Towards de novo design of transmembrane α-helical assemblies using structural modelling and molecular dynamics simulation

Abstract

This review discusses a potential new approach to de novo design of membrane proteins aided by advanced molecular dynamics simulations.

Three-Dimensional Interaction Homology: Deconstructing Residue–Residue and Residue–Lipid Interactions in Membrane Proteins

“…A relatively small number of experimental membrane protein structures deposited in the protein data bank or other databases include any or more than a small handful of structural lipids in the transmembrane region, let alone their native set [ 19 , 20 , 57 , 58 , 59 , 60 , 61 ]. This experimental deficit has led to a number of proposed methodologies to simulate the lipid structures [ 62 , 63 , 64 ] or to predict or simulate protein structure within implicit [ 65 , 66 , 67 , 68 ] or explicit [ 69 , 70 , 71 , 72 ] membrane environments.…”

Elucidating the complex membrane binding of a protein with multiple anchoring domains using extHMMM

“…The highly mobile membrane-mimetic (HMMM) model was first used to simulate membrane binding of the GLA domain of human coagulation factor VIIa (FVIIa) [ 31 ] and since then it has been employed for many membrane proteins [ 32 – 34 ]. The HMMM model has been considered as one of most efficient settings when included in membrane systems to simulate in order to study membrane binding mechanisms of membrane proteins, particularly peripheral membrane proteins, and their complex formation on the membrane at atomic scale [ 87 – 89 ]. The HMMM system of the GLA domain [ 31 ] was not very large (~0.07M atoms with a ~90x90Å 2 membrane patch), and minor problems unnoticeable in such smaller systems may become evident when the HMMM model is employed for larger membrane systems, such as those in the current study (~0.3M atoms with a ~140x140Å 2 membrane).…”

Functionally Active Synthetic α-Helical Pores

“…De novo computational protein design is a highly ambitious approach to designing sophisticated functional protein structures, enabling exploration of a broader range of structures and unique sequences. , Among the various structural motifs considered for de novo design, α-helical bundles have emerged as promising candidates as their helical structures offer advantages, including rapid folding and flexibility to adjust relative positions in the sequence. − Computational designs will provide a good understanding of sequence-to-structure relationships in α-helical assemblies and are critical to predicting protein folding of membrane-spanning designs. ,, In particular, advanced computational protein designs have introduced powerful tools such as Rosetta, ISAMBARD, OSPREY, Damietta, and dTERMen, facilitating the design of membrane-active proteins. In a fascinating study, Baker’s group computationally designed membrane-spanning α-helical pores .…”

“…Challenges persist in improving the computational efficiency of de novo design, mainly due to the lack of high-quality structure data to construct better protein designs . Determining the optimal relationship between sequence and structure is vital, and even a minor alteration in the backbone conformation can significantly impact the energy and stability of the protein. ,, As designed peptides have been limited to simple structures until now, naturally occurring α-helical motifs of trans-membrane proteins form a great starting point for developing sophisticated synthetic α-helical peptide pores . In particular, short synthetic peptides based on natural α-helical transmembrane pores can be a good candidate.…”