Letter pubs.acs.org/ac
Protective Effects of Dimethyl Sulfoxide on Labile Protein Interactions during Electrospray Ionization Michael Landreh,*,† Gunvor Alvelius,† Jan Johansson,‡,§ and Hans Jörnvall† †
Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, S-171 77, Sweden KI-Alzheimer’s Disease Research Center, NVS Department, Karolinska Institutet, Stockholm, S-141 86, Sweden § Department of Anatomy, Physiology and Biochemistry, Swedish University of Agricultural Sciences, Uppsala, S-751 23, Sweden ‡
S Supporting Information *
ABSTRACT: Electrospray ionization mass spectrometry is a valuable tool to probe noncovalent interactions. However, the integrity of the interactions in the gas-phase is heavily influenced by the ionization process. Investigating oligomerization and ligand binding of transthyretin (TTR) and the chaperone domain from prosurfactant protein C, we found that dimethyl sulfoxide (DMSO) can improve the stability of the noncovalent interactions during the electrospray process, both regarding ligand binding and the protein quaternary structure. Low amounts of DMSO can reduce in-source dissociation of native protein oligomers and their interactions with hydrophobic ligands, even under destabilizing conditions. We interpret the effects of DMSO as being derived from its enrichment in the electrospray droplets during evaporation. Protection of labile interactions can arise from the decrease in ion charges to reduce the contributions from Coulomb repulsions, as well as from the cooling effect of adduct dissociation. The protective effects of DMSO on labile protein interactions are an important property given its widespread use in protein analysis by electrospray ionization mass spectrometry (ESI-MS).
E
electrospray process, while higher concentrations lead to supercharging.11 The onset of supercharging ranges from 2% to over 10%, dependent on the protein system, up to a certain threshold without significantly impacting the solution structure.11 However, caution must be taken when using DMSO in quantitative protein interaction studies, since it can reduce the amount of protein−ligand complexes observable by ESI-MS and then lead to a significant underestimation of binding affinities.12,13 We have investigated the effects of DMSO on the noncovalent protein−protein and protein−ligand interactions of transthyretin (TTR) and the C-terminal chaperone domain from lung surfactant proprotein C (CTC) in ESI-MS. TTR is a plasma carrier for the thyroid hormone thyroxine (T4) and is associated with several human amyloid diseases. The natively tetrameric TTR cooperatively binds two T4 molecules in a hydrophobic channel between two dimeric subunits, which prevents its dissociation and subsequent misfolding.14,15 CTC contains a conserved BRICHOS domain, which is a potent antiamyloid chaperone that binds hydrophobic protein segments and in this manner prevents aggregation.16,17 Due to the hydrophobic nature of natural and engineered TTR and CTC ligands, DMSO and similar compounds are commonly used as
lectrospray ionization mass spectrometry (ESI-MS) has evolved into an important tool for the study of protein structure and function. A growing body of evidence suggests that many proteins retain a near-native fold during the “soft” ionization from evaporating electrospray droplets and can be maintained in the gas-phase long enough to allow mass spectrometric analysis.1,2 ESI-MS is routinely applied to study protein interactions and has been used successfully for the determination of binding constants.3,4 Nevertheless, the desolvation of a protein can affect its structure, as hydrophobic interactions appear weakened and charge interactions strengthened.5,6 In addition, the charge state correlates to some extent with folding states in the gas-phase. Low charges and narrow charge distributions are usually observed for compactly folded states and more stable complexes,7,8 but even supercharged protein complexes can retain a native-like structure,7 although not in all cases.9 The charge states of a protein complex also dictate its dissociation pathways when subjected to collisioninduced dissociation (CID), with lower charge states being more resistant to CID, as they require a higher energy for dissociation.8,10 Ion charge states can be modulated chemically to obtain information about protein interactions. Dimethyl sulfoxide (DMSO), a common solvent for hydrophobic molecules and therefore often used in protein characterization and binding studies, is a two-step charge modulator. The presence of low concentrations of DMSO can lower ion charges during the © 2014 American Chemical Society
Received: March 8, 2014 Accepted: April 22, 2014 Published: April 22, 2014 4135
dx.doi.org/10.1021/ac500879c | Anal. Chem. 2014, 86, 4135−4139
Analytical Chemistry
Letter
Figure 1. DMSO-mediated charge state reduction and increased TTR tetramer stability in ESI-MS. (A) ESI-MS spectra of 30 μM TTR in the presence and absence of 3% DMSO and at 70 and 200 eV cone voltage. Increasing the cone voltage promotes the dissociation of TTR tetramers. The addition of 3% DMSO reduces ion charges and prevents in-source dissociation. (B) ESI-MS spectra of 30 μM at pH 4.4 in the presence of 0− 5% DMSO. With increasing DMSO concentration, tetramers become the dominant oligomeric species (highlighted in gray).
All TTR spectra at pH 4.4 were therefore recorded using aliquots that were frozen following a 24 h incubation at 4 °C. Lyophilized T4 (free acid) was dissolved to a final concentration of 1 mM in DMSO, dimethylformamide (DMF), isopropanol, acetonitrile, or methanol and stored at −20 °C. CTC (theoretical average molecular mass: 18 263 Da) was expressed and purified as described.16 Amidated and acetylated triple-valine was purchased from Thermo electron (Darmstadt, Germany). Immediately before analysis, samples were mixed to final concentrations of 30 μM TTR and 30 μM ligand. Cosolvents were added to v/v %. Spectra were acquired on a Q-ToF Ultima API mass spectrometer (Waters, Milford, MA) equipped with a Z-spray source in positive-ion mode. Samples were sprayed from metal-plated borosilicate glass capillary needles (Proxeon, Denmark). The source temperature was 70 °C. The capillary voltage was kept at 1.6 kV, and the cone and RF lens 1 potentials were 90 and 50 eV, respectively, unless otherwise noted. The mass spectrometer was operated in single-reflector mode for sufficient resolution and sensitivity
cosolvents in binding studies. Our results show that low concentrations of DMSO mitigate the dissociation of labile interactions during ionization and enhance their detection by ESI-MS.
■
EXPERIMENTAL SECTION Recombinant, His6-tagged human TTR (theoretical average molecular mass: 15150.5 Da) was purchased as lyophilized powder from Wako Chemicals (Neuss, Germany). All chemicals were purchased from Sigma (St. Louis, MO). TTR was dissolved in 20 mM ammonium acetate, pH 4.4 or pH 7.5, to a final concentration of 100 μM and stored in 100 μL aliquots at −20 °C. ESI-MS spectra of freshly dissolved TTR at pH 4.4 showed a high amount of tetramers which were found to disappear after 24 h at 4 °C, after which the spectra did not change (data not shown). We observed that a TTR concentration of 30 μM was required to obtain spectra showing the changes in protein oligomerization at pH 4.4. All spectra in this study were acquired at the same concentration. 4136
dx.doi.org/10.1021/ac500879c | Anal. Chem. 2014, 86, 4135−4139
Analytical Chemistry
Letter
(10 000, full width half-maximum definition). The mass scale was calibrated in linear mode using horse myoglobin (Sigma, St. Louis, MO). Scans were acquired at a rate of 1 scan per 2 s between 1000 and 6000 m/z. Collision gas was argon at 3.5 × 10−5 mbar. Mass spectra were analyzed with the Waters MassLynx 4.1 software package.
■
RESULTS DMSO Reduces in-Source Dissociation of TTR Tetramers. Since TTR ligand binding studies are routinely performed in the presence of low amounts of DMSO, we investigated the effects of 0−7% DMSO on the charge pattern of TTR in ESI-MS. At pH 7.5, almost exclusively, tetrameric TTR, with a compact charge envelope of ∼4 charge states centered around the [M + 16H]16+ ion, was detected (Figure 1A). The addition of 0.1−7% DMSO leads to a pronounced charge reduction, with the main charge envelope centered around the [M + 14H]14+ ion (Figure 1A and Supplementary Figure 1A, Supporting Information). We also observed signal broadening, indicative of the formation of DMSO adducts. The onset of supercharging occurs at 10% DMSO and is accompanied by a moderate broadening of the charge envelope, as noted before.11 Charge reduction was also observed in the presence of 0.1−7% DMF, while the presence of 0.1−7% acetonitrile, isopropanol, or methanol did not affect the charge state distribution to a similar extent (Supplementary Figure 1B−E, Supporting Information). Raising the cone voltage from 70 to 200 eV to facilitate in-source dissociation of protein− protein interactions,16 we found that, in the absence of DMSO, the tetramers dissociate into monomeric and trimeric TTR. In the presence of DMSO, however, the tetramers remain intact (Figure 1A). All tested DMSO concentrations between 0.1% and 7% exhibit the same pattern, while the higher charged complexes at 10% DMSO dissociate more easily (Supplementary Figure 2, Supporting Information). Another destabilizing factor for tetrameric TTR is low pH, which leads to local unfolding and reduced intersubunit contacts, and consequently, the subunits dissociate easily under these conditions.18 In line with the reduced tetramer stability at low pH, we find that TRR appears predominantly as monomers and dimers in ESI-MS spectra recorded at pH 4.4. However, the addition of 3−5% DMSO leads to the detection of intact tetramers, indicating that DMSO can in part prevent the dissociation of the pH-destabilized subunit contacts (Figure 1B). Again, similar effects were observed for DMF but not for acetonitrile, isopropanol, or methanol (Supplementary Figure 3, Supporting Information). These observations suggest that DMSO can facilitate the detection of labile protein interactions that may otherwise not be observable by ESI-MS. DMSO and DMF Stabilize TTR-T4 Interactions. Next, we asked whether DMSO might also affect the detection of TTR-T4 complexes. Due to their high hydrophobicity, TTR ligands usually are dissolved in DMSO, DMF, or related solvents. At pH 7.5 and pH 4.4, near-complete saturation with two T4 moieties and a minor peak corresponding to one T4 moiety per tetramer are observed in the presence of 3% DMSO and at a 2:1 excess of T4 over tetrameric TTR (Figure 2A, top panel, and Supplementary Figure 4, Supporting Information). We furthermore observed a shoulder at the high m/z side of the complex ions which can be attributed to the formation of unspecific DMSO adducts. Increasing the cone voltage from 70 to 200 eV results in an increase of 1:1 complexes between T4 and tetrameric TTR, but the 2:1 complex remains the
Figure 2. DMSO and DMF reduce in-source dissociation of TTR-T4 complexes. ESI-MS spectra of 30 μM TTR at pH 7.5 in the presence of 30 μM T4 added from DMSO, DMF, or methanol stock solutions. Green, yellow, and red bars indicate the tetrameric TTR with two, one, or no T4 molecules attached. Asterisks indicate DMSO and DMF adducts. In the presence of DMSO or DMF, TTR-T4 complexes have a lower charge state and remain intact even at increased cone voltage. In the presence of methanol, T4 and monomeric and trimeric TTR are dissociated from the complexes at increased cone voltage.
dominant species. The same behavior was observed when T4 was added together with 3% DMF instead of DMSO (Figure 2, middle panel). When T4 was added with 3% methanol, isopropanol, or acetonitrile as controls, no pronounced change in the charge state distribution was observed. In the absence of DMSO, the complexes between T4 and TTR tetramers are shifted from a 2:1 toward a 1:1 stoichiometry (Figure 3, lower panel, and Supplementary Figure 5, Supporting Information). These complexes are almost completely dissociated at a cone voltage of 200 eV, giving the free tetramer and minor fractions of trimeric and monomeric TTR (Figure 2, lower panel). Charge State Reduction Contributes to Complex Stabilization. To investigate the causes of the increased complex stability in the presence of DMSO, we subjected the TTR tetramers with 18−15 charges, observed in the absence or the presence of DMSO, to CID at increasing collision voltages. We found that higher charge states can readily be dissociated into monomeric and trimeric TTR, while lower charge states are more resistant to CID and start at higher collision voltages to release peptide fragments rather than to dissociate.10 However, the 16+ charge state observed in both the presence and the absence of DMSO exhibited the same gas-phase stability, indicating that the increased tetramer stability is due to the lower charge states rather than to any DMSO-induced conformational changes (Figure 3A and Supplementary Figure 7, Supporting Information). We observed a similar charge state-dependent dissociation behavior during in-source dissociation of TTR-T4 complexes at pH 7.5. Plotting the amounts of the intact TTR complexes with two T4 molecules against the total amount of tetrameric TTR for each charge state at a cone voltage of 200 eV reveals that T4 preferentially dissociates from higher charge states (Figure 3B). However, a moderate increase in intact complexes for the 16+ charge state is seen in the presence of DMSO (Figure 3B and Supplementary Figure 7, Supporting Information). DMSO Protects Protein−Protein and Protein−Ligand Interactions of CTC. To determine whether the results with DMSO and TTR, above, apply also to other protein systems, we investigated the effects of low amounts of DMSO on the 4137
dx.doi.org/10.1021/ac500879c | Anal. Chem. 2014, 86, 4135−4139
Analytical Chemistry
Letter
Figure 3. CID stability of TTR tetramers and in-source dissociation of TTR-T4 complexes. (A) CID of the [M + 18H]18+, [M + 17H]17+, and [M + 16H]16+ ions observed in the absence and the [M + 16H]16+ and [M + 15H]15+ ions in the presence of 3% DMSO. The fraction of intact tetramers is calculated from the peak intensity for tetramers divided by the sum of the intensities of all TTR peaks in the MS/MS spectra. The gas-phase stability of the [M + 16H]16+ ion present under both conditions remains unaffected by DMSO. The dashed line indicates the onset of protein fragmentation. (B) Charge state-dependent in-source dissociation of the T4-TTR complexes. The fraction of intact complexes at high cone voltage for each charge state is calculated from the peak intensity for tetramers with two T4 bound divided by the sum of the intensities of all TTR peaks of the charge state. While all complexes are increasingly dissociated at higher charges, the [M + 16H]16+ TTR-T4 complex is more stable in the presence of DMSO.
contributions from Coulomb repulsions.19 This has been confirmed for TTR tetramers, which require disproportionally more energy to dissociate with decreasing charge state.10 The origin of the charge-reducing effects of DMSO in the electrospray process remains debated and has been attributed to the basic nature of DMSO as well as compaction of the protein fold.11,12,20 However, the pH of ammonium acetatebuffered electrospray solutions is not affected by the addition of up to 8% DMSO,13 and the observation that the [M + 16H]16+ charge states of tetrameric TTR in the presence or absence of DMSO have comparable gas-phase stabilities argues against major contributions from DMSO-induced conformational changes in the TTR tetramer (Figure 3A). An additional protective mechanism has been identified for the stabilization of weakly bound trypsin−inhibitor complexes by imidazole. Although charge state reduction was observed also in these cases, the cooling effect from the dissociation of nonspecifically bound imidazole during desolvation was identified as the reason for reduced in-source complex dissociation.19,21,22 Similar mechanisms can contribute to the protective effects of DMSO. Thus, we found that DMSO causes signal broadening indicative of adduct formation that could be reduced by increasing the cone voltage or collision energy. The observation that DMSO reduces in-source fragmentation of the 16+ charge state of the TTR-T4 complex, but does not increase the gas-phase stability of the apo-tetramer of the same charge (Figure 3), suggests that both charge state reduction and adduct dissociation contribute to the protection of TTR interactions by DMSO. Similar effects were observed for the protein−protein and protein−hydrophobic ligand interactions of CTC, suggesting that these effects are not protein specific. DMSO Distorts Protein−Ligand Interactions. In a recent study, DMSO was shown to reduce the amounts of noncovalent complexes of lysozyme with NAG3, carbonic anhydrase with chlorothiazide, and trypsin with Pefabloc in ESI-MS, leading to artificially increased Kd values.13 Those ligands are not dependent on hydrophobic solvents13 and interact with their target proteins through contributions from hydrogen bonding and charge interactions likely to remain intact after gentle desolvation.23,24 Thyroxine, on the other
stability of the quaternary structure and ligand binding of the CTC chaperone domain. Like TTR, CTC forms homomultimers and binds with hydrophobic ligands. Using ESI-MS and X-ray crystallography, we have previously investigated the trimeric structure of CTC and its interactions with synthetic peptide ligands that mimic its natural target polyvaline.16,17 At pH 7.5, ESI-MS shows a mixture of CTC trimers and monomers. Increasing the cone voltage to 200 eV shifts the distribution toward the monomeric form and leads to an increase in higher-charged monomers, indicating partial unfolding (Supplementary Figure 8, Supporting Information). In the presence of 3% DMSO, the charge states of monomers were lowered by one charge and the trimers by approximately two charges (Supplementary Figure 8, Supporting Information). Increasing the cone voltage did not affect the monomer− trimer distribution. We also monitored the interactions of CTC with the VVV peptide. In the presence of 3% DMSO, strong signals, corresponding to heteromers of one CTC and one or two VVV molecules, were observed. In the presence of 3% methanol, only small signals corresponding to heteromers composed of one CTC and one VVV molecule were detected (Supplementary Figure 9, Supporting Information). These findings suggest that DMSO has comparable protective effects on weak interactions of TTR as well as of the CTC chaperone in ESI-MS.
■
DISCUSSION Mechanism of DMSO-Mediated Protection of Noncovalent Interactions. In this study, we show that low amounts of DMSO protect labile protein−protein and protein−ligand interactions of TTR in ESI-MS. DMSO and also DMF, for which similar effects were observed in this study, have higher boiling points than water (189 and 155 °C, respectively) and are therefore enriched in the electrospray droplets during evaporation. Our data suggest that DMSO exerts a two-part protective effect on labile interactions through reduction of complex charges and cooling effects from dissociation of nonspecific adducts. It is well-established that lower charge states increase the gasphase stability of protein complexes by reducing the 4138
dx.doi.org/10.1021/ac500879c | Anal. Chem. 2014, 86, 4135−4139
Analytical Chemistry
Letter
(17) Willander, H.; Askarieh, G.; Landreh, M.; Westermark, P.; Nordling, K.; Keranen, H.; Hermansson, E.; Hamvas, A.; Nogee, L. M.; Bergman, T.; Saenz, A.; Casals, C.; Åqvist, J.; Jörnvall, H.; Berglund, H.; Presto, J.; Knight, S. D.; Johansson, J. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 2325−2329. (18) Palaninathan, S. K.; Mohamedmohaideen, N. N.; Snee, W. C.; Kelly, J. W.; Sacchettini, J. C. J. Mol. Biol. 2008, 382, 1157−1167. (19) Sun, J.; Kitova, E. N.; Klassen, J. S. Anal. Chem. 2007, 79, 416− 425. (20) Hunter, E. P. L.; Lias, S. G. J. Phys. Chem. Ref. Data 1998, 27, 413−656. (21) Bagal, D.; Kitova, E. N.; Liu, L.; El-Hawiet, A.; Schnier, P. D.; Klassen, J. S. Anal. Chem. 2009, 81, 7801−7806. (22) Cubrilovic, D.; Biela, A.; Sielaff, F.; Steinmetzer, T.; Klebe, G.; Zenobi, R. J. Am. Soc. Mass Spectrom. 2012, 23, 1768−1777. (23) Temperini, C.; Cecchi, A.; Scozzafava, A.; Supuran, C. T. Bioorg. Med. Chem. 2009, 17, 1214−1221. (24) Von Dreele, R. B. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2005, 61, 22−32. (25) Wojtczak, A.; Cody, V.; Luft, J. R.; Pangborn, W. Acta Crystallogr., Sect. D: Biol. Crystallogr. 1996, 52, 758−765. (26) Johannesson, H.; Denisov, V. P.; Halle, B. Protein Sci. 1997, 6, 1756−1763.
hand, is lipid-soluble and binds to tetrameric TTR via hydrophobic interactions25 that are generally less wellpreserved in the gas-phase. Here, the protective effects of DMSO predominate by enhancing the direct detection of the labile complexes. However, even total protection from insource dissociation would not be able to ameliorate any insolution effects on ligand binding that arise from competition between DMSO and ligands for binding sites.8,13,26 Therefore, caution must be taken when using DMSO in quantitative binding experiments. Besides these limitations, DMSO provides a simple means to stabilize labile protein interactions during desolvation, an important fact considering its widespread use in ESI-MS.
■
ASSOCIATED CONTENT
* Supporting Information S
Supplementary Figures 1−9. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Author Contributions
M.L. and H.J. designed the study. M.L. and G.A. performed experiments. J.J. provided materials and expertise. M.L. wrote the manuscript with contributions of all authors. All authors have given approval to the final version of the manuscript. Notes
The authors declare no competing financial interest.
■ ■
ACKNOWLEDGMENTS This study was supported by the Swedish Research Council. REFERENCES
(1) Katta, V.; Chait, B. T. J. Am. Chem. Soc. 1991, 113, 8534−8535. (2) Breuker, K.; McLafferty, F. W. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 18145−18152. (3) Loo, J. A. Mass Spectrom Rev. 1997, 16, 1−23. (4) Boeri Erba, E.; Barylyuk, K.; Yang, Y.; Zenobi, R. Anal. Chem. 2011, 83, 9251−9259. (5) Daniel, J. M.; Friess, S. D.; Rajagopalan, S.; Wendt, S.; Zenobi, R. Int. J. Mass Spectrom. 2002, 216, 1−27. (6) Bich, C.; Baer, S.; Jecklin, M. C.; Zenobi, R. J. Am. Soc. Mass Spectrom. 2010, 21, 286−289. (7) Hall, Z.; Robinson, C. V. J. Am. Soc. Mass Spectrom. 2012, 23, 1161−1168. (8) Benesch, J. L.; Ruotolo, B. T.; Sobott, F.; Wildgoose, J.; Gilbert, A.; Bateman, R.; Robinson, C. V. Anal. Chem. 2009, 81, 1270−1274. (9) Sterling, H. J.; Kintzer, A. F.; Feld, G. K.; Cassou, C. A.; Krantz, B. A.; Williams, E. R. J. Am. Soc. Mass Spectrom. 2012, 23, 191−200. (10) Pagel, K.; Hyung, S. J.; Ruotolo, B. T.; Robinson, C. V. Anal. Chem. 2010, 82, 5363−5372. (11) Sterling, H. J.; Prell, J. S.; Cassou, C. A.; Williams, E. R. J. Am. Soc. Mass Spectrom. 2011, 22, 1178−1186. (12) Tjernberg, A.; Markova, N.; Griffiths, W. J.; Hallén, D. J. Biomol. Screening 2006, 11, 131−137. (13) Cubrilovic, D.; Zenobi, R. Anal. Chem. 2013, 85, 2724−2730. (14) Johnson, S. M.; Connelly, S.; Fearns, C.; Powers, E. T.; Kelly, J. W. J. Mol. Biol. 2012, 421, 185−203. (15) McCammon, M. G.; Scott, D. J.; Keetch, C. A.; Greene, L. H.; Purkey, H. E.; Petrassi, H. M.; Kelly, J. W.; Robinson, C. V. Structure 2002, 10, 851−863. (16) Fitzen, M.; Alvelius, G.; Nordling, K.; Jörnvall, H.; Bergman, T.; Johansson, J. Rapid Commun. Mass Spectrom. 2009, 23, 3591−3598. 4139
dx.doi.org/10.1021/ac500879c | Anal. Chem. 2014, 86, 4135−4139
Protective Effects of Dimethyl Sulfoxide on Labile Protein Interactions during Electrospray Ionization Michael Landreh,*,† Gunvor Alvelius,† Jan Johansson,‡,§ and Hans Jörnvall† †
Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, S-171 77, Sweden KI-Alzheimer’s Disease Research Center, NVS Department, Karolinska Institutet, Stockholm, S-141 86, Sweden § Department of Anatomy, Physiology and Biochemistry, Swedish University of Agricultural Sciences, Uppsala, S-751 23, Sweden ‡
S Supporting Information *
ABSTRACT: Electrospray ionization mass spectrometry is a valuable tool to probe noncovalent interactions. However, the integrity of the interactions in the gas-phase is heavily influenced by the ionization process. Investigating oligomerization and ligand binding of transthyretin (TTR) and the chaperone domain from prosurfactant protein C, we found that dimethyl sulfoxide (DMSO) can improve the stability of the noncovalent interactions during the electrospray process, both regarding ligand binding and the protein quaternary structure. Low amounts of DMSO can reduce in-source dissociation of native protein oligomers and their interactions with hydrophobic ligands, even under destabilizing conditions. We interpret the effects of DMSO as being derived from its enrichment in the electrospray droplets during evaporation. Protection of labile interactions can arise from the decrease in ion charges to reduce the contributions from Coulomb repulsions, as well as from the cooling effect of adduct dissociation. The protective effects of DMSO on labile protein interactions are an important property given its widespread use in protein analysis by electrospray ionization mass spectrometry (ESI-MS).
E
electrospray process, while higher concentrations lead to supercharging.11 The onset of supercharging ranges from 2% to over 10%, dependent on the protein system, up to a certain threshold without significantly impacting the solution structure.11 However, caution must be taken when using DMSO in quantitative protein interaction studies, since it can reduce the amount of protein−ligand complexes observable by ESI-MS and then lead to a significant underestimation of binding affinities.12,13 We have investigated the effects of DMSO on the noncovalent protein−protein and protein−ligand interactions of transthyretin (TTR) and the C-terminal chaperone domain from lung surfactant proprotein C (CTC) in ESI-MS. TTR is a plasma carrier for the thyroid hormone thyroxine (T4) and is associated with several human amyloid diseases. The natively tetrameric TTR cooperatively binds two T4 molecules in a hydrophobic channel between two dimeric subunits, which prevents its dissociation and subsequent misfolding.14,15 CTC contains a conserved BRICHOS domain, which is a potent antiamyloid chaperone that binds hydrophobic protein segments and in this manner prevents aggregation.16,17 Due to the hydrophobic nature of natural and engineered TTR and CTC ligands, DMSO and similar compounds are commonly used as
lectrospray ionization mass spectrometry (ESI-MS) has evolved into an important tool for the study of protein structure and function. A growing body of evidence suggests that many proteins retain a near-native fold during the “soft” ionization from evaporating electrospray droplets and can be maintained in the gas-phase long enough to allow mass spectrometric analysis.1,2 ESI-MS is routinely applied to study protein interactions and has been used successfully for the determination of binding constants.3,4 Nevertheless, the desolvation of a protein can affect its structure, as hydrophobic interactions appear weakened and charge interactions strengthened.5,6 In addition, the charge state correlates to some extent with folding states in the gas-phase. Low charges and narrow charge distributions are usually observed for compactly folded states and more stable complexes,7,8 but even supercharged protein complexes can retain a native-like structure,7 although not in all cases.9 The charge states of a protein complex also dictate its dissociation pathways when subjected to collisioninduced dissociation (CID), with lower charge states being more resistant to CID, as they require a higher energy for dissociation.8,10 Ion charge states can be modulated chemically to obtain information about protein interactions. Dimethyl sulfoxide (DMSO), a common solvent for hydrophobic molecules and therefore often used in protein characterization and binding studies, is a two-step charge modulator. The presence of low concentrations of DMSO can lower ion charges during the © 2014 American Chemical Society
Received: March 8, 2014 Accepted: April 22, 2014 Published: April 22, 2014 4135
dx.doi.org/10.1021/ac500879c | Anal. Chem. 2014, 86, 4135−4139
Analytical Chemistry
Letter
Figure 1. DMSO-mediated charge state reduction and increased TTR tetramer stability in ESI-MS. (A) ESI-MS spectra of 30 μM TTR in the presence and absence of 3% DMSO and at 70 and 200 eV cone voltage. Increasing the cone voltage promotes the dissociation of TTR tetramers. The addition of 3% DMSO reduces ion charges and prevents in-source dissociation. (B) ESI-MS spectra of 30 μM at pH 4.4 in the presence of 0− 5% DMSO. With increasing DMSO concentration, tetramers become the dominant oligomeric species (highlighted in gray).
All TTR spectra at pH 4.4 were therefore recorded using aliquots that were frozen following a 24 h incubation at 4 °C. Lyophilized T4 (free acid) was dissolved to a final concentration of 1 mM in DMSO, dimethylformamide (DMF), isopropanol, acetonitrile, or methanol and stored at −20 °C. CTC (theoretical average molecular mass: 18 263 Da) was expressed and purified as described.16 Amidated and acetylated triple-valine was purchased from Thermo electron (Darmstadt, Germany). Immediately before analysis, samples were mixed to final concentrations of 30 μM TTR and 30 μM ligand. Cosolvents were added to v/v %. Spectra were acquired on a Q-ToF Ultima API mass spectrometer (Waters, Milford, MA) equipped with a Z-spray source in positive-ion mode. Samples were sprayed from metal-plated borosilicate glass capillary needles (Proxeon, Denmark). The source temperature was 70 °C. The capillary voltage was kept at 1.6 kV, and the cone and RF lens 1 potentials were 90 and 50 eV, respectively, unless otherwise noted. The mass spectrometer was operated in single-reflector mode for sufficient resolution and sensitivity
cosolvents in binding studies. Our results show that low concentrations of DMSO mitigate the dissociation of labile interactions during ionization and enhance their detection by ESI-MS.
■
EXPERIMENTAL SECTION Recombinant, His6-tagged human TTR (theoretical average molecular mass: 15150.5 Da) was purchased as lyophilized powder from Wako Chemicals (Neuss, Germany). All chemicals were purchased from Sigma (St. Louis, MO). TTR was dissolved in 20 mM ammonium acetate, pH 4.4 or pH 7.5, to a final concentration of 100 μM and stored in 100 μL aliquots at −20 °C. ESI-MS spectra of freshly dissolved TTR at pH 4.4 showed a high amount of tetramers which were found to disappear after 24 h at 4 °C, after which the spectra did not change (data not shown). We observed that a TTR concentration of 30 μM was required to obtain spectra showing the changes in protein oligomerization at pH 4.4. All spectra in this study were acquired at the same concentration. 4136
dx.doi.org/10.1021/ac500879c | Anal. Chem. 2014, 86, 4135−4139
Analytical Chemistry
Letter
(10 000, full width half-maximum definition). The mass scale was calibrated in linear mode using horse myoglobin (Sigma, St. Louis, MO). Scans were acquired at a rate of 1 scan per 2 s between 1000 and 6000 m/z. Collision gas was argon at 3.5 × 10−5 mbar. Mass spectra were analyzed with the Waters MassLynx 4.1 software package.
■
RESULTS DMSO Reduces in-Source Dissociation of TTR Tetramers. Since TTR ligand binding studies are routinely performed in the presence of low amounts of DMSO, we investigated the effects of 0−7% DMSO on the charge pattern of TTR in ESI-MS. At pH 7.5, almost exclusively, tetrameric TTR, with a compact charge envelope of ∼4 charge states centered around the [M + 16H]16+ ion, was detected (Figure 1A). The addition of 0.1−7% DMSO leads to a pronounced charge reduction, with the main charge envelope centered around the [M + 14H]14+ ion (Figure 1A and Supplementary Figure 1A, Supporting Information). We also observed signal broadening, indicative of the formation of DMSO adducts. The onset of supercharging occurs at 10% DMSO and is accompanied by a moderate broadening of the charge envelope, as noted before.11 Charge reduction was also observed in the presence of 0.1−7% DMF, while the presence of 0.1−7% acetonitrile, isopropanol, or methanol did not affect the charge state distribution to a similar extent (Supplementary Figure 1B−E, Supporting Information). Raising the cone voltage from 70 to 200 eV to facilitate in-source dissociation of protein− protein interactions,16 we found that, in the absence of DMSO, the tetramers dissociate into monomeric and trimeric TTR. In the presence of DMSO, however, the tetramers remain intact (Figure 1A). All tested DMSO concentrations between 0.1% and 7% exhibit the same pattern, while the higher charged complexes at 10% DMSO dissociate more easily (Supplementary Figure 2, Supporting Information). Another destabilizing factor for tetrameric TTR is low pH, which leads to local unfolding and reduced intersubunit contacts, and consequently, the subunits dissociate easily under these conditions.18 In line with the reduced tetramer stability at low pH, we find that TRR appears predominantly as monomers and dimers in ESI-MS spectra recorded at pH 4.4. However, the addition of 3−5% DMSO leads to the detection of intact tetramers, indicating that DMSO can in part prevent the dissociation of the pH-destabilized subunit contacts (Figure 1B). Again, similar effects were observed for DMF but not for acetonitrile, isopropanol, or methanol (Supplementary Figure 3, Supporting Information). These observations suggest that DMSO can facilitate the detection of labile protein interactions that may otherwise not be observable by ESI-MS. DMSO and DMF Stabilize TTR-T4 Interactions. Next, we asked whether DMSO might also affect the detection of TTR-T4 complexes. Due to their high hydrophobicity, TTR ligands usually are dissolved in DMSO, DMF, or related solvents. At pH 7.5 and pH 4.4, near-complete saturation with two T4 moieties and a minor peak corresponding to one T4 moiety per tetramer are observed in the presence of 3% DMSO and at a 2:1 excess of T4 over tetrameric TTR (Figure 2A, top panel, and Supplementary Figure 4, Supporting Information). We furthermore observed a shoulder at the high m/z side of the complex ions which can be attributed to the formation of unspecific DMSO adducts. Increasing the cone voltage from 70 to 200 eV results in an increase of 1:1 complexes between T4 and tetrameric TTR, but the 2:1 complex remains the
Figure 2. DMSO and DMF reduce in-source dissociation of TTR-T4 complexes. ESI-MS spectra of 30 μM TTR at pH 7.5 in the presence of 30 μM T4 added from DMSO, DMF, or methanol stock solutions. Green, yellow, and red bars indicate the tetrameric TTR with two, one, or no T4 molecules attached. Asterisks indicate DMSO and DMF adducts. In the presence of DMSO or DMF, TTR-T4 complexes have a lower charge state and remain intact even at increased cone voltage. In the presence of methanol, T4 and monomeric and trimeric TTR are dissociated from the complexes at increased cone voltage.
dominant species. The same behavior was observed when T4 was added together with 3% DMF instead of DMSO (Figure 2, middle panel). When T4 was added with 3% methanol, isopropanol, or acetonitrile as controls, no pronounced change in the charge state distribution was observed. In the absence of DMSO, the complexes between T4 and TTR tetramers are shifted from a 2:1 toward a 1:1 stoichiometry (Figure 3, lower panel, and Supplementary Figure 5, Supporting Information). These complexes are almost completely dissociated at a cone voltage of 200 eV, giving the free tetramer and minor fractions of trimeric and monomeric TTR (Figure 2, lower panel). Charge State Reduction Contributes to Complex Stabilization. To investigate the causes of the increased complex stability in the presence of DMSO, we subjected the TTR tetramers with 18−15 charges, observed in the absence or the presence of DMSO, to CID at increasing collision voltages. We found that higher charge states can readily be dissociated into monomeric and trimeric TTR, while lower charge states are more resistant to CID and start at higher collision voltages to release peptide fragments rather than to dissociate.10 However, the 16+ charge state observed in both the presence and the absence of DMSO exhibited the same gas-phase stability, indicating that the increased tetramer stability is due to the lower charge states rather than to any DMSO-induced conformational changes (Figure 3A and Supplementary Figure 7, Supporting Information). We observed a similar charge state-dependent dissociation behavior during in-source dissociation of TTR-T4 complexes at pH 7.5. Plotting the amounts of the intact TTR complexes with two T4 molecules against the total amount of tetrameric TTR for each charge state at a cone voltage of 200 eV reveals that T4 preferentially dissociates from higher charge states (Figure 3B). However, a moderate increase in intact complexes for the 16+ charge state is seen in the presence of DMSO (Figure 3B and Supplementary Figure 7, Supporting Information). DMSO Protects Protein−Protein and Protein−Ligand Interactions of CTC. To determine whether the results with DMSO and TTR, above, apply also to other protein systems, we investigated the effects of low amounts of DMSO on the 4137
dx.doi.org/10.1021/ac500879c | Anal. Chem. 2014, 86, 4135−4139
Analytical Chemistry
Letter
Figure 3. CID stability of TTR tetramers and in-source dissociation of TTR-T4 complexes. (A) CID of the [M + 18H]18+, [M + 17H]17+, and [M + 16H]16+ ions observed in the absence and the [M + 16H]16+ and [M + 15H]15+ ions in the presence of 3% DMSO. The fraction of intact tetramers is calculated from the peak intensity for tetramers divided by the sum of the intensities of all TTR peaks in the MS/MS spectra. The gas-phase stability of the [M + 16H]16+ ion present under both conditions remains unaffected by DMSO. The dashed line indicates the onset of protein fragmentation. (B) Charge state-dependent in-source dissociation of the T4-TTR complexes. The fraction of intact complexes at high cone voltage for each charge state is calculated from the peak intensity for tetramers with two T4 bound divided by the sum of the intensities of all TTR peaks of the charge state. While all complexes are increasingly dissociated at higher charges, the [M + 16H]16+ TTR-T4 complex is more stable in the presence of DMSO.
contributions from Coulomb repulsions.19 This has been confirmed for TTR tetramers, which require disproportionally more energy to dissociate with decreasing charge state.10 The origin of the charge-reducing effects of DMSO in the electrospray process remains debated and has been attributed to the basic nature of DMSO as well as compaction of the protein fold.11,12,20 However, the pH of ammonium acetatebuffered electrospray solutions is not affected by the addition of up to 8% DMSO,13 and the observation that the [M + 16H]16+ charge states of tetrameric TTR in the presence or absence of DMSO have comparable gas-phase stabilities argues against major contributions from DMSO-induced conformational changes in the TTR tetramer (Figure 3A). An additional protective mechanism has been identified for the stabilization of weakly bound trypsin−inhibitor complexes by imidazole. Although charge state reduction was observed also in these cases, the cooling effect from the dissociation of nonspecifically bound imidazole during desolvation was identified as the reason for reduced in-source complex dissociation.19,21,22 Similar mechanisms can contribute to the protective effects of DMSO. Thus, we found that DMSO causes signal broadening indicative of adduct formation that could be reduced by increasing the cone voltage or collision energy. The observation that DMSO reduces in-source fragmentation of the 16+ charge state of the TTR-T4 complex, but does not increase the gas-phase stability of the apo-tetramer of the same charge (Figure 3), suggests that both charge state reduction and adduct dissociation contribute to the protection of TTR interactions by DMSO. Similar effects were observed for the protein−protein and protein−hydrophobic ligand interactions of CTC, suggesting that these effects are not protein specific. DMSO Distorts Protein−Ligand Interactions. In a recent study, DMSO was shown to reduce the amounts of noncovalent complexes of lysozyme with NAG3, carbonic anhydrase with chlorothiazide, and trypsin with Pefabloc in ESI-MS, leading to artificially increased Kd values.13 Those ligands are not dependent on hydrophobic solvents13 and interact with their target proteins through contributions from hydrogen bonding and charge interactions likely to remain intact after gentle desolvation.23,24 Thyroxine, on the other
stability of the quaternary structure and ligand binding of the CTC chaperone domain. Like TTR, CTC forms homomultimers and binds with hydrophobic ligands. Using ESI-MS and X-ray crystallography, we have previously investigated the trimeric structure of CTC and its interactions with synthetic peptide ligands that mimic its natural target polyvaline.16,17 At pH 7.5, ESI-MS shows a mixture of CTC trimers and monomers. Increasing the cone voltage to 200 eV shifts the distribution toward the monomeric form and leads to an increase in higher-charged monomers, indicating partial unfolding (Supplementary Figure 8, Supporting Information). In the presence of 3% DMSO, the charge states of monomers were lowered by one charge and the trimers by approximately two charges (Supplementary Figure 8, Supporting Information). Increasing the cone voltage did not affect the monomer− trimer distribution. We also monitored the interactions of CTC with the VVV peptide. In the presence of 3% DMSO, strong signals, corresponding to heteromers of one CTC and one or two VVV molecules, were observed. In the presence of 3% methanol, only small signals corresponding to heteromers composed of one CTC and one VVV molecule were detected (Supplementary Figure 9, Supporting Information). These findings suggest that DMSO has comparable protective effects on weak interactions of TTR as well as of the CTC chaperone in ESI-MS.
■
DISCUSSION Mechanism of DMSO-Mediated Protection of Noncovalent Interactions. In this study, we show that low amounts of DMSO protect labile protein−protein and protein−ligand interactions of TTR in ESI-MS. DMSO and also DMF, for which similar effects were observed in this study, have higher boiling points than water (189 and 155 °C, respectively) and are therefore enriched in the electrospray droplets during evaporation. Our data suggest that DMSO exerts a two-part protective effect on labile interactions through reduction of complex charges and cooling effects from dissociation of nonspecific adducts. It is well-established that lower charge states increase the gasphase stability of protein complexes by reducing the 4138
dx.doi.org/10.1021/ac500879c | Anal. Chem. 2014, 86, 4135−4139
Analytical Chemistry
Letter
(17) Willander, H.; Askarieh, G.; Landreh, M.; Westermark, P.; Nordling, K.; Keranen, H.; Hermansson, E.; Hamvas, A.; Nogee, L. M.; Bergman, T.; Saenz, A.; Casals, C.; Åqvist, J.; Jörnvall, H.; Berglund, H.; Presto, J.; Knight, S. D.; Johansson, J. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 2325−2329. (18) Palaninathan, S. K.; Mohamedmohaideen, N. N.; Snee, W. C.; Kelly, J. W.; Sacchettini, J. C. J. Mol. Biol. 2008, 382, 1157−1167. (19) Sun, J.; Kitova, E. N.; Klassen, J. S. Anal. Chem. 2007, 79, 416− 425. (20) Hunter, E. P. L.; Lias, S. G. J. Phys. Chem. Ref. Data 1998, 27, 413−656. (21) Bagal, D.; Kitova, E. N.; Liu, L.; El-Hawiet, A.; Schnier, P. D.; Klassen, J. S. Anal. Chem. 2009, 81, 7801−7806. (22) Cubrilovic, D.; Biela, A.; Sielaff, F.; Steinmetzer, T.; Klebe, G.; Zenobi, R. J. Am. Soc. Mass Spectrom. 2012, 23, 1768−1777. (23) Temperini, C.; Cecchi, A.; Scozzafava, A.; Supuran, C. T. Bioorg. Med. Chem. 2009, 17, 1214−1221. (24) Von Dreele, R. B. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2005, 61, 22−32. (25) Wojtczak, A.; Cody, V.; Luft, J. R.; Pangborn, W. Acta Crystallogr., Sect. D: Biol. Crystallogr. 1996, 52, 758−765. (26) Johannesson, H.; Denisov, V. P.; Halle, B. Protein Sci. 1997, 6, 1756−1763.
hand, is lipid-soluble and binds to tetrameric TTR via hydrophobic interactions25 that are generally less wellpreserved in the gas-phase. Here, the protective effects of DMSO predominate by enhancing the direct detection of the labile complexes. However, even total protection from insource dissociation would not be able to ameliorate any insolution effects on ligand binding that arise from competition between DMSO and ligands for binding sites.8,13,26 Therefore, caution must be taken when using DMSO in quantitative binding experiments. Besides these limitations, DMSO provides a simple means to stabilize labile protein interactions during desolvation, an important fact considering its widespread use in ESI-MS.
■
ASSOCIATED CONTENT
* Supporting Information S
Supplementary Figures 1−9. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Author Contributions
M.L. and H.J. designed the study. M.L. and G.A. performed experiments. J.J. provided materials and expertise. M.L. wrote the manuscript with contributions of all authors. All authors have given approval to the final version of the manuscript. Notes
The authors declare no competing financial interest.
■ ■
ACKNOWLEDGMENTS This study was supported by the Swedish Research Council. REFERENCES
(1) Katta, V.; Chait, B. T. J. Am. Chem. Soc. 1991, 113, 8534−8535. (2) Breuker, K.; McLafferty, F. W. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 18145−18152. (3) Loo, J. A. Mass Spectrom Rev. 1997, 16, 1−23. (4) Boeri Erba, E.; Barylyuk, K.; Yang, Y.; Zenobi, R. Anal. Chem. 2011, 83, 9251−9259. (5) Daniel, J. M.; Friess, S. D.; Rajagopalan, S.; Wendt, S.; Zenobi, R. Int. J. Mass Spectrom. 2002, 216, 1−27. (6) Bich, C.; Baer, S.; Jecklin, M. C.; Zenobi, R. J. Am. Soc. Mass Spectrom. 2010, 21, 286−289. (7) Hall, Z.; Robinson, C. V. J. Am. Soc. Mass Spectrom. 2012, 23, 1161−1168. (8) Benesch, J. L.; Ruotolo, B. T.; Sobott, F.; Wildgoose, J.; Gilbert, A.; Bateman, R.; Robinson, C. V. Anal. Chem. 2009, 81, 1270−1274. (9) Sterling, H. J.; Kintzer, A. F.; Feld, G. K.; Cassou, C. A.; Krantz, B. A.; Williams, E. R. J. Am. Soc. Mass Spectrom. 2012, 23, 191−200. (10) Pagel, K.; Hyung, S. J.; Ruotolo, B. T.; Robinson, C. V. Anal. Chem. 2010, 82, 5363−5372. (11) Sterling, H. J.; Prell, J. S.; Cassou, C. A.; Williams, E. R. J. Am. Soc. Mass Spectrom. 2011, 22, 1178−1186. (12) Tjernberg, A.; Markova, N.; Griffiths, W. J.; Hallén, D. J. Biomol. Screening 2006, 11, 131−137. (13) Cubrilovic, D.; Zenobi, R. Anal. Chem. 2013, 85, 2724−2730. (14) Johnson, S. M.; Connelly, S.; Fearns, C.; Powers, E. T.; Kelly, J. W. J. Mol. Biol. 2012, 421, 185−203. (15) McCammon, M. G.; Scott, D. J.; Keetch, C. A.; Greene, L. H.; Purkey, H. E.; Petrassi, H. M.; Kelly, J. W.; Robinson, C. V. Structure 2002, 10, 851−863. (16) Fitzen, M.; Alvelius, G.; Nordling, K.; Jörnvall, H.; Bergman, T.; Johansson, J. Rapid Commun. Mass Spectrom. 2009, 23, 3591−3598. 4139
dx.doi.org/10.1021/ac500879c | Anal. Chem. 2014, 86, 4135−4139
