Moltenprot: A High-Throughput Analysis Platform to Assess Thermodynamic Stability of Membrane Proteins and Complexes

Abstract

Acidic human fibroblast growth factor 1 (hFGF1) is a major signaling molecule that is heavily involved in cell proliferation, angiogenesis, tumor invasion and metastatic progression. Previous experimental studies have demonstrated that hFGF1 is naturally unstable and that it has a near-physiological denaturation temperature. Heparin (a linear sulfated polysaccharide) is known to stabilize hFGF1 and protect it from thermal and proteolytic degradation. Our study used experimental data to set up a rigorous computational investigation of the hFGF1-heparin hexasaccharide complex. Three models were simulated for 4.8 microseconds each. Our equilibrium-MD simulations confirmed that the heparin-free monomer is less stable than the heparin-bound monomer. The decreased stability of the heparin-free monomer is due to a conformational change in the heparin-binding region. This conformational change was not observed in the heparin-bound systems. Important interactions that contribute to the stability of the complex were also characterized. K113 and S117 were identified as important residues of the heparin-binding region that contribute to intramolecular hydrogen bonding. Strong intermolecular hydrogen bonding was also observed between R123 and IdoA(4) of the heparin hexasaccharide. We then used a combination of non-equilibrium pulling and bias exchange umbrella sampling simulations to determine the binding free energy of heparin. Thus far, all published computational studies of the hFGF-1 heparin complex have been based on nanosecond-level simulations. For the very first time, we have used a combination of microsecond-level MD simulations and largescale enhanced sampling techniques to carry out a more realistic and biochemically relevant assessment of the hFGF1-heparin complex. Aggregation of prion proteins causes neurodegenerative disorders. The misfolded form of monomeric cellular prion protein leads to toxic protein aggregates rich in b-sheets. Pathological scrapie form of prion protein also catalyzes the conversion of cellular protein to the infectious conformation. Therefore it is very important to understand the mechanism of structural transition of a prion protein from a cellular functional form to the scrapie form. To probe the structures of transient intermediates populated in the folding pathways of prion protein, which can be probable precursors to the pathological misfolded forms, we performed molecular dynamics simulations to probe the folding pathways of prion protein using coarse-grained protein models. The protein populates an intermediate, which is globular and resembles a molten globule like structure where only the b 1 strand detaches from the rest of the folded protein structure. This observation also supports the NMR studies, which report the partial unfolding of the b 1 strand. This conformation might possibly be a precursor to the misfolded or the scrapie prion form. Experiments on lysozyme folding [Dobson et al., 1994; Kiefhaber, 1995] show that it folds in parallel pathways: a slow kinetic pathway wi...

In‐depth interrogation of protein thermal unfolding data with MoltenProt

“…Application of the concepts outlined in the previous section to a high‐throughput assay such as NanoDSF is challenging, because it requires a robust nonlinear curve fitting procedure and an easy way to view the results. This prompted us to develop a software package MoltenProt 44 (Figure 1(c), Figure S1c), which we successfully used to characterize a large NanoDSF dataset 40 . Our previous work, however, did not compare the thermodynamic parameters obtained by the combination of NanoDSF and MoltenProt with other methods.…”

High-throughput stability screening for detergent-solubilized membrane proteins

“…Data processing and visualization was done in MoltenProt 59 , which is written in Python 60 using the following modules: scipy 57 , numpy 61 , pandas 62 , matplotlib 63 .…”