It looks like I see what the problem is. You probably don't quite understand that in quantum chemistry you cannot fix charge on any atom (unlike molecular mechanics). When you remove (or add) an electron from one atom of the entire system, the electron density redistributes in a uniform way. So, you may speak only about overall charge of the system. You may further decompose the resulting electron density into atomic contributions using some procedure (say, Mulliken, Lowdin, NBO, or Bader), but these will be partial charges. Imagine communicating vessels: you cannot take some water from one of them while leaving the water level in the other one intact; the levels immediately become even. Think of electron density as some sort of liquid in a vessel made of nuclear potential.