Andrew J. Doig

Amyloid β Protein and Alzheimer’s Disease: When Computer Simulations Complement Experimental Studies

Simulations Complement Experimental Studies Jessica Nasica-Labouze,† Phuong H. Nguyen,† Fabio Sterpone,† Olivia Berthoumieu,‡ Nicolae-Viorel Buchete, Sebastien Cote, Alfonso De Simone, Andrew J. Doig, Peter Faller,‡ Angel Garcia, Alessandro Laio, Mai Suan Li, Simone Melchionna, Normand Mousseau, Yuguang Mu, Anant Paravastu, Samuela Pasquali,† David J. Rosenman, Birgit Strodel, Bogdan Tarus,† John H. Viles, Tong Zhang,†,▲ Chunyu Wang, and Philippe Derreumaux*,†,□ †Laboratoire de Biochimie Theorique, Institut de Biologie Physico-Chimique (IBPC), UPR9080 CNRS, Universite Paris Diderot, Sorbonne Paris Cite, 13 rue Pierre et Marie Curie, 75005 Paris, France ‡LCC (Laboratoire de Chimie de Coordination), CNRS, Universite de Toulouse, Universite Paul Sabatier (UPS), Institut National Polytechnique de Toulouse (INPT), 205 route de Narbonne, BP 44099, Toulouse F-31077 Cedex 4, France School of Physics & Complex and Adaptive Systems Laboratory, University College Dublin, Belfield, Dublin 4, Ireland Deṕartement de Physique and Groupe de recherche sur les proteines membranaires (GEPROM), Universite de Montreal, C.P. 6128, succursale Centre-ville, Montreal, Quebec H3C 3T5, Canada Department of Life Sciences, Imperial College London, London SW7 2AZ, United Kingdom Manchester Institute of Biotechnology, University of Manchester, 131 Princess Street, Manchester M1 7DN, United Kingdom Department of Physics, Applied Physics, & Astronomy, and Department of Biology, Rensselaer Polytechnic Institute, Troy, New York 12180, United States The International School for Advanced Studies (SISSA), Via Bonomea 265, 34136 Trieste, Italy Institute of Physics, Polish Academy of Sciences, Al. Lotnikow 32/46, 02-668 Warsaw, Poland Institute for Computational Science and Technology, SBI Building, Quang Trung Software City, Tan Chanh Hiep Ward, District 12, Ho Chi Minh City, Vietnam Instituto Processi Chimico-Fisici, CNR-IPCF, Consiglio Nazionale delle Ricerche, 00185 Roma, Italy School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, 637551 Singapore Department of Chemical and Biomedical Engineering, Florida A&M University-Florida State University (FAMU-FSU) College of Engineering, 2525 Pottsdamer Street, Tallahassee, Florida 32310, United States National High Magnetic Field Laboratory, 1800 East Paul Dirac Drive, Tallahassee, Florida 32310, United States Institute of Complex Systems: Structural Biochemistry (ICS-6), Forschungszentrum Julich GmbH, 52425 Julich, Germany School of Biological and Chemical Sciences, Queen Mary University of London, London E1 4NS, United Kingdom Institut Universitaire de France, 75005 Paris, France

Inhibition of Toxicity in the β-Amyloid Peptide Fragment β-(25–35) Using N-Methylated Derivatives A GENERAL STRATEGY TO PREVENT AMYLOID FORMATION

β-(25–35) is a synthetic derivative of β-amyloid, the peptide that is believed to cause Alzheimers disease. As it is highly toxic and forms fibrillar aggregates typical of β-amyloid, it is suitable as a model for testing inhibitors of aggregation and toxicity. We demonstrate thatN-methylated derivatives of β-(25–35), which in isolation are soluble and non-toxic, can prevent the aggregation and inhibit the resulting toxicity of the wild type peptide.N-Methylation can block hydrogen bonding on the outer edge of the assembling amyloid. The peptides are assayed by Congo red and thioflavin T binding, electron microscopy, and a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) toxicity assay on PC12 cells. One peptide (Gly25 N-methylated) has properties similar to the wild type, whereas five have varying effects on prefolded fibrils and fibril assembly. In particular, β-(25–35) with Gly33 N-methylated is able to completely prevent fibril assembly and to reduce the toxicity of prefolded amyloid. With Leu34 N-methylated, the fibril morphology is altered and the toxicity reduced. We suggest that the use of N-methylated derivatives of amyloidogenic peptides and proteins could provide a general solution to the problem of amyloid deposition and toxicity.

Properties and identification of human protein drug targets

MOTIVATION We analysed 148 human drug target proteins and 3573 non-drug targets to identify differences in their properties and to predict new potential drug targets. RESULTS Drug targets are rare in organelles; they are more likely to be enzymes, particularly oxidoreductases, transferases or lyases and not ligases; they are involved in binding, signalling and communication; they are secreted; and have long lifetimes, shown by lack of PEST signals and the presence of N-glycosylation. This can be summarized into eight key properties that are desirable in a human drug target, namely: high hydrophobicity, high length, SignalP motif present, no PEST motif, more than two N-glycosylated amino acids, not more than one O-glycosylated Ser, low pI and membrane location. The sequence features were used as inputs to a support vector machine (SVM), allowing the assignment of any sequence to the drug target or non-target classes with an accuracy in the training set of 96%. We identified 668 proteins (23%) in the non-target set that have target-like properties. We suggest that drug discovery programmes would be more likely to succeed if new targets are chosen from this set or their homologues.

Is the hydrophobic effect stabilizing or destabilizing in proteins? The contribution of disulphide bonds to protein stability.

It has been recently concluded that the hydrophobic effect, hitherto regarded as a major driving force in the folding of proteins, destabilizes the folded state relative to the unfolded state. We summarize the properties of the hydrophobic effect obtained from solvent transfer experiments and show that the recent conclusion is an artifact of crosslinking in the unfolded state, caused by disulphide bonds, metals or cofactors. We show that, for the proteins in the data set, crosslinks surprisingly destabilize folded structures entropically, but stabilize them enthalpically to a greater extent. We also calculate non-polar surface areas of these unfolded proteins. These surface areas are decreased by crosslinks. The unfolded state of proteins lacking constraints, such as myoglobin, is well approximated by a mixture of residues containing alpha-helical and beta-sheet dihedral angles. Surface areas of unfolded proteins cannot be obtained by summing the surface areas of individual residues, since this ignores any unavoidable side-chain-side-chain interactions.

Distinguishing enzyme structures from non-enzymes without alignments.

The ability to predict protein function from structure is becoming increasingly important as the number of structures resolved is growing more rapidly than our capacity to study function. Current methods for predicting protein function are mostly reliant on identifying a similar protein of known function. For proteins that are highly dissimilar or are only similar to proteins also lacking functional annotations, these methods fail. Here, we show that protein function can be predicted as enzymatic or not without resorting to alignments. We describe 1178 high-resolution proteins in a structurally non-redundant subset of the Protein Data Bank using simple features such as secondary-structure content, amino acid propensities, surface properties and ligands. The subset is split into two functional groupings, enzymes and non-enzymes. We use the support vector machine-learning algorithm to develop models that are capable of assigning the protein class. Validation of the method shows that the function can be predicted to an accuracy of 77% using 52 features to describe each protein. An adaptive search of possible subsets of features produces a simplified model based on 36 features that predicts at an accuracy of 80%. We compare the method to sequence-based methods that also avoid calculating alignments and predict a recently released set of unrelated proteins. The most useful features for distinguishing enzymes from non-enzymes are secondary-structure content, amino acid frequencies, number of disulphide bonds and size of the largest cleft. This method is applicable to any structure as it does not require the identification of sequence or structural similarity to a protein of known function.

Inhibition of protein aggregation and amyloid formation by small molecules.

For decades, drug after drug has failed to slow the progression of Alzheimers disease in human trials. How compounds reducing fibril formation in vitro and toxicity in transgenic mice and flies bind to the Aβ toxic oligomers, is unknown. This account reviews recent drugs mainly targeting Aβ, how they were identified and report their successes from in vitro and in vivo experimental studies and their current status in clinical trials. We then focus on recent in vitro and simulation results on how inhibitors interact with Aβ monomers and oligomers, highly desirable knowledge for predicting new efficient drugs. We conclude with a perspective on the future of the inhibition of amyloid formation by small molecules.

Design strategies for anti-amyloid agents.

Numerous diseases have been linked to a common pathogenic process called amyloidosis, whereby proteins or peptides clump together in the brain or body to form toxic soluble oligomers and/or insoluble fibres. An attractive strategy to develop therapies for these diseases is therefore to inhibit or reverse protein/peptide aggregation. A diverse range of small organic ligands have been found to act as aggregation inhibitors. Alternatively, the wild-type peptide can be derivatised so that it still binds to the amyloid target, but prevents further aggregation. This can be achieved by adding a bulky group or charged amino acid to either end of the peptide, or by incorporating proline residues or N-methylated amide groups.

Recent advances in helix-coil theory.

Peptide helices in solution form a complex mixture of all helix, all coil or, most frequently, central helices with frayed coil ends. In order to interpret experiments on helical peptides and make theoretical predictions on helices, it is therefore essential to use a helix-coil theory that takes account of this equilibrium. The original Zimm-Bragg and Lifson-Roig helix-coil theories have been greatly extended in the last 10 years to include additional interactions. These include preferences for the N-cap, N1, N2, N3 and C-cap positions, capping motifs, helix dipoles, side chain interactions and 3(10)-helix formation. These have been applied to determine energies for these preferences from experimental data and to predict the helix contents of peptides. This review discusses these newly recognised structural features of helices and how they have been included in helix-coil models.

Effect of the N1 residue on the stability of the alpha-helix for all 20 amino acids.

N1 is the first residue in an α‐helix. We have measured the contribution of all 20 amino acids to the stability of a small helical peptide CH3CO‐XAAAAQAAAAQAAGY‐NH2 at the N1 position. By substituting every residue into the N1 position, we were able to investigate the stabilizing role of each amino acid in an isolated context. The helix content of each of the 20 peptides was measured by circular dichroism (CD) spectroscopy. The data were analyzed by our modified Lifson‐Roig helix‐coil theory, which includes the n1 parameter, to find free energies for placing a residue into the N1 position. The rank order for free energies is Asp−, Ala > Glu− > Glu0 > Trp, Leu, Ser > Asp0, Thr, Gln, Met, Ile > Val, Pro > Lys+, Arg, His0 > Cys, Gly > Phe > Asn, Tyr, His+. N1 preferences are clearly distinct from preferences for the preceding N‐cap and α‐helix interior. pKa values were measured for Asp, Glu, and His, and protonation‐free energies were calculated for Asp and Glu. The dissociation of the Asp proton is less favorable than that of Glu, and this reflects its involvement in a stronger stabilizing interaction at the N terminus. Proline is not energetically favored at the α‐helix N terminus despite having a high propensity for this position in crystal structures. The data presented are of value both in rationalizing mutations at N1 α‐helix sites in proteins and in predicting the helix contents of peptides.

Inhibitors of protein aggregation and toxicity

The aggregation of numerous peptides or proteins has been linked to the onset of disease, including Abeta (amyloid beta-peptide) in AD (Alzheimers disease), asyn (alpha-synuclein) in Parkinsons disease and amylin in Type 2 diabetes. Diverse amyloidogenic proteins can often be cut down to an SRE (self-recognition element) of as few as five residues that retains the ability to aggregate. SREs can be used as a starting point for aggregation inhibitors. In particular, N-methylated SREs can bind to a target on one side, but have hydrogen-bonding blocked on their methylated face, interfering with further assembly. We applied this strategy to develop Abeta toxicity inhibitors. Our compounds, and a range of compounds from the literature, were compared under the same conditions, using biophysical and toxicity assays. Two N-methylated D-peptide inhibitors with unnatural side chains were the most effective and can reverse Abeta-induced inhibition of LTP (long-term potentiation) at concentrations as low as 10 nM. An SRE in asyn (VAQKTV) was identified using solid-state NMR. When VAQKTV was N-methylated, it was able to disrupt asyn aggregation. N-methylated derivatives of the SRE of amylin are also able to inhibit amylin aggregation.