Studies to determine subcellular localization and translocation of proteins are important because subcellular localization of proteins affects every aspect of cellular function. Such studies frequently utilize mutagenesis to alter amino acid sequences hypothesized to constitute subcellular localization signals. These studies often utilize fluorescent protein tags to facilitate live cell imaging. These methods are excellent for studies of monomeric proteins, but for multimeric proteins, they are unable to rule out artifacts from native protein subunits already present in the cells. That is, native monomers might direct the localization of fluorescent proteins with their localization signals obliterated. We have developed a method for ruling out such artifacts, and we use glucose 6-phosphate dehydrogenase (G6PD) as a model to demonstrate the method's utility. Because G6PD is capable of homodimerization, we employed a novel approach to remove interference from native G6PD. We produced a G6PD knockout somatic (hepatic) cell line using CRISPR-Cas9 mediated genome engineering. Transfection of G6PD knockout cells with G6PD fluorescent mutant proteins demonstrated that the major subcellular localization sequences of G6PD are within the N-terminal portion of the protein. This approach sets a new gold standard for similar studies of subcellular localization signals in all homodimerization-capable proteins. Copyright © 2015 Elsevier Inc. All rights reserved.