Cooperativity among short amyloid stretches in long amyloidogenic sequences

Abstract

Amyloid fibrillar aggregates of polypeptides are associated with many neurodegenerative diseases. Short peptide segments in protein sequences may trigger aggregation. Identifying these stretches and examining their behavior in longer protein segments is critical for understanding these diseases and obtaining potential therapies. In this study, we combined machine learning and structure-based energy evaluation to examine and predict amyloidogenic segments. Our feature selection method discovered that windows consisting of long amino acid segments of ~30 residues, instead of the commonly used short hexapeptides, provided the highest accuracy. Weighted contributions of an amino acid at each position in a 27 residue window revealed three cooperative regions of short stretch, resemble the β-strand-turn-β-strand motif in A-βpeptide amyloid and β-solenoid structure of HET-s(218-289) prion (C). Using an in-house energy evaluation algorithm, the interaction energy between two short stretches in long segment is computed and incorporated as an additional feature. The algorithm successfully predicted and classified amyloid segments with an overall accuracy of 75%. Our study revealed that genome-wide amyloid segments are not only dependent on short high propensity stretches, but also on nearby residues.

Figure 1. Context dependent behavior of amyloid formation can be shown from the change of prediction rate with window size.

(A) Accuracy of prediction increases with the length of window size and maximized at a 27 residue segment. (B) Random forest (RF) algorithm removed the redundancy among the features and increase prediction accuracy.

Figure 2. Feature analysis revealed important factor for amyloid formation.

(A) The ratio of each feature category occurred in the selected 446 features in the optimal set compared to the ratio of 48.6% which is the ratio of selected features out of the total number. the disordered factors contribute most to the fibril formation followed by the secondary structure factors, amino acid volume factors and pssm factors. (B) the pssm features of each amino acid contained in the selected 446 features.

Figure 3. The cooperativity among the short amyloid stretches is consistent with the common motif in amyloidogenic structure.

(A) Weighted contributions of an amino acid at each position in the 27 residue segment revealed three regions. The contribution of each position is measured by the number of features in each position. The average contribution from all positions is 16.5. The positions with contributions higher than average are in green, and the red bars are position with contribution less than average. The 14^th residue in the center is highlighted as black. The arrangement of the three regions are similar to the common motifs of amyloid structures of β-strand-turn-β-strand motif in A-β peptide amyloid (B) and β-solenoid structure of HET-s(218–289) prion (C).

Figure 4. Amyloid interaction energy can be searched by the summation of residue interactions between two short amyloid stretches.

The βstrand-turn-βstrand motif is defined as two six-residue β-strands connected with a flexible turn with a length up to 15 residues, with total window length of 27 residues. When there is no linker (L = 0) or the linker is very short (for example, L = 1−2), the motif may be classified as triangular shape observed for β-solenoid structure in Figure 3.

Figure 5. The distribution of different features in the optimal feature set with 82 features indicated the protein-protein interaction energy dominate the amyloid formation.

Pssm_C describes the likelihood that the amino acid in the sequence mutates to the cystine (C), Pssm_H describes the likelihood that the amino acid in the sequence mutates to the Histidine (H), and so forth.