Communication pubs.acs.org/JACS
Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Characterization of Residue Specific Protein Folding and Unfolding Dynamics in Cells Xiangfei Song,†,‡,§ Tianhang Lv,†,‡,§ Jingfei Chen,†,‡ Jia Wang,†,‡,§ and Lishan Yao*,†,‡ †
Key Laboratory of Biofuels and ‡Shandong Provincial Key Laboratory of Synthetic Biology, Qingdao Institute of Bioenergy and Bioprocess Technology, Chinese Academy of Sciences, Qingdao 266101, China § University of Chinese Academy of Sciences, Beijing 100049, China
Downloaded via KEAN UNIV on July 17, 2019 at 11:13:17 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
(Figure S1b). Two mutants, L5V-K10H-K19E-F30L (MutX) and L5V-K10H-T18A-K19E (MutY) were generated (Figure 1a). The 1H−15N HSQC spectrum of each mutant is similar to
ABSTRACT: In this work, we measured the millisecond residue specific protein folding and unfolding dynamics in E. coli cells for two protein GB3 mutants using NMR. The results show that the protein folding and unfolding dynamics in cells is different from that in buffer. Through a two-site exchange model, it is shown that both the population and the exchange rate are changed by the E. coli cellular environment. Further investigation suggests that the change is likely due to the quinary interaction with crowded molecules in the cell. Our work underlines the importance of cellular environment to protein folding kinetics and thermodynamics although this environmental effect may not be large enough to change the protein structure.
T
he interior of a cell is crowded with biomolecules, metabolites, and ions which can affect protein folding1−6 and binding properties.7,8 The change of these properties is generally attributed to the excluded volume effect9 or the quinary interaction.10,11 The former is mainly an entropic effect caused by the steric repulsion of molecules, whereas the latter is a weak interaction with the cellular environment.3,12−14 The 3D structures of two well folded proteins, protein GB1 and TTHA1718, have been solved in cells and are essentially the same as those in the buffered solution.15−17 In a few other studies,18−20 folded proteins display fingerprint NMR 1H−15N heteronuclear single quantum coherence (HSQC) spectra in cells similar to those in buffer, suggesting that the overall protein structures are unchanged, although conformational distribution differences have been observed for disordered proteins in cells.21 Although the protein stability and folding thermodynamics in cells have been studied,2−6 reporting on the residue specific folding dynamics is rare. Conformational dynamics is an intrinsic property of proteins.22 Differences and similarities of protein dynamics have been observed in cell lysate compared to that in buffer.23−25 But the real cellular environment is far more complex. In this work, we attempt to study the residue specific folding and unfolding dynamics of a model protein, GB3,26 in E. coli cells. GB3 is a small protein that binds to the constant region of IgG. The wild-type GB3 in E. coli shows a poor NMR 1 H−15N HSQC spectrum (Figure S1a). The in-cell spectrum is dramatically improved after mutating K19 to a glutamate (E) © XXXX American Chemical Society
Figure 1. Mutants MutX and MutY maintain the overall structure of GB3 but create the millisecond conformational dynamics. (a) Mutated side chains in the GB3 structure (pdb 2OED). (b) Fitting of experimental backbone 1H−15N RDCs of the two mutants to the GB3 wild-type structure. (c) Global fitting of 15N CPMG profiles for D40 at 600 and 900 MHz magnetic fields. (d) Absolute value of Δω for residues (38 for MutX and 33 for MutY) displaying millisecond conformational dynamics. The values of Δω are listed in Tables S1 and S2.
that of the wild-type GB3 although differences in certain regions can be seen (Figure S2). The mutants do not change the overall backbone structure of the protein, as confirmed by the good fitting of their backbone N−H residual dipolar couplings (RDCs) to the wild-type GB3 structure (Figure 1b).27 Unlike the wild-type protein, the two mutants display millisecond dynamics, which can be detected by the 15N Carr− Purcell−Meiboom−Gill (CPMG) NMR relaxation dispersion experiment (Figure 1c).28,29 The dynamics of the proteins in a buffered solution (90 mM bis−tris propane/90 mM HEPES, at Received: April 24, 2019
A
DOI: 10.1021/jacs.9b04435 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Communication
Journal of the American Chemical Society
The in-cell spectra of the two mutants, albeit with lower signal-to-noise (s/n) ratios and broader peaks due to high viscosity and quinary interactions, agree well with those in the buffered solution (Figure 2a), implying that the overall protein structure of the folded state A is unchanged in E. coli cells. The pH inside cells was measured indirectly, by using the H10 backbone 15N chemical shift, which is correlated with pH (Figure S5),12 yielding a value of 6.3, the same as that in buffer. The Rex was determined for 24 residues of MutX and 17 residues of MutY in E. coli cells using the CPMG experiment. The results show that Rex is considerably larger than that in the buffer for both proteins (Figure 2b), indicating that the cellular environment has modulated the folding and unfolding dynamics. The fitting of the CPMG data to the two-site exchange model confirms that the population of the unfolded state increases from 2.6% (1.6%) to 3.7% (2.8%) for MutX (MutY) (Table 1). Furthermore, the forward rate k1 deceases by ∼22% for MutX and remains unchanged for MutY, whereas the reverse rate k−1 decreases by ∼45% for both proteins. Apparently, both thermodynamic and kinetic properties are perturbed by the cellular environment. Essentially, the dynamics is promoted by shifting the protein population toward the unfolded state, whereas the kinetics is hindered by decreasing the exchange rates. When comparing the Δω values in cells to those in the buffer, a linear correlation was observed for MutX and MutY (Figure 2c) although the measurement error for proteins in cells is larger (Figure 2c,d). This correlation implies that the cellular environmental effect on the chemical shift difference between the folded and unfolded states is small. In other words, the unfolded state conformation is unchanged in cells, just like the folded state. The preservation of the protein conformation is consistent with the published studies.15−17 The observed in-cell folding dynamics change might be caused by the excluded volume effect due to crowding of macromolecules or quinary interactions with molecules in E. coli cells. To probe the excluded volume effect on dynamics, the 15N CPMG experiment was performed again for MutX and MutY in the buffer with the addition of a polyglucan, dextran 70 (a final concentration of 100 g/L), which decreases the
pH = 6.3 to match that in cells, see discussion below) was characterized by Rex, the conformational exchange contribution to the backbone amide 15N transverse relaxation rate R2, which was extracted from the CPMG data. A total of 38 and 33 backbone amides show conformational exchanges for MutX and MutY, respectively. Using a two-site exchange model,28 A ⇔ B, the data were fitted, yielding the following parameters (Table 1 and Figure 1d): the populations of two Table 1. Folding Dynamics Parameters for MutX and MutY in Different Environments k1 (s−1)
k−1 (s−1)
buffer dextran lysate cell
55 58 63 43
± ± ± ±
3 6 6 3
2084 2384 1779 1103
buffer dextran lysate cell
32 33 50 31
± ± ± ±
2 2 5 4
2015 2174 2110 1084
MutX ± 90 ± 119 ± 124 ± 90 MutY ± 77 ± 70 ± 129 ± 148
pA (%)
pB (%)
97.4 97.6 96.6 96.3
± ± ± ±
0.1 0.1 0.1 0.2
2.6 2.4 3.4 3.7
± ± ± ±
0.1 0.1 0.1 0.2
98.4 98.5 97.7 97.2
± ± ± ±
0.1 0.1 0.1 0.2
1.6 1.5 2.3 2.8
± ± ± ±
0.1 0.1 0.1 0.2
conformations, pA and pB (pA + pB = 1), the forward (A to B) and reverse (B to A) exchange rates, k1 and k−1, and the absolute value of chemical shift difference between the two conformations, Δω. The robustness of the fitting is confirmed by systematically changing pB or kex (kex = k1 + k−1) where a single minimum is identified (Figure S3). The results show that the population of A is 97.4% and 98.4% for MutX and MutY, respectively. Thus, A is the major conformation, whereas B is the minor conformation. To further investigate this exchange process, the 15N chemical shifts of the unfolded protein (MutY) were measured (Figure S2d). Δω is linearly correlated with the chemical shift difference between A and the unfolded state (in 8 M urea) with a slope of 0.85 ± 0.04 (Figure S4). It appears that the minor state B is the unfolded state. Thus, the conformational exchange corresponds to the protein folding and unfolding process.
Figure 2. Conformational dynamics of MutX and MutY in E. coli cells. Overlay of 1H−15N HSQC spectra (a), comparison of Rex at the 600 MHz field (b), and correlation of the absolute value of Δω (c) for the two proteins in cells and in the buffer. In panel b, only residues that display dynamics in cells are drawn for the direct comparison. (d) Fitting of 15N CPMG profiles for D40 in cells at 600 and 900 MHz magnetic fields. B
DOI: 10.1021/jacs.9b04435 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Communication
Journal of the American Chemical Society
The quinary interaction was also probed using the 15N R1R2 rate product.30,31 R2 was derived from the R1ρ measurement where the spin lock effectively eliminates the exchange contribution (Figure S6). R1R2 increases in the order of dextran < lysate < cell (Figure S7), indicating that the quinary interaction increases in the same order. Apparently, quinary interactions with macromolecules can be an important factor for the folding dynamics change in cells. It is well-known that the interior of a cell is far more crowded and complicated than a buffered solution where biophysical and biochemical experiments are typically performed. Here, we show that the folding dynamics of GB3 mutants in E. coli cells is different from that in buffer. The quinary interaction is likely responsible for the dynamics change, although the interaction is not strong enough to change the protein structure. Our work emphasizes that one has to take the cellular environment into consideration to understand protein folding dynamics and thus functions in cells.
accessible volume for the proteins. It appears that the excluded volume effect has a very minor effect on the dynamics of both proteins (Table 1). The unfolded state population decreases by 0.2% for MutX and 0.1% for MutY. The forward rate changes slightly, and the reverse rate increases by ∼14% (MutX) and ∼8% (MutY). The role of quinary interaction on protein folding dynamics was probed by measuring the 15N CPMG rate in the buffer and in the presence of 100 g/L E. coli cell lysate, which only includes soluble macromolecules (greater than ∼3.5 kDa). The results show that the population of the unfolded state B in lysate increases to 3.4% and 2.3% for MutX and MutY, respectively (Table 1), similar to the change in cells. But when comparing the exchange rates, differences appear: k1 increases in lysate but decreases in cells; k−1 decreases for MutX but increases slightly for MutY in lysate, while k−1 decreases for both proteins in cells. It is worth noting that both excluded volume effects and quinary interactions exist in lysate. But in dextran where only excluded volume effects are present, the change is much smaller. These results highlight the importance of quinary interactions to folding dynamics. The effect on the rate and population can be better illustrated in a free energy diagram (Figure 3). The folding free energy was calculated
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.9b04435. Experimental methods, HSQC spectra, χ2 as function of kex and pB in CPMG fitting, correlation between Δω and Δω1, correlation of 15N shift and pH, relaxation rates, and absolute values of Δω (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*
[email protected] ORCID
Figure 3. Folding free energy (kJ/mol) landscapes of the two proteins in different environments. The folded state A is used as the reference state (TS, the transition state; B, the unfolded state). The relative free energies of the proteins in the buffer, dextran (100 g/L), lysate (100 g/L), or cells were calculated from the rates and populations in Table 1. The free energy error is ∼0.2 kJ/mol.
Lishan Yao: 0000-0003-1797-922X Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by National Key R&D Program of China (Grant No. 2017YFA0505400 to L.Y.), the National Natural Science Foundation of China (Grant Nos. 21773280 and 31661143036 to L.Y.), Shandong Provincial Natural Science Foundation (Grant No. ZR2018ZB0207 to L.Y.), and the Taishan Scholars Program of Shandong Province. A portion of this work was performed on the National Facility for Protein Science Shanghai, Zhangjiang Lab. The authors thank Zhan Wei for arranging the NMR machine time and Zhijun Liu for helping set up the CPMG experiment.
from the population ratio, whereas the barrier was calculated using the Eyring equation. Using the folded state A as the reference state, the free energy of the unfolded state B decreases in cells and in lysate but increases slightly in dextran for both proteins. The exchange barriers show a more complex picture. For both the forward and reverse exchanges, the barriers increase in cells. The barrier changes are very likely caused by the quinary interaction with states A and B, rather than that with the transition state. In other words, the quinary interaction in cells stabilizes A and B states to slow down the exchange. The transient nature of quinary interaction implies that its effect on the transition state with a short lifetime is probably small. In lysate, the forward exchange barrier decreases for both proteins, whereas the reverse exchange barrier increases for MutX but decreases slightly for MutY. The changes of the barriers in lysate are different from those in cells. It is known that the composition of lysate is simpler than that of a real cell. For example, it has been suggested that RNAs may play an important role in quinary interactions.19 But in the preparation of E. coli cell lysate, many RNAs are lost due to their instability. Using lysate as the mimic, only the population shift is captured, suggesting that direct measurement in cells is needed to study the cellular dynamic effect.
■
REFERENCES
(1) Latham, M. P.; Kay, L. E. A Similar In Vitro and In Cell Lysate Folding Intermediate for the FF Domain. J. Mol. Biol. 2014, 426, 3214−3220. (2) Danielsson, J.; Mu, X.; Lang, L.; Wang, H. B.; Binolfi, A.; Theillet, F. X.; Bekei, B.; Logan, D. T.; Selenko, P.; Wennerstrom, H.; Oliveberg, M. Thermodynamics of protein destabilization in live cells. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 12402−12407. (3) Monteith, W. B.; Cohen, R. D.; Smith, A. E.; Guzman-Cisneros, E.; Pielak, G. J. Quinary structure modulates protein stability in cells. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 1739−1742. (4) Smith, A. E.; Zhou, L. Z.; Gorensek, A. H.; Senske, M.; Pielak, G. J. In-cell thermodynamics and a new role for protein surfaces. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 1725−1730. C
DOI: 10.1021/jacs.9b04435 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Communication
Journal of the American Chemical Society (5) Gruebele, M.; Dave, K.; Sukenik, S. Globular Protein Folding In Vitro and In Vivo. Annu. Rev. Biophys. 2016, 45, 233−251. (6) Sukenik, S.; Salam, M.; Wang, Y. H.; Gruebele, M. In-Cell Titration of Small Solutes Controls Protein Stability and Aggregation. J. Am. Chem. Soc. 2018, 140, 10497−10503. (7) Guseman, A. J.; Pielak, G. J. Cosolute and Crowding Effects on a Side-By-Side Protein Dimer. Biochemistry 2017, 56, 971−976. (8) Phillip, Y.; Kiss, V.; Schreiber, G. Protein-binding dynamics imaged in a living cell. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 1461− 1466. (9) Reiss, H.; Frisch, H. L.; Lebowitz, J. L. Statistical Mechanics of Rigid Spheres. J. Chem. Phys. 1959, 31, 369−380. (10) Anfinsen, C. B. Principles That Govern Folding of Protein Chains. Science 1973, 181, 223−230. (11) Mcconkey, E. H. Molecular Evolution, Intracellular Organization, and the Quinary Structure of Proteins. Proc. Natl. Acad. Sci. U. S. A. 1982, 79, 3236−3240. (12) Cohen, R. D.; Guseman, A. J.; Pielak, G. J. Intracellular pH modulates quinary structure. Protein Sci. 2015, 24, 1748−1755. (13) Kyne, C.; Ruhle, B.; Gautier, V. W.; Crowley, P. B. Specific ion effects on macromolecular interactions in Escherichia coli extracts. Protein Sci. 2015, 24, 310−318. (14) Zhang, N.; An, L.; Li, J.; Liu, Z.; Yao, L. Quinary Interactions Weaken the Electric Field Generated by Protein Side-Chain Charges in the Cell-like Environment. J. Am. Chem. Soc. 2017, 139, 647−654. (15) Sakakibara, D.; Sasaki, A.; Ikeya, T.; Hamatsu, J.; Hanashima, T.; Mishima, M.; Yoshimasu, M.; Hayashi, N.; Mikawa, T.; Walchli, M.; Smith, B. O.; Shirakawa, M.; Guntert, P.; Ito, Y. Protein structure determination in living cells by in-cell NMR spectroscopy. Nature 2009, 458, 102−105. (16) Muntener, T.; Haussinger, D.; Selenko, P.; Theillet, F. X. InCell Protein Structures from 2D NMR Experiments. J. Phys. Chem. Lett. 2016, 7, 2821−2825. (17) Pan, B. B.; Yang, F.; Ye, Y. S.; Wu, Q.; Li, C. G.; Huber, T.; Su, X. C. 3D structure determination of a protein in living cells using paramagnetic NMR spectroscopy. Chem. Commun. 2016, 52, 10237− 10240. (18) Bertrand, K.; Reverdatto, S.; Burz, D. S.; Zitomer, R.; Shekhtman, A. Structure of Proteins in Eukaryotic Compartments. J. Am. Chem. Soc. 2012, 134, 12798−12806. (19) Majumder, S.; Xue, J.; DeMott, C. M.; Reverdatto, S.; Burz, D. S.; Shekhtman, A. Probing Protein Quinary Interactions by In-Cell Nuclear Magnetic Resonance Spectroscopy. Biochemistry 2015, 54, 2727−2738. (20) Luchinat, E.; Banci, L. In-cell NMR: a topical review. IUCrJ 2017, 4, 108−118. (21) Banci, L.; Barbieri, L.; Bertini, I.; Luchinat, E.; Secci, E.; Zhao, Y. G.; Aricescu, A. R. Atomic-resolution monitoring of protein maturation in live human cells by NMR. Nat. Chem. Biol. 2013, 9, 297−299. (22) Lewandowski, J. R.; Halse, M. E.; Blackledge, M.; Emsley, L. Direct observation of hierarchical protein dynamics. Science 2015, 348, 578−581. (23) Latham, M. P.; Kay, L. E. Is Buffer a Good Proxy for a Crowded Cell-Like Environment? A Comparative NMR Study of Calmodulin Side-Chain Dynamics in Buffer and E. coli Lysate. PLoS One 2012, 7, No. e48226. (24) Latham, M. P.; Kay, L. E. Probing non-specific interactions of Ca2+-calmodulin in E. coli lysate. J. Biomol. NMR 2013, 55, 239−247. (25) Latham, M. P.; Sekhar, A.; Kay, L. E. Understanding the mechanism of proteasome 20S core particle gating. Proc. Natl. Acad. Sci. U. S. A. 2014, 111, 5532−5537. (26) Derrick, J. P.; Wigley, D. B. The 3rd IgG-binding Domain from Streptococcal Protein-G - an Analysis by X-ray Crystallography of the Structure Alone and in a Complex with Fab. J. Mol. Biol. 1994, 243, 906−918. (27) Ulmer, T. S.; Ramirez, B. E.; Delaglio, F.; Bax, A. Evaluation of backbone proton positions and dynamics in a small protein by liquid crystal NMR spectroscopy. J. Am. Chem. Soc. 2003, 125, 9179−9191.
(28) Palmer, A. G.; Kroenke, C. D.; Loria, J. P. Nuclear magnetic resonance methods for quantifying microsecond-to-millisecond motions in biological macromolecules. Methods Enzymol. 2001, 339, 204−238. (29) Long, D.; Liu, M. L.; Yang, D. W. Accurately probing slow motions on millisecond timescales with a robust NMR relaxation experiment. J. Am. Chem. Soc. 2008, 130, 2432−2433. (30) Kneller, J. M.; Lu, M.; Bracken, C. An effective method for the discrimination of motional anisotropy and chemical exchange. J. Am. Chem. Soc. 2002, 124, 1852−1853. (31) Wang, Y. Q.; Li, C. G.; Pielak, G. J. Effects of Proteins on Protein Diffusion. J. Am. Chem. Soc. 2010, 132, 9392−9397.
D
DOI: 10.1021/jacs.9b04435 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
