Search

Oligomeric amyloid-β is currently of interest in amyloid-β mediated toxicity and the pathogenesis of Alzheimer's disease. Mapping the amyloid-β interaction partners could help to discover novel pathways in disease pathogenesis. To discover the amyloid-β interaction partners, we applied a protein array with more than 8100 unique recombinantly expressed human proteins. We identified 324 proteins as potential interactors of oligomeric amyloid-β. The Gene Ontology functional analysis of these proteins showed that oligomeric amyloid-β bound to multiple proteins with diverse functions both from extra and intracellular localizations. This undiscriminating binding phenotype indicates that multiple protein interactions mediate the toxicity of the oligomeric amyloid-β. The most highly impacted cellular system was the protein translation machinery. Oligomeric amyloid-β could bind to altogether 24 proteins involved in translation initiation and elongation. The binding of amyloid-β to purified rat hippocampal ribosomes validated the protein array results. More importantly, in vitro translation assays showed that the oligomeric amyloid-β had a concentration dependent inhibitory activity on translation. Our results indicate that the inhibited protein synthesis is one of the pathways that can be involved in the amyloid-beta induced neurotoxicity.

Amyloid oligomers are considered to play causal roles in the pathogenesis of amyloid-related degenerative diseases including Alzheimer's disease. Using MD simulation techniques, we explored the contributions of the different structural elements of trimeric and pentameric full-length Abeta1-42 aggregates in solution to their stability and conformational dynamics. We found that our models are stable at a temperature of 310 K, and converge toward an interdigitated side-chain packing for intermolecular contacts within the two beta-sheet regions of the aggregates: beta1 (residues 18-26) and beta2 (residues 31-42). MD simulations reveal that the beta-strand twist is a characteristic element of Abeta-aggregates, permitting a compact, interdigitated packing of side chains from neighboring beta-sheets. The beta2 portion formed a tightly organized beta-helix, whereas the beta1 portion did not show such a firm structural organization, although it maintained its beta-sheet conformation. Our simulations indicate that the hydrophobic core comprising the beta2 portion of the aggregate is a crucial stabilizing element in the Abeta aggregation process. On the basis of these structure-stability findings, the beta2 portion emerges as an optimal target for further antiamyloid drug design.

Polyunsaturated fatty acids (PUFA) like stearidonic acid (SDA;18:4 n-3) eicosapentaenoic acid (EPA; 20:5 n-3), and docosahexaenoic acid (DHA; 22:6 n-3) and its chain fragment models were studied at B3LYP/6-31G(d) levels of theory. Significant conformations for the cis and trans isomers were selected to obtained the thermodynamic functions (DeltaH, DeltaS, DeltaG) for the cis-trans isomerization and for folding using the B3LYP/6-311+G(2d,p)//B3LYP/6-31G(d) level of theory. The structural analysis shows that there are significant differences in thermodynamic function of the trans- and cis-PUFAs. The trans-cis isomerization energy values reinforce the consistency and the relative accuracy of theoretical model calculations. The observed flexibility of naturally cis PUFAs could be explained by a very special "smooth basin" PES of the motif of sp(2)-sp(3)-sp(2) hybrid states as reported previously (J. Phys. Chem. A 2005, 109, 520-533). We assumed that intrinsic thermodynamic functions may describe this flexible folding process. The folding enthalpy as well as the folding entropy suggests that there is a new role of the cis-PUFAs in membranes: these cis isomers may have a strong influence on membrane stability and permeability. The average length of the cis helix and beta PUFA was approximated. The difference between the lengths of these two structures is approximately 10 A.

Bulky, flexible molecules such as peptides and peptidomimetics are often used as lead compounds during the drug discovery process. Pathophysiological events, e.g., the formation of amyloid fibrils in Alzheimer's disease, the conformational changes of prion proteins, or beta-secretase activity, may be successfully hindered by the use of rationally designed peptide sequences. A key step in the molecular engineering of such potent lead compounds is the prediction of the energetics of their binding to the macromolecular targets. Although sophisticated experimental and in silico methods are available to help this issue, the structure-based calculation of the binding free energies of large, flexible ligands to proteins is problematic. In this study, a fast and accurate calculation strategy is presented, following modification of the scoring function of the popular docking program package AutoDock and the involvement of ligand-based two-dimensional descriptors. Quantitative structure-activity relationships with good predictive power were developed. Thorough cross-validation tests and verifications were performed on the basis of experimental binding data of biologically important systems. The capabilities and limitations of the ligand-based descriptors were analyzed. Application of these results in the early phase of lead design will contribute to precise predictions, correct selections, and consequently a higher success rate of rational drug discovery.

Amyloid fibrils are self-associating filamentous structures formed from the 39- to 42-residue-long amyloid beta peptide (Abeta peptide). The deposition of Abeta fibrils is one of the most important factors in the pathogenesis of Alzheimer's disease. Abeta25-35 is a fibril-forming peptide that is thought to represent the biologically active, toxic form of the full-length Abeta peptide. We have recently shown that beta sheets can be mechanically unzipped from the fibril surface with constant forces in a reversible transition, and the unzipping forces differ in fibrils composed of different peptides. In the present work, we explored the effect of epsilon-amino acetylation of the Lys28 residue on the magnitude of the unzipping force of Abeta25-35 fibrils. Although the gross structure of the Lys28-acetylated (Abeta25-35_K28Ac) and wild-type Abeta25-35 (Abeta25-35wt) fibrils were similar, as revealed by atomic force microscopy, the fundamental unzipping forces were significantly lower for Abeta25-35_K28Ac (20 +/- 4 pN SD) than for Abeta25-35wt (42 +/- 9 pN SD). Simulations based on a simple two-state model suggest that the decreased unzipping forces, caused most likely by steric constraints, are likely due to a destabilized zippered state of the fibril.

Computational and experimental determinations were carried out in parallel on the conformational probability of N-Acetyl-Phenylalanine-NH2 (NAPA). Ab initio computations were completed at the BLYP/6-311G(df,p), B3LYP/6-31G(d), B3LYP/6-31G(d,p), and B3LYP/6-31+G(d) levels of theory, labeled L/61fp, B/6, B/6p, and B/6+, respectively. Three experimentally identified conformers were compared with theoretical data, confirming their identities as the betaLanti, gammaLgauche+, and gammaLgauche- (BACKBONESIDECHAIN) conformers. Evidence comes from matching experimental and theoretical data for all three constituent N-H stretches of NAPA, with a Delta(Experimental-Theoretical) = approximately 1-3 cm(-1), approximately 0-5 cm(-1), and approximately 1-6 cm(-1), at the L/61fp and B/6+ levels, respectively. Corrected-ZPE relative energies were computed to be 0.14, 0.00, 0.26 and 0.00, 0.67, 0.57 (kcal*mol(-1)) for the betaLanti, gammaLgauche+, and gamma(Lgauche- conformers, respectively, at the L/61fp and B/6+ levels, respectively. The MP2/6-31+G(d) level of theory was subsequently found to give similar relative energies. Characterization of the intramolecular interactions responsible for red and blue shifting of the N-H stretches showed the existence of the following intramolecular interactions: C=O[i]- - -HN[i], (Ar[i])-Cgamma- - -HN[i+1], (Ar[i])-Cdelta-H- - -O=C[i-1] for betaLanti; C=O[i-1]- - -HN[i+1], (Ar[i])-Cgamma- - -HN[i+1], (Ar[i])-C-H- - -O=C[i] for gammaLgauche+; and C=O[i-1]- - -HN[i+1] for gammaLgauche-. Each of these interactions were further investigated and subsequently characterized by orbital population and Atoms-In-Molecules (AIM) analyses, with the identity of overlap and bond critical points (BCP) serving as 'scoring criteria', respectively. Experimental and theoretical carbonyl stretches were also compared and showed good agreement, adding further strength to the synergy between experiment and theory.

Ab initio molecular orbital computations were carried out at three levels of theory: RHF/3-21G, RHF/6-31G(d), and B3LYP/6-31G(d), on four model systems of the amino acid proline, HCO-Pro-NH2 [I], HCO-Pro-NH-Me [II], MeCO-Pro-NH2 [III], and MeCO-Pro-NH-Me [IV], representing a systematic variation in the protecting N- and C-terminal groups. Three previously located backbone conformations, gammaL, epsilonL, and alphaL, were characterized together with two ring-puckered forms syn (gauche+ = g+) or "DOWN" and anti (gauche- = g-) or "UP", as well as trans-trans, trans-cis, cis-trans, and cis-cis peptide bond isomers. The topologies of the conformational potential energy cross-sections (PECS) of the potential energy hypersurfaces (PEHS) for compounds [I]-[IV] were explored and analyzed in terms of potential energy curves (PEC), and HCO-Pro-NH2 [I] was also analyzed in terms of potential energy surfaces (PESs). Thermodynamic functions were also calculated for HCO-Pro-NH2 [I] at the CBS-4M and G3MP2 levels of theory. The study confirms that the use of the simplest model, compound [I] with P(N) = P(C) = H, along with the RHF/3-21G level of theory, is an acceptable practice for the analysis of peptide models because only minor differences in geometry and stability are observed.

We present a family of human sst(4)-selective, high-affinity (IC(50) = 2-4 nM) cyclic somatostatin (SRIF) octapeptides. These peptides result from the substitution of dTrp(8) in H-c[Cys(3)-Phe(6)-Phe(7)-DTrp(8)-Lys(9)-Thr(10)-Phe(11)-Cys(14)]-OH (SRIF numbering) (ODT-8) by one of the four conformationally biased stereoisomers of beta-methyl-3-(2-naphthyl)alanine (beta-Me2Nal). Whereas H-c[Cys-Phe-Phe-DNal-Lys-Thr-Phe-Cys]-OH (ODN-8, 2) has high affinity and marginal selectivity for human sst(3) (Reubi et al., Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 13973-13978), H-c[Cys-Phe-Tyr-D-threo-beta-Me2Nal-Lys-Thr-Phe-Cys]-OH (5) has high affinity for all sst's except for sst(1); H-c[Cys-Phe-Tyr-L-threo-beta-Me2Nal-Lys-Thr-Phe-Cys]-OH (6) has high affinity for sst(4) (IC(50) = 2.1 nM), with more than 50-fold selectivity toward the other receptors. Analogues 7 and 8, containing d- and l-erythro-beta-Me2Nal instead of the corresponding threo derivatives at position 8, are essentially inactive at all receptors. Substitution of Tyr(7) in 5 and 6 by Aph(7) resulted in 9 and 10 with similar affinity patterns overall yet lowered affinity. The substitution of DCys(3) for Cys(3) in 5 and 6 yielded H-c[DCys-Phe-Tyr-D-threo-beta-Me2Nal-Lys-Thr-Phe-Cys]-OH (11) and H-c[DCys-Phe-Tyr-L-threo-beta-Me2Nal-Lys-Thr-Phe-Cys]-OH (12), with biological profiles almost identical to those of their parents 5 and 6 (i.e., high affinity for sst(2-5) for 11 and high affinity and selectivity for sst(4) for 12). Analogue 12, with high sst(4) affinity combined with the highest sst(4) selectivity among all tested compounds, is an agonist in the cAMP accumulation assay (EC(50) = 1.29 nM). Cold monoiodination of 12 yielded 14, with loss of sst(4) selectivity and loss of high affinity (IC(50) = 21 nM). Introduction of Tyr(2) in 9 and 10 and substitution of Cys(3) by dCys(3), to yield 15 and 16 (IC(50) = 9.8 and 61 nM, respectively, for sst(4) and limited selectivity), failed to generate a high-affinity (125)iodinatable sst(4)-selective ligand. Substitution of Phe by Tyr at position 11 in H-c[DCys-Phe-Phe-L-threo-beta-Me2Nal-Lys-Thr-Phe-Cys]-OH yielded 18 (IC(50) = 11.8 nM at sst(4)), with limited sst(4) selectivity (30-fold or greater at the other receptors) yet only slightly improved affinity over that of 14. Cold monoiodination of 18 yielded 20 (IC(50) = 30 nM at sst(4) and high selectivity). Whereas we were able, in this study, to identify a new family of sst(4)-selective, high-affinity compounds, our additional goal, to identify highly potent and sst(4)-selective ligands amenable to (125)iodination, could not be achieved satisfactorily. On the other hand, some of the diastereomers identified in this study, such as 5, 11, 17, and 19, are very potent ligands at all receptors but sst(1).

The enantiospecific synthesis of (R)-Boc-(Fmoc)-aminoglycine 7 was achieved. (S)-Cbz-serine 1 was reacted with diphenylphosphoryl azide in the presence of triethylamine to yield cyclic (S) carbamate 2. The ring nitrogen of 2 was protected with a Boc group (3). The cyclic carbamate of 3 was hydrolyzed with benzyltrimethylammonium hydroxide to yield the (R)-enantiomer of alcohol 4. The oxidation of 4 with pyridinium dichromate yielded the enantiomerically pure (97% ee) (R)-Boc-(Cbz)-aminoglycine 5, which was converted to 7 with retention of optical purity. Similarly, starting from (S)-Boc-serine 9, cyclic (S) carbamate 10 was obtained. The ring nitrogen of 10 was protected with a Cbz group (11) with retention of configuration. The cyclic carbamate of 11 was base hydrolyzed to yield 12, the (S)-enantiomer alcohol. Independently, Boc-(Fmoc)-aminoglycine amide 13 and Boc-(Fmoc)-aminoglycine methyl ester 14 were resolved using papain. The stereochemistry of the isolated acid was determined to be (R) by coelution on HPLC of its derivative with Marfey's reagent and that of an authentic sample (7) obtained by enantiospecific synthesis.

Nine new analogues of acetyl-CCK-heptapeptide (Ac-Tyr(SO3H)2-Met3-Gly4-Trp5-Met6-Asp7-Phe8-NH2 ) were synthesized by solid-phase methodology. In a first series, the Asp7 residue was replaced by hydroxy amino acid sulfate esters. In another series, Gly4 was substituted by D-Ala, while Trp5 and Met6 were replaced by their D enantiomer. The introduction of the sulfate ester was performed with a new, mild, crystalline, and stable reagent, pyridinium acetyl sulfate. Each analogue that contained Tyr(SO3H)2 and a hydroxy amino acid sulfate ester [Ser(SO3H), Thr(SO3H), or Hyp(SO3H)] in position 7 proved to be more potent (1.9, 1.7, and 3.0 times, respectively) than CCK-8 in vitro (isolated gallbladder strips). While devoid of gastrin-like activity in vivo, these analogues had potent anticonvulsive activity. The analogues containing a D-amino acid residue were less potent than the parent compound in vitro. The D-Ala4 replacement, however, yielded a compound that was 40% as potent as CCK-8 in the in vitro test but showed prolonged duration of action on sphincter Oddi. While the 7-substituted Ac-CCK heptapeptides are among the most potent CCK analogues reported so far, the D-Ala4 replacement resulted, for the first time, in prolonged activity in vivo.