Dedicated to Professor Teruaki Mukaiyuma, a dear friend and revered colleague on the occasion of his 70th birthday(1 3. VIII. 96)Structural prerequisites for the stability of the 3, helix of p-peptides can be defined from inspection of models (Figs. 1 and 2 ) : lateral non-Hi-substituents in 2-and 3-position on the 3-amino-acid residues of the helix are allowed, axial ones are forbidden. To be able to test this prediction, we synthesized a series of hepta- (1) and of the p-heptapeptide H-P-HVal-P-HAla-p-HLeu-(S,S)-~-HAla(a Me)-p-HVal-p-HAla-p-HLeu-OH (22), with a central (2S,3S)-3-amino-2-methylbutanoic-acid residue, confirm the helical structure of such 8-peptides (previously discovered in pyridine solution) (Fig. 3 and Tables 1-5). The CD spectra of helicalp-peptides, the residues of which were prepared by (retentive) Arndt-Eistert homologation of the (S)-or ~-a-amino acids, show a trough at 215 nm. Thus, this characteristic pattern of the CD spectra was taken as an indicator for the presence of a helix in methanol solutions of compounds 1 S 2 2 and 25 (including partially and fully deprotected forms) (Figs. 4 4 ) . The results fully confirm predicted structural effects: incorporation of a single 'wrong' residue ((R)-P-HAla, p-HAib, (R,S)-B-HAla(aMe), or N-Me-P-HAla) in the central positioh of thep-heptapeptide derivatives A (see 17, 18,20, or 21, resp.) causes the CD minimum to disappear. Also, the P-heptadepsipetide 25 (missing H-bond) and the /I-heptapeptide analogs with a single a-amino-acid moiety in the middle (13 and 14) are not helical, according to this analysis. An interesting case is the heptapeptide 15 with the central achiral, unsubstituted 3-aminopropanoicacid moiety: helical conformation appears to depend upon the presence or absence of terminal protection and upon the solvent (MeOH vs. MeOH/H20).

The 8-hexapeptide (H-8-HVal-8-HAla-8-HLeu)2-OH (2) was prepared from the component L-8-amino acids by conventional peptide synthesis, including fragment coupling. A cyclo-8-tri-and a cyclo-8-hexapeptide were also prepared. The 8-amino acids were obtained from a-amino acids by Amdf-Eisfert homologation. All reactions leading to the 8-peptides occur smoothly and in high yields. The 8-peptides were characterized by their CD and NMR spectra (COSY, ROESY, TOCSY, and NOE-restricted modelling), and by an X-ray crystal-structure analysis. 8-Sheet-type structures (in the solid state) and a compact, left-handed or ( M ) 3 , helix of 5-A pitch (in solution) were discovered. Comparison with the analogous secondary structures of a-peptides shows fundamental differences, the most surprising one at this point being the greater stability of$peptide helices. There are structural relationships of 8-peptides with oligomers of 8-hydroxyalkanoic acids, and dissimilarities between the two classes of compounds are a demonstration of the power of H-bonding. Thep-hexapeptide 2 is stable to cleavage by pepsin at pH 2 in H 2 0 for at least 60 h at 37", while the corresponding a-peptide H-(Val-Ala-Leu),-OH is cleaved instantaneously under these conditions. The implication of the described results are discussed.Peptides and proteins are molecules central to life on our planet. When proteinogenic or ribosomal, they consist of a-amino acids linked together by amide bonds. We have recently embarked on a project aimed at the synthesis of oligomers from /?-amino acids (P-peptides), in order to be able to compare the structures and the properties of these unnatural peptide analogs with those of the natural products. Another point which attracted our interest in the /?-oligopeptides is their resemblance with poly(/?-hydroxyalkanoates) (PHA), an ubiquitous class of biopolymers which have been the subject of research in our group for many years [ 11.

Immobilized lanthanide ions offer the opportunity to refine structures of proteins and the complexes they form by using restraints obtained from paramagnetic NMR experiments. We report the design, synthesis, and spectroscopic evaluation of the lanthanide chelator, Caged Lanthanide NMR Probe 5 (CLaNP-5) readily attachable to a protein surface via two cysteine residues. The probe causes tunable pseudocontact shifts, alignment, paramagnetic relaxation enhancement, and luminescence, by chelating it to the appropriate lanthanide ion. The observation of single shifts and the finding that the magnetic susceptibility tensors obtained from shifts and alignment analyses are highly similar strongly indicate that the probe is rigid with respect to the protein backbone. By placing the probe at various positions on a model protein it is demonstrated that the size and orientation of the magnetic susceptibility tensor of the probe are independent of the local protein environment. Consequently, the effects of the probe are readily predictable using a protein structure only. These findings designate CLaNP-5 as a protein probe to deliver unambiguous high quality structural restraints in studies on protein-protein and protein-ligand interactions.

A lanthanide ligand is reported, amenable to be immobilized on a protein surface via cysteine residues. This probe yields large single shifts of the protein nuclei resonances in HSQC NMR spectra and causes significant alignment of the protein in the magnetic field, providing additional structural restraints. Anchoring the probe by two cysteine residues instead of one is highly beneficial for the size of its effects.