Estimating Inter-protein Pairwise Interaction ... - ACS Publications

Subscriber access provided by University of Winnipeg Library

Letter

Estimating Inter-protein Pairwise Interaction Energies in Cell Lysates from a Single Native Mass Spectrum Jelena Cveticanin, Ravit Netzer, Galina Arkind, Sarel J. Fleishman, Amnon Horovitz, and Michal Sharon Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b02349 • Publication Date (Web): 14 Aug 2018 Downloaded from http://pubs.acs.org on August 18, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Estimating Inter-protein Pairwise Interaction Energies in Cell Lysates from a Single Native Mass Spectrum

Jelena Cveticanin2, Ravit Netzer2, Galina Arkind2, Sarel J. Fleishman2 Amnon Horovitz1,* and Michal Sharon2,*

Affiliations: Departments of 1Structural Biology and 2Biomolecular Sciences, Weizmann Institute of Science, Rehovot 761001 Israel

*Correspondence to: Amnon Horovitz

Michal Sharon

[email protected]

[email protected]

Tel: 972-8-9343399

Tel: 972-8-9343947

1

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract A powerful method to determine the energetic coupling between amino acids is doublemutant cycle analysis.

In this method, two residues are mutated separately and in

combination and the energetic effects of the mutations are determined. A deviation of the effect of the double mutation from the sum of effects of the single mutations indicates that the two residues are interacting directly or indirectly. Here, we show that doublemutant cycle analysis by native mass spectrometry can be carried out for interactions in crude Escherichia coli cell extracts, thereby obviating the need for protein purification and generating binding isotherms. Our results indicate that inter-molecular hydrogen bond strengths are not affected by the more crowded conditions in cell lysates.

2


Page 2 of 15

Page 3 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Reactions in vivo take place under conditions of macromolecular crowding, i.e. in the presence of a high weight per volume concentration of ‘background’ macromolecules that are not directly involved in the reaction of interest.1 Given that such crowding can impact the rates and equilibria of reactions, much effort has been devoted in recent years to thermodynamic and kinetic characterization of biochemical reactions in intact cells, cell lysates or in the presence of macromolecules that may mimic crowding in the cell.2 Little is known, however, about whether crowding affects the strength of the various types of pairwise interactions (e.g. hydrogen bonds, salt-bridges, etc.) that govern complex formation. A classical approach for determining the strengths of pairwise interactions involves invoking the double-mutant cycle (DMC) method.3,4 In this method, two residues of interest are mutated separately and together and the energetic effects of the mutations on some process such as binding or folding are measured. The deviation of the effect of the double mutation from the sum of effects of the two single mutations indicates that an interaction, either direct or indirect, exists between the two residues. The deviation from additivity, termed the coupling energy, can often be used to estimate the strength of intermolecular pairwise interactions in proteins or protein-ligand complexes.5-7 A DMC for detecting and quantifying an inter-protein pairwise interaction between residue i in protein X and residue j in protein Y consists of the four possible complexes that can be formed by the two wild-type proteins, Xi and Yj, and the two corresponding single mutants X0 and Y0 (Figure 1). The coupling energy between residues i and j can be expressed in terms of binding constants as follows: ∆2Gint(ij) = RTln(KijK00/Ki0K0j)

(1)

where Kij is the binding constant for the two wild-type proteins, Xi and Yj, K00 is the binding constant for the two corresponding single mutants X0 and Y0 and Ki0 and Koj are the respective binding constants for Xi with Y0 and X0 with Yj. Recently, we described a native mass-spectrometry (MS) approach for DMC analysis that circumvents the need for the determination of the individual binding constants and provides a

3



way to measure ∆2Gint(ij) directly from a single spectrum.8 This approach is based on the realization that substituting each of the binding constants in eq 1 by the appropriate ratio of free protein and complex concentrations (e.g. Kij = [XiYj]/[Xi][Yj]) leads to the following expression: ∆2Gint(ij) = RTln([XiYj][X0Y0]/[XiY0][X0Yj])

(2)

where the concentrations of all the free species cancel out. Hence, by mixing the two wildtype proteins, Xi and Yj, and the two corresponding single mutants X0 and Y0 and then measuring the peak areas corresponding to all four possible co-existing complexes using native MS, one can obtain a direct estimate of ∆2Gint(ij). Importantly, this procedure does not require knowing the concentrations of the free species ([Xi], [Yj], [X0] and [Y0]). This MS approach for DMC analysis is, therefore, less labor-intensive because it does not involve generating binding isotherms for the four complexes in the cycle using methods such as isothermal calorimetry or surface plasmon resonance. It is also less error-prone because the estimate it provides for the value of the coupling energy, ∆2Gint(ij), is not affected by errors in protein concentration determination. Remarkably, we show here that DMC analysis by native MS can be carried out using crude cell lysates, thereby providing estimates for coupling energies under more crowded conditions (i.e. 10-100 fold more crowded than in typical buffer conditions but 100-fold less than in vivo) and also obviating the need for any protein purification. Native MS is a method based on the ability to transfer protein complexes to the gas phase without disrupting them.9-12 Recently, we showed that native MS analysis can be carried out for proteins in crude Escherichia coli cell lysates without their purification,13 thereby enabling us to carry out DMC analysis in cell lysates. The two wild-type proteins and the two mutant proteins that combine to form the four different complexes comprising each cycle were, therefore, co-expressed in E. coli using the pRSFDuet-1 expression plasmid. The relative concentrations in a cell lysate of the four co-existing complexes comprising each cycle were then determined from a single mass spectrum obtained using an Orbitrap EMR platform.

Determining the concentrations of the complexes (without knowing the

concentrations of the individual proteins) allowed us to calculate the coupling energies 4


Page 4 of 15

Page 5 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(according to eq 2) in the cell lysates. We also determined the coupling energies using purified proteins in a buffer solution as a control. Results and Discussion Three DMCs were constructed for measuring coupling energies between residues in designed versions,14 termed des2 and des3, of the 1:1 complex between E2 colicin endonuclease (colE) and the Im2 immunity protein (Im)15 (Figure S1). The des2 and des3 complexes are part of a recently designed network of interacting pairs of homologous proteins in which some proteins are specific (i.e. they bind strongly to only a few targets) and others are promiscuous.

In all three cycles, all the mutations were to alanine.

One cycle was

constructed for the interaction between Gln34 in Im and Asn78 in colE of the des3 version of the complex (Figure 1A). The crystal structure of this complex (PDB accession code 6ERE) indicates that there is a hydrogen bond between the Nε atom of Gln34 and the Oδ atom of Asn78, which are separated by about 3.0 Å (Figure 1A). A second cycle was constructed for the interaction between Asn31 in Im and Asn83 in colE of the des3 version of the complex (Figure 1B). The crystal structure indicates that the Nδ atom of Asn31 and the Oδ atom of Asn83 are separated by about 2.7 Å and form a hydrogen bond (Figure 1B). A third cycle was constructed for the interaction between Gln31 in Im and Asn78 in colE of the des2 version of the complex (Figure 1C). A computational model14 indicates that these residues form a double hydrogen bond involving the Nε atom of Gln31 with the Oδ atom of Asn78 and the Oε atom of Gln31 with the Nδ atom of Asn78, which are separated by 2.7 and 3.1 Å, respectively (Figure 1C). The charge series of the individual complexes forming each of the cycles could be well resolved in the crude lysates13 due to the overproduction of these proteins in comparison to the endogenous bacterial proteins and because of reduced binding of adducts (Figure S2). In parallel, spectra of the four purified complexes in each cycle were also collected as controls (Figures 2, 3 and S3). The coupling energies in the lysates for the interactions of Gln34 in Imdes3 with Asn78 in colE des3 (Figure 2) and Asn31 in Imdes3 with Asn83 in colEdes3 (Figure S3) were found to be 0.11 (± 0.05) and -0.14 (± 0.03) kcal mol-1, respectively. Very similar values of 0.11 (± 0.07) and -0.02 (± 0.02) kcal mol-1 were determined for these respective interactions using purified proteins in buffer. The coupling energy in the lysate

5



Page 6 of 15

for the interaction of Gln31 in Imdes2 with Asn78 in colEdes2 (Figure 3) was found to be 1.16 (± 0.09) kcal mol-1. A very similar value of 1.13 (± 0.10) kcal mol-1 was determined for this interaction using purified proteins in buffer. The values of the coupling energies determined for all three pairwise interactions are, therefore, found to be very similar in lysates and in buffer, thereby indicating that crowding has little effect on the strength of inter-molecular hydrogen bonds. There is still uncertainty regarding the strengths of hydrogen bonds and their contribution to protein stability and complex formation. The total number of hydrogen bonds usually does not change during folding or binding since intra- or inter-protein hydrogen bonds replace those made with water. It has, therefore, been argued that the enthalpic contribution of hydrogen bonding to protein stability is small if any16. The enthalpy of a hydrogen bond depends, however, on whether it is in a low-dielectric medium (e.g. the protein interior) or in a less favorable high-dielectric medium such as bulk water. Entropic effects also determine the strengths of hydrogen bonds.

These include a favorable

contribution due to the release of protein-bound water molecules into bulk solution and a penalty that arises from the loss of side-chain configurational entropy.

Multi-dentate

hydrogen bonds are, therefore, cooperative and contribute to stability because formation of one hydrogen bond reduces the configurational entropy price of forming the other hydrogen bonds. The above-mentioned enthalpic and entropic effects account for the strengths of the hydrogen bonds analyzed in this work. The hydrogen bonds of Gln34 in Imdes3 with Asn78 in colE des3 (Figure 1A) and Asn31 in Imdes3 with Asn83 in colEdes3 (Figure 1B) are solventexposed and mono-dentate and, therefore, contribute little to complex stability. By contrast, the hydrogen bonds between Gln31 in Imdes2 and Asn78 in colEdes2 contribute about 1 kcal mol-1 to complex stability because they are more buried and bi-dentate (Figure 1C). It is noteworthy that designing such a stabilizing hydrogen bond is not trivial because of the various above-mentioned energetic tradeoffs that determine the net strength of such an interaction. Importantly, given that the values of the coupling energies are similar in lysate and buffer, our findings indicate that weak crowding has little effect on the factors that govern the strength of inter-protein hydrogen bonds.

6


Page 7 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Acknowledgements AH is an incumbent of the Carl and Dorothy Bennett Professorial Chair in Biochemistry. MS is grateful for the support of a Starting Grant from the European Research Council (ERC) (Horizon 2020)/ERC Grant Agreement no. 636752 and a Israel Science Foundation grant no. 300/17. MS is an incumbent of the Aharon and Ephraim Katzir Memorial Professorial Chair.

Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: Supporting Information: Methods of molecular biology, protein expression and purification, mass spectrometry analysis, assignment of peaks and area calculations and supporting figures.

7



References (1) Rivas, G.; Minton, A. P. Trends Biochem Sci 2016, 41, 970-981. (2) Phillip, Y.; Schreiber, G. FEBS Lett 2013, 587, 1046-1052. (3) Carter, C. W., Jr. Annu Rev Biophys 2017, 46, 433-453. (4) Horovitz, A. Fold Des 1996, 1, R121-126. (5) Hidalgo, P.; MacKinnon, R. Science 1995, 268, 307-310. (6) Post, M. R.; Lester, H. A.; Dougherty, D. A. Biochemistry 2017, 56, 1836-1840. (7) Schreiber, G.; Fersht, A. R. J Mol Biol 1995, 248, 478-486. (8) Sokolovski, M.; Cveticanin, J.; Hayoun, D.; Korobko, I.; Sharon, M.; Horovitz, A. Nat Commun 2017, 8, 212. (9) Chen, F.; Gulbakan, B.; Weidmann, S.; Fagerer, S. R.; Ibanez, A. J.; Zenobi, R. Mass Spectrom Rev 2016, 35, 48-70. (10) Liko, I.; Allison, T. M.; Hopper, J. T.; Robinson, C. V. Curr Opin Struct Biol 2016, 40, 136-144. (11) Lossl, P.; van de Waterbeemd, M.; Heck, A. J. EMBO J 2016, 35, 2634-2657. (12) Sharon, M. Science 2013, 340, 1059-1060. (13) Gan, J.; Ben-Nissan, G.; Arkind, G.; Tarnavsky, M.; Trudeau, D.; Noda Garcia, L.; Tawfik, D. S.; Sharon, M. Anal Chem 2017, 89, 4398-4404. (14) Netzer, R.; Listov, D.; Dym, O.; Albeck, S.; Knop, O.; Kleanthous, C.; Fleishman, S. J. submitted 2018. (15) Wojdyla, J. A.; Fleishman, S. J.; Baker, D.; Kleanthous, C. J Mol Biol 2012, 417, 79-94. (16) Fersht, A. R. Trends Biochem Sci 1987, 12, 301-304.

8


Page 8 of 15

Page 9 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figures

Fig. 1. Double-mutant cycles for measuring the strengths of inter-protein hydrogen bonds. A doublemutant cycle (left) was constructed for each of the following three pairs of residues that form inter-molecular hydrogen bonds (right): (A) N78 and Q34 in the des3 version of E2 colicin endonuclease (colE) and Im immunity protein, respectively; (B) N83 and N31 in the des3 version of ColE and Im, respectively; and (C) N78 and Q31 in the des2 version of ColE and Im, respectively. Each double-mutant cycle consists of the four possible complexes that can be formed by the two wild-type proteins and the two corresponding single mutants. The hydrogen bonds in the crystal structure (PDB accession code 6ERE) and model14 of the respective des3 and des2 complexes were drawn using PyMOL.

9



Fig. 2. Determination of the interaction energy of Asn78 in colEdes3 with Gln34 in Imdes3 using native MS. Representative mass spectra of the +10 charge state are shown for the purified complexes

colE Im

(pink), colE Im (purple), colE Im (blue), and colE Im (green) in buffer. Also shown are the spectra of the four co-existing complexes that form (i) when these purified wildtype and mutant versions of colEdes3 and Imdes3 are mixed together in buffer; or (ii) in a crude cell lysate after co-expression and without purification. Shown under the spectra are the calculated values of the coupling energies. The values ± SD are averages calculated from at least four independent experiments.

10


Page 10 of 15

Page 11 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 3. Determination of the interaction energy of Asn78 in colEdes2 with Gln31 in Imdes2 using native MS. Representative mass spectra of the +10 charge state are shown for the purified complexes colE Im (blue), colE Im (green), colE Im (pink), and colE Im (purple) in buffer. Also shown are the spectra of the four co-existing complexes that form (i) when these purified wildtype and mutant versions of colEdes2 and Imdes2 are mixed together in buffer; or (ii) in a crude cell lysate after co-expression and without purification. Shown under the spectra are the calculated values of the coupling energies. The values ± SD are averages calculated from at least four independent experiments.

11



figure 1 201x223mm (300 x 300 DPI)


Page 12 of 15

Page 13 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


figure 2 206x228mm (300 x 300 DPI)



figure 3 207x249mm (300 x 300 DPI)


Page 14 of 15

Page 15 of 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC 79x42mm (300 x 300 DPI)


Estimating Inter-protein Pairwise Interaction ... - ACS Publications

Recommend Documents