass of neutrons and protons, as well as loss/gain of mass to binding energy. However, since mass defect due to binding energy is minimal compared to the mass of a nucleon, rounding the atomic mass of an isotope tells you the total nucleon count. Neutron count can then be derived by subtracting the atomic number.

http://en.wikipedia.org/wiki/Atomic_weight

This change in mass due to binding energy is minimal in chemical reactions, but is there...