Molecular models for the interaction of parathyroid hormone (PTH) with its G-protein-coupled receptors (PTH1 and PTH2) have been developed. The proposed ligand-receptor complex is based on experimental data from spectroscopic investigations of the hormone and receptor fragments as well as theoretical structure predictions based on homology analysis with proteins of known structure. From the insight afforded by the models, biochemical and pharmacological observations can be correlated with specific molecular or atomic interactions. The ligand selectivity of PTH2, specifically the lack of binding of His5-containing analogues, can be ascribed to unfavorable steric interactions (the binding pocket is markedly smaller in PTH2 than PTH1) as well as repulsive Coulombic forces between amino acids of like-charge (a positively charged H384 is located in the binding pocket in PTH2). The model of PTH1 suggests that the constitutive activity observed from the incorporation of a positively charged amino acid at position 223, found at the cytoplasmic end of TM2, is caused by a Coulombic attraction to E465, at the cytoplasmic end of TM7, leading to an association of TM2 and TM7 and thereby ligand-free activation. Additionally, a number of important interactions in the ligand-receptor complex are described along with predictions of the pharmacological profile which will result from specific modifications at these sites. In this regard, the models described here allow for atomic insight into the biochemical data currently available and allow targeting of future mutations to probe specific ligand/receptor interactions and thereby further our understanding of the functioning of this important hormone system.

No abstract

Tumor cell migration and metastasis in cancer are facilitated by interaction of the serine protease urokinase type plasminogen activator (uPA) with its receptor uPAR (CD 87). Overexpression of uPA and uPAR in cancer tissues is associated with a high incidence of disease recurrence and early death. In agreement with these findings, disruption of the protein-protein interaction between uPAR present on tumor cells and its ligand uPA evolved as an attractive intervention strategy to impair tumor growth and metastasis. For this, the uPAR antagonist cyclo[19,31][D-Cys(19)]-uPA(19)(-)(31) was optimized to efficiently interrupt binding of uPA to cellular uPAR. First, the disulfide bridge of this lead compound was shifted and then the modified peptide was shortened from the amino and carboxy terminus to generate cyclo[21,29][Cys(21,29)]-uPA(21)(-)(30). Next, cyclo[21,29][D-Cys(21)Cys(29)]-uPA(21)(-)(30) was yielded by changing the chirality of Cys(21) to D-Cys(21). For analysis of uPAR binding activity, we employed competitive flow cytofluorometric receptor binding assays, using FITC-uPA as the ligand and U937 promyeloid leukemia cells as the cellular source of uPAR. As demonstrated for cyclo[21,29][D-Cys(21)Cys(29)]-uPA(21)(-)(30), the achieved peptide modifications maintained receptor binding activity (IC(50) = 0.04 microM), which is close in order to that of the parent protein ligand, uPA (IC(50) = 0.01 microM). A detailed NMR analysis with restrained and free molecular dynamics calculations in explicit H(2)O exhibits a well-defined structure with characteristic features such as an omega-loop with two betaI-turns about Lys(3), Tyr(4), Ser(6), and Asn(7). Hydrophobic clustering of the side chains of Tyr(4), Phe(5), Ile(8), and Trp(10) is observed. Side chain mobility is analyzed with time-dependent distance restraints. The NMR structure of cyclo[21,29][D-Cys(21)Cys(29)]-uPA(21)(-)(30) is very similar to the previously reported structure of the amino terminal fragment of uPA. Systematic point mutations led to cyclo[21,29][D-Cys(21)Nle(23)Cys(29)]-uPA(21)(-)(30), which still binds to uPAR but is resistant to proteolytic cleavage, e.g., by the tumor-associated serine proteases uPA and plasmin, and is stable in blood serum or plasma. In conclusion, small cyclic peptides were created, which mimic the structure and activity of the binding epitope of uPA to uPAR and which may serve as novel therapeutic agents in cancer metastasis.