Polyglutamine induced misfolding of huntingtin exon1 is modulated by the flanking sequences

Abstract

Polyglutamine (polyQ) expansion in exon1 (XN1) of the huntingtin protein is linked to Huntington's disease. When the number of glutamines exceeds a threshold of approximately 36-40 repeats, XN1 can readily form amyloid aggregates similar to those associated with disease. Many experiments suggest that misfolding of monomeric XN1 plays an important role in the length-dependent aggregation. Elucidating the misfolding of a XN1 monomer can help determine the molecular mechanism of XN1 aggregation and potentially help develop strategies to inhibit XN1 aggregation. The flanking sequences surrounding the polyQ region can play a critical role in determining the structural rearrangement and aggregation mechanism of XN1. Few experiments have studied XN1 in its entirety, with all flanking regions. To obtain structural insights into the misfolding of XN1 toward amyloid aggregation, we perform molecular dynamics simulations on monomeric XN1 with full flanking regions, a variant missing the polyproline regions, which are hypothesized to prevent aggregation, and an isolated polyQ peptide (Q(n)). For each of these three constructs, we study glutamine repeat lengths of 23, 36, 40 and 47. We find that polyQ peptides have a positive correlation between their probability to form a beta-rich misfolded state and their expansion length. We also find that the flanking regions of XN1 affect its probability to form a beta-rich state compared to the isolated polyQ. Particularly, the polyproline regions form polyproline type II helices and decrease the probability of the polyQ region to form a beta-rich state. Additionally, by lengthening polyQ, the first N-terminal 17 residues are more likely to adopt a beta-sheet conformation rather than an alpha-helix conformation. Therefore, our molecular dynamics study provides a structural insight of XN1 misfolding and elucidates the possible role of the flanking sequences in XN1 aggregation.

Figure 1. Constructs studied.

Diagrams of the organization of the sequence regions in the (a) XN1 and (b) XN1-P₁₁-P₁₀ constructs are shown. The number of residues in each region is indicated below the region. For all three constructs (XN1, XN1-P₁₁-P₁₀ and the homopolymer Q_n, not shown) we vary the number of repeats modeled: n = 23, 36, 40, 47. In (a) XN1, the first N-terminal 17 residues are collectively referred to as the Nt17 region. Following Nt17 is the polyQ region, which contains a variable n number of glutamine repeats. P₁₁ and P₁₀ are the regions of 11 and 10 proline repeats respectively; they are referred to as the polyP regions. A region of 17 residues tethers the polyP regions together. Finally, there are 12 residues in the C-terminus of XN1. The (b) XN1-P₁₁-P₁₀ construct is identical to XN1 with the exception that it does not contain the polyP regions. The XN1 sequence is explicitly written in Fig. 4d. We use the title of “construct” to refer to either XN1, XN1-P₁₁-P₁₀ or Q_n. A “model” is a specific polypeptide, such as XN1Q₂₃, which is a polypeptide of XN1 with 23 glutamine repeats. Thus a total of 12 models were studied, divided into 3 constructs with 4 different glutamine lengths.

Figure 2. Thermodynamics of the peptides.

Heat capacity and temperature curves are calculated with WHAM analysis of the simulation trajectories. (a) Calculations of the Q₂₃, Q₃₆, Q₄₀ and Q₄₇ models indicate the folding transition peaks become taller and narrower for longer glutamine lengths. (b) For the XN1Q₂₃, XN1Q₃₆ and XN1Q₄₇ models, the transition temperatures are nearly identical (308K, 308K and 304K respectively). However, the XN1Q₄₀ model has a larger transition temperature around 325K. (c) The XN1Q₂₃-P₁₁-P₁₀, XN1Q₃₆-P₁₁-P₁₀, XN1Q₄₀-P₁₁-P₁₀ and XN1Q₄₇-P₁₁-P₁₀ models have varied transition peaks. The transition temperatures are 365K, 335K, 317K and 343K respectively. Detailed data on the peak positions are found in Table S1.

Figure 3. Secondary structure probabilities.

Selected secondary structure probabilities of residues in the (a) polyQ region, the (b) Nt17 region and (c) each residue of Q₄₇. Lastly, (d) representative structures of Q_n. (a) In the XN1 and XN1-P₁₁-P₁₀ models, polyQ residues have an almost constant β-strand probability for varying number of glutamine repeats. For all lengths of polyQ in XN1, the polyQ residues have a 31%±4% β-strand probability. For all lengths of polyQ in XN1-P₁₁-P₁₀, the polyQ residues have a 42%±1% β-strand probability. In the context of Q_n, however, the glutamine residues for long Q_n lengths have an increase in β-strand probability and a decrease in the random coil probability (data not shown). On average, for n = 36, 40 and 47, residues in the Q_n polypeptides are 9% more likely to adopt β-strand conformations than the polyQ residues in the XN1 polypeptides. For the same polyQ lengths, polyQ residues in XN1-P₁₁-P₁₀ and all the residues in Q_n have an average difference of less than 1% in β-strand probability. (b) We show the α-helix and β-strand probabilities of the Nt17 residues as the length of the neighboring polyQ region increases. For XN1Q₂₃, we find the probabilities of forming an α-helix or β-strand are similar; the difference is less than 1%. However, when the polyQ length increases (XN1Q_36–47), the difference becomes more than 20% in favor of a β-strand. Contrarily, we find that the Nt17 residues in XN1-P₁₁-P₁₀ models consistently prefer β-strand dihedral angles over α-helix dihedral angles; the difference is over 20% for each length of polyQ. (c) As an example, we present the probability of each residue in Q₄₇ to have a β-sheet conformation. There are continuous stretches of high probabilities: residues 2–8, 11–17, 21–29 and 34–45. These stretches are likely the locations of continuous β-strands. The residues with surprisingly low β-sheet conformation are the location of turns between β-strands. The periodic shape of the graph indicates a β-sheet similar to the one in panel d). The average of all these probabilities is 42%; it is one data point in panel a). (d) These are representative structures of the Q_n polypeptides determined through clustering. Q₂₃ shows a β-hairpin, and although this structure is from the largest cluster, it represents only 6% of all the structures used for clustering (see methods). Therefore, it is rare, but possible, for Q₂₃ to adopt a β-hairpin conformation. The longer homopolymers are more likely to adopt a conformation similar to the ones depicted here; Q₃₆, Q₄₀ and Q₄₇ represent 50%, 23% and 40% of their respective clustered structures (Table S3). Most strands are between 6 and 9 residues long. Intra-backbone hydrogen bonds are shown only for those residues forming β-strands. Secondary structures are automatically calculated by PyMOL and are not used to calculate probabilities.

Figure 4. Representative structures of XN1.

Structures that are representative of (a) XN1Q₂₃ and (b) XN1Q₄₇ are shown. A detailed view of the polyP regions from the XN1Q₂₃ structure is also presented (c). Pink residues are glutamines; blue residues are prolines, and all other residues are colored green. The sequence for XN1 is shown (d) for n = 23 as in panel a). The coloring in the sequence matches the coloring of the polypeptide backbones in both structures. The (a) XN1Q₂₃ and (b) XN1Q₄₇ structures respectively represent 37% of 1036 clustered structures and 17% of 1032 clustered structures (see methods and Table S3). From these example conformations, we see some structural correlations that complement the secondary structure probability calculations (Fig. 3). First, no drastic change in the β-strand structure of the polyQ region is seen between the XN1Q₂₃ and XN1Q₄₇ structures. Second, the α-helix in the Nt17 region of (a) XN1Q₂₃ has transformed into a β-strand in (b) XN1Q₄₇. Third, the two polyP regions form (c) PPII helices. The probability data (not shown) indicates that these PPII helices exist in 100% of the partially folded structures. The intra-main chain hydrogen bonds are shown for those residues forming β-strands. As in Fig. 3d, the secondary structures are assigned by PyMOL. (d) The sequence is divided into the same regions as outlined in Fig. 1a.