Using mass spectrometry for probing protein structure & protein-protein interaction

1. Using mass spectrometry forprobing protein structure & protein-protein interaction Katalin Medzihradszky PC 204

2. I will cover • Surface-labeling studies • Active-site labeling studies • H/D exchange studies • Direct measurement of protein complexes

3. Protein surface labeling • Protein surface topology-probing by selective chemical modification and mass spectrometric peptide mapping. Suckau D, Mak M, Przybylski M. Proc Natl Acad Sci U S A. 1992 Jun 15;89(12):5630-4. → acetylation of Lys → 1,2-cyclohexanedione-derivatization of Arg

4. Issues • solubility of the label • solubility of the protein • labeling efficiency • side reactions • a delicate balance between labeling and denaturing the protein • finding the labels, and nail the sites

5. Active-site/binding-site mapping • with a highly specific probe → binding specificity obviously is required • specificity in reactivity may be important • it should be highly reactive too • producing probes in situ? • efficiency is an issue • finding the label and nail the sites could be tricky

6. Active-site/binding-site mapping • Trifluoromethyldiazirinylphenyldiazenes:  New Hemoprotein Active-Site Probes. Richard A. Tschirret-Guth, Katalin F. Medzihradszky, and Paul R. Ortiz de Montellano J. Am. Chem. Soc., 1999, 121 (20), pp 4731–4737

7. A cautionary tale: each method measures only what it was designed to detect Project: identifying binding site residues of heme-coordinating proteins by linking them to the porphyrin ring Step 1 Oxidative shift Step 2. activate the UV-reactive group on the aromatic ring → forms radical & binds to amino acid residues nearby

8. Probes used

9. Everything is hunky-dory UV spectrum of native Mb ( ), the Mb Fe−aryl complex (- - -), and the Mb Fe−aryl complex after photolysis (− −):  (A) complex formed with (meta-azidophenyl)diazene and (B) complex formed with (para-azidophenyl)diazene.

10. What did mass spectrometry find ? • On the protein level – expected MW before the oxidative shift; multiple side chain oxidations detected after that; complete mess after UV-activation • On the peptide level – all kind of unexpected side-reactions; porphyrin-ring fragment-attachments; proton extraction i.e. double bond formation – all over the protein → based on UV-absorbance labeled fractions were collected

11. Post source decay (PSD) mass spectrum of labeled peptides (A) Val17-Arg31, (B) His64-Lys77, (C) Ala57-Lys77; the modified fragments are labeled with an asterisk.

12. Side view of the distal side of the active site of the Mb-phenyl complex. The phenyl group bound to the iron is shown, and the residues are labeled.

13. An antibiotic factory caught in action Adrian T Keatinge-Clay, David A Maltby, Katalin F Medzihradszky, Chaitan Khosla & Robert M Stroud Nature Structural & Molecular Biology11, 888 - 893 (2004).

14. It took almost the entire crystal to get this spectrum…

15. Application of amide proton exchange mass spectrometry for the study of protein-protein interactions. Mandell JG, Baerga-Ortiz A, Croy CH, Falick AM, Komives EA. Curr Protoc Protein Sci. 2005 Jun;Chapter 20:Unit20.9.

16. Flow chart diagram of the two types of amide proton exchange experiments used to study protein-protein interfaces. • In the on-exchange experiment, the protein-protein complex shows a region in which less deuterium is incorporated when compared with control samples in which each protein is deuterated separately. • In the off-exchange experiment, each protein is allowed to incorporate deuterium separately, the deuterated proteins are allowed to complex, and then deuterium atoms are off-exchanged with hydrogen atoms by dilution in H2O; residues located at the protein-protein interface are characterized by the retention of deuterons in the protein-protein complex as compared with a lack of retention in control experiments in which only one of the two proteins is present.

17. Example for on-exchange In a complex • Example of a peptide mass envelope resulting from pepsin digestion of (i) nondeuterated CheB protein; (ii) a deuterated protein complex involving CheB; and (iii) CheB alone after being subjected to the same deuteration period as the complex in (ii). The mass envelope broadens somewhat upon deuteration, and there is less deuteration of the peptide in the protein-protein complex than in the protein alone. • Graph of number of deuterons incorporated into the peptide shown in panel A versus deuteration time. Circles, uncomplexed CheB; squares, complexed CheB.

18. Example for off-exchange (A) Nondeuterated peptide fragment from the uncomplexed protein. (B) Peptide fragment from the uncomplexed protein after on-exchange for 10 min. (C) Peptide fragment from the uncomplexed protein after on-exchange for 10 min followed by off-exchange for 10 min. Some residual deuteration is seen, due to the fact that some D2O remains in the H2O buffer used for off-exchange. (D) Peptide fragment from the complexed protein after on-exchange (performed on each component of the complex separately) for 10 min followed by off-exchange (performed on the bound complex) for 10 min. In the bound complex, the peptide retains deuterium throughout off-exchange, whereas in the uncomplexed state, it does not. This indicates that the peptide is probably part of the protein-protein interface region. In a complex

19. What can we learn from H/D exchange? Structure of PPACK thrombin with loops colored according to the change in amide H/2H exchange seen upon activation of thrombin. The catalytic triad residues (H57CT, D102CT, and S195CT) are shown as sticks and in green. Loops in red become more dynamic upon thrombin activation. Loops in blue become less dynamic upon thrombin activation. PPACK thrombin =(D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone- treated alpha-thrombin) Its X-ray structure was determined Amide H/2H exchange reveals a mechanism of thrombin activation. Koeppe JR, Komives EA. Biochemistry. 2006 Jun 27;45(25):7724-32.

20. The impact of mass spectrometry on the study of intact antibodies: from post-translational modifications to structural analysis Natalie J. Thompson ,  Sara Rosati ,  Rebecca J. Rose and Albert J. R. Heck Chem. Commun., 2013,49, 538-548

23. ESI-mass spectra of intact humanized mAbs before (A) and after (B) treatment with PNGaseF. The inset shows the mass spectrum acquired under denaturing conditions. Maxent deconvolution resulted in the masses of each component. The mass differences between species in (A) are a mass of 162 Da corresponding to the addition of multiple hexose units. The deglycosylated antibody (B) results in only a single peak indicating that the heterogeneity in the glycosylated version stems solely from the heterogeneity of the glycan chain. Beck A, Wagner-Rousset E, Bussat MC, Lokteff M, Klinguer-Hamour C, Haeuw JF, Goetsch L, Wurch T, Van Dorsselaer A, Corvaïa N. Curr Pharm Biotechnol. 2008 Dec;9(6):482-501. Review.

24. ESI-mass spectra of intact antibodies. (a) An intact antibody spectrum acquired under denaturing conditions showing highly charged ion signals at relatively low m/z. (b) The same antibody sprayed under native conditions with the inset showing the various glycoforms present. (c) The native mass spectrum of a non-covalently bound hinge-deleted IgG4 molecule, revealing both half (low m/z, ca. 75 kDa) and whole (high m/z, ca. 150 kDa) antibodies present. Charge states are indicated above the peaks.

25. Antibody–antigen binding monitored by native MS using the rV antigen from Y. pestis at various Ab:Ag ratios. Addition of the antigen in substoichiometric amounts converts the unbound antibody (u) to the antibody–antigen complex (c). Upon the addition of an equimolar amount of antigen, the entire antibody population is converted to the complex, and excess monomeric antigen (m) is observed. (a) 1:0.1 molar ratio Ab:Ag; (b) 1:0.2 molar ratio of Ab:Ag; (c) 1:1 molar ratio of Ab:Ag. Probing molecular interactions in intact antibody: antigen complexes, an electrospray time-of-flight mass spectrometry approach. Tito MA, Miller J, Walker N, Griffin KF, Williamson ED, Despeyroux-Hill D, Titball RW, Robinson CV. Biophys J. 2001 Dec;81(6):3503-9.

26. Orbitrap-based analysis of intact, native, glycosylated mAb. (a) The complete mass spectrum of an intact mAb acquired under native conditions. The inset shows the FWHM to be approximately 1.1 Th. (b) Glycosylation pattern of an intact antibody with baseline-resolved glycan peaks. (c) Complex glycosylation patterns of a half-antibody, also with baseline-resolved glycan peaks. Individual glycoforms were assigned based on the differences in m/z between peaks, corresponding to 162 Da (hexose/galactose, H), 203 Da (GlcNAc, G), 146 Da (fucose, F) or 291 Da (sialic acid, S), as indicated by the lists on the right of each spectrum.

27. Exploring an Orbitrap Analyzer for the Characterization of Intact Antibodies by Native Mass Spectrometry Rosati S, Rose RJ, Thompson NJ, van Duijn E, Damoc E, Denisov E, Makarov A, Heck AJ. Angew Chem Int Ed Engl. 51, 12992-6. doi: 10.1002/anie.201206745. (2012) Modified Exactive Plus instrument (ThermoFisher Scientific) Modifications included: i) storing ions in the HCD cell, rather than trapping in the C trap, thus allowing more efficient trapping and increased desolvation, ii) manual regulation of the Xe pressure in the HCD cell, iii) altering the voltage offset on the flatapoles and transfer multipole, iv) applying in-source dissociation energy, v) modifying the software and applying maximum RF voltages to all RF multipoles (including the C trap) to enable m/z values up to 30 000 to be measured.

28. Chem Rev. 2007 Aug;107(8):3544-67. Epub 2007 Jul 25. Protein complexes in the gas phase: technology for structural genomics and proteomics. Benesch JL, Ruotolo BT, Simmons DA, Robinson CV.

29. Mass spectrometry is used at every level Pyramid of protein organization states.

30. Conventional and nanoelectrospray MS of a GroEL complex complex. The nESI spectrum displays a series of peaks around 11500 m/zwhich correspond to the 800 kDa tetradecamer. Conventional ESI of the same solution results in poorly resolved “humps” centered on 12500,16000, and 18500 m/z, identified as the tetradecamer, a dimer of tetradecamers, and a trimer of tetradecamers, respectively. There is also a signal at low m/z which corresponds to the GroEL monomer. The charge state distribution is broader and bimodal, and the formation of nonspecific oligomers is increased. These results indicate some of the benefits of using nESI. A smaller initial droplet size leads to less nonspecific aggregation (both protein−protein and protein−salt), and the gentler interface conditions possible, while still allowing adequate desolvation, lead to less dissociation and disruption of oligomeric structure. Solution conditions were 200 mM ammonium acetate, pH 6.9, and a protein concentration of 2 μM tetradecamer. Spectra were obtained on a modified Q-ToF 2 (Waters/Micromass). nanoESI normal ESI

31. Nonspecific aggregation during ion formation by nESI. Initial droplets undergo asymmetric fission, offspring droplets containing none, one, or several of the molecules of interest are formed (A). These droplets are on the order of 18 nm. A large proportion of the droplets formed are vacant, but at higher concentrations, more occupied droplets are formed (B). The relative proportions of the droplets which contained none, one, two, and three molecules vary according to the concentration. Droplets containing multiple copies of proteins eventually will lead to the formation of nonspecific aggregates. (simulation; charged residue model)

32. Measured values @ different Vacc Theoretical mass Adduction of solvent molecules and buffer ions to proteins. (A) The theoretical deconvoluted spectrum of a pure protein complex would appear at a mass representing the primary sequence, with a peak width defined by the isotopic distribution and instrumental resolution The masses of protein complexes observed are higher (B) Examination of the 68+ charge state of GroEL at different activation voltages. The amount of accelerating voltage (low to high) required to achieve the indicated peak width is indicated by the color of the points in panel B.

33. A Q-ToF type instrument customized for the transmission and analysis of protein complexes. (A) This analyzer is operates at a reduced RF frequency, allowing the selection and transmission of high m/z ions. The dissociation of protein complex ions requires higher collision cell gas pressures and accelerating voltages relative to “normal” parameters. (B) Improved transmission is achieved by adjusting the velocity of the ions, via altering the pressures in the instrument. C). As the pressure is increased from 9.3 μbar, the total signal intensity (width of the bars) increases. Examining how the ion signal divides between the four segments of the MCP plate suggests that, at pressures in the source region below and above the optimum, ions over- and undershoot the detector, respectively.

34. Number of collisions experienced and time spent in the collision cell. Upper panel: the calculated number of collisions relative to mass, by cytochrome c monomer (violet), transthyretin tetramer (blue), MjHSP16.5 24mer (green), GroEL tetradecamer (orange), and the 70S ribosome from Thermus thermophilus (red). The inset shows the linear dependence of the number of collisions on pressure for MjHSP16.5. Lower panel: Time spent in the collision cell vs. mass. The inset shows the accelerating voltage and pressure dependency of this time (for MjHSP16.5, 47+ charge state). These calculations are based on a collision cell length of 18.5 cm, a gas pressure of 30 μbar argon, and an accelerating voltage of 200 V.

35. Dependence on mass @ constant Ar pressure Energy conversion during collisional activation. Upper panel Simulations conversion of energy from kinetic to internal modes cytochrome c monomer (violet) transthyretin tetramer (blue) MjHSP16.5 24mer (green) GroEL tetradecamer (orange) 70S ribosome from Thermus thermophilus (red) Middle panel The effect of varying pressure on this energy conversion process for a single species, MjHSP16.5. Different pressures (labeled from 50 μbar, violet, to 10 μbar, red) of argon are simulated. Lower panel The effect of the mass of the gas (the noble gases from radon, violet, to neon, red) (-using the same ion as above) Inset into these panels are the dependencies for 50% conversion. These simulations show that in order for sufficient conversion of kinetic energy into internal modes to occur, in this length of collision cell, the use of higher pressures and/or heavier target gas is preferable.

36. Dissociation pathway of a multiprotein complex. (A) CID of the 32+ charge state of the TaHSP16.9 dodecamer (violet) results in the formation of complementary monomers at low m/zand 11mers at high m/z (both blue). At higher accelerating voltages (>100 V), a second distribution of monomers as well as decamers is observed (both cyan). This is indicative of a sequential dissociation reaction (white arrows). (B) Plotting the relative abundance of the different oligomeric species as a function of accelerating voltage (C) How long it takes the protein-complexes to dissociate Experiments were performed on a modified Q-ToF 2.105 The solution of TaHSP16.9 was infused by nESI at a concentration of 1.4 μM in 200 mM ammonium acetate, pH 6.9.14

37. Applications of collisional activation to the study of protein complexes. Information as to oligomeric composition (green) can be obtained from the identity of dissociation products (purple), as well as from beneficial effects regarding the parent ion (pink). A detailed examination of the reaction pathway can reveal certain parameters (blue) which can allow the determination of interaction strengths (red) and information as to oligomeric organization and size (orange).

38. PNAS Proceedings of the National Academy of Sciences of the United States of America Quaternary dynamics and plasticity underlie small heat shock protein chaperone function 1. Florian Stengela, 2. Andrew J. Baldwinb, 3. Alexander J. Paintera, 4. Nomalie Jayac, 5. Eman Bashac, 6. Lewis E. Kayb, 7. Elizabeth Vierlingc, 8. Carol V. Robinsona,1, and 9. Justin L. P. Benescha,1

39. Temperature-induced changes in HSP18.1 oligomerization. At the lowest temperatures HSP18.1 exists almost exclusively as a 216 kDa dodecamer, with charge states centered around 6,350 m/z At higher temperatures monomers and dimers are observed at low m/z, and higher-order oligomers at high m/z, demonstrating the temperature-dependent dissociation and augmentation of the dodecamers. B A plot of the relative amount of HSP18.1 subunits existing in different oligomeric states. CClose examination of the abundance of the different species at 46° C shows a clear preference relative to a Gaussian distribution (p < 0.01) for oligomers with an even number of subunits. D Plots of the difference in free energy between a subunit free in solution or incorporated into either a dimer or dodecamer follow linear trends, suggesting that neither undergo significant structural change upon thermal activation. In all graphs the data and error bars represent the mean and standard deviation of three independent experiments.

40. Time-course of the formation of complexes between HSP18.1 and unfolding Luc. ( Luc=firefly luciferase, extremely thermo-sensitive ) B Quantifying the relative abundances of the different species shows the initial rapid binding of Luc by HSP18.1, followed by further incorporation of HSP18.1 into the resultant complexes. The average mass of these complexes (Red, right-hand y-axis) mirrors this behavior, revealing the presence of distinct “binding” and “augmentation” steps in the chaperone action of HSP18.1.

41. Identification and quantification of HSP18.1:Luc complexes by means of tandem-MS. A Spectrum of a complex formed between HSP18.1 and Luc at a 1:0.1 ratio. Unbound dodecamer is observed around 6,350 m/z, and a broad area of signal, characteristic of a polydisperse ensemble, is observed above 8,000 m/z. B Selection of ions at a particular m/z, gray, and activation of selected ions leads to the removal of highly charged HSP18.1 subunits (≈1,800 m/z) from the parent oligomers. Complementary stripped complexes appear at high m/z. C Relative quantification of the different complexes from the heights of the peaks reveals a number of different stoichiometries of HSP18.1 bound to one Luc, and a dominance (p < 0.01) of those containing an even number of HSP18.1 subunits.

42. AThe dodecameric form of HSP18.1 represents a “storage” form, which is in equilibrium with suboligomeric species, and higher-order oligomers. These higher-order species are themselves continually recycling through the loss and reincorporation of dimers and monomers. Upon heat-shock the equilibria shift (Red) from the dodecamer to dissociated species, and unfolding clients are bound to form sHSP:client complexes. B From the number of different species we observe here for the different states of HSP18.1 (Orange) we can extrapolate to systems incorporating multiple sHSPs (Bars). Based on the ability of related sHSPs to hetero-oligomerise, the number of potential species (NComb) can be calculated using the inset equation, where NsHSP is the number of compatible sHSPs, i the number of HSP18.1 subunits in the oligomer, and NClient the number of different bound states for a particular i. This reveals a remarkably extensive potential sHSP network, presumably catering for the array of different unfolding clients requiring protection from aggregation during cellular stress. Conclusion for small HSP-containing complexes