From the RHS of the every FD given for the relation, it can be observed that we can find C,D and E attribute values using the FD. But Remaining attributes AB will not derived using any of the FD, hence these should be part of the CK. An attribute set which does not contain AB will not be candidate key, hence no need to check for (AC)+ and others you have mentioned.
For any attribute or attribute set , if we can able to get all the attributes by applying closure set of attributes, then that attribute or attribute set will become CK.
