Quantifying Protein-Fatty Acid Interactions Using Electrospray Ionization Mass Spectrometry
The application of the direct electrospray ionization mass spectrometry (ESI-MS) assay to quantify interactions between bovine β-lactoglobulin (Lg) and a series of fatty acids (FA), CH3(CH2)xCOOH, where x = 6 (caprylic acid, CpA), 8 (capric acid, CA), 10 (lauric acid, LA), 12 (myristic acid, MA), 14 (palmitic acid, PA) and 16 (stearic acid, SA), is described. Control ESI-MS binding measurements performed on the Lg-PA interaction revealed that both the protonated and deprotonated gas phase ions of the (Lg + PA) complex are prone to dissociate in the ion source, which leads to artificially small association constants (Ka). The addition of imidazole, a stabilizing solution additive, at high concentration (10 mM) increased the relative abundance of (Lg + PA) complex measured by ESI-MS in both positive and negative ion modes. The Ka value measured in negative ion mode and using sampling conditions that minimize in-source dissociation is in good agreement with a value determined using a competitive fluorescence assay. The Ka values measured by ESI-MS for the Lg interactions with MA and SA are also consistent with values expected based on the fluorescence measurements. However, the Ka values measured using optimal sampling conditions in positive ion mode are significantly lower than those measured in negative ion mode for all of the FAs investigated. It is concluded that the protonated gaseous ions of the (Lg + FA) complexes are kinetically less stable than the deprotonated ions. In-source dissociation was significant for the complexes of Lg with the shorter FAs (CpA, CA, and LA) in both modes and, in the case of CpA, no binding could be detected by ESI-MS. The affinities of Lg for CpA, CA, and LA determined using the reference ligand ESI-MS assay, a method for quantifying labile protein–ligand complexes that are prone to in-source dissociation, were found to be in good agreement with reported values.
Key wordsFatty acids Protein-ligand complexes Affinity Lactoglobulin Electrospray ionization mass spectrometry
The direct electrospray ionization mass spectrometry (ESI-MS) assay has emerged as a useful technique for measuring the binding stoichiometry and affinity (Ka) of protein–ligand interactions, as well as other noncovalent biological complexes, in solution. The assay is based on the relative abundance of free and ligand-bound protein ions measured by ESI-MS for a solution of known initial protein and ligand concentration [1, 2, 3]. The ESI-MS assay is fast (measurements normally can be completed in less than ~1 min), consumes little sample (typically ~ pmol per analysis when performed using nanoflow ESI), provides direct insight into binding stoichiometry and can be used to measure multiple binding equilibria simultaneously. Furthermore, the assay is very versatile as there is no requirement for labeling or immobilization of analyte.
Absolute Ka values, in the 103–107 M−1 range, have been determined for a variety of protein–ligand complexes, including antibody-antigen, lectin-carbohydrate, enzyme-inhibitor complexes, and shown to be in reasonable agreement with values determined by other analytical methods [4, 5, 6, 7, 8, 9, 10]. To date, however, there have been few reported examples of the application of ESI-MS to characterize protein-hydrophobic ligand binding [11, 12, 13, 14] and, to our knowledge, no reports of absolute affinity measurements using the direct ESI-MS assay. The two principal challenges to quantifying interactions between water soluble proteins and hydrophobic ligands using ESI-MS are the low solubility of the ligand in aqueous solution, and the propensity of the desolvated complexes to dissociate during analysis. Low ligand solubility prevents the ESI-MS assay from being applied to very weak interactions (<104 M−1) because high ligand concentrations are needed to produce detectable levels of complex. Protein-ligand complexes stabilized predominantly by nonpolar interactions also tend to have low kinetic stabilities in the gas phase and are prone to dissociate in the ion source (i.e., in-source dissociation) [11, 12]. In-source dissociation reduces the relative abundance of the ligand-bound protein ions and, consequently, leads to artificially low binding constants [11, 15, 16, 17]. In extreme cases, in-source dissociation results in false negatives, wherein no ligand-bound protein ions are detected by ESI-MS . However, using gentle sampling conditions, along with stabilizing solution or gas phase additives, it is sometimes possible to preserve the labile protein–ligand complexes during ESI-MS analysis and to quantify their association constants [17, 18].
Here, we describe the application of ESI-MS to quantify the affinities of bovine β-lactoglobulin (Lg) for a series of linear, saturated fatty acids (FA), CH3(CH2)xCOOH. Lg, a member of the lipocalin protein family, possesses a large hydrophobic cavity and is known to bind a variety of hydrophobic ligands, including FAs [19, 20]. Association constants for the interaction between Lg and several FAs [x = 6 (caprylic acid), 10 (lauric acid), and 14 (palmitic acid)], at 25 °C and various pH values (6.2, 7.2, and 8.3) have been determined using a competitive fluorescence assay . Notably, these measurements revealed that at a given pH, the logarithm of Ka increases linearly with the length of the FA acyl chain. Based on this dependence, it is possible to estimate the Ka values for related FAs, vide infra. To test the reliability of the direct ESI-MS assay for quantifying protein–hydrophobic ligand interactions, affinity measurements were carried out on Lg and a series of FAs (x = 6, 8, 10, 12, 14, and 16) at pH 8.5 and 25 °C. Importantly, it is shown that the Ka values for Lg interactions with the longer FAs (x = 12, 14, and 16) measured in negative ion mode and using gentle sampling conditions in conjunction with a stabilizing solution additive are in good agreement with the reported values. In contrast, in-source dissociation of the gaseous ions of the complexes of Lg and the shorter FAs is extensive, which precludes the determination of reliable affinities using the direct ESI-MS assay. However, the affinities determined for the shorter FAs using the reference ligand ESI-MS assay , a method for quantifying labile protein–ligand complexes that are prone to in-source dissociation, are in good agreement with reported values.
2.1 Protein and Ligand
Bovine β-lactoglobulin (Lg, monomer MW 18281 Da), caprylic acid (CpA, 144.2 Da), capric acid (CA, 172.3 Da), lauric acid (LA, 200.3 Da), myristic acid (MA 228.4 Da), palmitic acid (PA, MW 256.4 Da), and stearic acid (SA, 284.8 Da) were purchased from Sigma-Aldrich Canada (Oakville, Canada). The Lg was dissolved and exchanged directly into Milli-Q water, using an Amicon microconcentrator with a molecular weight cutoff of 10 kDa, followed by filtration using an Amicon ultra centrifugal filter with a molecular weight cutoff of 100 kDa to remove Lg aggregates in solution. The concentration of the Lg solution was determined by lyophilizing a known volume of the filtrate and measuring the mass of the protein. The protein stock solution was stored at −20 °C. The ligand stock solutions were prepared by dissolving each FA into aqueous ammonium acetate (25 mM). The ESI solutions were prepared from aqueous stock solutions of protein and ligand. Aqueous ammonium hydroxide was added to adjust the pH of the ESI solution to 8.5. Where indicated, imidazole (10 mM) was also added in order to stabilize the (Lg + FA) complexes during ESI-MS analysis.
2.2 Mass Spectrometry
All experiments were performed on a 9.4 tesla Apex II Fourier transform ion cyclotron resonance (FTICR) mass spectrometer (Bruker, Billerica, MA) equipped with a nanoflow ESI ion source. NanoES was performed using borosilicate tubes (1.0 mm o.d., 0.68 mm i.d.), pulled to ~5 μm o.d. at one end using a P-2000 micropipette puller (Sutter Instruments, Novato, CA). Details of the instrumental parameters employed in positive ion mode are given below. For measurements in negative ion mode, the voltage polarity was switched. The electric field required to spray the solution in positive ion mode was established by applying a voltage of 800 V to a platinum wire inserted inside the glass tip. The solution flow rate was ~20 nL/min. The gaseous ions produced by nanoES were introduced into the mass spectrometer through a stainless steel capillary (i.d. 0.43 mm) maintained at an external temperature of 66 °C. The gas flow rate into the instrument was measured to be 0.6 L min−1. The ions sampled by the capillary (50 V) were transmitted through a skimmer (0 V) and accumulated electrodynamically in an rf hexapole. Ions were then ejected from the hexapole and accelerated (−2700 V) into the superconducting magnet, decelerated, and introduced into the ion cell. The trapping plates of the cell were maintained at a constant potential of 1.4 V throughout the experiment. The typical base pressure for the instrument was ~5 × 10−10 mbar. Data acquisition was controlled by an SGI R5000 computer running the Bruker Daltonics
XMASS software, ver. 5.0. Mass spectra were obtained using standard experimental sequences with chirp broadband excitation. The time domain signal, consisting of the sum of 50 transients containing 128 K data points per transient, was subjected to one zero-fill prior to Fourier transformation.
2.3 Determining Protein–Ligand Affinities Using the Direct ESI-MS Assay
2.4 Determining Protein–Ligand Affinities Using the Reference Ligand ESI-MS Method
The reference ligand ESI-MS method for determining protein–ligand affinities is based on the direct ESI-MS assay and competitive ligand binding . A reference ligand (Lref), which binds specifically to the protein, at the same binding site as the ligand of interest, with known affinity and forms a stable protein-ligand complex in the gas phase, is added to the solution. The fraction of P bound to Lref, which is determined directly from the ESI mass spectrum, is sensitive to the fraction of P bound to L in solution and enables the affinity of P for L to be determined.
3 Results and Discussion
3.1 Association Constant for the (Lg + PA) Complex Determined by the Direct ESI-MS Assay
Recently, it was shown that the presence of imidazole, at high concentration (~10 mM), can protect protein-ligand complexes from in-source dissociation during ESI-MS analysis . It was proposed that the stabilizing effect is due, at least in part, to enhanced evaporative cooling, which arises from the loss of multiple nonspecifically bound imidazole molecules from the gaseous protein–ligand complexes in the ion source . The stabilizing effect of imidazole is evident in the ESI mass spectra shown in Figure 1c and d. These mass spectra were acquired under identical conditions as those shown in Figure 1a and b, respectively, but with the addition of 10 mM imidazole to the solution. The presence of imidazole results in two obvious changes in the appearance of the mass spectra. The charge state distribution for the Lgn+ and (Lg + PA)n+ ions broadens from 7–8 to 6–8. This change is due to the high gas phase basicity of imidazole (217 kcal mol−1)  and its ability to strip protons from multiply protonated protein ions in the gas phase . Additionally, the relative abundance of the (Lg + PA)n+ ions is larger than that measured in the absence of imidazole. In fact, the R values measured at 0.4 and 0.8 s hexapole accumulation times are nearly indistinguishable. However, the corresponding Ka value, (2.7 ± 0.2) × 105 M−1, is still a factor of two smaller than the reported value, suggesting the possibility that in-source dissociation was not completely eliminated under these conditions.
Association constants (Ka) for Lg binding to fatty acids (FA = CpA, CA, LA, MA, PA, and SA), determined by the direct ESI-MS assay and the reference ligand ESI-MS method at 25 °C and pH 8.5, and by competitive fluorescence at 25 °C and pH 8.3a
ESI-MS (ref ligand)
(2.3 ± 0.3) × 104
(2.6 ± 0.5) × 104
(4.5 ± 0.1) × 103 d
(4.6 ± 0.5) × 104
(4.1 ± 0.1) × 104 d
(1.1 ± 0.2) × 105
(1.3 ± 0.1) × 105
(1.9 ± 0.1) × 105
(3.8 ± 0.1) × 105
(5.0 ± 0.2) × 105
(1.6 ± 0.1) × 106
The larger R values measured in negative ion mode compared with those measured in positive ion mode confirm that, even with the addition of imidazole to the solution, a fraction of the protonated (Lg + PA)n+ ions are lost due to in-source dissociation. At present, the origin of the differential stabilities of the protonated and deprotonated ions of the (Lg + PA) complex is not known but presumably reflects differences in the structure of the Lg ions arising from differences in the nature of the charges and their locations. Our laboratory is currently carrying out a comparative study, employing thermal dissociation rate constant measurements and molecular dynamics simulations, of the structures and stabilities of protonated and deprotonated (Lg + FA) ions. The results of this study will be reported in a forthcoming manuscript.
3.2 Association Constants for the (Lg + FA) Complexes Determined by the Direct ESI-MS Assay
3.3 Association Constants for the (Lg + FA) Complexes Determined by the Reference Ligand ESI-MS Assay
The reference ligand ESI-MS method was also used to establish the binding constants for the (Lg + CA) and (Lg + LA) complexes. Again, PA was used as Lref for these measurements. Shown in Figure S1, which can be found in the electronic version of this article, are illustrative ES mass spectra acquired for solutions of Lg and PA, in the absence and presence of either CA or LA. As outlined above, data analysis was carried out differently than was the case for the (Lg + CpA) complex because ions corresponding to the (Lg + CA) and (Lg + LA) complexes could still be detected . The binding constant determined for (Lg + LA) complex using the reference ligand ESI-MS assay, (1.1 ± 0.2) × 105 M−1, is in excellent agreement with the reported value, (1.3 ± 0.1) × 105 M−1. The value measured for (Lg + CA) complex, (4.6 ± 0.5) × 104 M−1, is also in reasonable agreement with the value (6 × 104 M−1) estimated from the linear dependence of log Ka on acyl chain length (Figure 4).
In summary, the use of ESI-MS to quantify protein–fatty acid complexes in aqueous solution is described. The interactions between Lg and a series of linear, saturated FAs (CpA, CA, LA, MA, PA, and SA) were investigated using both the direct ESI-MS assay and the reference ligand ESI-MS method. The binding constant measured for PA at 25 °C and pH 8.5 using the direct ESI-MS assay, performed in negative ion mode and utilizing a stabilizing solution additive (imidazole), is in good agreement with the value measured using a competitive fluorescence assay. The affinities measured directly for MA and SA are also consistent with values expected based on the linear dependence of the log Ka values with acyl chain length established from the fluorescence-derived binding data for related FAs. However, the Ka values measured directly for LA and CA were significantly smaller than the expected values. In the case of CpA, no binding was detected by direct ESI-MS measurement. These results indicate the occurrence of in-source dissociation, which reduces the relative abundance of the gaseous (Lg + FA)n- ions, for these shorter FAs. Notably, in-source dissociation was prevalent in the ESI-MS binding measurements performed in positive ion mode for all of the (Lg + FA) complexes investigated. The reference ligand ESI-MS assay was used to measure binding constants for the Lg interactions with the shorter FAs. Importantly, the measured Ka values for CpA and LA are in excellent agreement with values measured using the fluorescence assay.
- 7.Rademacher, C., Shoemaker, G.K., Kim, H.-S., Zheng, R.B., Taha, H., Liu, C., Nacario, R.C., Schriemer, D.C., Klassen, J.S., Peters, T., Lowary, T.L.: Ligand specificity of CS-35, a monoclonal antibody that recognizes mycobacterial lipoarabinomannan. A model system for oligofuranoside-protein recognition. J. Am. Chem. Soc. 129, 10489–10502 (2007)CrossRefGoogle Scholar
- 8.Jorgensen, T.J.D., Roepstorff, P., Heck, A.J.R.: Direct determination of solution binding constants for noncovalent complexes between bacterial cell wall peptide analogues and vancomycin group antibiotics by electrospray ionization mass spectrometry. Anal. Chem. 70, 4427–4432 (1998)CrossRefGoogle Scholar
- 9.Jecklin, M.C., Touboul, D., Bovet, C., Wortmann, A., Zenobi, R.: Which electrospray-based ionization method best reflects protein-ligand interactions found in solution? A comparison of ESI, nanoESI, and ESSI for the determination of dissociation constants with mass spectrometry. J. Am. Soc. Mass Spectrom. 19, 332–343 (2008)CrossRefGoogle Scholar
- 14.Hooper, A.M.; Dufour, S.; He, X.; Muck, A.; Zhou, J.; Almeida, R.; Field, L.M.; Svatos, A.; Pickett, J.A. High-throughput ESI-MS analysis of binding between the Bombyx mori pheromone-binding protein BmorPBP1, its pheromone components, and some analogues. Chem. Commun. 5725–5727 (2009)Google Scholar
- 22.El-Hawiet, A., Kitova, E.N., Liu, L., Klassen, J.S.: Quantifying labile protein–ligand interactions using electrospray ionization mass spectrometry. J. Am. Soc. Mass Spectrom. 21, 1893–1899 (2010)Google Scholar