Despite being different cell lines, you select a constitutive expression gene such as actin, which corrects the expression levels of the protein of interest. Any change you observe in the gene of interest without changes in your control may be due to a difference in the level of expression of your study protein. Obviously you must place the same amount of genetic material or proteins per lane of the gel, to be able to establish valid comparisons.
I agree with J Ramon Vielma. On the other hand I wonder why you need to normalize becuase comparing results form two different cells lines is tricky business.