Thermal Decomposition of a Gaseous Multiprotein Complex Studied

Anal. Chem. 2001, 73, 4647-4661

Thermal Decomposition of a Gaseous Multiprotein Complex Studied by Blackbody Infrared Radiative Dissociation. Investigating the Origin of the Asymmetric Dissociation Behavior Natalia Felitsyn, Elena N. Kitova, and John S. Klassen* Department of Chemistry, University of Alberta, Edmonton, Alberta Canada T6G 2G2

The blackbody infrared radiative dissociation technique was used to study the thermal decomposition of the gaseous B5 pentamer of the Shiga-like toxin I and its complexes with the Pk trisaccharide and a decavalent Pkbased oligosaccharide ligand (STARFISH, S). Dissociation of the protonated pentamer, (B5 + nH)n+ t B5n+ where n ) 11-14, proceeds almost exclusively by the loss of a single subunit (B) with a disproportionately large fraction (30-50%) of the parent ion charge. The degree of charge enrichment of the leaving subunit increases with increasing parent ion charge state. For n ) 12-14, a distribution of product ion charge states is observed. The yields of the complementary pairs of product ions are sensitive to the reaction temperature, with higher temperatures favoring greater charge enrichment of the leaving subunit for +13 and +14, and the opposite effect for +12. These results indicate that some of the protons are rapidly exchanged between subunits in the gas phase. Dissociation of B514+‚ S proceeds exclusively by the loss of one subunit, although the ligand increases the stability of the complex and also reduces the degree of charge enrichment in the ejected monomer. For B512+(Pk)1-3, the loss of neutral Pk competes with loss of a subunit at low temperatures. Linear Arrhenius plots were obtained from the temperaturedependent dissociation rate constants measured for the loss of B from B5n+ and B514+‚S. The magnitude of the Arrhenius parameters is highly dependent on the charge state of the pentamer: Ea ) 35 kcal/mol and A ) 1019 s-1 (+14), 46 kcal/mol and 1023 s-1 (+13), 50 kcal/mol and 1026 s-1 (+12), and 80 kcal/mol and 1039 (+11). The Ea and A for B514+‚S are 59 kcal/mol and 1030 s-1, respectively. The reaction pathways leading to greater charge enrichment of the subunit lost from the B514+ and B513+ ions correspond to higher energy processes, however, these pathways are kinetically preferred at higher temperatures due to their large A factors. A simple electrostatic model, whereby charge enrichment leads to Coulombic repulsion-induced denaturation of the subunits and disruption of the intersubunit interactions, provides an explanation for the magnitude of the Arrhenius parameters and the origin of the asymmetric dissociation behavior of the complexes. * Fax: 780-492-8231. E-mail: [email protected]. 10.1021/ac0103975 CCC: $20.00 Published on Web 08/30/2001

© 2001 American Chemical Society

Proteomics is the characterization of the structure, interactions and physicochemical properties of the proteome, the complete set of proteins of a cell or organism, and its functional roles in coordinating cellular activity and responses. Many important cellular functions involve complexes composed of two or more protein subunits. These complexes may be “nonobligatory”, being made and broken according to the cellular environment, and involve proteins that can also exist as monomers. Alternatively, the complexes may be “permanent”, in which the individual proteins do not exist as monomers under normal physiological conditions, and biological function is achieved by the assembly. Permanent oligomeric complexes are believed to constitute the majority of soluble and membrane-bound proteins.1,2 To realize the goals of proteomics, analytical techniques capable of rapidly and accurately characterizing the composition, structure, and function of cellular proteins are required. Many of the established analytical techniques used to study the structure and interactions of isolated monomeric proteins and their complexes are not well-suited to the demands of proteomics in which rapid screening of large numbers of diverse protein complexes is required. Traditional structural techniques for the analysis of higher-order structure of monomeric proteins, such as X-ray diffraction and NMR, are not as attractive for the rapid screening of the oligomeric complexes, which may be present in very small amounts. The yeast two-hybrid genetic method,3 in which transcription factors are activated by specific protein-protein interactions, has been the most widely used technique for identifying binary interactions. With the advent of the two-bait interaction trap,4 it is now possible to gain some insight into the topology and order of interactions in multiprotein complexes. These genetic methods have the advantage of screening protein interactions in vivo. However, they are labor-intensive and may not be sufficiently sensitive to identify the weak binary interactions responsible for the assembly of higher-order oligomeric complexes.5 Mass spectrometry (MS), with its inherent speed, sensitivity and accurate mass capabilities, has emerged as a powerful tool for characterizing multiprotein complexes. MS and tandem MS (1) Goodsell, D. S.; Olson, A. J. Annu. Rev. Biophys. Struct. 2000, 29, 105153. (2) Jones, S.; Thornton, J. M. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 13-20. (3) Xu, C. W.; Mendelsohn, A. R.; Brent, R. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 12473-2478. (4) Serebriiskii, I.; Khazak, V.; Golemis, E. A. J. Biol. Chem. 1999, 274, 1708017087. (5) Mendelsohn, A. R.; Brent, R. Science 1999, 284, 1948-1950.

Analytical Chemistry, Vol. 73, No. 19, October 1, 2001 4647

(i.e., MS/MS), in conjunction with matrix-assisted laser desorption/ionization (MALDI)6 or electrospray ionization (ES),7 are widely employed for identifying the composition of protein complexes.8 The experimental protocol generally involves the isolation and digestion of the individual components of the protein complex followed by mass analysis. MS/MS may also be employed to yield sequence information in cases in which the peptide molecular weights are not sufficient to identify the protein. ES and tandem mass spectrometry also hold considerable promise as a tool for the “direct” analysis of the intact multiprotein complex, providing information about the composition without the need for denaturation and digestion of the complex prior to MS analysis and the nature and strength of specific interactions responsible for protein assembly. Early ES-MS studies of multiprotein complexes established the experimental conditions (solution and instrumental) necessary for the preservation and detection of specific interactions.9-11 For example, most oligomeric complexes were found to be stable in a narrow range of pH, around 7, and control of ionic strength was needed in some cases to preserve protein folding.10,11 Under the appropriate solution conditions, oligomeric complexes appear in the mass spectrum at high mass-to-charge ratios (m/z), relative to monomeric proteins, and with a narrow charge distribution. The higher m/z associated with the oligomers has been ascribed, in part, to the reduced accessibility of strongly basic or acidic sites and to Coulombic repulsion constraints on the proximity of charge sites in the complex related to the compactness of the structure and has been considered as indirect evidence for at least partial preservation of their solution structure.8,12-19 Recent technological developments, such as nanoflow electrospray, with its higher sensitivities for noncovalent complexes, coupled to high m/z mass analyzers, time-of-flight (TOF), and high-field FT-ICR instruments, have greatly facilitated the application of ES-MS to the detection of these multiprotein species. In addition to providing accurate masses for the intact complexes, it may be possible to use MS to dissect gaseous multiprotein assemblies in a sequential fashion to identify all of the protein components and possibly obtain information regarding topology, quaternary structure, and stability. At present, however, the application of ES and multiple stages of MS (i.e., MSn) for the direct determination of the composition and quaternary structure of the oligomeric complexes is hindered by an incomplete understanding of the effects of desolvation and the influence of charge on the structure of the complexes and their dissociation mechanisms. A number of mass spectrometric studies of the decomposition pathways of multiprotein complexes and their relationship to original quaternary structure in solution have been reported.12-19 A general feature of the decomposition patterns of (6) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science 1989, 246, 64-71. (7) Hillenkamp, F.; Karas, M.; Beavis, R. C.; Chait, B. T. Anal. Chem. 1991, 63, 1193A. (8) Loo, J. A. Mass. Spectrom. Rev. 1997, 16, 1-23. (9) Smith, R. D.; Bruce, J. E.; Wu, Q. Y.; Lei, Q. P. Chem. Soc. Reviews 1997, 26, 191-202. (10) Vis, H.; Heinemann, U.; Dobson, C. M.; Robinson, C. V. J. Am. Chem. Soc. 1998, 120, 6427-6428. (11) Potier, N.; Donald, L. J.; Chernushevich, I.; Ayed, A.; Ens, W.; Arrowsmith, C. H.; Standing, K. G.; Duckworth, H. W. Protein Sci. 1998, 7, 1388-1395. (12) Light-Wahl, K. J; Schwartz, B. L.; Smith, R. D. J. Am. Chem. Soc. 1994, 116, 5271-527.

4648

Analytical Chemistry, Vol. 73, No. 19, October 1, 2001

the tetrameric assemblies concanavalin A, adult human hemoglobin, and avidin12 was the formation of mostly monomer and trimer, with a relatively high charge on the monomer and a much lower charge per subunit on the trimer. The resulting trimer was also found to be resistant to subsequent dissociation. Notably, both hemoglobin and concanavalin A are known to be composed of two dimers, heterogeneous for hemoglobin and homogeneous in the case of concanavalin A. The dissociation of the streptavidin tetramer also follows the monomer/trimer decomposition pattern in which the monomer carries a disproportionately large fraction of the charge.13 The authors suggested that the asymmetric dissociation process is analogous to the fission of charged droplets in the ES process in which the smaller offspring droplets are believed to form with much higher z/m than the parent droplet. The stability of the trimer product ions was attributed to the much lower charge-to-subunit ratio. Similarly, several other studies of protein asssemblies have indicated an asymmetric charge distribution in favor of the monomer.15-19 In all cases, no correlation was found between the solution structure of the assemblies and their gas-phase dissociation pathways. The general nature of the asymmetric dissociation is intriguing, yet no adequate model has been developed for this process. The relationship between the decomposition behavior and the quaternary structure of the complex has to be understood in order to provide structural information. The aforementioned dissociation studies were performed with ill-defined internal energies20 and, in some cases, with short experimental windows that can lead to large kinetic shifts that obscure the actual dissociation energetics.21 Interpretation of the experimental data and elucidation of the mechanistic details under these conditions are difficult. The blackbody infrared radiative dissociation (BIRD) technique,22 whereby ions are thermalized by the exchange of blackbody (13) Schwartz, B. L.; Bruce, J. E.; Anderson, G. A.; Hofstadler, S. A.; Rockwood, A. L.; Smith, R. D.; Chilkoti, A.; Stayton, P. S. J. Am. Soc. Mass. Spectrom. 1995, 6, 459-465. (14) Loo, J. A. J. Mass. Spectrom. 1995, 30, 180-183. (15) Zhang, Z.; Krutchinsky, A.; Endicott, S.; Realini, C.; Rechsteiner, M.; Standing, K. G. Biochemistry 1999, 38, 5651-5658. (16) Fitzgerald, M. C.; Chernushevich, I.; Standing, K. G.; Whitman, C. P.; Kent, S. B. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 6851-6856. (17) Rostom, A. A.; Sunde, M.; Richardson, S. J.; Schreiber, G.; Jarvis, S.; Bateman, R.; Dobson, C. M.; Robinson, C. V. Proteins 1998, (Suppl. 2) 3-11. (18) Bruce, J. E.; Smith, V. F.; Liu, C.; Randall, L. L.; Smith, R. D. Protein Sci. 1998, 7, 1180-1185. (19) Versluis, C.; van der Staaij, A.; Stokvis, E.; de Craene, B.; Heck, A. J. R. J. Am. Soc. Mass Spectrom. 2001, 6, 459-465. (20) Vekey, K. J. Mass. Spectrom. 1996, 31, 445-463. (21) Armentrout, P. B.; Baer, T. J. J. Phys. Chem. 1996, 100, 12866-12877. (22) (a) Tholmann, D.; Tonner, D. S.; McMahon, T. B. J. Phys. Chem. 1994, 98, 2002-2004. (b) Tonner, D. S.; Tholmann, D.; McMahon, T. B. Chem. Phys. Lett. 1995, 233, 324-330. (c) Dunbar, R. C. J. Phys. Chem. 1994, 98, 87058712. (d) Dunbar, R. C.; McMahon, T. B.; Tholmann, D.; Tonner, D. S.; Slahub, D. R.; Wei, D. J. Am. Chem. Soc. 1995, 117, 12819-12825. (d) Lin, C. Y.; Dunbar, R. C. J. Phys. Chem. 1996, 100, 655-659. (e) Price, W. D.; Schnier, P. D.; Williams, E. R. Anal. Chem. 1996, 68, 859-866. (f) Schnier, P. D.; Price, W. D.; Jockusch, R. A.; Williams, E. R. J. Am. Chem. Soc. 1996, 118, 7178-7189. (g) Gross, D. S.; Williams, E. R. J. Am. Chem. Soc. 1995, 117, 883-890. (h) Price, W. D.; Schnier, P. D.; Jockusch, R. A.; Strittmatter, E. F.; Williams, E. R. J. Am. Chem. Soc. 1996, 118, 10640-10644. (i) Price, W. D.; Williams, E. R. J. Phys. Chem. A 1997, 101, 8844-8852, (j) Klassen, J. S.; Schnier, P. D.; Williams, E. R. J. Am. Soc. Mass Spectrom. 1998, 9, 1117-1124, (k) Rodriguez-Cruz, S. E.; Klassen, J. S.; Williams, E. R. J. Am. Soc. Mass Spectrom 1999, 10, 958-68, (l) Strittmatter, E. F.; Schnier, P. D.; Klassen, J. S.; Williams, E. R. J. Am. Soc. Mass Spectrom. 1999, 10, 1095-104, (m) Dunbar, R. C.; McMahon, T. B. Science 1998, 279, 194197.

photons with the surroundings (i.e., the FT-ICR ion cell), allows the measurement of the temperature dependence of dissociation rate constants and determination of the Arrhenius activation energy (Ea) and preexponential factor (A) for large gaseous ions, such as biopolymers. To date, BIRD has been used to determine Arrhenius activation parameters for the dissociation of a variety of biopolymers, including protonated peptides,22e,22f small protonated proteins such as ubiquitin,22h single and double stranded DNA anions,22j,22l and loss of heme from hemoglobin.23 Until recently, BIRD was the only technique available to study the thermal decomposition of large, isolated gas-phase ions. However, McLuckey and co-workers24 have demonstrated the thermal dissociation of bradykinin ions in a heated quadrupole ion trap. Techniques such as infrared multiphoton dissociation (IRMPD)25 and SORI dissociation26 have been used to determine relative binding energies for structurally similar biological compounds. However, these methods do not provide absolute Ea and A factors, both of which provide important information about the nature of the dissociation mechanism. Here, we report on the dissociation pathways, kinetics, and Arrhenius parameters for a homooligomeric protein complex determined using BIRD and FT-ICR/MS. To our knowledge, this is the first study of the thermal dissociation of a gaseous multiprotein complex. The B5 homopentamer of the lethal Shigalike toxin I (SLT-I) was chosen as a model system for this study. We also investigated the dissociation of B5 bound to the Pk trisaccharide glycoside and to a decavalent Pk- based oligosaccharide ligand (STARFISH, S), both analogues of the natural trisaccharide receptor globotriaoside. Our interest in the dissociation kinetics and energetics of SLT-1(B5) and its oligosaccharide complexes arose from a recent study of the binding stoichiometry and affinity of B5 with Pk and two multivalent oligosaccharide ligands.27 In this work, we demonstrated that the B5‚Pkn complexes are readily observed in mass spectrum, despite the weak solution binding affinity (Kd ) 1.5 mM). Furthermore, excellent agreement was found between the binding affinity determined from the ion intensity of the bound and unbound forms of the pentamer and Pk in the nanoES spectrum and the microcalorimetry-derived value, despite the high ligand concentrations (>300 µM) required to produce a significant concentration of complex in solution. We attributed the agreement between the solution and the gas-phase composition to the retention of specific solution interactions (protein-protein and protein-oligosaccharide) of the B5‚Pkn complexes in the gas phase and the stability of these interactions. The BIRD data obtained in the present work provide important new insights into the stability of the B5 pentamer and its oligosaccharide complexes in the gas phase. The measured Arrhenius parameters also provide insights into the dissociation mechanism of the complexes and the origin of the asymmetric charge and mass distributions and the important roles played by the charge. (23) Gross, D. S.; Zhao, Y.; Williams, E. R. J. Am. Soc. Mass Spectrom. 1997, 8, 519-524. (24) Butcher, D. J.; Asano, K. G.; Goeringer, D. E.; McLuckey, S. A. J. Phys. Chem. A 1999, 103, 8664-8671. (25) Freitas, M. A.; Hendrickson, C. L.; Marshall, A. G. J. Am. Chem. Soc. 2000, 122, 7768-7775. (26) Gao, J.; Wu, Q.; Carbeck, J.; Lei, P. Q.; Smith, R. D.; Whitesides, G. M. Biophys. J. 1999, 76, 3253-3260. (27) Kitova, E. N.; Kitov, P. I.; Bundle, D. R.; Klassen, J. S. Glycobiology 2001, 11, 605-611.

EXPERIMENTAL SECTION Mass spectra were obtained using an ApexII 47e Fourier transform ion cyclotron resonance (FT-ICR) mass spectrometer (Bruker, Billerica) equipped with an external electrospray source (Analytica, Branford). Nanoelectrospray was performed using an aluminosilicate capillary (0.68 mm i.d.), pulled to ∼20 µm o.d. and 1-5 µm i.d. at one end using a micropipet puller (Sutter Instrument Co.). The nanospray tips were positioned ∼1 mm from the sampling capillary using a microelectrode holder (Warner Instrument Inc.). The electric field required to spray the solution was established by applying a voltage of 800-1000 V to a platinum wire inserted inside the glass tip. The solution flow rate ranged from 5 to 60 nL/min, depending on the diameter of the nanoelectrospray tip, the electrospray voltage, and the composition of the solution. The droplets formed at atmospheric pressure were introduced into the vacuum system of the mass spectrometer through a stainless steel capillary (0.43 mm i.d.) operated at a temperature of 150 °C to assist with solvent evaporation from the droplets and microsolvated complexes. The ion/gas jet sampled by the capillary (52 V) was transmitted through a skimmer (4 V, 1.0 mm i.d.) and stored electrostatically in the hexapole. Ions were accumulated in the hexapole for 2-15 s, depending on the ion intensities, then were ejected and accelerated to ∼2700 V through the fringing field of the 4.7 T magnet, decelerated, and introduced into the ion cell. The trapping plates of the cell were maintained at a constant potential of 1.3 V throughout the experiments. The typical base pressure for the instrument was 3 × 10-10 mbar. Data acquisition was controlled by an SGI R5000 computer running the Bruker Daltonics XMASS software, version 5.0. Mass spectra were obtained using standard experimental sequences with chirp broadband excitation. Isolation of the parent ions for the BIRD experiments was achieved using broadband rf sweep. The ions were stored inside the heated cell for variable reaction times prior to excitation and detection. The excitation pulse length was varied between 10 and 15 µs, and the power of the excitation pulse was varied so as to maximize the intensity of the ion signal. The time-domain spectra, consisting of the sum of 10-30 transients containing 128 K data points per transient, were subjected to one zero-fill prior to Fourier transformation. The temperature of the ion cell, which established the reaction temperature, was controlled by applying a voltage to the two external flexible heating blankets placed around the vacuum tube in the vicinity of the ion cell. The heating blankets, 24 cm in length, were centered around the ion cell (6.5 cm in length). The presence of internal thermocouples in the vicinity of the ion cell resulted in a significant decrease in ion transmission and an increase in electrical noise in the spectra. Therefore, in a separate experiment, the temperature inside the cell, measured by a thermocouple placed temporarily inside the cell, was calibrated against the temperature measured at several points on the outside of the vacuum tube. Because of the unavoidable contact between the end of the vacuum tube and the room temperature shield of the magnet, a temperature gradient existed across the vacuum tube between the cell and the end flange. To properly account for the gradient, the internal cell temperature was calibrated against eight thermocouples on the outside of the vacuum tube. Using this approach, calibration plots were generated for cell temperatures ranging from 25 to 175 °C. The plots were found to be linear, Analytical Chemistry, Vol. 73, No. 19, October 1, 2001

4649

Figure 1. (a) Crystal structure of pentamer SLT-1 B5, (b) structures of the Pk trisaccharide glycoside and the STARFISH ligand.

with the cell temperature 10 to 20 °C below that of the external temperature, depending on the external temperature. To validate the calibration plot approach for determining the internal cell temperature, Arrhenius parameters were measured for the decomposition of the deprotonated oligonucleotide, A72- and the duplex (A7‚T7)23-. The Arrhenius parameters (Ea ) 25.5 ( 1.2 kcal/mol, A ) 1011.2(1.1 s-1 for A72- and 38.0 ( 3.7 kcal/mol, 1018.5(1.8 s-1 for (A7‚T7)23-) were in agreement with parameters reported by Williams and co-workers22j,28 (Ea ) 23.7 ( 1.2 kcal/ mol, A ) 1010.3(0.6 s-1 for A72- and 38.6 ( 2.5 kcal/mol, 1019.2(1.4 s-1 for (A7‚T7)23-) and which were determined using BIRD and internal thermocouples to establish the reaction temperature. The samples of SLT-1(B5) and oligosaccharide ligands were generously provided by D. Bundle (University of Alberta). The purified protein was dissolved in either 1 mM aqueous ammonium bicarbonate buffer (pH ) 7.2) or 0.2 mM aqueous acetic acid (pH ≈ 4) at a concentration of ≈10 µM. The intensities observed for the B5 and B5-oligosaccharide complexes were sensitive to electrospray and source conditions: nanospray tip dimensions, solution flow rate, temperature of the metal capillary, nanospray voltage, source voltages (capillary, skimmer and hexapole), and hexapole accumulation times. These parameters were varied to maximize the intensity of the protein complexes. The strongest ion intensities for the noncovalent complexes were observed at a relatively low electrospray voltage (∼800 V), which resulted in a solution flow rate of 5-10 nL/min. (28) Schnier, P. D.; Klassen, J. S.; Strittmatter, E. F.; Williams, E. R. J. Am. Chem. Soc. 1998, 120, 9605-9613.

4650 Analytical Chemistry, Vol. 73, No. 19, October 1, 2001

RESULTS AND DISCUSSION Shiga-like toxin 1 (SLT-1) is produced by enteropathogenic Escherichia coli and belongs to the AB5 class of bacterial toxins. These toxins consist of an enzymatic A subunit, noncovalently bound to a B5 oligomer, that is responsible for the recognition of the cell-surface receptors.29 From solution NMR studies,30,31 it is known that the B subunits assemble into a pentameric ring structure in aqueous solution at pH 7 (Figure 1a). Each B subunit (7.7 kDa) is composed of 69 amino acids that form two threestranded antiparallel β sheets and an R helix.32,33 The tertiary structure of each subunit is stabilized by a disulfide bond between cysteine 4 located in the first β strand and cysteine 57 located in the γ turn after the fifth β strand. The second β strand (β2) of each monomer interacts with the sixth β strand (β6) of the following monomer, thus joining two three-stranded β sheets from adjacent monomers into one six-stranded antiparallel β sheet. The quaternary structure might be additionally stabilized by the hydrophobic interactions of the R-helixes located on the inside of the ring. Each subunit contains 8 basic amino acids (2× arginine, 5× lysine, and 1× histidine). On the basis of the crystal structure,34 (29) Kitov, P. I.; Sadovska, J. M.; Mulvey, G.; Armstrong, G. D.; Ling, H.; Pannu, N. S.; Read, R. J.; Bundle, D. R. Nature 2000, 403, 669-672. (30) Shimizu, H.; Field, R. A.; Homans, S. W.; Donohue-Rolfe, A. Biochemistry 1998, 37, 11078-11082. (31) Richardson, J. M.; Evans, P. D.; Homans, S. W.; Donohue-Rolfe, A. Nature Struct. Biol. 1997, 4, 190-193. (32) Stein, P. E.; Boodhoo, A.; Tyrrell, G. J.; Brunton, J. L.; Read, R. J. Nature 1992, 355, 748-750. (33) Ling, H.; Boodhoo, A.; Hazes, B.; Cummings, M. D.; Armstrong, G. D.; Brunton, J. L.; Read, R. J. Biochemistry 1998, 37, 1777-1788.

each subunit can engage in as many as five strong interactions (one ionic and four H-bonds) with each neighboring subunit. The intersubunit hydrogen bonds are expected to strengthen upon removal of the solvent and are expected to be retained in the gas phase.35 A salt bridge between the guanidine group of arginine (residue 33) and the carboxylic group of aspartic acid (residue 18) of the neighboring subunit was identified from the crystal structure. Generally, ionic interactions are energetically unstable in the absence of solvent and are neutralized by proton transfer.35 A number of studies of protonated and cationized peptides and amino acids have suggested that in some instances, salt bridges can be preserved in the gas phase, provided that additional stabilizing ionic interactions are present.22f,36,37 Because according to the crystal structure, no basic amino acids other than arginine 33 are in proximity to aspartic acid 18, we believe the ionic interactions are neutralized in the gas phase, possibly resulting in the formation of hydrogen bonds. Dissociation Pathways of B5n+ and B5n+-Oligosaccharide Complexes. B5 Homopentamer. As described previously,27 nanoelectrospray of aqueous solutions of SLT-1(B5) at pH 7 results almost exclusively in the production of protonated pentamer ions, (B5 + nH)n+ t B5n+, where n ) 11-13, with the base peak at n ) 12. Acidification of the electrospray solution with acetic acid to pH 4 results in a slight increase in the charge state of pentamer, with the base peak at n ) 14. In addition, a significant fraction of the ions observed in the mass spectrum correspond to monomer (B7+, B6+, B5+, and B4+) and smaller oligomer ions (e.g., B29+ and B27+). The higher charge states and the appearance of monomer and dimer ions were attributed to the partial denaturation of the subunits, leading to reduced stability in the complex. The dissociation pathways and kinetics for the protonated pentamers (B511+, B512+, B513+, and B514+), were investigated at temperatures ranging from 105 to 172 °C. Representative BIRD spectra for each parent ion are shown in Figure 2. The B511+ and B512+ pentamers were produced from solution at neutral pH, but B514+ was generated from an acidified solution, pH 4. The B513+ ion was observable at neutral and acidic pH and kinetic data were obtained for pentamers generated under both conditions. The protonated pentamers dissociate almost exclusively by the loss of one subunit (eq 1), which is in agreement with the dissociation behavior exhibited by other multimeric protein complexes.12-19

B5n+ f Bp+ + B4(n-p)+

(1)

The dominant product ion charge states observed are shown in Scheme 1. The charge state distribution of the product ions was found to be sensitive to the charge state of the parent ion and to temperature. In all cases, the monomer was ejected with a significant fraction of the overall charge, and the degree of enrichment increased with the parent charge state. For example, the B4+ monomer, ejected from B511+, retained 36% of the total (34) http://www.rcsb.org/pdb/entry 1BOS. (35) Jarrold, F. J. Annu. Rev. Phys. Chem. 2000, 51, 179-207. (36) Freitas, M. A.; Marshall, A. G. Int. J. Mass Spectrom. 1999, 182/183, 221231. (37) Wyttenbach, T.; Bushnell, J. E.; Bowers, M. T. J. Am. Chem. Soc. 1998, 120, 5098-5103.

Scheme 1. Decomposition Pathways Observed for the B5+n Complexesa

a This pathway was significant only for B +13 generated from 5 acidic solution.

charge, but for B514+, the B6+ and B5+ ions, the dominant monomer products, retained 43 and 36%, respectfully. For B513+ and B514+, higher reaction temperatures were found to favor greater charge enrichment of the ejected subunit (Figure 2d,e,g,h). For example, at 107 °C, the ion intensities of the complementary product ions B6+/B48+, B5+/B49+ and B4+/B410+, formed from B514+, are quite similar. However, at 142 °C, the B6+/B48+ ion pair dominates, and even a small amount of B7+/B47+ is observed. A similar trend was observed for B513+. In contrast, for B512+, the higher reaction temperature favored the ejection of the subunit with less charge. These results are illustrated in Figure 3, where the relative yields of the pairs of complementary product ions are plotted versus reaction temperature for different pentamer charge states. The B4 ions resulting from the loss of a single subunit were found to be resistant to further dissociation over the temperature range studied. For example, no dissociation products were observed for the isolated B410+ ion, produced from B514+ after 5 s at a temperature of 149 °C. Longer reaction delays were not feasible because of a poor S/N ratio. In comparison, the dissociation half-life of B514+ is 0.4 and 1.2 s for B513+ at this temperature. Therefore, on the time scale of the dissociation experiments performed on the B514+ ion, the B410+ product ion does not undergo subsequent dissociation. Another tetramer, B48+, was found to be less reactive than the most stable pentamer, B511+. At 172 °C, no reaction was observed for B48+ over a 10-s reaction delay, but the dissociation half-life of B511+ was 1 s. The robustness of the tetramers appears to be the result of their low charge-to-subunit ratio (vide infra). For the B514+ and B513+ ions, generated from acidic solution, a second, albeit minor (