Targeting Breast Cancer Using Diphtheria Toxin

Institution: Salk Institute for Biological Studies
Investigator(s): Senyon Choe, Ph.D. -
Award Cycle: 1996 (Cycle II) Grant #: 2CB-0063 Award: $375,052
Award Type: ITaMoCAs
Research Priorities
Detection, Prognosis and Treatment>Innovative Treatment Modalities: search for a cure



Initial Award Abstract (1996)
Normally, the growth of human breast cells is closely controlled by the levels of hormones (such as estrogen) and growth-promoting proteins (such as heregulin). Heregulins are a class of proteins known to promote the unwanted growth of breast cancer cells under certain conditions. The overabundance of heregulin on the surface of breast cancer cells often indicates that the cancer has progressed to the stage that the cells are no longer responsive to estrogen treatment using a drug like tamoxifen. The counter balancing activities of these two classes of molecules, steroid hormones and growth-promoting proteins, are important in determining the progression of breast cancer. The ultimate goal of this proposal is to disrupt this balance in order to stop growth of breast cancer cells. In practice, our specific aim is to design a toxin protein that will kill cells carrying the growth-promoting protein heregulin.

Diphtheria toxin is known to kill certain types of cells carrying a specific receptor protein on their surface. This toxin will serve as the starting material in our design of a new toxin for killing breast cancer cells. Our strategy is to create a toxin variant that is capable of recognizing and killing only those cancer cells carrying heregulin. We hypothesize that the targeted delivery of this toxin variant to these cells will effectively reduce the overall growth of breast cancer.

Our experimental approach is to custom design the toxin by using a three-dimensional picture as a guide. With the help of x-rays, we have recently obtained this three-dimensional picture. This picture shows in detail how atoms of the toxin fit together with atoms of the receptor, like pieces of a jigsaw puzzle. This detailed view will allow us to identify key atoms dictating the fit between toxin and receptor. In our designed toxin, these key atoms will be modified using molecular biology techniques with the aim of targeting the toxin to heregulin-bearing cancer cells. Since the molecular makeup of heregulin is similar to that of the natural toxin receptor, our design strategy appears to be feasible, needing only a small change on the toxin to change the target. The function of this designed toxin against breast cancer will first be tested in test tubes. Once successful, it will provide an alternate therapeutic means of controlling the progression of breast cancer.


Final Report (1998)
The normal growth of human breast cancer cells is closely controlled by steroid hormones (e.g., estrogen) and growth-promoting proteins, such as heregulin. Our aim is to develop a novel method to stop the production of heregulin by designing a toxin protein that will kill heregulin-overproducing breast cancer cells.

Diphtheria toxin (DT) was used as a starting material. Normally, the cytotoxic activity of DT is cell-specific, because it recognizes a cell surface receptor protein called ‘EGF precursor’. The toxin has two separate elements that cause killing and recognition. Thus, if we can alter the recognition module of DT, then the toxic effect could be ‘designed’ to be delivered specifically to breast cancer cells that produce heregulin. The key to this design is the similarity between the ‘EGF precursor’ and heregulin. Thus, our plan is first to gain more information about how DT bind to the EGF precursor, and then to modify the DT structure so that this recognition event is displaced to be specific to heregulin.

Using X-ray crystallography, we have completed the three-dimensional picture of DT bound to an ‘EGF precursor’ fragment. This shows in detail how atoms of the toxin fit together with atoms of the receptor- like pieces of a jigsaw puzzle. Using a three-dimensional picture as a guide, we have identified key atoms dictating the fit between toxin and receptor. We have chosen six distinct positions from the natural toxin receptor to modify amino acids by molecular biology techniques. The function of this recombinant toxin will be tested on breast cancer cells to develop a novel therapy.

Crystal structure of the complex of diptheria toxin with an extracellular fragment of its receptor
Periodical:Molecular Cell
Index Medicus: Mol Cell
Authors: Louie GV, Yang W, Bowman ME, and Choe S
Yr: 1997 Vol: 1 Nbr: Abs: Pg:67-78