Posted in: information

Integrated structure mass spectrometry and computational simulation to explore phosphorylation mediated protein allosteric regulation and conformational dynamics in glycogen phosphorylase

Hello, everyone. This week, I will introduce an article published by our research group in ACS chem Biol. Insights into phosphorylation induced protein allostery and conventional dynamics of glycogen phosphorylase via integrated structural mass spectroscopy and in silicon modelingone

Allosteric regulation exists widely in nature and can be used to regulate cellular processes. Glycogen phosphorylase (GP) is the first identified phosphorylated protein related to allosteric regulation. GP is a homodimer protein with a molecular weight of about 196kd. It is an important component in glucose metabolism and a target for type 2 diabetes and cancer. Amp binding and phosphorylation of ser14 mediate allosteric regulation of GP, transforming its conformation from non activated T-state GPB (non phosphorylated state) to activated R-state GPA (phosphorylated state). Even though the atomic protein structure of GP has been resolved by X-ray crystallography, it is difficult to detect its structural dynamics due to the large molecular weight, so the structural dynamic change process related to GP allosteric regulation is still vague.

Nuclear magnetic resonance (NMR) spectroscopy and molecular dynamics (MD) simulation are common methods to explore the dynamics of protein structure. However, NMR analysis has an upper limit of molecular weight, and the sample consumption is large. The time scale and force field accuracy of MD simulation are limited. Mass spectrometry (MS) is a powerful technique for the analysis of protein structure, dynamics and conformational changes. Hydrogen deuterium exchange mass spectrometry (hdx-ms) reflects the conformational dynamics of proteins by monitoring the exchange of amide hydrogen atoms in protein skeleton with deuterium in solution. Therefore, it is suitable for exploring the conformational changes of proteins caused by ligands, protein binding or covalent modification. At the same time, several software have realized the calculation of protective factors (PFS) and Gibbs free energy from hdx-ms data, so as to extract the protein dynamic information at the residue level. In addition, in the previous work 2 and 3, we integrated native MS and top-down methods (native top-down, NTD MS technology) to successfully detect the primary sequence to high-order structure and other information of multiple protein complexes (including sequencing, post-translational modification, ligand binding, structural stability, orientation, etc.). Integration of multiple structure mass spectrometry (integrated structure mass spectrometry) can effectively fill the gap between structure and dynamics in traditional biophysical methods, and better represent the phenomenon of allosteric regulation. In this paper, hdx-ms, NTD MS, PF analysis, MD simulation and allosteric signal analysis were integrated to detect the structural and dynamic basis of phosphorylation mediated allosteric regulation of GP, providing insights into the allosteric regulation process of GP.

According to the X-ray crystallographic structure report (Fig. 1a), when T-state GPB is converted to R-state GPA, the N-terminal tail in the dimer interface α 2. Cap ‘(Fig. 1b) and tower tower helices (Fig. 1c) have obvious structural rearrangement, resulting in the opening of catalytic sites, so that the substrate pyridoxal phosphate (PLP) can bind. Despite crystallographic reports, the conformational dynamics associated with allosteric regulation remains to be explored.

Figure 1 (a) Phosphorylation mediates the conformational transition from T-state GPB (pdb:8gpb) to R-state GPA (pdb:1gpa); Subunit interaction interface:(b) C-terminal region and (c) tower tower helices, GPB is blue and GPA is green.

First, we tested it by NTD Ms. As shown in Fig. 2a and B, monomer and dimer signals of GPB are observed in the spectrum, in which dimer is the main form; In addition to monomers and dimers, GPA also has a small amount of tetramers in the spectrum, but the main form is still dimer. When the sampling cone (SC) voltage is increased, GPB and GPA retain the dimer form (Fig. 2C and D). Subsequently, we selected ions (29+) and fragmented them in the trap cell (Fig. 2E, F, G, H). The fragment relative peak intensity of GPA in the low mass charge ratio region of the spectrum was higher than that of GPB, indicating that the dimer interaction interface of GP was more stable, and the structure of GPB subunit was more stable than that of GPA. NTD MS can not only explore the structural differences between GPB and GPA, but also do a good job in the early sample quality inspection for the next hdx-ms experiment.

Figure 2 NTD MS spectra of GPB and GPA under different activation conditions. (a、b)SC=40V; (c、d)SC=150V; (e、f)SC=150V、trap=100eV; (g,h)SC=150V、trap=200eV。 GPB on the left and GPA on the right.

Then we carried out hdx-ms experiment. Figure 3A shows the HDX heat map at five time points. Figure 3B shows the PF value calculated by pyhdx software. The N-terminal (1-22) and the loop region (256-261) before the tower helix have high deuterium values and low pf values, indicating that these regions are flexible or disordered in structure. In addition, we found that the deuterium value in the tower tower helices (262-276) region is low and the PF value is high, indicating that the rotation of helices may be triggered by the front-end plastic hinge region, rather than the deformation and remodeling of helices itself. These findings are consistent in the crystal structure data. In addition to these two regions, GPA and GPB basically maintain a stable overall structure. And from 1 μ In the root mean square fluctuation (RMSF) and solvent accessible surface (SASA) calculated by atomic level MD simulation (Fig. 3C), we also found that the data of these two regions are consistent with the information of hdx-ms, but some regions in MD simulation that are not consistent with hdx-ms may be related to insufficient sequence coverage.

Figure 3 (a, d) deuterium thermograms of GPB and GPA at different labeling times and mapped to the structure (pdb:1GPa). (B, e) pf values calculated based on hdx-ms data and mapped to the crystal structure. (C, f) RMSF and Sasa values in MD simulation and mapped to structure.

As can be seen from the deuterium difference diagram (FIG. 4A), four regions show a deuterium decreasing trend (red box), and many regions show a deuterium increasing trend (blue box) (GPA GPB). The variation trend of PF difference is basically consistent with that of deuterium (Fig. 4b). According to the data, the changes of N-terminal and tower tower helices indicate that phosphorylation mediated allosteric stabilizes these two regions, α 1-cap- α The dynamics of zone 2 decreased slightly. In addition, many regions (especially the region after the tower tower helices sequence) showed a decrease in PF value, indicating that compared with GPB, the regional dynamics near the catalytic site of GPA was enhanced. Next, we classify them according to the characteristics of HDX kinetic plot, and discuss in detail the changes of their regions.

Figure 4 (a) Deuterium difference diagram of GPA GPB hdx-ms. (b) Change from GPB to GPA PF.

The first is the change of N-terminal and C-terminal (Fig. 5). N-terminal residues 1-22 show deuterization decrease, which indicates that the N-terminal has certain plasticity. Affected by phosphorylation and structural changes in the N-terminal region, the C-terminal region also changed. In addition, residues 30-50 (cap region) and residues 111-117( α 4back loop) region shows deuterization decrease, while 103-109( α 4front) shows deuterium rising. According to the crystal structure, the cap region and α The deuterization change of back loop is affected by the change of N-terminal, the original residue interaction is broken, and a new residue interaction is formed. At the same time, the two regions also undergo structural rearrangement, so they show obvious deuterization change. Residues 88-99( β 2- α 3) And residues 125-141( β 3-L- α 6) Deuterium rise. In general, phosphorylation makes cap ′/α At the same time, the electrostatic interaction between phosphorylated groups and arginine residues is the main reason for the change of cap region α 1 and α 2 plays an anchoring role, and its relative position remains basically unchanged.

Figure 5 Local structure and HDX dynamic curve (c) of N-terminal and C-terminal regions of GPB (a) and GPA (b).

In addition, tower tower helices( α 7. The change of residues 262-278) is also noteworthy (Fig. 6). The 250s loop is a surface exposed area, which is not in contact with other areas. The deuterization decrease may be due to the contraction of its own structure. The decrease of deuterization of peptide 262-267 and 268-274 suggests that low turnover or strong interaction may occur in this region. Deuterium value in 280s loop region increased. These changes indicate that the change of the angle of tower tower helix not only affects the interface structure of the dimer, but also affects the surrounding region near the catalytic site. Therefore, based on the crystal structure, we speculate that phosphorylation and the change of the relative position of the N-terminal make the structure of the 250s loop shrink, thus breaking the interaction between tyr262′-pro281 and tyr262-tyr280 ‘, resulting in the relative sliding of the tower helices of the two subunits and the increase of the tilt angle.

Figure 6 The local structure of GPB (a) and GPA (b) tower helix region and HDX dynamic curve (c).

Finally, the changes of catalytic sites, PLP binding sites and glycogen storage sites (Fig. 7). Most of the regions around the catalytic sites showed an upward trend of deuterization. We speculate that as the interaction between pro281, ile165 and asn133 is broken, the interaction between arg569 and ile165, pro281 and asn133 is also broken. Therefore, the solvent accessibility of residues around the catalytic site and PLP binding site is increased, the local regional structure becomes more flexible, and the catalytic site is opened and transformed into an activated conformation. The glycogen storage site is located on the surface of GP, 30 Å from the catalytic site, except α 23 (residues 699 − 708), no significant changes were observed in the glycogen storage region by hdx-ms.

Figure 7 The local structure and HDX kinetic curve of GPB (a) and GPA (b) catalytic sites and PLP (Orange) binding sites (c).

Combined with all the above data, we speculated on the dynamic mechanism of phosphorylation regulation (flow chart 1). After phosphorylation, the interaction between N-terminal tail residues and acid patch is broken, which also leads to the ordering of N-terminal tail, the disorder of C-terminal tail and other structural changes. By forming a new salt bridge between pser14 and arg69 and arg43 ‘, the N-terminal residues were relocated, resulting in a salt bridge fracture between asp838 and his36’. With the transformation of the tertiary and quaternary structures, the 250s loop shrinks and plays a role similar to the”doorring”. When it shrinks, the interaction between tyr262 ′ -pro281 and tyr262-tyr280 ′ and the hydrogen bond between 276-279 and 162-164 regions are also broken, resulting in the relative sliding of tower helix, the breaking of the role between tower helices, and the transmission of structural changes to the catalytic site. Finally, the residues near the 280s loop, catalytic site and PLP binding site were loosened, the channels leading to the catalytic site and substrate phosphate recognition site were opened, and the enzyme was activated.

Flow chart 1 During the allosteric regulation of GP, the broken (blue) or newly formed (red) key residues interact.

In this paper, NTD MS, HDX MS, PF analysis and MD simulation were integrated to detect the structure and dynamic basis of GP phosphorylation allosteric regulation process. Through this integrated structure method, the conformational flexibility, local dynamics and the information worthy of attention in the long-range allosteric regulation of conformational change of GP were revealed. Each method has its own advantages, but it also has limitations at a certain level. We look forward to integrating hdx-ms information with computational simulation information to achieve more accurate analysis of protein structure.

Written by:luoyuxiang


Original text:insights into phosphorylation induced protein allostery and conventional dynamics of glycogen phosphorylase via integrated structural mass spectroscopy and in silicon modeling

Lihuilin research group website:


one Huang, J.;   Chu, X.;   Luo, Y.;   Wang, Y.;   Zhang, Y.;   Zhang, Y.; Li, H., Insights into Phosphorylation-Induced Protein Allostery and Conformational Dynamics of Glycogen Phosphorylase via Integrative Structural Mass Spectrometry and In Silico Modeling. ACS Chem. Biol. 2022.

2. Li, H.;  Nguyen, H. H.;  Ogorzalek Loo, R. R.;  Campuzano, I. D. G.; Loo, J. A., An integrated native mass spectrometry and top-down proteomics method that connects sequence to structure and function of macromolecular complexes. Nat. Chem. 2018, 10 (2), 139-148.

3. Li, H.;  Wongkongkathep, P.;  Van Orden, S. L.;  Ogorzalek Loo, R. R.; Loo, J. A., Revealing ligand binding sites and quantifying subunit variants of noncovalent protein complexes in a single native top-down FTICR MS experiment. J. Am. Soc. Mass Spectrom. 2014, 25 (12), 2060-8.