Cancer is one of the main reasons of mortality all over the world. Over the time, the major ways for cancer-therapy were based on radiotherapy, chemotherapy and surgery. These methods are not specific enough for that purpose, therefore, new ideas for design of new drugs with higher specificity are considered. Chimeric protein toxins are hybrid proteins consisting of a targeting portion and a toxic one which specifically bind and kill the target cancer cells. The main purpose of this study was designing a recombinant chimeric toxin with biding capability to one of the most key receptors namely claudin-4 which is over-expressed in almost all cancer cells. To design it, we utilized the last 30 C-terminal amino acids of Clostridium perfringens enterotoxin (CPE) as a binding module for claudin-4 and the toxic module which is the A-domain of Shiga toxin from Shigella dysenteriae. Using molecular modeling and docking methods, appropriate binding affinity of the recombinant chimeric toxin to its specific receptor was demonstrated. In the next step, the stability of this interaction was investigated by molecular dynamics simulation. Although partial instability was detected at some time points, however, sufficient stable situation of hydrogens bonds and high binding affinity between the chimeric toxin and receptor were observed in the in silico studies which in turn suggested that this complex could be formed successfully.