Are long-range structural correlations behind the aggregration phenomena of polyglutamine diseases?

Abstract

We have characterized the conformational ensembles of polyglutamine Qn peptides of various lengths n (ranging from 6 to 40), both with and without the presence of a C-terminal polyproline hexapeptide. For this, we used state-of-the-art molecular dynamics simulations combined with a novel statistical analysis to characterize the various properties of the backbone dihedral angles and secondary structural motifs of the glutamine residues. For Q40 (i.e., just above the pathological length ≃36 for Huntington's disease), the equilibrium conformations of the monomer consist primarily of disordered, compact structures with non-negligible α-helical and turn content. We also observed a relatively small population of extended structures suitable for forming aggregates including β- and α-strands, and β- and α-hairpins. Most importantly, for Q40 we find that there exists a long-range correlation (ranging for at least 20 residues) among the backbone dihedral angles of the Q residues. For polyglutamine peptides below the pathological length, the population of the extended strands and hairpins is considerably smaller, and the correlations are short-range (at most 5 residues apart). Adding a C-terminal hexaproline to Q40 suppresses both the population of these rare motifs and the long-range correlation of the dihedral angles. We argue that the long-range correlation of the polyglutamine homopeptide, along with the presence of these rare motifs, could be responsible for its aggregation phenomena.

Figure 1. (a) Schematic of amino acid backbone dihedrals

and , and (b) a corresponding Ramachandran plot. In a typical Ramachandran plot of a glutamine residue, each pixel represents a bin, whose intensity represents its relative population, ranging from 1,2,,9, and 10 or more conformations, sampled in our simulations. Blue, yellow, grey, and pink clusters identify PPII, , , and regions, respectively.

Figure 2. , PPII and content of selected polyQ peptides.

Here, given are the contents (as a percentage) of individual glutamine residues found in: (a,b) -region (c,d) PPII-region (e,f) . These percentages are plotted against the Glu residue numbers for (a,c,e) [red], [blue] and (b,d,f) [red], [blue]. These percentages are obtained from clustering the conformations based on their dihedral angles in the Ramachandran plot.

Figure 3. Helical, turn and coil content of selected polyQ peptides.

Here, given are the contents (as a percentage) of individual glutamine residues found in the following conformations: (a,b) helical (,) (c,d) turn (H-bonded,bend) (e,f) coil. These percentages are plotted against the Glu residue numbers for (a,c,e) [red],[blue] and (b,d,f) [red], [blue]. These percentages are obtained from the DSSP , analysis code.

Figure 4. Sample conformations of

and . Cartoon representation of sample conformations of (a) and (b) . Purple, blue, cyan, and orange represent -helix, -helix, turn, and coil secondary structural motifs, respectively. The licorice-like representation of the proline segment of is given in (b). These structures are plotted by VMD using STRIDE for secondary structure prediction.

Figure 5. Selected extended conformations of

peptides. Here, we give (a) cartoon and (b) licorice-like representation of select conformations of the peptide with (,,,) and (,,,) strands. (a) The coloring is similar to Fig. 4 with yellow and green representing and strands respectively. We used a dihedral angle-based algorithm to detect the strands and for other secondary structures in these plots we used STRIDE distributed with VMD . (b) The residues involved in () -hairpin, () isolated -strand, () -harpin, and (i) isolated -strand are highlighted. The rest of residues are grey and all the side chains are represented by thin lines.

Figure 6. Correlation analysis results for selected polyQ peptides.

Here is given the (a) odds ratio based between any two glutamine residues ( and ) of [red] and [blue] in terms of (). From each side of the peptide ending residues are omitted in the calculations to reduce the end effects. (b) Similar to (a) for [red], [blue], and [black]. Here residues from each end are omitted. (c,d) Correlation coefficient between dihedral angles of any two glutamine residues ( and ) in terms of () for (c) [red], [blue] and (d) [red], [blue], and [black]. The end residues were omitted according to the same protocol used for odds ratio analysis. (e,f) Similar to (a,b) but with the odds ratio calculated using the probabilities that residues belong or not to an repeat region.

Figure 7. Correlation analysis results for selected polyQ peptides.

Specifically, we give for (a) (b) and (c) based on OR()[red] OR(PPII)[blue] and OR()[black]. (d) To compare the linear and OR-based results we plotted (r) versus the correlation coefficient corr(r) for that suggests an almost linear behavior with a correlation coefficient of 0.97.

Figure 8. Distribution of radius of gyration of polyQ peptides.

(a) The estimated distribution for [red] and [blue]. (b) The estimated distribution for [red] and [blue]. The blue curve can be estimated as the sum [black] of three Gaussian distributions [dotted]. (c) The estimated distribution for , considering only the structures with an all-trans proline segment [green]. Similarly the green curve can be estimated as the sum [black] of four Gaussian distributions [dotted]. Considering only the structures that at least have one cis-proline results in the magenta curve for the distribution. All the histograms are obtained using a window of width . (d) The exponent in relation estimated from select pairs of (x axis) and ( for blue circles and for yellow squares). Inset: The average (in ) of Q peptides for .