English

• Protein-small ligand interactions are universal and of crucial importance in cellular processes [1–3]. Flavonoids serve as specialized plant metabolites and often act as effective pharmaceutical agents [4–9]. Previous studies on the interactions of flavonols and proteins, such as hemoglobin, bovine serum albumin, β-lactogolobulin, and others, reveal that flavonols form complexes with proteins, generating diverse structural and physicochemical characteristics [10–15]. However, limited information can be obtained regarding how the intrinsic property of protein is affected by flavonols.

  Mass spectrometry (MS) has become an essential method for exploring non-covalent interactions [16–22]. It only requires minimal sample quantities, and can be easily conducted without the need for extensive purification or labeling [23–26]. By preserving either covalent or non-covalent interactions during the transition from solution to the gas phase, MS provides initial stoichiometric information on the composition of the complexes [27], their average charge states, and excels in extracting stability from protein mixtures [28]. The combination of MS with other techniques such as equilibrium dialysis, ion mobility (IM) and Förster resonance energy transfer (FRET) can also provide more information of the interactions [29–31]. However, the correlation between solution and gas-phase stabilities remains an unanswered question, despite extensive efforts from researchers to comprehend the behavior of proteins [32–34] in both gas and solution phases.

  Employing MS detection to track changes in protein structures, temperature-controlled electrospray ionization (TC-ESI), also called variable-temperature electrospray ionization (vt-ESI), has been emergingly used as an equivalent to conventional solution-phase temperature annealing methods [35]. Measuring melting temperature (T_m) using calorimetry or spectroscopy is inherently slow, requires a substantial amount of purified material (0.1–2 mg), and poses challenges in terms of multiplexing. This creates a critical bottleneck for extracting stability measurements from protein mixtures [36–38]. However, TC-ESI is also useful to depict protein T_m values with MS-grade sensitivity and specificity, which proves to be in line with traditional technologies [39–48]. Moreover, its strength lies in its ability to unequivocally identify and quantify each of the multiple coexisting stoichiometries in a mixture. Structural information on ligand binding has also been primarily investigated by these methods [49–51].

  Tandem MS (MSⁿ), usually using collision-induced dissociation (CID), has been widely employed for native protein complexes [52]. MSⁿ for protein-ligand complexes in the gas phase may offer valuable and quantitative insights into protein-ligand interactions [53,54]. Nevertheless, the degree of ion dissociation following the implementation of CID is contingent on the dissociation rate constant for the gaseous ion [55]. Therefore, the gas-phase stability of protein-ligand complexes relies on the dissociation thermochemistry and kinetics of the complex in the gas phase. This is not anticipated to accurately mirror solution-phase binding affinities [55–57].

  The development of TC-ESI primarily focused on the temperature control methods (Scheme 1) [58], but it was seldom combined with other MS techniques. The integration of TC-ESI with MSⁿ greatly expands the application field of this ion source, enabling more effective exploration of interactions between biomacromolecules and ligands. Furthermore, due to the capability of block heating to effectively control parameters such as heating time, flow rate, and temperature, it is highly suitable for compensating MSⁿ method to study subtle changes in protein-small molecule interactions. Combining TC-ESI with MSⁿ allows for a rapid and simultaneous assessment of the stability of the complex in both gas and liquid phases under various temperature conditions. Additionally, it facilitates the examination of how these conditions affect protein structure, highlighting structural alterations in proteins affected by small molecules that often go undetected in traditional ESI-MSⁿ. Temperature-controlled ESI-MS in combination with tandem mass spectrometry (TC-ESI-MSⁿ) experiments allow for the observation of differences in dissociation patterns and affinities of complexes across various temperature dimensions in the gas phase. By integrating with TC-ESI experiments, it can facilitate a more comprehensive comparison of flavonol-protein interactions between gas and solution phases. Furthermore, this combined strategy will provide new insights into proteins and enable careful tracking of changes in thermodynamic data, and the characteristics of structural intermediates. Compared with traditional ESI-MS, this method can offer unprecedented insights and an additional layer of detail into protein-small molecule interactions. A schematic drawing of the experimental setup is shown in Fig. S1 in Supporting information.

  Scheme 1

  

  Scheme 1.  Timeline of the development of TC-ESI-MSⁿ.

  DownLoad: Full-Size Img PowerPoint

  In this work, a prototype protein ubiquitin (U), one of the most investigated proteins with +5 to +8 charge states under native MS conditions, is firstly selected [59]. Then, TC-ESI-MSⁿ is used to study the interaction of ubiquitin with two typical flavonols (Scheme 2), quercetin (Q) and rutin (R), respectively.

  Scheme 2

  

  Scheme 2.  Structures of (a) quercetin (Q, left) and (b) rutin (R, right).

  DownLoad: Full-Size Img PowerPoint

  At room temperature, the spray of ubiquitin and quercetin mixture in methanol/water solution produces ions in U, U-Q₁ to U-Q₃ at +5 and +6 charge states (Fig. 1a). The 1:1 U-Q₁ complex is to date the most abundant in the spectrum while the 1:2 U-Q₂ complex is of lower abundance. Similarly, rutin and ubiquitin complex solution sprays to forms U, U-R₁ to U-R₂ at +5 and +6 charge states (Fig. 1b).

  Figure 1

  

  Figure 1.  Mass spectra of ubiquitin with different flavonols.

  DownLoad: Full-Size Img PowerPoint

  Weighted average charge state of ubiquitin (Fig. 2) and mass spectra (inset of Fig. 2) versus solution temperature of ubiquitin obtained in methanol/water mixture are in line with temperature-dependent unfolding transitions of ubiquitin observed by El-Baba et al. [42]. Notably, the melting temperature derived in Fig. 2 is 45 ℃, which is different from previous results, such as in aqueous solution with pH 3 [60]. This result is consistent with earlier studies and is primarily attributed to changes in solvent conditions within the solution (Figs. S2 and S3 in Supporting information) [61,62].

  Figure 2

  

  Figure 2.  Weighted average charge state of ubiquitin versus solution temperature with midpoint melting temperature T_m = 45.4 ℃. Inset shows mass spectra of ubiquitin in 20% (v/v) methanol in water in an acidified aqueous solution collected at 25, 50, and 70 ℃ at flow rate of 5 µL/min. Weighted average charge state is calculated using Eq. S1 in Supporting information. The values are presented as mean ± standard deviation (n = 3).

  DownLoad: Full-Size Img PowerPoint

  As shown in Fig. 3a, investigation on temperature-dependent effects of flavonols binding to ubiquitin has been conducted. As temperature grows, binding of quercetin keeps the same binding stoichiometries. A zoomed mass spectrum for ubiquitin-quercetin (Fig. 3b) shows the formation of the noncovalent U-Q₁ to U-Q₃ complex during thermal denaturation. However, the peaks for +5 and +6 ions decrease against increasing temperature, making it challenging to intuitively compare the binding affinities between ligands and ubiquitin. Similarly, thermal denaturation of ubiquitin-rutin complex (Figs. 3c and d) has similar trend mass spectra during the entire heating process and no obvious new binding peaks of U-R complex are observed.

  Figure 3

  

  Figure 3.  (a) Mass spectra of the ubiquitin-quercetin complex. (b) Zoomed-in mass spectra (m/z 1450–1650) of the [U-Q_1–3]⁶⁺ ions. (c) Mass spectra of the ubiquitin-rutin complex and (d) zoomed-in mass spectra (m/z 1500–1700) of the [U-R_1–2]⁶⁺ ions, all at different temperatures, respectively.

  DownLoad: Full-Size Img PowerPoint

  To better understand the complex mechanism, Eq. 1 has been utilized to assess the relative binding affinities of flavonols [63], where I(M) and I(1:n) represent the relative abundance of each charge state of the free ubiquitin and various binding stoichiometries of ubiquitin-flavonol complexes. It is assumed that the relative intensities of the proteins or complexes in the mass spectra are directly proportional to their relative abundances in the solution [63,64]. The higher fraction value of ubiquitin in the TC-ESI-MS experiment indicates greater binding affinities.

  (1)

  As shown in Fig. 4, the relative ligand affinities of flavonol-ubiquitin complexes at different temperatures can be calculated from their extractive mass spectra, respectively. It illustrates two distinctly different trends. The binding affinity between quercetin and ubiquitin consistently decreases as temperature rises. In contrast, the binding affinity between rutin and ubiquitin increases with temperature elevation.

  Figure 4

  

  Figure 4.  Binding affinities based on TC-ESI-MS measurements of flavonol-ubiquitin complexes at different temperatures. The values are presented as mean ± standard deviation (n = 3).

  DownLoad: Full-Size Img PowerPoint

  The discrepancy in the interaction of proteins with quercetin and rutin is proposed to be attributed to the structural differences between the flavonoid ligands [65,66]. The U/Q complex becomes less stable at elevated temperatures, indicating an exothermic nature of their interaction as in soy protein [67]. In contrast, the binding affinity of U/R complexes exhibits notable increments as the temperature rises, indicating that these complexes are formed through endothermic processes. These results align with those observed in the interactions between rutin and soybean protein isolates [68]. Hydrogen bonds and hydrophobic interactions are reported to have key roles in the non-covalent binding of flavonoids and proteins [69]. The strengthening of the hydrophobic interactions in the U/R complexes at higher temperatures could potentially enhance their stability [70], while the binding between ubiquitin and quercetin may primarily be mediated by hydrogen bonds.

  In order to further explore the interactions between the complexes in gas phase, the gas phase stability of the two ligand complexes by TC-ESI-MSⁿ has been introduced. What deserves expecting is that combination of CID to quantify the stability of the protein-ligand complexes in the gas phase can offer meaningful and quantitative insights into the interactions.

  Fig. 5 shows the dissociation profiles of the TC-ESI-MSⁿ experiment performed on the +6 charge state of the ubiquitin-flavonols complex, with molar ratios of 1:1 and 1:2 at various temperatures. As will be shown, structural changes of proteins disturbed by small drug molecules are unseen in traditional ESI-MSⁿ, but the introduction of temperature dimension in this study can reveal subtle interactions of flavonols on ubiquitin.

  Figure 5

  

  Figure 5.  Dissociation profiles of (a) U-Q₁, (b) U-Q₂, (c) U-R₁ and (d) U-R₂ at different temperatures. The values are presented as mean ± standard deviation (n = 3).

  DownLoad: Full-Size Img PowerPoint

  The half-wave collision energy (E_1/2) is defined as the collision energy at which the relative abundance of the complex precursor ion, reaches 0.5 [71]. E_1/2 corresponds to T_m in a ubiquitin melting curve. The larger the E_1/2 value, the more stable the complex in the gas phase. As depicted in Figs. 5a and b, quercetin gives similar E_1/2 values for its complexes with ubiquitin at different temperatures. The similarity in the dissociation profiles suggests that the noncovalent complexes adopt very similar gas-phase conformations even when they transition into an unfolded state. Meanwhile, as the molar ratio of the complex increases from 1:1 to 1:2, the E_1/2 value increase significantly. The observed greater stability of U-Q₂ complex may be due to the stronger and more opportunities of intramolecular interactions between the two ligands. This pattern is consistent with the data shown in Figs. 5c and d. Rutin exhibits comparable E_1/2 values for its interactions with ubiquitin across various temperatures.

  It is also worth noting that when ubiquitin with one ligand is activated, it dissociates to give back the ubiquitin. When ubiquitin with two ligands is activated, it initially loses one ligand before reverting to ubiquitin itself (Figs. S4 and S5 in Supporting information). The CID results of corresponding ubiquitin-flavonol complexes fail to reflect the solution behavior. This can be attributed to the destruction of the specific interaction between the flavonols and the ubiquitin in the protonated complex. A possible explanation is that the charge state of U-Q or U-R depends on the aquas structure. Thus, U-Q or U-R with the same charge state but produced from different temperatures have the same or similar conformations, resulting in unchanged gas phase stability. Therefore, the addition of small molecule ligands under different temperature conditions does not alter the gas-phase stability of the complex, in contrast to the alterations observed in the liquid phase. However, it should be noted that the comparison between gas phase with solution stabilities does not provide insights into the stability changes of ubiquitin itself under the influence of quercetin and rutin.

  To even further understand the thermal denaturation of ubiquitin in the presence of quercetin and rutin, thermal stability is investigated [72]. The average charge states of ubiquitin in the mixed solutions are then calculated from the intensities, and plotted as a function of temperature (Fig. 6a).

  Figure 6

  

  Figure 6.  (a) Weighted average charge state versus solution temperature for ubiquitin (blue triangles), ubiquitin + rutin (red circles), ubiquitin + quercetin (black squares). The midpoint melting temperatures (T_m) determined from sigmoidal curve are shown with colored arrows. (b) ΔG values for ubiquitin (blue rectangle), ubiquitin + rutin (red rectangle), and ubiquitin + quercetin (black rectangle) are calculated at various temperatures. The values are presented as mean ± standard deviation (n = 3).

  DownLoad: Full-Size Img PowerPoint

  In the presence of quercetin or rutin, the melting temperatures of ubiquitin are both increased from 45.4 ℃ to 53.4 ℃ or 52.8 ℃, respectively. The results indicate that the binding of quercetin and rutin significantly enhances the stability of ubiquitin. Yet the shape of the melting curve appears to be similar to ubiquitin individually, suggesting no big changes happen during the ubiquitin unfolding process. One common approach to assess ligand binding affinities is by calculating the ΔT_m values obtained from experiments conducted with and without the ligand. Based on MS-melting studies, the two flavonols obtain similar ΔT_m values and show their significant stabilization.

  Gibbs free energy (ΔG) derived from the MS-melting experiment is used to simulate the direction of unfolding process of proteins during heating. As the solution temperature increases, the folded ubiquitin transforms into the unfolded form, and the unfolding process can be described and calculated (Eqs. S2 and S3 in Supporting information). As shown in Fig. 6b, in the presence of quercetin and rutin, ΔG increases significantly as the temperature increases from 25 ℃ to 55 ℃, demonstrating that ubiquitin necessitates higher energy input for denaturation. However, ΔG does not change significantly at above 55 ℃, implying that ubiquitin has the similar tendency to unfold. Notably, at 55 ℃, ΔG values of isolated ubiquitin are negative, but in the presence of quercetin and rutin, ΔG values remain positive. This demonstrates that in a solution containing only ubiquitin, a spontaneous unfolding process occurs at this temperature.

  In summary, TC-ESI-MSⁿ is used to study the interactions of ubiquitin with two typical flavonols: quercetin and rutin in both solution and gas phases. The interactions at different temperature conditions can be obtained and determined from the MS data. The capability to distinguish and quantify distinct binding stoichiometries, binding affinities and stabilities in both solution and gas phases at various temperatures offers detailed insights into the complexes formed by ubiquitin and flavonols.

  These results further demonstrate that TC-ESI-MSⁿ also has the capacity to investigate protein T_m and ΔG that aligns with conventional technologies. In the presence of quercetin and rutin, the T_m and ΔG of ubiquitin are significantly higher than ubiquitin alone, indicating an enhanced stability of ubiquitin. As compared to conventional ESI-MSⁿ experiment, TC-ESI-MSⁿ has yielded more valuable insights into the protein-small molecule complexes. It simultaneously and rapidly compares the stability of complexes in both the gas and condensed phases, and further reveals the intrinsic property of proteins. Since there are quite few studies on ubiquitin-flavonol complexes using TC-ESI-MSⁿ previously, this combined strategy will provide new insights into proteins and inspire further studies on a wider-range of protein-small molecule interactions.

  
  
• 1. [1]

     F. Chen, B. Gülbakan, S. Weidmann, et al., Mass Spectrom. Rev. 35 (2016) 48–70. doi: 10.1002/mas.21462

  2. [2]

     J.A. Loo, Mass Spectrom. Rev. 16 (1997) 1–23. doi: 10.1002/(SICI)1098-2787(1997)16:1<1::AID-MAS1>3.0.CO;2-L

  3. [3]

     J.L. Bennett, G.T. Nguyen, W.A. Donald, Chem. Rev. 122 (2022) 7327–7385. doi: 10.1021/acs.chemrev.1c00293

  4. [4]

     C.F. Lourenço, B. Gago, R.M. Barbosa, et al., J. Agric. Food Chem. 56 (2008) 3798–3804. doi: 10.1021/jf0733259

  5. [5]

     J.J. Peterson, J.T. Dwyer, P.F. Jacques, et al., Nutr. Rev. 70 (2012) 491–508. doi: 10.1111/j.1753-4887.2012.00508.x

  6. [6]

     P. Xin, S. Han, J. Huang, et al., Chin. Chem. Lett. 34 (2023) 108125. doi: 10.1016/j.cclet.2022.108125

  7. [7]

     D. Treutter, Environ. Chem. Lett. 4 (2006) 147–157. doi: 10.1007/s10311-006-0068-8

  8. [8]

     J. Laoué, C. Fernandez, E. Ormeño, Plants 11 (2022) 172. doi: 10.3390/plants11020172

  9. [9]

     X. Zhang, J. Li, K.Z. Lu, et al., Chin. Chem. Lett. 35 (2024) 109456. doi: 10.1016/j.cclet.2023.109456

  10. [10]

      C.D. Kanakis, I. Hasni, P. Bourassa, et al., Food Chem. 127 (2011) 1046–1055. doi: 10.1016/j.foodchem.2011.01.079

  11. [11]

      Y. Yan, J. Hu, P. Yao, Langmuir 25 (2009) 397–402. doi: 10.1021/la8030123

  12. [12]

      V. de Freitas, N. Mateus, Curr. Org. Chem. 16 (2012) 724–746. doi: 10.2174/138527212799958002

  13. [13]

      A. Papadopoulou, R.A. Frazier, Trends Food Sci. Technol. 15 (2004) 186–190. doi: 10.1016/j.tifs.2003.09.017

  14. [14]

      J. Xiao, G. Kai, Crit. Rev. Food Sci. 52 (2012) 85–101. doi: 10.1080/10408398.2010.499017

  15. [15]

      J. Xi, R. Guo, Int. J. Biol. Macromol. 40 (2007) 305–311. doi: 10.1016/j.ijbiomac.2006.08.011

  16. [16]

      J.D. Eschweiler, R. Kerr, J. Rabuck-Gibbons, et al., Annu. Rev. Anal. Chem. 10 (2017) 25–44. doi: 10.1146/annurev-anchem-061516-045414

  17. [17]

      A.J. Heck, Nat. Methods 5 (2008) 927–933. doi: 10.1038/nmeth.1265

  18. [18]

      N.P. Barrera, S.C. Isaacson, M. Zhou, et al., Nat. Methods 6 (2009) 585–587. doi: 10.1038/nmeth.1347

  19. [19]

      J. Marcoux, C.V. Robinson, Structure 21 (2013) 1541–1550. doi: 10.1016/j.str.2013.08.002

  20. [20]

      N.P. Barrera, C.V. Robinson, Annu. Rev. Biochem. 80 (2011) 247–271. doi: 10.1146/annurev-biochem-062309-093307

  21. [21]

      L. Ma, S. Liu, N.S. Xu, et al., Chin. Chem. Lett. 25 (2014) 1179–1184. doi: 10.1016/j.cclet.2014.03.049

  22. [22]

      Z. Guo, Y. Liu, N. He, et al., Chin. Chem. Lett. 32 (2021) 40–47. doi: 10.1016/j.cclet.2020.11.061

  23. [23]

      E.B. Erba, R. Zenobi, Annu. Rep. Prog. Chem. Sect. C 107 (2011) 199–228. doi: 10.1039/c1pc90006d

  24. [24]

      D. Rathore, A. Faustino, J. Schiel, et al., Expert Rev. Proteomics 15 (2018) 431–449. doi: 10.1080/14789450.2018.1469982

  25. [25]

      C. Guo, X. Zhang, X. Hong, et al., Chin. Chem. Lett. 35 (2024) 108867. doi: 10.1016/j.cclet.2023.108867

  26. [26]

      X. Chang, N. Wang, D. Jiang, et al., Chin. Chem. Lett. 34 (2023) 107522. doi: 10.1016/j.cclet.2022.05.036

  27. [27]

      H. Hernández, C.V. Robinson, Nat. Protoc. 2 (2007) 715–726. doi: 10.1038/nprot.2007.73

  28. [28]

      D.D. Vallejo, C. Rojas Ramiŕez, K.F. Parson, et al., Chem. Rev. 122 (2022) 7690–7719. doi: 10.1021/acs.chemrev.1c00857

  29. [29]

      R. Wu, J.B. Metternich, A.S. Kamenik, et al., Nat. Commun. 14 (2023) 2913. doi: 10.1038/s41467-023-38463-z

  30. [30]

      J.F. Van Dyck, J.R. Burns, K.I.P. Le Huray, et al., Nat. Commun. 13 (2022) 3610. doi: 10.1038/s41467-022-31029-5

  31. [31]

      K.G. Hicks, A.A. Cluntun, H.L. Schubert, et al., Science 379 (2023) 996–1003. doi: 10.1126/science.abm3452

  32. [32]

      Y. Zhong, L. Han, B.T. Ruotolo, Angew. Chem. Int. Ed. 126 (2014) 9363–9366. doi: 10.1002/ange.201403784

  33. [33]

      A. Al-jabiry, M. Palmer, J. Langridge, et al., Chemistry 27 (2021) 13783–13792. doi: 10.1002/chem.202101857

  34. [34]

      L. McAlary, J.A. Harrison, J.A. Aquilina, et al., Anal. Chem. 92 (2020) 1702–1711. doi: 10.1021/acs.analchem.9b01699

  35. [35]

      S.K. Chowdhury, V. Katta, B.T. Chait, J. Am. Chem. Soc. 112 (1990) 9012–9013. doi: 10.1021/ja00180a074

  36. [36]

      B. Ibarra-Molero, A.N. Naganathan, J.M. Sanchez-Ruiz, et al., Modern analysis of protein folding by differential scanning calorimetry, in: A. Feig (Ed.), Methods in Enzymology, Academic Press, New York, 2016, pp. 281–318.

  37. [37]

      F.E. Torres, M.I. Recht, J.E. Coyle, et al., Curr. Opin. Struct. Biol. 20 (2010) 598–605. doi: 10.1016/j.sbi.2010.09.001

  38. [38]

      R. Mahran, N. Vello, A. Komulainen, et al., Sci. Rep. 13 (2023) 20066. doi: 10.1038/s41598-023-46720-w

  39. [39]

      G. Wang, R.R. Abzalimov, I.A. Kaltashov, Anal. Chem. 83 (2011) 2870–2876. doi: 10.1021/ac200441a

  40. [40]

      A. Pruska, A. Marchand, R. Zenobi, Angew. Chem. Int. Ed. 60 (2021) 15518–15526.

  41. [41]

      G. Li, S. Zheng, Y. Chen, et al., Anal. Chem. 90 (2018) 7997–8001. doi: 10.1021/acs.analchem.8b00859

  42. [42]

      T.J. El-Baba, D.W. Woodall, S.A. Raab, et al., J. Am. Chem. Soc. 139 (2017) 6306–6309. doi: 10.1021/jacs.7b02774

  43. [43]

      S.A. Raab, T.J. El-Baba, D.W. Woodall, et al., J. Am. Chem. Soc. 142 (2020) 17372–17383. doi: 10.1021/jacs.0c05365

  44. [44]

      A. Laganowsky, D.E. Clemmer, D.H. Russell, Annu. Rev. Biophys. 51 (2022) 63–77. doi: 10.1146/annurev-biophys-102221-101121

  45. [45]

      J.L.P. Benesch, F. Sobott, C.V. Robinson, Anal. Chem. 75 (2003) 2208–2214. doi: 10.1021/ac034132x

  46. [46]

      J.S. Jordan, E.R. Williams, Anal. Chem. 94 (2022) 16894–16900. doi: 10.1021/acs.analchem.2c04204

  47. [47]

      A. Marchand, M.F. Czar, E.N. Eggel, et al., Nat. Commun. 11 (2020) 566. doi: 10.1038/s41467-019-14179-x

  48. [48]

      A. Pruška, J.A. Harrison, A. Granzhan, et al., Anal. Chem. 95 (2023) 14384–14391. doi: 10.1021/acs.analchem.3c02742

  49. [49]

      A. Marchand, F. Rosu, R. Zenobi, et al., J. Am. Chem. Soc. 140 (2018) 12553–12565. doi: 10.1021/jacs.8b07302

  50. [50]

      T.E. Walker, M. Shirzadeh, H.M. Sun, et al., J. Am. Chem. Soc. 144 (2022) 2667–2678. doi: 10.1021/jacs.1c11341

  51. [51]

      X. Cong, Y. Liu, W. Liu, et al., J. Am. Chem. Soc. 138 (2016) 4346–4349. doi: 10.1021/jacs.6b01771

  52. [52]

      K.J. Light-Wahl, B.L. Schwartz, R.D. Smith, J. Am. Chem. Soc. 116 (1994) 5271–5278. doi: 10.1021/ja00091a035

  53. [53]

      K.J. Light-Wahl, B.E. Winger, R.D. Smith, J. Am. Chem. Soc. 115 (1993) 5869–5870. doi: 10.1021/ja00066a083

  54. [54]

      L. Dai, N. Guo, Y. Liu, et al., Chin. Chem. Lett. 30 (2019) 103–106. doi: 10.1016/j.cclet.2017.12.023

  55. [55]

      L. Deng, E.N. Kitova, J.S. Klassen, J. Am. Soc. Mass Spectrom. 24 (2013) 988–996. doi: 10.1007/s13361-013-0651-8

  56. [56]

      M.C. Jecklin, D. Touboul, R. Jain, et al., Anal. Chem. 81 (2009) 408–419. doi: 10.1021/ac801782c

  57. [57]

      S. Yin, Y. Xie, J.A. Loo, J. Am. Soc. Mass Spectrom. 19 (2011) 1199–1208.

  58. [58]

      J.A. Harrison, A. Pruška, I. Oganesyan, et al., Chem. Eur. J. 27 (2021) 18015–18028. doi: 10.1002/chem.202102474

  59. [59]

      T. Wyttenbach, M.T. Bowers, J. Phys. Chem. B 115 (2011) 12266–12275. doi: 10.1021/jp206867a

  60. [60]

      T.J. El-Baba, D.R. Fuller, D.W. Woodall, et al., Chem. Commun. 54 (2018) 3270–3273. doi: 10.1039/C7CC09829D

  61. [61]

      K.R. Babu, A. Moradian, D. Douglas, J. Am. Soc. Mass Spectrom. 12 (2001) 317–328. doi: 10.1016/S1044-0305(00)00226-9

  62. [62]

      M. Jourdan, M.S. Searle, Biochemistry 39 (2000) 12355–12364. doi: 10.1021/bi000718r

  63. [63]

      C. Wan, M. Cui, F. Song, et al., Int. J. Mass Spectrom. 283 (2009) 48–55. doi: 10.1016/j.ijms.2009.01.007

  64. [64]

      C.L. Mazzitelli, Y. Chu, J.J. Reczek, et al., J. Am. Soc. Mass Spectrom. 18 (2007) 311–321. doi: 10.1016/j.jasms.2006.09.021

  65. [65]

      Z. Jiang, T. Li, L. Ma, et al., Food Res. Int. 131 (2020) 109006. doi: 10.1016/j.foodres.2020.109006

  66. [66]

      Y. Jia, X. Yan, Y. Huang, et al., Food Chem. 396 (2022) 133685. doi: 10.1016/j.foodchem.2022.133685

  67. [67]

      C. Ren, W. Xiong, J. Li, et al., Food Hydrocolloid 92 (2019) 155–162. doi: 10.1016/j.foodhyd.2019.01.053

  68. [68]

      J. Ye, L. Deng, Y. Wang, et al., Food Chem. 362 (2021) 130238. doi: 10.1016/j.foodchem.2021.130238

  69. [69]

      J. Li, R. Tian, G. Liang, et al., Food Chem. 355 (2021) 129617. doi: 10.1016/j.foodchem.2021.129617

  70. [70]

      I.J. Joye, G. Davidov-Pardo, R.D. Ludescher, et al., Food Chem. 185 (2015) 261–267. doi: 10.1016/j.foodchem.2015.03.128

  71. [71]

      K.X. Wan, M.L. Gross, T. Shibue, J. Am. Soc. Mass Spectrom. 11 (2000) 450–457. doi: 10.1016/S1044-0305(00)00095-7

  72. [72]

      A. Jarzab, N. Kurzawa, T. Hopf, et al., Nat. Methods 17 (2020) 495–503. doi: 10.1038/s41592-020-0801-4

• 

  Scheme 1  Timeline of the development of TC-ESI-MSⁿ.

  下载: 全尺寸图片 幻灯片
  

  Scheme 2  Structures of (a) quercetin (Q, left) and (b) rutin (R, right).

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Mass spectra of ubiquitin with different flavonols.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Weighted average charge state of ubiquitin versus solution temperature with midpoint melting temperature T_m = 45.4 ℃. Inset shows mass spectra of ubiquitin in 20% (v/v) methanol in water in an acidified aqueous solution collected at 25, 50, and 70 ℃ at flow rate of 5 µL/min. Weighted average charge state is calculated using Eq. S1 in Supporting information. The values are presented as mean ± standard deviation (n = 3).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Mass spectra of the ubiquitin-quercetin complex. (b) Zoomed-in mass spectra (m/z 1450–1650) of the [U-Q_1–3]⁶⁺ ions. (c) Mass spectra of the ubiquitin-rutin complex and (d) zoomed-in mass spectra (m/z 1500–1700) of the [U-R_1–2]⁶⁺ ions, all at different temperatures, respectively.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Binding affinities based on TC-ESI-MS measurements of flavonol-ubiquitin complexes at different temperatures. The values are presented as mean ± standard deviation (n = 3).

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Dissociation profiles of (a) U-Q₁, (b) U-Q₂, (c) U-R₁ and (d) U-R₂ at different temperatures. The values are presented as mean ± standard deviation (n = 3).

  下载: 全尺寸图片 幻灯片
  

  Figure 6  (a) Weighted average charge state versus solution temperature for ubiquitin (blue triangles), ubiquitin + rutin (red circles), ubiquitin + quercetin (black squares). The midpoint melting temperatures (T_m) determined from sigmoidal curve are shown with colored arrows. (b) ΔG values for ubiquitin (blue rectangle), ubiquitin + rutin (red rectangle), and ubiquitin + quercetin (black rectangle) are calculated at various temperatures. The values are presented as mean ± standard deviation (n = 3).

  下载: 全尺寸图片 幻灯片
•