Steric hindrance as a basis for structure-based design of selective inhibitors of protein-tyrosine phosphatases.

Utilizing structure-based design, we have previously demonstrated that it is possible to obtain selective inhibitors of protein-tyrosine phosphatase 1B (PTP1B). A basic nitrogen was introduced into a general PTP inhibitor to form a salt bridge to Asp48 in PTP1B and simultaneously cause repulsion in PTPs containing an asparagine in the equivalent position [Iversen, L. F., et al. (2000) J. Biol. Chem. 275, 10300-10307]. Further, we have recently demonstrated that Gly259 in PTP1B forms the bottom of a gateway that allows easy access to the active site for a broad range of substrates, while bulky residues in the same position in other PTPs cause steric hindrance and reduced substrate recognition capacity [Peters, G. H., et al. (2000) J. Biol. Chem. 275, 18201-18209]. The current study was undertaken to investigate the feasibility of structure-based design, utilizing these differences in accessibility to the active site among various PTPs. We show that a general, low-molecular weight PTP inhibitor can be developed into a highly selective inhibitor for PTP1B and TC-PTP by introducing a substituent, which is designed to address the region around residues 258 and 259. Detailed enzyme kinetic analysis with a set of wild-type and mutant PTPs, X-ray protein crystallography, and molecular modeling studies confirmed that selectivity for PTP1B and TC-PTP was achieved due to steric hindrance imposed by bulky position 259 residues in other PTPs.

Antisense peptides: a critical mini-review.

Antisense peptides are defined as those generated from the non-coding strand of DNA, and represent a peptide analog to antisense RNA technologies. Peptides generated from both parallel and anti-parallel readings of the non-coding strand of DNA have displayed biological activity, although considerable controversy exists concerning the mechanism(s) by which these "anti-peptides" exert their effects. This paper provides a critical review of some of the key data and issues defining this emerging field and focuses on contradictions and discrepancies in the current studies. We also suggest some directions for future research such as more physico-chemical studies and the use of combinatorial chemistry techniques combined with solid phase binding studies to test, once and for all, the generality and specificity of antisense peptide interactions.

Augmentation of aortic ring contractions by angiotensin II antisense peptide.

Previous biochemical experiments have revealed two antisense peptide antagonists to human angiotensin II (Ang II), one encoded in the cDNA in the antiparallel reading, the other in the parallel reading. Neither peptide's ability to produce physiological antagonism has been demonstrated previously. Both peptides were tested for their ability to antagonize Ang II-induced contractions on rabbit aorta smooth muscle. Neither peptide had any direct contractile activity. The antiparallel Ang II peptide had physiological antagonism to Ang II contractions at a lower sensitivity than reported in biochemical studies, and its antagonist activity was partially blocked by Ang II antiserum, suggesting that it is not an antipeptide but an Ang II homologue. The parallel Ang II antipeptide also required high concentrations for physiological inhibition. Its contractile inhibition was not affected by Ang II antiserum and diminished the Ang II contraction at high micromolar concentrations, findings consistent with physicochemical data showing that it is an Ang II complement. The concentration of either peptide required to produce an antagonistic physiological effect was too high to predict any pharmacological usefulness. The parallel antipeptide, however, significantly increased the force of muscle contractions at high nanomolar concentrations, thus displaying a unique dual augmentation/antagonist activity. This antipeptide seems to have highly sequence-specific activity because other similar parallel antipeptides had no activity. The parallel antipeptide augmentation mimics the shift in the Ang II dose-response curve produced in hypertension studies of the slow pressor effect of Ang II and may be useful in deducing the currently unknown cause of the slow pressor effect. It may also have some uses in migraine studies.

Antisense-designed peptides: a comparative study focusing on possible complements to angiotensin II.

A comprehensive study of antisense peptides possibly complementary to angiotensin II (AII) is described. Antisense peptides of AII were designed using two different procedures outlined by Blalock and Root-Bernstein. Also, peptide complements designed to interact as homologs of AII were investigated. Three methods were used to detect binding between these peptides and AII. Several antisense-designed peptides were studied with unprotected termini to compare the effects of protected vs. unprotected termini. It was determined that the protected antisense-designed peptides derived from Root-Bernstein's methods interacted (high micro-molar range) directly with AII, while those protected antisense peptides derived from Blalock's method interacted only with the AII receptor. Two novel AII antagonists were discovered using this technology, a Root-Bernstein derived unprotected complementary peptide (H2N-K-G-V-Y-M-H-A-L-CO2H) and a Blalock derived unprotected antisense peptide (H2N-E-G-V-Y-V-H-P-V-CO2H), which exhibited 5 microM and 70 nM affinity toward the AII receptor, respectively.

Use of electrospray ionization mass spectrometry to probe antisense peptide interactions.

Electrospray ionization mass spectrometry with a magnetic sector instrument has been used to test for non-covalent interactions between human angiotensin II (M(r) 1046) and eight synthetic octapeptides that are considered complementary peptides (encoded by DNA sequences complementary to the DNA sequence that codes for human angiotensin II) or analogues of these antisense peptides. The relative abundance of the doubly charged heterodimer complex broadly correlates to the trend observed with solution-phase studies such as 1H nuclear magnetic resonance. Dissociation constants for the reaction in solution are in the high micromolar range. Electrospray ionization can potentially be a sensitive method for rapidly screening weak molecular interactions. Further work is necessary to study the possible gas-phase contributions to the observed binding interactions indicated in the mass spectrometry data.