Reconstruction of ARNT PAS-B Unfolding Pathways by Steered Molecular Dynamics and Artificial Neural Networks

Abstract

Several experimental studies indicated that large conformational changes, including partial domain unfolding, have a role in the functional mechanisms of the basic helix loop helix Per/ARNT/SIM (bHLH-PAS) transcription factors. Recently, single-molecule atomic force microscopy (AFM) revealed two distinct pathways for the mechanical unfolding of the ARNT PAS-B. In this work we used steered molecular dynamics simulations to gain new insights into this process at an atomistic level. To reconstruct and classify pathways sampled in multiple simulations, we designed an original approach based on the use of self-organizing maps (SOMs). This led us to identify two types of unfolding pathways for the ARNT PAS-B, which are in good agreement with the AFM findings. Analysis of average forces mapped on the SOM revealed a stable conformation of the PAS-B along one pathway, which represents a possible structural model for the intermediate state detected by AFM. The approach here proposed will facilitate the study of other signal transmission mechanisms involving the folding/unfolding of PAS domains.

Figure 1

Typical PAS-domain fold is here represented using the three-dimensional structure of the ARNT PAS-B domain (PDBID: 1X0O). Secondary structure elements are a five-stranded antiparallel β-sheet (the N-terminal Aβ, Bβ and the C-terminal Gβ, Hβ, and Iβ) flanked by a long α-helix (Fα, called “helical-connector”) and several shorter α-helices (Cα, Dα, and Eα, often called “helical bundle”).

Figure 2

3D structures of the neuron centrotypes represented on the SOM. Colors assigned to neurons refer to cluster analysis that is treated later in the text.

Figure 3

Distribution of dRMSD values calculated between structures from the same neuron (blue) and between structures assigned to neighboring neurons (red). The neurons numbering is shown in Figure S2.

Figure 4

Self-organizing map. Conformations sampled in the different replicas are first assigned to different neurons (tiles of the map), and then neurons are further grouped by hierarchical clustering (colors of the tiles). See Methods for a detailed description.

Figure 5

Tracing of two replicas representative of pathways 1 and 2 on the SOM.

Figure 6

Graph representation of the transition matrix for the SMD unfolding of the PAS-B domain. Nodes are colored according to the cluster colors of Figure 2.

Figure 7

Three-dimensional representation of unfolding pathways. Pathway 1 starts with the detachment of the N-terminal region (a), then unfolding of Iβ (b), elongation of the N-terminal region (c), unfolding of Fα (d), and finally unfolding of Gβ and Hβ (e). Pathway 2 starts with detachment and unfolding of the N-terminal region (f) before the unfolding of the Iβ (g). Then the pathway proceeds as in the case of pathway 1 (h and i). Secondary structure colors are consistent with Figure 1.

Figure 8

Average SMD forces mapped on the SOM. Neuron 41, discussed in text, is labeled. Cluster boundaries are highlighted for cluster B (green) and C (magenta).

Figure 9

Three-dimensional representation of the putative intermediate state during pathway 1 unfolding. The two insets show the backbone hydrogen bonds in the C-terminal lobe (left) and the group of residues that mostly contribute to stabilizing the N-terminal lobe (right). Secondary structures are colored according to Figure 1.

Figure 10

Work profiles for replicas at higher (0.2 nm ns^–1 on the left) and lower (0.02 nm ns^–1 on the right) pulling speeds.

Figure 11

Evolution of low pulling speed (0.02 nm ns^–1) SMD simulations, plotted over the 0.2 nm ns⁻¹ SOM. For each replica, frames were assigned to the nearest neuron and represented as circles colored in a blue–white–red color scale according to their time in the simulation. Consecutive frames are connected by solid lines.