Bombesin is an endogenous peptide involved in a wide spectrum of physiological activities ranging from satiety, control of circadian rhythm and thermoregulation in the central nervous system, to stimulation of gastrointestinal hormone release, activation of macrophages and effects on development in peripheral tissues. Actions of the peptide are mediated through the two high affinity G-protein coupled receptors BB1 and BB2. Under pathophysiological conditions, these receptors are overexpressed in many different types of tumors, such as prostate cancer, breast cancer, small and non-small cell lung cancer and pancreatic cancer. This knowledge has been used for designing cell markers, but it has not been yet exploited for therapeutical purposes. Despite the enormous biological interest of the peptide, little is known about the stereochemical features that contribute to their activity. On the one hand, mutagenesis studies identified a few receptor residues important for high bombesin affinity and on the other, a few studies focused on the relevance of diverse residues of the peptide for receptor activation. Models of the peptide bound to BB1 and BB2 can be helpful to improve our understanding of the stereochemical features granting bombesin activity. Accordingly, the present study describes the computational process followed to construct such models from models of the peptide and its receptors by means of Steered Molecular Dynamics. Present results provide new insights into the structure-activity relationships of bombesin and its receptors, as well as render an explanation for the differential binding affinity observed towards the BB1 and BB2 receptors. Finally, these models can be further exploited to help for designing novel small molecule peptidomimetics with improved pharmacokinetics profile.