Structure propensities in mutated polyglutamine peptides.

Polyglutamine is a naturally occurring peptide found within several proteins in neuronal cells of the brain, and its aggregation has been implicated in several neurodegenerative diseases, including Huntington's disease. The resulting aggregates have been demonstrated to possess ß-sheet structure, and experimental evidence has demonstrated that aggregation begins with a nucleus composed of a single peptide. In this paper, we computationally examined the structural tendencies of mutant polyglutamine peptides that were studied experimentally, and found to aggregate with varying efficiencies. Low-energy structures were generated for each peptide by simulated annealing molecular dynamics, and were analyzed quantitatively by various geometry-based methods. In all simulations, the carboxy-terminal end of each peptide was constrained to a ß-turn-ß-strand structure to simulate a situation in which ß-structure formation has initiated due to interaction with a seed or a growing oligomer/aggregate. Our results suggest the experimentally-observed inhibition of aggregation to be due to localized conformational restraint on the peptide backbone, which in turn confines the peptide to native coil structure, discouraging transition towards the ß-sheet structure required for aggregation.

Procrustean rotation in concert with principal component analysis of molecular dynamics trajectories: Quantifying global and local differences between conformational samples.

Given the principal component analysis (PCA) of a molecular dynamics (MD) conformational trajectory for a model protein, we perform orthogonal Procrustean rotation to "best fit" the PCA squared-loading matrix to that of a target matrix computed for a related but different molecular system. The sum of squared deviations of the elements of the rotated matrix from those of the target, known as the error of fit (EOF), provides a quantitative measure of the dissimilarity between the two conformational samples. To estimate precision of the EOF, we perform bootstrap resampling of the molecular conformations within the trajectories, generating a distribution of EOF values for the system and target. The average EOF per variable is determined and visualized to ascertain where, locally, system and target sample properties differ. We illustrate this approach by analyzing MD trajectories for the wild-type and four selected mutants of the beta1 domain of protein G.

Water-mediated interactions in the CRP-cAMP-DNA complex: does water mediate sequence-specific binding at the DNA primary-kink site?

The cyclic AMP receptor protein (CRP) of Escherichia coli binds preferentially to DNA sequences possessing a T:A base pair at position 6 (at which the DNA becomes kinked), but with which it does not form any direct interactions. It has been proposed that indirect readout is involved in CRP-DNA binding, in which specificity for this base pair is primarily related to sequence effects on the energetic susceptibility of the DNA to kink formation. In the current study, the possibility of contributions to indirect readout by water-mediated hydrogen bonding of CRP with the T:A base pair was investigated. A 1.0 ns molecular dynamics simulation of the CRP-cAMP-DNA complex in explicit solvent was performed, and assessed for water-mediated CRP-DNA hydrogen bonds; results were compared to several X-ray crystal structures of comparable complexes. While several water-mediated CRP-DNA hydrogen bonds were identified, none of these involved the T:A base pair at position 6. Therefore, the sequence specificity for this base pair is not likely enhanced by water-mediated hydrogen bonding with the CRP.

Patterns of invasion and colonization of the sea lamprey (Petromyzon marinus) in North America as revealed by microsatellite genotypes.

Invasions by exotic organisms have had devastating affects on aquatic ecosystems, both ecologically and economically. One striking example of a successful invader that has dramatically affected fish community structure in freshwater lakes of North America is the sea lamprey (Petromyzon marinus). We used eight microsatellite loci and multiple analytical techniques to examine competing hypotheses concerning the origins and colonization history of sea lamprey (n = 741). Analyses were based on replicated invasive populations from Lakes Erie, Huron, Michigan, and Superior, populations of unknown origins from Lakes Ontario, Champlain, and Cayuga, and populations of anadromous putative progenitor populations in North America and Europe. Populations in recently colonized lakes were each established by few colonists through a series of genetic bottlenecks which resulted in lower allelic diversity in more recently established populations. The spatial genetic structure of invasive populations differed from that of native populations on the Atlantic coast, reflecting founder events and connectivity of invaded habitats. Anadromous populations were found to be panmictic (theta(P) = 0.002; 95% CI = -0.003-0.006; P > 0.05). In contrast, there was significant genetic differentiation between populations in the lower and upper Great Lakes (theta(P) = 0.007; P < 0.05; 95% CI = 0.003-0.009). Populations in Lakes Ontario, Champlain, and Cayuga are native. Alternative models that describe different routes and timing of colonization of freshwater habitats were examined using coalescent-based analyses, and demonstrated that populations likely originated from natural migrations via the St Lawrence River.