Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2023 Feb 22;95(9):4421–4428. doi: 10.1021/acs.analchem.2c05003

Apparatus for Automated Continuous Hydrogen Deuterium Exchange Mass Spectrometry Measurements from Milliseconds to Hours

Joseph Anacleto †, Cristina Lento †, Vladimir Sarpe ‡, Ayesha Maqsood †, Banafsheh Mehrazma †, David Schriemer ‡, Derek J Wilson †,*
PMCID: PMC9996604  PMID: 36880265

Abstract

graphic file with name ac2c05003_0007.jpg

Hydrogen deuterium exchange mass spectrometry (HDX-MS) is a rapidly growing technique for protein characterization in industry and academia, complementing the “static” picture provided by classical structural biology with information about the dynamic structural changes that accompany biological function. Conventional hydrogen deuterium exchange experiments, carried out on commercially available systems, typically collect 4–5 exchange timepoints on a timescale ranging from tens of seconds to hours using a workflow that can require 24 h or more of continuous data collection for triplicate measurements. A small number of groups have developed setups for millisecond timescale HDX, allowing for the characterization of dynamic shifts in weakly structured or disordered regions of proteins. This capability is particularly important given the central role that weakly ordered protein regions often play in protein function and pathogenesis. In this work, we introduce a new continuous flow injection setup for time-resolved HDX-MS (CFI-TRESI-HDX) that allows automated, continuous or discrete labeling time measurements from milliseconds to hours. The device is composed almost entirely of “off-the-shelf” LC components and can acquire an essentially unlimited number of timepoints with substantially reduced runtimes compared to conventional systems.

Virtually all aspects of protein function and pathogenesis require conformational changes or shifts in conformational dynamics.1−5 Hydrogen deuterium exchange mass spectrometry (HDX-MS) has emerged as a leader among a small number of techniques that allow for characterization of conformational and dynamic shifts with structural resolution, enabling deep explorations of protein activity at the molecular level.6−9 In industry, HDX-MS has found “high demand” niches in biotherapeutic characterization10−12 and in the determination of protein interaction surfaces, particularly epitopes in antibody/antigen interactions.13−16 Interest in this approach has risen to the point where specialized commercial systems are available for automated HDX-MS experiments.17 These systems are essentially variations of classical liquid chromatography–mass spectrometry (LC-MS), combining sophisticated robotic autosamplers for complex mixing and injection programs with refrigerated columns for (acid protease) digestion and reverse-phase separation. Commercial platforms have dramatically improved the robustness of HDX-MS measurements but have several drawbacks, including minimum HDX labeling times of around 10 s, which limit these systems to well-structured proteins and tight (or at least slow off-rate) protein interactions.18

A recurrent theme in recent HDX-MS development has been the need to measure exchange on the “complete” biologically relevant timescale, with labeling times from milliseconds to hours.19 Millisecond labeling times are of particular interest since they allow for the characterization of dynamic shifts in less-ordered regions of proteins and even “intrinsically disordered” proteins.20 Several groups have developed devices that combine millisecond rapid mixing with hydrogen deuterium exchange mass spectrometry, including simple capillary setups,21−23 microfluidic chips,24−26 and more recently a stopped-flow setup.27 However, none of these approaches, with the exception of the microfluidic chip that is the foundation of the current work,24 has yet been used extensively to explore novel biological phenomena beyond model systems.5 Part of the reason for this may be that these setups tend to require highly specialized equipment and expertise and do not incorporate all of the capabilities associated with commercial systems. For instance, none (with the exception of the nascent stopped-flow system) have demonstrated LC-based, automated workflows.

In this work, we introduce an LC-automated system that overcomes many of the barriers to the broader adoption of millisecond HDX-MS and “full timescale” HDX experiments. The system is composed almost entirely of “off-the-shelf” LC components, uses commercial autosampler control software and is capable of multiple modes of automated operation, including discrete and continuous data collection over timescales ranging from milliseconds to hours. Continuous data acquisition also drastically reduces experiment times, with triplicate measurements up to 1 h labeling times requiring just under 4 h of acquisition, although continuous acquisition does not allow for the incorporation of reverse-phase chromatography. Data analysis can be carried out using existing commercial or noncommercial HDX software and has been automated in a new module incorporated into the Mass Spec Studio28 software suite.

This system can measure HDX kinetics from milliseconds to hours, with the capacity to collect an (effectively) unlimited number of timepoints within the accessible time window. The result is the ability to explore the full range of protein conformational dynamics with unprecedented precision, from disordered regions and weak binding interactions to structured regions and strong binding interactions.

Experimental Section

Materials

Cytochrome c (C7752) from equine heart (≥99%), deuterium oxide (D2O, 99.99%), ammonium acetate (≥98%), and formic acid (≥98%) were purchased from Sigma-Aldrich (St. Louis, MO). Tau protein was expressed from Escherichia coli BL21 cells containing a pET-29b vector encoding the htau40 isoform. Purification was carried out as previously reported20 and protein was stored in 20% glycerol at −80 °C. Both cytochrome c and tau were resuspended or buffer exchanged into 50 mM ammonium acetate prior to MS analysis.

Full Timescale HDX Experiments

The CFI-TRESI-HDX experiments were completed according to the experimental setup shown in Figure 1. Construction of the capillary-based sub-second mixing device has been previously described.29,30 PerkinElmer Series 200 pumps and autosamplers were used to deliver solvents. Pepsin (porcine gastric mucosa, Sigma-Aldrich) or Protease XIII (Aspergillus saitoi, Sigma-Aldrich) was cross-linked in-house onto NHS-activated agarose (PierceTM, Thermo Fisher Scientific). Digestion columns were constructed in-house using PEEK tubing with an ID of 0.040” and a 2 μm pore-size frit upstream of the ESI emitter. All data were acquired on a Waters Synapt G2-S, and IMS was employed in the TriWave cell to improve the spatial resolution of peptides in the digested samples. Peptide identification was performed using Proteome Discoverer (Thermo Fisher Scientific) after LC-MS/MS analysis with an Orbitrap Elite Hybrid Ion Trap-Orbitrap Mass Spectrometer.

Figure 1.

Figure 1

Open in a new tab

Schematic of the basic automated CFI-TRESI-HDX setup. Continuous solvent delivery occurs via LC pumps, with protein and D2O plugs delivered via autosamplers. The capillary mixing device remains unchanged from the original setup,29 with the inner capillary sealed at the end and notched to allow for efficient protein mixing with D2O (inset). Low millisecond to second HDX labeling times can be achieved based on the position of the inner capillary within the outer capillary, size of capillaries, and flow rates.

Full Timescale HDX Experiments with LC Separation

For CFI-TRESI-HDX coupled to LC separation, samples were directly loaded into a cooled HDX-UPLC system (Waters, Milford, MA) with digestion on an in-house built pepsin/protease XII column at 15 °C and desalting at 0 °C. The flowrate during digestion was 8 μL/min protein, 8 μL/min D2O, and 64 μL/min acid quench. Desalting took place at 80 μL/min for 1 min and ramped up to 140 μL/min for 2 min. Peptides were separated by reverse-phase chromatography (Waters Acquity BEH C18, 1.7 μm, 1 mm × 100 mm) with a 7 min gradient from 5 to 35% using acetonitrile with 0.1% formic acid at 35 μL/min.

Data Analysis

Continuous uptake measurement data were analyzed using a new “continuous measurement” HX module for Mass Spec Studio 2.0.28,31 This module was built as an extension to the classic peptide HX-MS module (HX-DEAL) to reuse the existing model-based deuterium uptake calculations and apply them on data acquired using the continuous pullback method. A custom processing routine was created (“Continuous HX-MS”), which aggregates MS1 data over the predefined ion-mobility (IM) range of each peptide and measures deuterium uptake at continuous intervals in time. During data processing, the measured retention times in the data file are transformed to the true HX reaction times by applying the experimental and hardware specifications in the parameters section of the new processing routine (e.g., time increment, flowrate, notch position, start/stop infusion time, etc.). The continuous % D values for each peptide as well as their transformed reaction times are exportable in.csv format via the “raw” output option inside the Export Wizard.

Results and Discussion

Setup Design and Optimization

The current setup is built around a capillary mixing system that was first introduced for mass spectrometry applications in 2003.29 In 2012, this system was incorporated into a microfluidic chip capable of supporting all of the mixing and reaction steps necessary for “bottom-up” hydrogen deuterium exchange mass spectrometry.24 That setup did not allow for automated operation, making continuous data collection and automated triplicate measurements impossible. The current setup uses the capillary mixer, but the microfluidic chip has been replaced by conventional LC mixing tees and a short PEEK microcolumn (column volumes range from 8 to 100 μL, optimized to achieve adequate digestion and reduce carryover) for immobilized acid protease-mediated digestion (Figure 1).

The workflow enabled by this device is highly similar to that described in ref (24), corresponding to a classic “bottom-up” HDX approach incorporating acid quenching of the HDX reaction, followed by acid protease-based digestion of the labeled protein. Briefly, protein and D2O are supplied via separate LC pumps with independent autosamplers. The protein mixture passes through a fused silica “inner capillary” that is blocked at the distal end. Fluid escapes from this capillary through a small notch cut 2 mm from the distal end, forcing the protein solution to mix with D2O in the narrow intercapillary space (Figure 1, inset). The volume between the end of the inner capillary and the subsequent acid mixing tee can be adjusted by moving the inner capillary within the outer capillary, which allows for the acquisition of “labeling times” from 140 ms up to 18 s. This adjustment can be made in a continuous manner using a syringe-pump mounted mechanism described previously29 or in discrete increments by withdrawing the inner capillary from the quench mixer, corresponding to “continuous mode” and “discrete mode” data acquisition, respectively.

After labeling, the HDX reaction is acid-quenched using a conventional microLC mixing tee, and the quenched solution is passed through an acid protease column to generate the labeled peptides whose deuterium uptake will be analyzed. In continuous mode, the peptides are injected directly into the mass spectrometer, resulting in a minimum “quench-to-ESI” delay of 8–60 s, depending on the digest column volume. This short period is generally insufficient for significant back exchange to occur, with typical back exchange levels staying below 5%. In discrete mode, peptides can either be directly injected into the mass spectrometer or passed into a reverse-phase separation module. The latter approach significantly enhances coverage and redundancy but also drastically increases the experiment time and back exchange, bringing both to levels consistent with commercial systems.

A series of experiments were conducted to determine the optimal solutions and flow conditions for this device, with the aim of minimizing back exchange and carryover while maximizing quench conditions and sequence coverage. The details of this optimization are provided in Table S1. In discrete mode, conventional quenching solutions, like 5–10% acetic acid in water, were found to allow significant carryover of peptides from previous replicates on the agarose beads of the acid protease column. In continuous mode, this carryover was observed as a linear increase in deuterium uptake superimposed on the “true” uptake kinetics. Ultimately, it was determined that carryover could be essentially eliminated by adding ammonium formate to the quench buffer (while maintaining an appropriate quench pH with formic acid), resulting in the following optimized buffer conditions for cytochrome C and Tau proteins: 100 mM ammonium acetate protein buffer (pH 7.0) and 3% formic/40 mM ammonium formate as the quench buffer (pH 2.3).

Millisecond HDX, Discrete Mode

Millisecond discrete mode is used to acquire a limited number of HDX timepoints, typically between 150 ± 20 and 1000 ± 20 ms, with automated technical replicates. The main advantage of this mode is that it allows for indefinite-length acquisitions at individual HDX timepoints (subject to limitations on sample consumption) and can incorporate LC separation. In terms of workflow, the defining characteristic of this mode is that the inner capillary is held at a fixed position for multiple injections, each corresponding to a technical replicate of a given timepoint. When the desired number of technical replicates have been acquired, the inner capillary is manually withdrawn at a fixed distance within the outer capillary to transition to a new HDX timepoint. To facilitate autosampler programming in this mode, we created an excel worksheet tool called CFIset that estimates sample elution times and visualizes the experiment. A typical millisecond discrete mode experiment is shown in Figure 2, with predicted and actual elutions using cytochrome c as a model.

Figure 2.

Figure 2

Open in a new tab

Discrete timepoint mode of operation on the millisecond to second timescale. (a) Predicted injection profile for four plugs of protein and one plug of D2O, both flowing at 5 μL/min %. (b) Experimental total ion chromatogram depicting four 8 μL injections of protein at 240 ms, a representative peptide (a.a. 96–105) is shown to indicate reproducibility between injections, labeled 1–4. (c) Kinetic profiles of representative Tau peptides with timepoints ranging from 140 ms to 9.4s. Error bars represent two standard deviations from four replicates.

The CFIset profile (Figure 2a) is idealized in the sense that it assumes perfect “plug-flow” and no peak broadening due to mixing in the labeling and quenching steps and thus narrow square-pulse-shaped injections. Nonetheless, the tool successfully predicts the signal onset time and approximate peak width for various flow conditions. Actual injection profiles, like the one shown in Figure 2b, exhibit the expected normal-skewed square distributions reflecting laminar flow and mixing within the device. For this system and under our conditions, we found that 2 min acquisitions provided the best balance between the “competing interests” of high signal-to-noise (resulting in good uptake measurement repeatability) and low sample consumption. Figure 2b also shows a sample peptide (a.a. 96–105) extracted from these replicates for the 240 ms exchange timepoint, demonstrating inter-injection reproducibility.

Exchange kinetics can be acquired in this mode by manually withdrawing the inner capillary within the outer capillary after the desired number of replicate injections at the current timepoint have been performed. For labeling times on the ms to s timescale, complete exchange kinetics profiles for all peptides (e.g., Figure 2c) can typically be acquired in triplicate within 3 h for six timepoints (i.e., 30 min per timepoint). The experimental rapidity of this technique is attributed to (i) the HDX short labeling times, (ii) automated replicate acquisition, and (iii) the lack of LC separation. We have noted in previous work that short (ms) HDX labeling times are sensitive to the types of interactions covered by conventional timescale HDX and also offer substantial advantages in characterizing disordered regions, weak binding interactions, and allosteric effects.5,32 However, the inability to include LC separation in these experiments has been a substantial limitation, particularly since it has prevented the use of the nonvolatile buffers in which some proteins are maximally soluble. The “injection pulse” nature of millisecond discrete mode CFI-TRESI-HDX allows for automated online LC separation, which is carried out while the HDX module of the apparatus is being washed. The incorporation of LC into CFI-HDX experiments is discussed in detail in a subsequent section.

Millisecond HDX, Continuous Mode

In most recent conventional HDX studies, uptake kinetics are treated as a “secondary” result, with the most important number being the total uptake difference between protein states (e.g., “bound” and “unbound”) over all timepoints measured. This is understandable given the unequivocal nature of the “summed difference” parameter in defining binding sites or regions that undergo substantial dynamic shifts. However, the HDX kinetics themselves can provide critical information, including, for example, whether an observed difference in deuterium uptake between two protein states is due to a change in “structure” or a change in “dynamics” or both.33 There is also evidence that careful analysis of HDX kinetics can reveal kinetic and thermodynamic parameters (i.e., Kd and koff) for binding interactions.34

One reason that careful analysis of HDX kinetics is not widely practiced may be that the acquisition of sufficient timepoints to accurately determine the observed uptake rate constant would be an arduous process. In millisecond continuous mode acquisition, the inner capillary is continuously withdrawn from the end of the outer capillary at a given rate while HDX data are continuously collected. This allows for an essentially unlimited number of HDX timepoints to be acquired, typically from 150 ms to 10 s, with the rate of withdrawal, scan time, and width of the TIC segments selected by the user ultimately determining the number and spacing of “timepoints” used in the analysis. An example of this type of analysis is provided in Figure 3, using Tau protein as a model. By fitting the data to a single exponential expression, we are able to precisely extract “observed” (phenomenological) rate constants, identifying regions that are weakly protected from exchange due to loci of the residual structure in the Tau conformational ensemble.20

Figure 3.

Figure 3

Open in a new tab

Continuous timepoint mode of operation on the millisecond to the second timescale. Five representative Tau peptides were analyzed from a continuous pullback experiment spanning 140 ms to 10 s. Peptide sequences and rate constants are listed in the insets, with a zoomed-in view of the first 3.5 s, where most of the exchange is taking place. The PDB file contains the selected peptides, colored from high (red) to low (blue) rate constants. Error bars represent one standard deviation from two replicates.

The other main advantage of millisecond continuous mode data acquisition is the exceedingly short experiment time, owing to the fact that individual timepoints are not independent runs in this mode. In principle, a complete triplicate dataset with unlimited timepoints over the full 150 ms to 10 s timecourse could be acquired in just over 1.5 h, with user intervention required only between replicates to reset the position of the capillary mixer. Manual analysis of these data can be taxing, requiring the user to select n equally broad segments of the “chromatogram”—corresponding to the desired number of timepoints—and then individually analyze each of them for every peptide’s uptake. To assist with this process, we have generated a new module for the Mass Spec Studio 2.0 software package28 that automates data extraction and generation of kinetic plots for each identified peptide directly from continuous-mode data. Screenshots and outputs from this module are provided in Figure S1.

Conventional Timescale Discrete and Continuous Modes

The “conventional” timescale for HDX experiments, corresponding to 10 s to several hours of labeling time, is useful for characterizing tight binding interactions and conformational dynamics in “structured” regions of proteins. The setup described here is capable of automated conventional timescale HDX measurements, with labeling times from 4 min to tens of hours, which is achieved by mixing and then allowing the protein/D2O solution to age in the autosampler. Injection of the mixed solution can be continuous, allowing for unlimited timepoint acquisition, or pulsed (discrete mode), allowing for LC separation. Equivalent continuous mode experiments have occasionally been implemented by other groups,35 while the discrete mode with LC separation is essentially equivalent to the conventional commercial systems currently in widespread use. Typical data from discrete and long timescale measurements are shown in Figure 4. For the discrete mode experiment (Figure 4a), deuterium uptake on either end of the peak profiles is impacted by dilution of the labeling solution with carrier solution. To avoid this, data are analyzed only from where the peak hits 80% of its maximum height, where dilution effects are negligible.

Figure 4.

Figure 4

Open in a new tab

Discrete and continuous modes of operation for long-timepoint HDX. (a) TIC and mass spectra of a representative Cyt c peptide (a.a. 96–105) during discrete long-timepoint acquisition. (b) TIC and mass spectra of a representative Cyt c peptide (a.a. 96–105) during continuous long-timepoint acquisition. In both modes, the exchange reaction is initiated at the start of the acquisition file.

Note that since conventional timescale measurements can be carried out without any modification of the apparatus, these experiments can be initiated (with limited user intervention) immediately following a millisecond run, allowing for the acquisition of the “full” HDX timecourse in a single experiment. The same benefits and limitations apply as for millisecond studies: Continuous measurements require the use of “electrospray friendly” salts (e.g., NH4Ac, (NH4)2CO3) and have lower sensitivity but generate an unlimited number of timepoints in a comparatively very short experiment time (corresponding to slightly longer than 3x the longest labeling time for triplicate data collection). Discrete measurements require substantially more experiment time but are compatible with LC, which has a number of advantages and limitations (see below).

Incorporation of Liquid Chromatography Separations

Liquid chromatography has long been used in “bottom-up” HDX workflows and is incorporated into commercially available HDX systems. This has several substantial advantages over HDX experiments that do not incorporate LC separation, notably increased sensitivity and sequence coverage, and a much-improved tolerance for “biologically relevant” nonvolatile buffer salts. The latter advantage can be critically important when dealing with proteins that aggregate in “ESI-friendly” solvent alternatives like NH4Ac “buffers.” However, LC separation also comes with notable disadvantages resulting from the length of time required for effective separation/elution on a reverse-phase column (typically 5–20 min). In particular, this delay between HDX quenching and ionization substantially increases back exchange, whose minimization has recently been a major focus of conventional HDX method development.36−38

The use of LC separation also precludes continuous mode data acquisition workflows and, even for the shortest gradients, greatly increases the experiment time. For example, a triplicate, 5-timepoint HDX run with a maximum 2 h labeling time could easily require 24 h of continuous measurement. In long-timepoint discrete mode, CFI-HDX reduces this considerably; a similar experiment to the one described above would require roughly 12 h. In millisecond discrete mode, the total experiment time corresponds to the width of the HDX timepoint pulse/wash sequence, times the number of replicates, times the number of timepoints, or roughly 4 h for the experiment described above. In spite of these drawbacks, the advantages of LC separation in terms of sequence coverage/redundancy and the broad applicability of the method are substantial (Figure S2). A typical CFI-TRESI HDX analysis with LC, incorporating timepoints from ms to hours and using Tau as a model, is shown in Figure 5.

Figure 5.

Figure 5

Open in a new tab

Full timescale HDX experiment from ms to hours in LC-integrated discrete mode. (a) The number of observable peptides for Tau increased to 78.7% coverage with 2.50 redundancy. (b) Uptake curves for Tau peptides and their respective rates obtained from CFI-TRESI-HDX with LC separation and discrete timepoints ranging from 150 ms to 4 h. The exchange data was normalized to 100%, with back exchange ranging between 30 and 40%.

While this experiment provides an accurate measurement of uptake rates across the entire HDX window (including tens of hours if so desired), the increase in back exchange was substantial and ranged considerably for each peptide (from ∼30–40%) in a way that did not correlate with retention time. The high initial uptake measurements, resulting from populations of backbone amides that exchange within the 150 ± 20 ms deadtime also represent some lost (kinetic) information (see Figure S4). Both of these issues can be minimized by further optimizing geometries and flow conditions within the device, which can be undertaken should it be required for particular applications. In principle, the capillary mixer used in the CFI-TRESI system is capable of deadtimes at or below 10 ms.29

On the other hand, sequence coverage improved from 47 to 79% and redundancy from 0 to 2.5 (compare Figures 5a to S2, note that Tau is a challenging target for pepsin and pXIII due to its high positive charge) with the incorporation of LC. Assessing the quantitative accuracy of the “rapid regime” uptake rates measured using CFI-TRESI with LC in “real” systems is challenging given the role that residual structure may play in attenuating the intrinsic rates of exchange (and the lack of comparable data from other sources). Nonetheless, the measured rates are reasonable in the sense that they result in approximated “segment-averaged” protection factors (intrinsic rate/observed rate) between 0.8–3.1 (Table S2), which is consistent with the residual structure in an intrinsically disordered protein and with previous rapid HDX uptake measurements of Tau.20 A comparison between theoretical and measured uptake for a sample Tau peptide is shown in Figure S4. There is also good agreement between the rates measured by CFI-TRESI-LC (Figure 5b) and millisecond continuous mode CFI-TRESI (Figure 3), at least for the two peptides that are in common for these very different implementations of the CFI-TRESI apparatus.

Conclusions

We have introduced a new automated continuous flow system for HDX mass spectrometry, which we call continuous flow injection hydrogen deuterium exchange (CFI-HDX) or CFI-TRESI-HDX when the millisecond mixer is incorporated. In some modes, the system can also accommodate automated LC separation, improving sensitivity and allowing a wide range of buffers. Millisecond to hour HDX timepoints can be acquired continuously or discretely in a single experiment, allowing for the detection of conformational/dynamic shifts in target proteins over the full range of structural stabilities and/or binding affinities.

The CFI-TRESI-HDX device is built almost exclusively from off-the-shelf components, is straightforward to implement, and uses existing (or developed) control and analysis software while significantly increasing the speed and power of HDX-MS as a bioanalytical tool. It is our hope that this work will enable an increasing number of researchers to take advantage of the information available in the “whole” HDX timecourse in a high(er) throughput manner and with a transformative impact on HDX-MS.

Acknowledgments

The authors would like to thank Esther Wolf and Dineesha Subasingha for helpful discussions. Emily Anacleto and Alexander Sever carried out early work related to this project. This work was funded by the Natural Sciences and Engineering Research Council of Canada Discovery Grant (RGPIN 480432) and the Collaborative Research and Development Grant (CRD-PJ 58776).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.analchem.2c05003.

  • Conditions used for the optimization of “standard” running conditions (Table S1); observed deuterium uptake rate constants and protection factors for all Tau peptides (Table S2); demonstration of continuous mode data analysis on (a) millisecond-to-second and (b) minute timescales using a custom-built Mass Spec Studio 2.0 module (Figure S1); sequence coverage on Tau protein using CFI-TRESI-HDX without LC (Figure S2); additional uptake curves with HDX rates determined from full timescale HDX with LC separation (Figure S3); comparison of intrinsic vs observed uptake profiles for a sample Tau peptide (Figure S4) (PDF)

  • CFIset autosampler programing aid excel worksheet (xls) (XLSX)

Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

The authors declare no competing financial interest.

Supplementary Material

ac2c05003_si_001.pdf (750.2KB, pdf)
ac2c05003_si_002.xlsx (70KB, xlsx)

References

  1. Guo J.; Zhou H.-X. Protein Allostery and Conformational Dynamics. Chem. Rev. 2016, 116, 6503–6515. 10.1021/acs.chemrev.5b00590. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Henzler-Wildman K.; Kern D. Dynamic Personalities of Proteins. Nature 2007, 450, 964–972. 10.1038/nature06522. [DOI] [PubMed] [Google Scholar]
  3. Wei G.; Xi W.; Nussinov R.; Ma B. Protein Ensembles: How Does Nature Harness Thermodynamic Fluctuations for Life? The Diverse Functional Roles of Conformational Ensembles in the Cell. Chem. Rev. 2016, 116, 6516–6551. 10.1021/acs.chemrev.5b00562. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Motlagh H. N.; Wrabl J. O.; Li J.; Hilser V. J. The Ensemble Nature of Allostery. Nature 2014, 508, 331–339. 10.1038/nature13001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Lento C.; Wilson D. J. Subsecond Time-Resolved Mass Spectrometry in Dynamic Structural Biology. Chem. Rev. 2022, 122, 7624–7646. 10.1021/acs.chemrev.1c00222. [DOI] [PubMed] [Google Scholar]
  6. Trabjerg E.; Nazari Z. E.; Rand K. D. Conformational Analysis of Complex Protein States by Hydrogen/Deuterium Exchange Mass Spectrometry (HDX-MS): Challenges and Emerging Solutions. TrAC, Trends Anal. Chem. 2018, 106, 125–138. 10.1016/j.trac.2018.06.008. [DOI] [Google Scholar]
  7. Oganesyan I.; Lento C.; Wilson D. J. Contemporary Hydrogen Deuterium Exchange Mass Spectrometry. Methods 2018, 144, 27–42. 10.1016/j.ymeth.2018.04.023. [DOI] [PubMed] [Google Scholar]
  8. Deng B.; Lento C.; Wilson D. J. Hydrogen Deuterium Exchange Mass Spectrometry in Biopharmaceutical Discovery and Development – A Review. Anal. Chim. Acta 2016, 940, 8–20. 10.1016/j.aca.2016.08.006. [DOI] [PubMed] [Google Scholar]
  9. James E. I.; Murphree T. A.; Vorauer C.; Engen J. R.; Guttman M. Advances in Hydrogen/Deuterium Exchange Mass Spectrometry and the Pursuit of Challenging Biological Systems. Chem. Rev. 2022, 122, 7562–7623. 10.1021/acs.chemrev.1c00279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Wei H.; Mo J.; Tao L.; Russell R. J.; Tymiak A. A.; Chen G.; Iacob R. E.; Engen J. R. Hydrogen/Deuterium Exchange Mass Spectrometry for Probing Higher Order Structure of Protein Therapeutics: Methodology and Applications. Drug Discovery Today 2014, 19, 95–102. 10.1016/j.drudis.2013.07.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Pan J.; Zhang S.; Borchers C. H. Comparative Higher-Order Structure Analysis of Antibody Biosimilars Using Combined Bottom-up and Top-down Hydrogen-Deuterium Exchange Mass Spectrometry. Biochim. Biophys. Acta - Proteins Proteomics 2016, 1864, 1801–1808. 10.1016/j.bbapap.2016.08.013. [DOI] [PubMed] [Google Scholar]
  12. Brown K. A.; Rajendran S.; Dowd J.; Wilson D. J. Rapid Characterization of Structural and Functional Similarity for a Candidate Bevacizumab (Avastin) Biosimilar Using a Multipronged Mass-Spectrometry-Based Approach. Drug Test. Anal. 2019, 11, 1207–1217. 10.1002/dta.2609. [DOI] [PubMed] [Google Scholar]
  13. Sun H.; Ma L.; Wang L.; Xiao P.; Li H.; Zhou M.; Song D. Research Advances in Hydrogen–Deuterium Exchange Mass Spectrometry for Protein Epitope Mapping. Anal. Bioanal. Chem. 2021, 413, 2345–2359. 10.1007/s00216-020-03091-9. [DOI] [PubMed] [Google Scholar]
  14. Masson G. R.; Jenkins M. L.; Burke J. E. An Overview of Hydrogen Deuterium Exchange Mass Spectrometry (HDX-MS) in Drug Discovery. Expert Opin. Drug Discovery 2017, 12, 981–994. 10.1080/17460441.2017.1363734. [DOI] [PubMed] [Google Scholar]
  15. Deng B.; Zhu S.; Macklin A. M.; Xu J.; Lento C.; Sljoka A.; Wilson D. J. Suppressing Allostery in Epitope Mapping Experiments Using Millisecond Hydrogen / Deuterium Exchange Mass Spectrometry. mAbs 2017, 9, 1327–1336. 10.1080/19420862.2017.1379641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Brown K. A.; Lento C.; Rajendran S.; Dowd J.; Wilson D. J. Epitope Mapping for a Preclinical Bevacizumab (Avastin) Biosimilar on an Extended Construct of Vascular Endothelial Growth Factor A Using Millisecond Hydrogen–Deuterium Exchange Mass Spectrometry. Biochemistry 2020, 59, 2776–2781. 10.1021/acs.biochem.0c00308. [DOI] [PubMed] [Google Scholar]
  17. Swartz M. HPLC Systems and Components Introduced at Pittcon 2011: A Brief Review. LCGC North Am. 2011, 29, 414–428. [Google Scholar]
  18. Jørgensen T. J. D.; Gårdsvoll H.; Danø K.; Roepstorff P.; Ploug M. Dynamics of Urokinase Receptor Interaction with Peptide Antagonists Studied by Amide Hydrogen Exchange and Mass Spectrometry. Biochemistry 2004, 43, 15044–15057. 10.1021/bi048706j. [DOI] [PubMed] [Google Scholar]
  19. Hamuro Y. Quantitative Hydrogen/Deuterium Exchange Mass Spectrometry. J. Am. Soc. Mass Spectrom. 2021, 32, 2711–2727. 10.1021/jasms.1c00216. [DOI] [PubMed] [Google Scholar]
  20. Zhu S.; Shala A.; Bezginov A.; Sljoka A.; Audette G.; Wilson D. J. Hyperphosphorylation of Intrinsically Disordered Tau Protein Induces an Amyloidogenic Shift in Its Conformational Ensemble. PLOS One 2015, 10, e0120416 10.1371/journal.pone.0120416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Keppel T. R.; Weis D. D. Mapping Residual Structure in Intrinsically Disordered Proteins at Residue Resolution Using Millisecond Hydrogen/Deuterium Exchange and Residue Averaging. J. Am. Soc. Mass Spectrom. 2015, 26, 547–554. 10.1007/s13361-014-1033-6. [DOI] [PubMed] [Google Scholar]
  22. Keppel T. R.; Weis D. D. Analysis of Disordered Proteins Using a Simple Apparatus for Millisecond Quench-Flow H/D Exchange. Anal. Chem. 2013, 85, 5161–5168. 10.1021/ac4004979. [DOI] [PubMed] [Google Scholar]
  23. Simmons D. A.; Dunn S. D.; Konermann L. Conformational Dynamics of Partially Denatured Myoglobin Studied by Time-Resolved Electrospray Mass Spectrometry with Online Hydrogen–Deuterium Exchange. Biochemistry 2003, 42, 5896–5905. 10.1021/bi034285e. [DOI] [PubMed] [Google Scholar]
  24. Rob T.; Liuni P.; Gill P. K.; Zhu S.; Balachandran N.; Berti P. J.; Wilson D. J. Measuring Dynamics in Weakly Structured Regions of Proteins Using Microfluidics-Enabled Subsecond H/D Exchange Mass Spectrometry. Anal. Chem. 2012, 84, 3771–3779. 10.1021/ac300365u. [DOI] [PubMed] [Google Scholar]
  25. Svejdal R. R.; Dickinson E. R.; Sticker D.; Kutter J. P.; Rand K. D. Thiol-Ene Microfluidic Chip for Performing Hydrogen/Deuterium Exchange of Proteins at Subsecond Time Scales. Anal. Chem. 2019, 91, 1309–1317. 10.1021/acs.analchem.8b03050. [DOI] [PubMed] [Google Scholar]
  26. Rob T.; Kamal Gill P.; Golemi-Kotra D.; J Wilson D. An Electrospray Ms-Coupled Microfluidic Device for Sub-Second Hydrogen/Deuterium Exchange Pulse- Labelling Reveals Allosteric Effects in Enzyme Inhibition. Lab Chip 2013, 13, 2528–2532. 10.1039/C3LC00007A. [DOI] [PubMed] [Google Scholar]
  27. Kish M.; Smith V.; Subramanian S.; Vollmer F.; Lethbridge N.; Cole L.; Bond N. J.; Phillips J. J. Allosteric Regulation of Glycogen Phosphorylase Solution Phase Structural Dynamics at High Spatial Resolution. BioRxiv 2019, 16, 654665 10.1101/654665. [DOI] [Google Scholar]
  28. Rey M.; Sarpe V.; Burns K.; Buse J.; Baker C. A. H.; van Dijk M.; Wordeman L.; Bonvin A. M. J. J.; Schriemer D. C. Mass Spec Studio for Integrative Structural Biology. Structure 2014, 22, 1538–1548. 10.1016/j.str.2014.08.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Wilson D. J.; Konermann L. A Capillary Mixer with Adjustable Reaction Chamber Volume for Millisecond Time-Resolved Studies by Electrospray Mass Spectrometry. Anal. Chem. 2003, 75, 6408–6414. 10.1021/ac0346757. [DOI] [PubMed] [Google Scholar]
  30. Lento C.; Zhu S.; Brown K. A.; Knox R.; Liuni P.; Wilson D. J. Time-Resolved ElectroSpray Ionization Hydrogen-Deuterium Exchange Mass Spectrometry for Studying Protein Structure and Dynamics. J. Vis. Exp. 2017, 122, 55464 10.3791/55464. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Raval S.; Sarpe V.; Hepburn M.; Crowder D. A.; Zhang T.; Viner R.; Schriemer D. C. Improving Spectral Validation Rates in Hydrogen-Deuterium Exchange Data Analysis. Anal. Chem. 2021, 93, 4246–4254. 10.1021/acs.analchem.0c05045. [DOI] [PubMed] [Google Scholar]
  32. Lento C.; Audette G. F.; Wilson D. J. Time-Resolved Electrospray Mass Spectrometry — a Brief History. Can. J. Chem. 2015, 93, 7–12. 10.1139/cjc-2014-0260. [DOI] [Google Scholar]
  33. Narang D.; James D. A.; Balmer M. T.; Wilson D. J. Protein Footprinting, Conformational Dynamics, and Core Interface-Adjacent Neutralization “Hotspots” in the SARS-CoV-2 Spike Protein Receptor Binding Domain/Human ACE2 Interaction. J. Am. Soc. Mass Spectrom. 2021, 32, 1593–1600. 10.1021/jasms.0c00465. [DOI] [PubMed] [Google Scholar]
  34. Zhang Z.; Vachet R. W. Kinetics of Protein Complex Dissociation Studied by Hydrogen/Deuterium Exchange and Mass Spectrometry. Anal. Chem. 2015, 87, 11777–11783. 10.1021/acs.analchem.5b03123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Wang H.; Shu Q.; Rempel D. L.; Frieden C.; Gross M. L. Continuous and Pulsed Hydrogen–Deuterium Exchange and Mass Spectrometry Characterize CsgE Oligomerization. Biochemistry 2015, 54, 6475–6481. 10.1021/acs.biochem.5b00871. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Fang M.; Wang Z.; Cupp-Sutton K. A.; Welborn T.; Smith K.; Wu S. High-Throughput Hydrogen Deuterium Exchange Mass Spectrometry (HDX-MS) Coupled with Subzero-Temperature Ultrahigh Pressure Liquid Chromatography (UPLC) Separation for Complex Sample Analysis. Anal. Chim. Acta 2021, 1143, 65–72. 10.1016/j.aca.2020.11.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Venable J. D.; Okach L.; Agarwalla S.; Brock A. Subzero Temperature Chromatography for Reduced Back-Exchange and Improved Dynamic Range in Amide Hydrogen/Deuterium Exchange Mass Spectrometry. Anal. Chem. 2012, 84, 9601–9608. 10.1021/ac302488h. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sheff J. G.; Rey M.; Schriemer D. C. Peptide–Column Interactions and Their Influence on Back Exchange Rates in Hydrogen/Deuterium Exchange-MS. J. Am. Soc. Mass Spectrom. 2013, 24, 1006–1015. 10.1007/s13361-013-0639-4. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ac2c05003_si_001.pdf (750.2KB, pdf)
ac2c05003_si_002.xlsx (70KB, xlsx)

Articles from Analytical Chemistry are provided here courtesy of American Chemical Society

RESOURCES