The toolbox comes along with some very nice and informative help files. For every solver included you can find the problem formulation and the details of the specific inputs. They are fairly explanatory, You will need to write a couple of scripts to provide the evaluation of the objective function and analytical expressions of the derivatives if you don't want to use the option given for approximating from the solver. You will need to determine the category of problem you have (linear, nonlinear, constrained, unconstrained, smooth, nonsmooth, convex, nonconvex, integer etc) and then select the solver.

Just try to solve a small problem with GA or other method using optimization tool box , You can find number of videos on using GA in Matlab on you tube. 

it's very simple but before using you have to formulate your optimization problem by specifying the objective function in a function file where you express this function and nonlinear constraints function in another file for the rest it is the mater of optimtool homepage where the two files function will be preceded by an @ sign