| BCAII and AZM sample conditions for apparent K(d) and in-droplet HDX-MS measurements | BCAII was dissolved in 100 mM ammonium acetate to make a concentration of 10 µM; the stock was diluted using 100 mM AA as required (10 nM) for the nMS apparent equilibrium dissociation constant (K(d)) determinations, while the 10 µM stock was employed directly for in-droplet HDX-MS experiments; AZM stock was prepared in DMSO at 10 mM and diluted as required to comprise no more than 3 % DMSO of the total volume of the protein-ligand sample; For each sample, the ligand was incubated with the protein for approximately 2 minutes before being sprayed into the MS inlet |
|---|---|
| Multidevice capillary vibrating sharp-edge spray ionization conditions for in-droplet HDX | A waveform generator (Tektronix AFG-1062, Beaverton, OR, USA) and an amplifier (Krohn-Hite 7500, Brockton, MA, USA) were used to actuate two cVSSI devices; analyte and reagent solutions were infused into the appropriate devices using the same syringe pump at 10 µL/min; emitter points were at a printed "90°" angle; the analyte emitter tip was positioned comparatively farther away (~4 mm) than the D₂O-reagent emitter tip (~1 mm), facing directly into the mass spectrometer inlet; +1300 V bias voltage was applied to the analyte solution; No voltage was applied to the reagent (D₂O) solution. First, the cVSSI device generating the D₂O reagent plume was turned off and the analyte device was turned on; mass spectra were recorded for analyte ions that were not deuterated. Then, both devices were turned on to initiate in-droplet HDX. Data recorded for the deuterated ions as well |
| Mass-spectrometry instrument, acquisition, and signal-averaging settings | Q-Exactive hybrid quadrupole orbitrap mass spectrometer from Thermo Fisher Scientific (San Jose, CA, U.S.A.); to identify biomolecules in a low charge state, the scan range was increased to 5000 m/z; ion transfer tube temperature was set to 300 °C in the positive ion mode experiments; AGC target 1 × 10⁶; complete scan resolution 140000; for K(d) and deuterium uptake, signal-averaged mass spectra were generated using xCalibur Qual Browser |
| Unbound-protein contribution correction for the AZM-bound hydrogen-deuterium exchange spectrum | The HDX spectrum of the untreated protein was scaled to match the intensity of the corresponding AZM-treated HDX spectrum, and the scaled unbound protein contribution was subtracted from the AZM-treated HDX spectrum to obtain a corrected spectrum for the AZM-bound protein. Scaling factor S = I(Unbound, AZM treated) / I(Unbound, AZM untreated). The scaling factor was then multiplied by every intensity value in the unbound-protein HDX spectrum. At each m/z data point, I(HDX, AZM bound corrected, i) = I(HDX, AZM-treated, i) − S × I(HDX, Unbound, i), where I(HDX, AZM-treated, i) is the measured intensity at the i’th m/z point in the AZM-treated HDX spectrum, containing contributions from both unbound and AZM-bound protein, I(HDX, Unbound, i) is the protein-without-AZM intensity at the same m/z point, S is the scaling factor used to match the unbound-protein contribution in the two spectra, and I(HDX, AZM bound corrected, i) is the corrected intensity assigned to the AZM-bound protein |
| Number of incorporated deuteriums from corrected spectral intensities | Main: the number of deuterium exchanges was determined by measuring the shift in weighted isotope m/z value, m/z(w), upon HDX reaction using Equation (1). SI: the corrected intensity values were subsequently used to calculate the intensity-weighted mean m/z according to equation (1) provided in the main manuscript; the number of incorporated deuteriums was determined from the charge-corrected difference between the weighted mean m/z values of the HDX and non-HDX spectra; the observed m/z shift was multiplied by the corresponding charge state, z, to get the number of incorporated deuteriums; N(D), Number of incorporated deuterium; Z, corresponding charge state |
| Native mass spectrum at a 1:1 BCAII-to-AZM molar ratio | Distinct peaks for unbound and ligand-bound ion species were evident; for the dominant 8⁺ and 9⁺ ions, distinct peaks were observed for unbound and ligand-bound protein; labeled features for unbound protein (P) and AZM-bound protein (PL) at 10⁺, 9⁺, 8⁺, and 7⁺ |
| Apparent dissociation constant, K(d), determined using ligand-based titration and single-point approaches; Langmuir binding isotherm for ligand-based titration | 6.34 ± 1.45 nM (mean ± SD, (n=3)) from three independent titration experiments; AZM concentrations from 0.5 nM to 480 nM were titrated against 10 nM BCAII, the fraction of protein bound was determined at each ligand concentration, and cVSSI-MS data were recorded for each ligand concentration; best-fit K(d) values were obtained by minimizing the sum of squared residuals between experimental and predicted values |
| Single-point apparent dissociation constant, K(d) | With the single point approach, mass spectral data for protein : ligand molar ratio samples of 1:1, and 1:2 is used to determine the K(d); protein-to-ligand molar ratio 1:1: 9.2 ± 0.2 nM; protein-to-ligand molar ratio 1:2: 6.3 ± 0.01 nM. SI: relative ion intensities of unbound and ligand bound proteins are measured, and the initial concentrations of the protein and ligands are required; [L](o) and [P](o) represent initial concentrations of protein and ligand respectively, and the summed P(n⁺) and PL(n⁺) terms correspond to the sum of ion intensities of unbound and ligand bound protein in n⁺ charge states, respectively |
| Native mass-spectrum comparison of protein-only and protein-ligand samples | For the protein-ligand sample, a significant shift towards lower charge states was observed; this lower charge states production upon AZM binding indicates a more compact and rigid conformational state of the protein |
| In-droplet hydrogen-deuterium exchange mass-spectral envelopes | The mass spectral envelopes of unbound and AZM-bound BCAII shifted towards higher m/z, confirming incorporation of deuterium atoms; the shift was larger in unbound BCAII compared to the bound, implying more deuterium incorporation in the unbound BCAII; before and after HDX, AZM-bound BCAII showed peak shift and broadening across both protein (P) and protein-ligand (PL) complex; the 9⁺ charge state showed a shift in m/z after HDX |
| Deuterium atoms incorporated by unbound BCAII across the 7⁺ to 10⁺ charge states, respectively | Mean ± standard deviation from three independent measurements (n = 3): 101.94 ± 6.33, 92.25 ± 4.02, 90.88 ± 4.58, and 100.38 ± 1.54 |
| Deuterium atoms incorporated by AZM-bound BCAII under the same experimental conditions across the 7⁺ to 10⁺ charge states, respectively | Mean ± standard deviation from three independent measurements (n = 3): 83.22 ± 5.29, 80.25 ± 3.38, 74.72 ± 4.50, and 86.65 ± 2.15 |
| Charge-state 7⁺ replicates: replicate; unbound weighted m/z before HDX; unbound weighted m/z after HDX; unbound N(D); AZM-bound weighted m/z before HDX; AZM-bound weighted m/z after HDX; AZM-bound N(D); ΔD; reduction (%) | 1: 4156.57; 4171.63; 105.40; 4188.32; 4200.64; 86.20; 19.20; 22.27. 2: 4156.57; 4170.09; 94.64; 4188.32; 4199.33; 77.11; 17.53; 22.73. 3: 4156.57; 4171.68; 105.78; 4188.32; 4200.65; 86.34; 19.45; 22.53 |
| Charge-state 8⁺ replicates: replicate; unbound weighted m/z before HDX; unbound weighted m/z after HDX; unbound N(D); AZM-bound weighted m/z before HDX; AZM-bound weighted m/z after HDX; AZM-bound N(D); ΔD; reduction (%) | 1: 3637.00; 3649.09; 96.70; 3664.78; 3675.26; 83.80; 12.90; 15.39. 2: 3637.00; 3648.11; 88.89; 3664.78; 3674.42; 77.08; 11.80; 15.31. 3: 3637.00; 3648.40; 91.16; 3664.78; 3674.76; 79.87; 11.29; 14.14 |
| Charge-state 9⁺ replicates: replicate; unbound weighted m/z before HDX; unbound weighted m/z after HDX; unbound N(D); AZM-bound weighted m/z before HDX; AZM-bound weighted m/z after HDX; AZM-bound N(D); ΔD; reduction (%) | 1: 3232.89; 3243.57; 96.10; 3257.58; 3266.42; 79.60; 16.50; 20.73. 2: 3232.89; 3242.61; 87.53; 3257.58; 3265.44; 70.73; 16.80; 23.75. 3: 3232.89; 3242.78; 89.02; 3257.58; 3265.78; 73.82; 15.19; 20.58 |
| Charge-state 10⁺ replicates: replicate; unbound weighted m/z before HDX; unbound weighted m/z after HDX; unbound N(D); AZM-bound weighted m/z before HDX; AZM-bound weighted m/z after HDX; AZM-bound N(D); ΔD; reduction (%) | 1: 2909.60; 2919.77; 101.70; 2931.82; 2940.70; 88.70; 13.00; 14.66. 2: 2909.60; 2919.47; 98.68; 2931.82; 2940.26; 84.41; 14.27; 16.91. 3: 2909.60; 2919.68; 100.76; 2931.82; 2940.51; 86.85; 13.91; 16.02 |
| Binding-associated percent-reduction calculation | Weighted m/z values are calculated for before and after HDX to determine the number of incorporated deuterium; percent reductions are reported from three independent measurements. N(D) = Number of incorporated deuterium. Percent reduction = [N(D,Unbound) − N(D,Bound)] / N(D,Bound) × 100. All numerical values are presented to two decimal places. Positive ΔD values indicate difference in deuterium incorporation in the AZM-bound protein than in the unbound protein |
| Decrease in deuterium uptake upon AZM binding | AZM-binding to BCAII decreased the deuterium uptake by 18.71 ± 1.04, 11.99 ± 0.81, 16.19 ± 0.87, and 13.73 ± 0.66 deuterium’s for the 7⁺, 8⁺, 9⁺, and 10⁺ charge states, respectively. Overall, a reduction of 22.51 %, 14.95 %, 21.68 %, and 15.86 % was observed for the 7⁺, 8⁺, 9⁺, and 10⁺ charge states, respectively. Although the magnitude of the HDX reactivity response varied among different charged species, consistent reduction in reactivity was observed for all charge states. The relative reduction was highest for 7⁺ and 9⁺ ions and somewhat smaller for 8⁺ and 10⁺ ions. This consistent decrease in deuterium incorporation across the examined charge states indicates AZM-binding reduces the global HDX reactivity. The findings of the in-droplet HDX-MS are consistent with ligand induced conformational stabilization of the protein. The percentage reduction was calculated independently for each replicate as (U − B)/B × 100, where U and B are the mean numbers of incorporated deuterium atoms for unbound and AZM-bound BCAII, respectively; the mean percentage reduction from three independent measurements (n=3) is reported. In the model system of BCAII and its inhibitor AZM, a significant reduction in deuterium incorporation was observed, indicating conformational stabilization and reduced structural plasticity in the complex compared to apo protein |
| Ligand-concentration-series native mass spectra | The signal intensity of ligand-bound mass-spectral peaks is observed to increase with increasing ligand concentration; at approximately 40 nM ligand concentration, a protein-to-ligand molar ratio of 1:4, a near constant ratio of bound to unbound ion intensities is achieved |
| Native mass-spectral ligand titration at fixed 10 nM protein concentration | Caption mapping of ligand concentration to protein:ligand molar ratio: 0.5 nM, 20:1; 1 nM, 10:1; 5 nM, 2:1; 10 nM, 1:1; 20 nM, 1:2; 40 nM, 1:4; 80 nM, 1:8; 480 nM, 1:48. P and PL represent the peaks corresponding to unbound and ligand bound respectively. With the increase of ligand concentrations, intensity of the bound peaks increased compared to the unbound peaks |
| Fraction-bound calculation for apparent K(d) | Fraction bound is determined from the relative abundance of free and ligand-bound charge states using Equation 2 and the features from the mass spectrum of the protein-ligand sample: fraction bound = Σₙ I(PL) / [Σₙ I(P) + Σₙ I(PL)], where Σₙ I(P) is the sum of the intensity of the unbound protein ions and Σₙ I(PL) is the sum of the intensity of the ligand-bound protein ions for the n⁺ charge state, with n = 8 and 9; only the 8⁺ and 9⁺ charge states were considered because they were clearly resolved and consistently detected across the whole ligand concentrations in the titration |
| Single-site Langmuir binding-isotherm fit | Y = B(max)[L] / [K(d) + [L]], where Y = Predicted values, B(max) = Maximum fraction bound, [L] = Ligand concentration, and K(d) = Apparent binding affinity / Equilibrium dissociation constant; K(d) values are obtained by fitting the mass-spectrometry-derived fraction bound to the non-linear binding isotherm |
| Langmuir binding-isotherm analysis: ligand concentration, nM; experimental fraction bound; predicted Y; residual² | 0: 0; 0; 0. 0.5: 3.47E-02; 5.672E-02; 4.86E-04. 1: 1.37E-01; 1.059E-01; 9.91E-04. 5: 3.49E-01; 3.462E-01; 8.32E-06. 10: 4.74E-01; 4.833E-01; 9.04E-05. 20: 5.83E-01; 6.026E-01; 3.69E-04. 40: 7.20E-01; 6.874E-01; 1.08E-03. 80: 7.24E-01; 7.394E-01; 2.25E-04. 480: 7.93E-01; 7.892E-01; 1.07E-05 |
| Langmuir binding-isotherm parameters | B(max): 0.8; K(d): 6.55; sum of squared residuals: 3.26E-03 |
| Charge-state-resolved apparent K(d) values, mean ± SD from three independent measurements (n = 3) | 9⁺, titration: 6.13 ± 1.47 nM; 8⁺, titration: 6.55 ± 1.54 nM; 9⁺, single point (P: L = 1:1): 10.46 ± 0.23 nM; 8⁺, single point (P: L = 1:1): 8.60 ± 0.24 nM; 9⁺, single point (P: L = 1:2): 6.03 ± 0.06 nM; 8⁺, single point (P: L = 1:2): 6.45 ± 0.03 nM |
Sohag Ahmed, Mst Nigar Sultana, Md Reaz Uddin, Sultan Mahmud, Mohammad A. Rahman, Peng Li, and Stephen J. Valentine. “Studying protein-ligand interactions with in-droplet hydrogen-deuterium exchange mass spectrometry and molecular dynamics simulations.” ChemRxiv (). https://doi.org/10.26434/chemrxiv.15008038/v1
Molecular-dynamics simulation results were not retained. The Supporting Information is appended to the main PDF. No additional separate Supporting Information file was available on the ChemRxiv article page when acquired
| Native mass spectrum of 10 nm protein in 100 mM ammonium acetate | Narrow and lower charge-state distribution; labeled mass-spectral features: 11⁺, 10⁺, 9⁺, and 8⁺ |
|---|