Probing Mobility-Selected Saccharide Isomers: Selective Ion

Apr 1, 2015 - Zhixun Luo , A. W. Castleman , Jr. , and Shiv N. Khanna. Chemical ... Juan Camilo Molano-Arevalo , Walter Gonzalez , Kevin Jeanne Dit Fo...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/JPCA

Probing Mobility-Selected Saccharide Isomers: Selective Ion− Molecule Reactions and Wavelength-Specific IR Activation Oscar Hernandez,† Samantha Isenberg,‡ Vincent Steinmetz,† Gary L. Glish,‡ and Philippe Maitre*,† †

Laboratoire de Chimie Physique, Université Paris Sud, 91400 Orsay, France Department of Chemistry, University of North Carolina, 320 Caudill Laboratories, Chapel Hill, North Carolina 27599-3290, United States

‡

ABSTRACT: Differential Ion Mobility Spectrometry (DIMS) provides orthogonal separation to mass spectrometry, and DIMS combined with the high sensitivity of a quadrupole ion-trap is shown to be useful for the separation and identification of saccharides. A comprehensive analysis of the separation of anomers (α- and β-methylated glucose) and epimers (α-methylated glucose and mannose) ionized with Li+, Na+, and K+ is performed. DIMS separation is found to be better for saccharides cationized with the two latter species. The corresponding resolving power for the two glucose anomers with Na+ is found to be very close to the corresponding drift-tube IMS value. The lithiated complexes are investigated further using a combination of infrared spectroscopy integrated to ion-trap mass spectrometry and quantum chemical calculations. Together with DIMS, consistent results are obtained. It is found that two competing structural motifs might be at play, depending on the subtle balance between the maximization of the coordination of the metal cation and the intrinsic conformational energetics of the saccharide, which is for a large part driven by hydrogen bonding. The comparison of simulated and observed spectra clearly shows that a band at ∼3400 cm−1 is specific to a structural motif found in the lithiated glucose complexes, which could explain the trends observed in the DIMS spectra of the saccharide complexes. It is shown that DIMS-MS/MS using wavelength specific IR activation would provide a new orthogonal dimension to mass spectrometry.

■

collision cross section with a buffer gas7−10 allows for an enhancement of the resolving power of the experiment. Additionally, structural information can be derived via the comparison of measured and predicted mobilities11 of various isomeric forms using either exact (elastic) hard sphere scattering12 or trajectory method13 calculations. These two advantages were evidenced in pioneering studies on oligosaccharides using IMS-MS instruments featuring electrospray14 as well as MALDI15 ionization techniques. Tetraoses and hexaoses cationized with Na+ were studied by Bowers and co-workers.15 Deprotonated oligosaccharides were studied by Clemmer and co-workers,14 and an interesting aspect of this work was to exploit the fragmentation occurring at the entrance of the drifttube for examining the mobility separation of oligosaccharides and also their fragments. More systematic recent studies have shown that using low electric field mobility techniques, drifttube ion mobility spectrometry (DTIMS), and/or, more recently, traveling wave IMS (TWIMS),16 anomers, epimers, and, more generally, isomeric stereoisomers can be efficiently separated using the IMS-MS approach.17−21 A subsequent step

INTRODUCTION Separation and identification of oligosaccharides remains a challenge for natural and synthetic carbohydrates because there are no universal reaction conditions for oligosaccharide synthesis.1 The structural identification of complex carbohydrates is arduous because it consists in characterizing the anomeric configuration and the monomer stereochemistry, and the inter-residue linkage positions, and the branching features. In this context, tandem mass spectrometry (MS/MS) has been shown to be a promising analytical approach, which allows the selection and manipulation of isomeric species, if they are pure. Information is derived from the fragmentation pattern of massselected ions, which often are adducts of oligosaccharides and metal cations. Interestingly, it has been observed that the smaller the metal ion used for ionizing the sample, the larger is the fragmentation yield.2 Using collision induced dissociation (CID), it has been shown in particular that characteristic CID fragments related to linkage types can be obtained.34 It has also been observed that the degree of branching of the oligosaccharide, which is of importance for its biology, can be correlated with its fragmentation yield.2 An important advance in the structural characterization of oligosaccharides was achieved with the integration of ion mobility spectrometry (IMS) techniques to mass spectrometry (MS).5,6 Separating isomeric species as a function of their © 2015 American Chemical Society

Special Issue: Jean-Michel Mestdagh Festschrift Received: December 1, 2014 Revised: March 31, 2015 Published: April 1, 2015 6057

DOI: 10.1021/jp511975f J. Phys. Chem. A 2015, 119, 6057−6064

Article

The Journal of Physical Chemistry A

that selective ion−molecule reactions with water can be observed and that the yields of this reaction nicely correlate with the mobility results. Similarly, IR spectra recorded for mobility-selected ions are also consistent with the mobility results. It is shown in particular that the two glucose anomer complexes share the same IR signature, a band observed near ∼3400 cm−1. These results can be rationalized using quantum chemical modeling, and they suggest that structural differences between metal cationized monosaccharides are associated with the competition between intrasugar hydrogen bonding and coordination to the metal cation.

toward the structural characterization of mobility- and massselected ions can be achieved by integrating IMS and MS/MS, leading to IMS-MS/MS instruments. The principal added value is that mobility-selected precursor ions can be reacted or dissociated, followed by mass analysis of the product ions.22 More generally, combining multiple ion activation methods in MS/MS, including electron capture dissociation and photoinduced dissociation, is highly desirable.23−28 Spectroscopy has the ability to provide additional structural information on mobility-selected ions. Several studies of the coupling of UV laser with IMS based instruments have been reported recently.26−29 Rizzo and co-workers26 showed a UV spectroscopic study of mobility-selected isomers of the doubly protonated peptide bradykinin. This integrated setup is based on a cold ion-trap, which by itself already greatly simplifies the electronic spectra of large molecular ions.30,31 A high-field asymmetric waveform IMS (FAIMS) device was used to separate conformers before injecting them into a cold iontrap.26 An interesting result of this work was to provide spectroscopic evidence for the conformational equilibrium of the mobility-selected conformer. It should be noted that the gas phase equilibrium of mobility-selected conformers of bradykinin [M + 3H]3+ ions had also been studied in detail by Clemmer and co-workers32,33 using an IMS-IMS instrument, where mobility-selected conformers could be energized through collisions before their injection into the second drift-tube. Clemmer, Reilly and co-workers27,28 recently proposed two combinations of IMS and UV induced fragmentation for elucidating the structure of disaccharides. It was shown in particular that, in contrast with CID, irradiation at 157 nm generates unique fragments for each disaccharide ion that are useful for distinguishing them. Another interesting recent report, by Bieske and co-workers,29 describes an IMS+UV setup allowing monitoring of the mobility variation as a function of the excitation wavelength prior to the injection of molecular ions into a drift-tube. Photoelectron spectroscopy of mobility-selected anions developed by Kappes and co-workers34,35 is also particularly promising. Using this approach, conformer-selective photoelectron spectroscopy of α-lactalbumin has been performed on multianions, demonstrating the correlation between mobility (folded versus unfolded) and gas phase adiabatic detachment energies.24 IR spectroscopy of mobility-selected metal cationized monosaccharides is presented in this paper. A differential ion mobility spectrometer (DIMS) has been mounted on the inlet of the mass spectrometer, allowing for MS/MS study of DIMSand mass-selected isomeric ions. In contrast to DTIMS or TWIMS, structural information cannot be directly linked to the mobility parameter of DIMS. The DIMS technique relies on the difference of ion mobility between high and low electric fields, and the mobility under high electric field conditions is still not well-understood. Conversely, DIMS can provide high resolution separation,36 and coupled with an ion-trap offers high ion sensitivity. Perhaps more impotantly DIMS separates ions in space rather than in time like the other IMS techniques. Thus, DIMS can be readily coupled with any type of mass analyzer, not just time-of-flight. Separation of adducts of alkali cations (Li+, Na+, K+) and monosaccharide methyl glycosides has been investigated, and the DIMS performance in terms of resolving power is compared with recently reported data obtained with DTIMS and TWIMS devices.18 MS/MS experiments were performed in the case of Li+ adducts for probing the structure of DIMS-selected isomers. It is shown

■

METHODS Ion mobility tandem mass spectrometry (IM-MS/MS) experiments were performed using a modified Esquire 3000+ (Bruker) instrument equipped with electrospray ionization (ESI). The DIMS device is threaded to match the spray shield of a Bruker Esquire 3000+, such that the spray shield can be removed and the DIMS assembly can be screwed on in its place. It is comprised of two parallel plates which are 25 × 6 mm, separated by 0.5 mm. A 4 kV voltage is applied to the ESI emitter. Its position on the center axis is adjusted to optimize the ion signal. Ions are transported by a carrier gas stream at atmospheric pressure between the two parallel plates. In the DIMS design used for these experiments,37 the desolvation gas which is already implemented in the source for ESI desolvation purposes is redirected through the outer housing of the DIMS assembly. Thus, the desolvation gas serves as the carrier gas through the DIMS assembly as well as for desolvation when coupled to ESI. The desolvation gas thus also heats the DIMS assembly. For these experiments the temperature of the DIMS assembly was 86 °C. The standard glass transfer capillary of the source was replaced with a custom flared glass transfer capillary.38 The waveform used for the DIMS separation is a bisinusoidal waveform generated by applying a sinusoidal waveform at 1.7 MHz to one of the DIMS electrodes and a second sinusoidal waveform at the second harmonic, phase shifted approximately 90°, and at 50% of the amplitude of the first sinusoidal waveform.37 The V0‑p is defined as the dispersion voltage (DV); DV is typically tuned in the 1.5−2.0 kV range with a trade-off between ion transmission and DIMS resolution. A dc offset, termed the compensation voltage (CV), is applied to one of the electrodes allowing the transmission of ions of interest through the DIMS based on their differential ion mobility. DIMS filtered ions are then transmitted through the capillary to the quadrupole ion-trap where ions can be accumulated and trapped. DIMS spectra are generated by plotting the ion abundance as a function of CV value. DIMS spectra were recorded using a lab-built device controlled with a LabView program allowing the synchronization of the scan of the CV mobility parameter with the Bruker acquisition software under chromatogram mode. As described in details elsewhere,39 a hole (0.7 mm diameter) was drilled in the ion-trap ring electrode allowing for IR irradiation of the trapped ions. MS/MS experiments involving IR activation are performed using the commercial software by setting the collision voltage to zero. An output trigger of the mass spectrometer is used to synchronize a mechanical shutter with the time activation window.39 IR activation was provided by an IR OPO/OPA laser (LaserVision), pumped by a nonseeded 10 Hz Nd:YAG laser 6058

DOI: 10.1021/jp511975f J. Phys. Chem. A 2015, 119, 6057−6064

Article

The Journal of Physical Chemistry A

Figure 1. DIMS spectra of complexes of saccharides and Li+ (a), Na+ (b), and K+ (c). In each panel, the DIMS spectrum of the individual isomer (Sac = α-Me-glucose (green), β-Me-glucose (red), and α-Me-mannose (black)) is given along with the spectrum recorded for the equimolar mixture (blue).

probably related to the relative affinity of the methyl glycoside isomers to the alkali cation.15,43 A single peak was observed for each combination of cation and saccharide, suggesting that a single complex is formed in all cases. The peak shape observed for the individual ions in the cases of M = Li and Na (Figure 1a and 1b, respectively) is the same, with a fwhm value of ∼0.5 V. The larger peak width observed in the case of K+ (Figure 1c) may suggest that more flexible structures or mixtures of conformers are formed. Separation between anomers and epimers is an important goal.18,44 The comparison of the DIMS spectra recorded for the sugar mixture clearly shows that the most efficient separation could be obtained with Na+ as an ionizing agent. In particular, the complexes corresponding to the glucose anomers, and to the methyl-α-glucose and methyl-α-mannose epimers are well resolved. For the Na+ complexes, the peak separation between the two glucose anomers is 1.2 V with a peak fwhm of 0.5 V. For these anomers bound to Na+, the DIMS resolution is 1.41, which is very close to the corresponding DTIMS 1.12 value, and larger than the TWIMS 0.44 value recently reported by Hill and co-workers.18 Separation of anomers, epimers, and more generally isomers of methyl glycosides does not only depend on the stereochemistry of the compounds, but also on its coordination to the ionizing cation. This is clearly illustrated in Figure 1a, which shows that the two glucose anomers have identical CV values when complexed to Li+, while two well resolved peaks can be observed when they are complexed to Na+ or K+. Furthermore, it is interesting to observe that the relative CV values of the two glucose anomers switch when changing the cation from Na+ to K+ (Figure 1b and 1c, respectively). Even if there is no obvious correlation between the DIMS mobility parameter (CV) and the collision cross section, as in DTIMS or even TWIMS, this observation suggests a nontrivial change of coordination when the size of the alkali cation increases. As mentioned above, MS/MS experiments on mobility- and mass-selected ions can afford structural information, and could also be interesting for enhancing the analytical resolving power. The CID spectra of the isomers studied here are very similar, making it difficult to distinguish them. Two alternative MS/MS

(Continuum Surlite II, 550 mJ per pulse, 1064 nm, 4−6 ns pulse duration). Reference gas phase IR spectra of electrosprayed ions were recorded using a 7T hybrid FT-ICR (Bruker) coupled with a similar OPO/OPA laser (LaserVision), but a different (25 Hz Nd:YAG laser (Innolas Spitlight 600)) pump laser as described in details elsewhere.40 As with the DIMS-selected ions, the irradiation time was 1 s. Sample solutions were prepared by mixing saccharides (Carbosynth) and alkali metal salts (Sigma-Aldrich) at a ratio of 1:20 in H2O:CH3OH = 3:7 with a final concentration of 100 μM (saccharide) and 2 mM (metal salt). ESI conditions were as follow: Flow rate of 120 μL/h, desolvation gas flow of 5 L/min, nebulizer pressure of 10 psi, and desolvation gas temperature of 300 °C. The ESI voltage (4000 V) is applied to the electrospray emitter. Quantum chemical calculations were performed using the Gaussian09 package. Stationary points on the potential energy surface were characterized at the B3LYP/6-31G(d,p) level of theory. 298 K free enthalpies were determined at this same level. IR absorption spectra were derived from harmonic calculations. A scaling factor of 0.955 was applied to the harmonic frequencies, as used previously for simulating IR spectra in the X−H (X = C, N, O) stretching spectral range.40−42

■

RESULTS AND DISCUSSION DIMS spectra of adducts of saccharides with alkali cations M+, M(Sac)+ (M = Li, Na, K) and saccharides (Sac = α-Me-Glc, βMe-Glc, and α-Me-Man) obtained using our DIMS device coupled to a quadrupole ion-trap are displayed in Figure 1. For each alkali cation, in addition to the spectrum of each individual isomer, the DIMS spectrum of the isomer mixture was also recorded. The dispersion voltage DV values were 2.0 kV (40 kV/cm) for the three alkali cations. This value was found to correspond to the best trade-off between ion transmission and mobility separation. As can be seen in Figure 1, using equimolar mixtures, different ionization efficiencies are observed for the isomers and the relative efficiencies vary with cation. This is 6059

DOI: 10.1021/jp511975f J. Phys. Chem. A 2015, 119, 6057−6064

Article

The Journal of Physical Chemistry A approaches shown to provide unique structural information on the mobility-selected methyl glycosides/M+ complexes in the case of M = Li are presented below. In one alternative MS2 approach, ion/molecule reaction experiments were carried out on DIMS-selected Li(Sac)+ ions. Following their mass selection, ions were allowed to react with water seeded in the He buffer gas using a setup allowing for a control of the neutral flow.45 While the monohydrated Li(Sac)+ adducts is formed in large amount in the case of the two glucose anomers, the reaction of water with methyl-α-mannose is less favorable. Only ≈8% of the monohydrated adduct could be formed after 3 s reaction time (Figure SI1). These results are consistent with the shape of the DIMS spectra observed for the lithiated complexes. The fact that a single mobility peak corresponding to the two glucose anomers was observed may suggest that the two corresponding complexes have similar 3D structure which, in turn, would suggest that the Li + coordination number is the same. Considering the strong dependency of the H2O−Li+ binding energy on the Li+ coordination number,46 it could be anticipated that the complex formed with the methyl-α-mannose corresponds to a larger coordination number of the metal, and thus to a weaker binding and less efficient association reaction with water. Structural information on the Li(Sac)+ mobility-selected H2O adducts could be derived using MS3 experiments. The mass-selected monohydrated clusters were activated using a wavelength tunable IR laser. Reference gas phase IR specta of electrosprayed monohydrated Li(Sac)(H2O)+ complexes were first recorded using a hybrid FT-ICR, in which Li(Sac)+ complexes were mass-selected in a linear quadrupole before their accumulation in a linear hexapole ion-trap pressurized with water seeded in the Ar line.40 Li(Sac)(H2O)+ adducts could be formed, as in the quadrupole ion-trap, and then transferred in the ICR cell where they were mass-selected and interrogated with the tunable IR laser. The advantage of this setup over the quadrupole ion-trap is that under the low pressure conditions of the ICR cell, association/dissociation water reactions do not compete with the IR activation. As a result, well resolved IR spectra could be recorded for the Li(Sac)(H2O)+ adducts (Figure 2), especially in the case of the Sac = Methyl-α-mannose for which the association reaction efficiency is low. Gas phase IR spectra of DIMS-selected Li(Sac)(H2O)+ isomers were then recorded in the 3000−3700 cm−1 region using the quadrupole ion-trap equipped with the DIMS device, by setting the CV to a value to transmit the specific ion of interest and then scanning the laser wavelength. As discussed recently by Clemmer, Reilly, and co-workers,28 sensitivity is an important challenge for performing IR photodissociation. With this respect, DIMS is the method of choice since using a fixed CV value, DIMS-selected ions can be continuously accumulated in the quadrupole ion-trap, before being subjected to MS/MS experiments.47 The IRMPD spectra recorded for the DIMSselected Li(α-Me-Glc)(H2O)+ and Li(β-Me-Glc)(H2O)+ adducts are shown in Figure 2a and 2b, respectively. Using an irradiation time of 1s, up to 10% fragmentation efficiency could be observed, providing good signal/noise. Overall, the IR spectra of mobility-selected ions compare very well with reference spectra recorded without DIMS selection using our hybrid FT-ICR (Figure 2). Well resolved bands are observed above 3100 cm−1. They correspond to the OH stretches of sugar and water moieties. In addition, weak bands corresponding to CH stretches near 3000

Figure 2. IR spectra of Li(Sac)(H2O)+ ions. (a) Sac = α-Me-glucose; (b) Sac = β-Me-glucose; (c) Sac = α-Me-mannose. Black spectra correspond to reference IR spectra of mass-selected electrosprayed ions recorded using a hybrid FT-ICR. IR spectra of mobility- and mass-selected ions using a quadrupole ion-trap of Li(α-Me-glucose)(H2O)+ and Li(β-Me-glucose)(H2O)+ ions are given in panels a (green) and b (red), respectively.

cm−1 can also be distinguished. Considering the weak absorbance of the CH stretching modes, this observation provides evidence for the high sensitivity of IR spectroscopy. Since the laser power increases from 3100 to 3700 cm−1, it is thus likely that all weak IR absorption bands can be revealed through IR induced photodissociation under our experimental conditions. A detailed assignment of the observed bands is out of the scope of this paper. We rather would like to focus on the main difference between the spectra of the Li(Sac)(H2O)+ adducts of the two anomers of glucose (Figure 2a and 2b) and the one corresponding to Sac = Me-α-Man (Figure 2c). While no band can be observed in the 3100−3500 cm−1 range for the Me-α-Man complex, a band is observed near ∼3400 cm−1 for the two Glu isomers. On the basis of spectroscopic studies of water solvated metal complexes,48 the position of this band is characteristic of a hydrogen bonded OH stretch and it is likely to be structure-specific. Structures of the present Li(Sac)(H2O)+ complexes, and more generally of metal-saccharide complexes, are the result of a trade-off between the maximization of the coordination of the metal cation and the intrinsic conformational energetics of the saccharide, which is for a large part driven by hydrogen bonding. This is clearly illustrated by the work of Armentrout and co-workers49 on the thermochemistry and structures of Na+ coordinated pentose and hexose, which had also been investigated earlier by Wesdemiotis and co-workers.43 Density functional calculations were carried out on the Li(Sac)(H2O)+ (Sac = α-Me-Glc, β-Me-Glc, and α-Me-Man) complexes. The structures of the free saccharides were first considered assuming a chair (4C1 or 1C4) or a boat conformation. Multiple orientations of the hydroxyl and hydroxymethyl groups were used as guess structures for this 6060

DOI: 10.1021/jp511975f J. Phys. Chem. A 2015, 119, 6057−6064

Article

The Journal of Physical Chemistry A purpose. Then, the Li+ cation was added and different coordination sites were envisaged using various orientations of the hydroxyl and hydroxymethyl groups in order to find the best compromise between hydrogen bonding and Li + coordination. In a third step, the water molecule was added on the Li(Sac)+ optimized structures. A detailed theoretical study of the isomers of Li+ complexes, bare and monohydrated, is under progress. Within the large number of optimized structures, two types of low-energy structures can be distinguished. As shown previously for Na+-hexose,43,49 structures with tri- or even tetradentate coordination of the sugar unit to the Li+ were found (if Li−O ≤ 200 pm distances are considered). Such structures were found to be the most favorable ones for the Li(α-Me-Man)(H2O)+ system, and the lowest-energy structure is given in Figure 3a. The Methyl-α-

energy structure of the Li(α-Me-Man)(H2O)+ system (Figure 3a), are found higher in energy (∼10 kJ/mol) than the structure shown in Figure 3b. The lowest-energy structure found for the glucose containing complexes is characterized by a sugar unit in a 4C1 chair conformation bound in a bidentate fashion through the hydroxyl O(3) and O(4) oxygen atoms. The O(4)H is also involved in a strong hydrogen bond with the hydroxymethyl oxygen characterized by a O(4)H−O(5) distance of ∼180 pm. The predicted IR absorption spectrum for the lowest energy structure of Li(α-Me-Glc)(H2O)+ is given in Figure 3b along with the experimental spectrum (gray trace). The observed bands nicely match with the predicted IR absorption bands in the OH stretching region. In particular, an IR absorption band is predicted near ∼3400 cm−1, which could explain the structure-specific IR band observed for both Li(α-Me-Glc)(H2O)+ and Li(β-Me-Glc)(H2O)+ complexes. The corresponding normal mode involves the O(4)-H stretching mode (Scheme 1). The ∼200 cm−1 red-shift associated with this vibrational mode is characteristic of the Li+-O(4)-H-O(6) bonding motif where the O(4)H2 water molecule acts simultaneously as an electron donor to Li+ and hydrogen bond donor to the hydroxylmethyl OH group. Similar hydrogen bonding motifs have been characterized when progressive water nanosolvation of metal cation has been studied in the gas phase using IR spectroscopy.48 Bands significantly red-shifted as compared to the free OH stretching position could be observed upon formation of the second coordination shell. It should be noted, however, that these bands observed for the M(H2O)n+ systems, for example M=Ni,48 were significantly broader than the ∼3400 cm−1 band observed for the present Li(Glc)(H2O)+ complexes. The structures obtained for both Li(α-Me-Glc)(H2O)+ and Li(β-Me-Glc)(H2O)+ complexes, versus the structure of the Li(α-Me-Man)(H2O)+, together with the spectroscopic results provide a consistent picture of the DIMS and ion−molecule results for the lithiated systems. As evidenced by the determination of sequential water binding energies of water to Li+,50 the binding energy of water to Li+ decreases when the coordination of Li+ increases. One can thus expect that the binding energy of water to Li+ is stronger when it is only dicoordinated (glucose anomers complexes) than when it is tricoordinated (mannose) by the sugar. This would thus explain that the water association/dissociation equilibrium is more displaced toward the monohydrated complex in the former than in the latter case. Similarly, assuming that the sugar denticity remains unchanged in the Li(Sac)+ versus the Li(Sac)(H2O)+, the structures for Sac = α-Me-Glc and β-MeGlc would be similar, which could explain that the optimum CV value for the corresponding complexes are very similar. A different CV value, however, would be expected for the Li(αMe-Man)+ complex which corresponds to a tricoordinated Li+ cation. Our results are also consistent with those Armentrout and co-workers49 on the thermochemistry and structures of Na+ coordinated pentose and hexose, including glucose anomers. The lowest energy structures of the Na(hexose)+ involve multidentate sodium cation coordination, with the sugar in a sterically favored chair ring conformation. Of particular interest in the context of the present study is the case of the Na(glucose)+ complexes. The ground state geometries of the α and β complexes differ considerably. This result can be considered as consistent with the DIMS spectra of the sodiated

Figure 3. IR spectra and corresponding lowest-energy structure of (a) Li(α-Me-Mannose)(H2O)+ and (b) Li(α-Me-Glucose)(H2O)+ complex ions. In each case, the experimental IR spectrum, in gray, is compared with the IR absorption spectrum predicted for the lowestenergy structure. Stick bars represent the intensities associated with the harmonic frequencies (scaling factor = 0.955). The structure specific band observed near 3400 cm−1 is associated with the OH bond involved in the hydrogen bond with the hydroxymethyl group as shown in panel b.

Mannose has a 4C1 chair conformation, and it is bonded through axial O(2)H, hemiacetal O(5) and hydroxymethyl O(6)H oxygen atoms. The predicted IR absorption spectrum for this structure is given in Figure 3a. As can be seen in this Figure, the observed bands nicely match with the predicted IR absorption bands in the OH stretching region. In the case of the two glucose anomers, however, the bonding motif associated with the lowest energy structure is different. The lowest energy structure of Li(α-Me-Glc)(H2O)+ is shown in Scheme 1 and Figure 3b. The sugar is bound to Li+ through only two oxygen atoms, and the hydroxymethyl OH group is involved in an intrasugar hydrogen bond. Structures with a tridentate sugar coordination, i.e. similar to the lowest Scheme 1. Lowest Energy Structure of the Li(α-MeGlc)(H2O)+ Complex

6061

DOI: 10.1021/jp511975f J. Phys. Chem. A 2015, 119, 6057−6064

Article

The Journal of Physical Chemistry A systems presented in Figure 1b. In contrast to the case of the lithiated complexes (Figure 1a), where the two glucose anomers cannot be separated, it was found that the sodiated methylated glucose anomers can be efficiently separated with our DIMS device. The results of Armentrout and co-workers49 together with our result thus suggest that when going from Li+ to Na+, the bonding of the glucose anomers to the metal cation changes. In the case of the sodiated complexes, the metal-sugar interaction associated with the lowest energy structure of the α-glucose anomer is very similar to the one associated with the lowest energy structure of the Li(α-Me-Glc)(H2O)+ and Li(β-MeGlc)(H2O)+ complexes. The sugar binds the metal in a bidentate fashion, and the hydroxymethyl O(5)H group is involved in a hydrogen bond with the O(4)H donor group, which in turn is bound to the metal. It should be noted that in an earlier study of the thermochemistry and structure of Na+ coordinated pentose and hexose, 43 it was found the hydroxymethyl O(5)H group was bound to the Na+ cation in the lowest energy structures. It should be stressed, however, that the relative energies of structures associated with different metal-sugar bonding schemes is very sensitive to the theoretical levels, as found for Na+,49 and also for Li+ complexes. Density functional calculations are currently being performed to see if the evolution of the ion behavior in the DIMS spectra as a function of the nature of the alkali metal (vide supra) could be correlated to a change in coordination mode of the monosaccharide in the cases of the two glucose anomers.

Figure 4. DIMS spectra for an equimolar mixture of the three saccharides in Li+ solution. An MS3 sequence was used where DIMSand mass-selected m/z 201 ions were allowed to react with water and the resulting m/z 201 (blue) and 219 (red) ions were irradiated for 1s with the laser tuned at 3400 cm−1. Top panel: Laser on; Bottom panel: Laser off.

■

specific functional groups could also be done using the 800− 2000 cm−1 fingerprint spectral range as available with Free Electron Laser systems and more recently with table-top laser systems. From a fundamental point of view, we herein demonstrate that the present instrument allows for the IR spectroscopic characterization of mobility- and mass-selected ions. A DIMS device coupled with an ion-trap offers the advantage that DIMS-selected ions can be accumulated. As a result, ion signal is sufficient for performing MS/MS experiments and excellent signal/noise is achieved with the IR spectroscopy. As illustrated in Figure 3, the comparison of experimental IR spectra with the predicted IR spectra of the different isomers allows for a structural characterization of the ions of interest. Information on the hydrogen bonding network can be obtained using OPO/OPA lasers tunable in the NH−OH stretching regions as in the present case. Pure mobility separation relies on the collision cross section of the ion with the buffer gas. Using drift-tube instruments, structural information can be derived from the comparison of experimental collision cross section against theoretical ones. While this is true when ions are submitted to low electric field conditions, as of yet no clear correlation between the mobility parameter and the structure can be drawn when ions are submitted to high electric field, as under DIMS. IR spectroscopy of DIMS-selected ions could thus contribute to a better understanding of the ion mobility separation under asymmetric high electric field conditions.

CONCLUDING REMARKS AND PERSPECTIVES To conclude, integrating DIMS and IR activation on the same tandem mass spectrometer provides two additional dimensions to the analytical technique. Our motivation for developing an innovative instrument is 2-fold. It ranges from understanding the correlation of the shape of a molecule with intramolecular interactions to enhancing the resolving power of the method for separating and characterizing isomeric or isobaric molecules. From an analytical point of view, an interesting perspective for enhancing the resolving power of the analytical method would be to use a structure-specific activation wavelength for inducing the fragmentation of DIMS-selected ions. This idea is illustrated in Figure 4, where two DIMS spectra of a mixture sample of the three saccharides cationized with Li+ recorded with the laser off (a) and on (b) are given. The laser was tuned at 3400 cm−1, i.e. corresponding to the IR active mode specific of the glucose complexes. The DIMS spectra of the m/z 201 (Li(Sac)+) and 219 (Li(Sac)(H2O)+) ions given in Figure 4 were recorded using an MS3 sequence. DIMS- and massselected m/z 201 ions were allowed to react with water, and the resulting m/z 201 and 219 ions were irradiated for 1s with the laser tuned at 3400 cm−1. As compared to the DIMS spectrum recorded with the laser off, a depletion of m/z 219 signal and a simultaneous increase of the m/z 201 signal can be observed in the CV range (≈2.5 V) corresponding to the complexes of the glucose anomers. Conversely, the peak at CV ≈ 3.6 V, the mannose complex, does not change. This is expected since 3400 cm−1 is a structure-specific wavenumber, characteristic of the hydrogen bonding network of the Li(α-Me-Glc)(H2O)+ and Li(β-Me-Glc)(H2O)+ complexes. This approach could be used for identifying specific bonding motif or functional groups associated with the different peaks within a DIMS spectrum. As illustrated in Figure 4, an OPO/OPA table-top laser is ideally suited for probing the hydrogen bonding network. Identifying

■

AUTHOR INFORMATION

Corresponding Author

*(P. Maitre) Tel: +33 1 69153250; e-mail: [email protected]. Notes

The authors declare no competing financial interest. 6062

DOI: 10.1021/jp511975f J. Phys. Chem. A 2015, 119, 6057−6064

Article

The Journal of Physical Chemistry A

■

(18) Li, H. L.; Giles, K.; Bendiak, B.; Kaplan, K.; Siems, W. F.; Hill, H. H. Resolving structural isomers of monosaccharide methyl glycosides using drift tube and traveling wave ion mobility mass spectrometry. Anal. Chem. 2012, 84, 3231−3239. (19) Fasciotti, M.; Sanvido, G. B.; Santos, V. G.; Lalli, P. M.; McCullagh, M.; de Sa, G. F.; Daroda, R. J.; Peter, M. G.; Eberlin, M. N. Separation of isomeric disaccharides by traveling wave ion mobility mass spectrometry using CO2 as drift gas. J. Mass Spectrom. 2012, 47, 1643−1647. (20) Dwivedi, P.; Bendiak, B.; Clowers, B. H.; Hill, H. H., Jr. Rapid resolution of carbohydrate isomers by electrospray ionization ambient pressure ion mobility spectrometry-time-of-flight mass spectrometry (ESI-APIMS-TOFMS). J. Am. Soc. Mass. Spectrom. 2007, 18, 1163− 1175. (21) Li, H. L.; Bendiak, B.; Siems, W. F.; Gang, D. R.; Hill, H. H. Carbohydrate structure characterization by tandem ion mobility mass spectrometry (IMMS)2. Anal. Chem. 2013, 85, 2760−2769. (22) Hilderbrand, A. E.; Myung, S.; Barnes, C. A. S.; Clemmer, D. E. Development of LC-IMS-CID-TOFMS techniques: Analysis of a 256 component tetrapeptide combinatorial library. J. Am. Soc. Mass. Spectrom. 2003, 14, 1424−1436. (23) Bellina, B.; Compagnon, I.; MacAleese, L.; Chirot, F.; Lemoine, J.; Maitre, P.; Broyer, M.; Antoine, R.; Kulesza, A.; Mitric, R.; BonacicKoutecky, V.; Dugourd, P. Binding motifs of silver in prion octarepeat model peptides: a joint ion mobility, IR and UV spectroscopies, and theoretical approach. Phys. Chem. Chem. Phys. 2012, 14, 11433−11440. (24) Vonderach, M.; Winghart, M. O.; MacAleese, L.; Chirot, F.; Antoine, R.; Dugourd, P.; Weis, P.; Hampe, O.; Kappes, M. M. Conformer-selective photoelectron spectroscopy of alpha-lactalbumin derived multianions in the gas phase. Phys. Chem. Chem. Phys. 2014, 16, 3007−3013. (25) Le, T. N.; Poully, J. C.; Lecomte, F.; Nieuwjaer, N.; Manil, B.; Desfrancois, C.; Chirot, F.; Lemoine, J.; Dugourd, P.; van der Rest, G.; Gregoire, G. Gas-phase structure of amyloid-beta (12−28) peptide investigated by infrared spectroscopy, electron capture dissociation and ion mobility mass spectrometry. J. Am. Soc. Mass. Spectrom. 2013, 24, 1937−1949. (26) Papadopoulos, G.; Svendsen, A.; Boyarkin, O. V.; Rizzo, T. R. Spectroscopy of mobility-selected biomolecular ions. Faraday Discuss. 2011, 150, 243−255. (27) Lee, S.; Valentine, S. J.; Reilly, J. P.; Clemmer, D. E. Analyzing a mixture of disaccharides by IMS-VUVPD-MS. Int. J. Mass Spectrom. 2012, 309, 161−167. (28) Zucker, S. M.; Lee, S.; Webber, N.; Valentine, S. J.; Reilly, J. P.; Clemmer, D. E. An ion mobility/ion trap/photodissociation instrument for characterization of ion structure. J. Am. Soc. Mass. Spectrom. 2011, 22, 1477−1485. (29) Adamson, B. D.; Coughlan, N. J. A.; Continetti, R. E.; Bieske, E. J. Changing the shape of molecular ions: photoisomerization action spectroscopy in the gas phase. Phys. Chem. Chem. Phys. 2013, 15, 9540−9548. (30) Papadopoulos, G.; Svendsen, A.; Boyarkin, O. V.; Rizzo, T. R. Conformational distribution of bradykinin [bk+2H]2+ revealed by cold ion spectroscopy coupled with FAIMS. J. Am. Soc. Mass. Spectrom. 2012, 23, 1173−1181. (31) Wassermann, T. N.; Boyarkin, O. V.; Paizs, B.; Rizzo, T. R. Conformation-specific spectroscopy of peptide fragment ions in a lowtemperature ion trap. J. Am. Soc. Mass. Spectrom. 2012, 23, 1029−1045. (32) Pierson, N. A.; Valentine, S. J.; Clemmer, D. E. Evidence for a quasi-equilibrium distribution of states for bradykinin [M+3H]3+ ions in the gas phase. J. Phys. Chem. B 2010, 114, 7777−7783. (33) Pierson, N. A.; Chen, L. X.; Valentine, S. J.; Russell, D. H.; Clemmer, D. E. Number of solution states of bradykinin from ion mobility and mass spectrometry measurements. J. Am. Chem. Soc. 2011, 133, 13810−13813. (34) Vonderach, M.; Ehrler, O. T.; Matheis, K.; Weis, P.; Kappes, M. M. Isomer-selected photoelectron spectroscopy of isolated DNA oligonucleotides: Phosphate and nucleobase deprotonation at high negative charge states. J. Am. Chem. Soc. 2012, 134, 7830−7841.

ACKNOWLEDGMENTS This work is supported by a public grant from the “Laboratoire d’Excellence Physics Atom Light Mater” (LabEx PALM) overseen by the French National Research Agency (ANR) as part of the “Investissements d’Avenir” program (reference: ANR-10-LABX-0039). Financial support from the National FTICR network (FR 3624 CNRS) for conducting the research is gratefully acknowledged. Bruker Daltonics has licensed some of the DIMS IP developed at UNC.

■

REFERENCES

(1) Boltje, T. J.; Buskas, T.; Boons, G. J. Opportunities and challenges in synthetic oligosaccharide and glycoconjugate research. Nat. Chem. 2009, 1, 611−622. (2) Cancilla, M. T.; Penn, S. G.; Carroll, J. A.; Lebrilla, C. B. Coordination of alkali metals to oligosaccharides dictates fragmentation behavior in matrix assisted laser desorption ionization Fourier transform mass spectrometry. J. Am. Chem. Soc. 1996, 118, 6736− 6745. (3) Asam, M. R.; Glish, G. L. Tandem mass spectrometry of alkali cationized polysaccharides in a quadrupole ion trap. J. Am. Soc. Mass. Spectrom. 1997, 8, 987−995. (4) Hofmeister, G. E.; Zhou, Z.; Leary, J. A. Linkage position determination in lithium-cationized disaccharides − Tandem MassSpectrometry and semiempirical calculations. J. Am. Chem. Soc. 1991, 113, 5964−5970. (5) Both, P.; Green, A. P.; Gray, C. J.; Sardzik, R.; Voglmeir, J.; Fontana, C.; Austeri, M.; Rejzek, M.; Richardson, D.; Field, R. A.; Widmalm, G.; Flitsch, S. L.; Eyers, C. E. Discrimination of epimeric glycans and glycopeptides using IM-MS and its potential for carbohydrate sequencing. Nat. Chem. 2014, 6, 65−74. (6) Lanucara, F.; Holman, S. W.; Gray, C. J.; Eyers, C. E. The power of ion mobility-mass spectrometry for structural characterization and the study of conformational dynamics. Nat. Chem. 2014, 6, 281−294. (7) Hill, H. H.; Siems, W. F.; Stlouis, R. H.; McMinn, D. G. Ion mobility spectrometry. Anal. Chem. 1990, 62, A1201−A1209. (8) Clemmer, D. E.; Jarrold, M. F. Ion mobility measurements and their applications to clusters and biomolecules. J. Mass Spectrom. 1997, 32, 577−592. (9) Jarrold, M. F. Drift-tube studies of atomic clusters. J. Phys. Chem. 1995, 99, 11−21. (10) Vonhelden, G.; Hsu, M. T.; Kemper, P. R.; Bowers, M. T. Structures of carbon cluster ions from 3 to 60 atoms - Linears to rings to fullerenes. J. Chem. Phys. 1991, 95, 3835−3837. (11) Shvartsburg, A. A.; Mashkevich, S. V.; Baker, E. S.; Smith, R. D. Optimization of algorithms for ion mobility calculations. J. Phys. Chem. A 2007, 111, 2002−2010. (12) Shvartsburg, A. A.; Jarrold, M. F. An exact hard-spheres scattering model for the mobilities of polyatomic ions. Chem. Phys. Lett. 1996, 261, 86−91. (13) Mesleh, M. F.; Hunter, J. M.; Shvartsburg, A. A.; Schatz, G. C.; Jarrold, M. F. Structural information from ion mobility measurements: Effects of the long-range potential. J. Phys. Chem. 1996, 100, 16082− 16086. (14) Liu, Y. S.; Clemmer, D. E. Characterizing oligosaccharides using injected-ion mobility mass spectrometry. Anal. Chem. 1997, 69, 2504− 2509. (15) Lee, S.; Wyttenbach, T.; Bowers, M. T. Gas phase structures of sodiated oligosaccharides by ion mobility ion chromatography methods. Int. J. Mass Spectrom 1997, 167, 605−614. (16) Shvartsburg, A. A.; Smith, R. D. Fundamentals of traveling wave ion mobility spectrometry. Anal. Chem. 2008, 80, 9689−9699. (17) Clowers, B. H.; Dwivedi, P.; Steiner, W. E.; Hill, H. H.; Bendiak, B. Separation of sodiated isobaric disaccharides and trisaccharides using electrospray ionization-atmospheric pressure ion mobility-time of flight mass spectrometry. J. Am. Soc. Mass. Spectrom. 2005, 16, 660− 669. 6063

DOI: 10.1021/jp511975f J. Phys. Chem. A 2015, 119, 6057−6064

Article

The Journal of Physical Chemistry A (35) Vonderach, M.; Ehrler, O. T.; Weis, P.; Kappes, M. M. Combining ion mobility spectrometry, mass spectrometry, and photoelectron spectroscopy in a high-transmission instrument. Anal. Chem. 2011, 83, 1108−1115. (36) Shvartsburg, A. A.; Smith, R. D. High-resolution differential ion mobility spectrometry of a protein. Anal. Chem. 2013, 85, 10−13. (37) Isenberg, S. L.; Armistead, P. M.; Glish, G. L. Optimization of peptide separations by differential ion mobility spectrometry. J. Am. Soc. Mass. Spectrom. 2014, 25, 1592−1599. (38) Bushey, J. M.; Kaplan, D. A.; Danell, R. M.; Glish, G. L. Pulsed nano-electrospray ionization: characterization of temporal response and implementation with a flared inlet capillary. Instrum. Sci. Technol. 2009, 37, 257−273. (39) Mac Aleese, L.; Simon, A.; McMahon, T. B.; Ortega, J. M.; Scuderi, D.; Lemaire, J.; Maitre, P. Mid-IR spectroscopy of protonated leucine methyl ester performed with an FTICR or a Paul type ion-trap. Int. J. Mass Spectrom. 2006, 249, 14−20. (40) Bakker, J. M.; Sinha, R. K.; Besson, T.; Brugnara, M.; Tosi, P.; Salpin, J. Y.; Maitre, P. Tautomerism of uracil probed via infrared spectroscopy of singly hydrated protonated uracil. J. Phys. Chem. A 2008, 112, 12393−12400. (41) Bakker, J. M.; Salpin, J. Y.; Maitre, P. Tautomerism of cytosine probed by gas phase IR spectroscopy. Int. J. Mass Spectrom. 2009, 283, 214−221. (42) Bythell, B. J.; Hernandez, O.; Steinmetz, V.; Paizs, B.; Maitre, P. Tyrosine side-chain catalyzed proton transfer in the YG a2 ion revealed by theory and IR spectroscopy in the ’fingerprint’ and X-H (X = C, N, O) stretching regions. Int. J. Mass Spectrom. 2012, 316, 227−234. (43) Cerda, B. A.; Wesdemiotis, C. Thermochemistry and structures of Na+ coordinated mono- and disaccharide stereoisomers. Int. J. Mass Spectrom. 1999, 189, 189−204. (44) Li, H. L.; Bendiak, B.; Siems, W. F.; Gang, D. R.; Hill, H. H. Ion mobility mass spectrometry analysis of isomeric disaccharide precursor, product and cluster ions. Rapid Commun. Mass Spectrom. 2013, 27, 2699−2709. (45) Gronert, S. Estimation of effective ion temperatures in a quadrupole ion trap. J. Am. Soc. Mass. Spectrom. 1998, 9, 845−848. (46) Rao, J. S.; Dinadayalane, T. C.; Leszczynski, J.; Sastry, G. N. Comprehensive study on the solvation of mono- and divalent metal cations: Li+, Na+, K+, Be2+, Mg2+ and Ca2+. J. Phys. Chem. A 2008, 112, 12944−12953. (47) Hall, A. B.; Coy, S. L.; Kafle, A.; Glick, J.; Nazarov, E.; Vouros, P. Extending the dynamic range of the ion trap by differential mobility filtration. J. Am. Soc. Mass. Spectrom. 2013, 24, 1428−1436. (48) Walters, R. S.; Pillai, E. D.; Duncan, M. A. Solvation dynamics in Ni+(H2O)n clusters probed with infrared spectroscopy. J. Am. Chem. Soc. 2005, 127, 16599−16610. (49) Heaton, A. L.; Armentrout, P. B. Experimental and theoretical studies of sodium cation interactions with D-Arabinose, Xylose, Glucose, and Galactose. J. Phys. Chem. A 2008, 112, 10156−10167. (50) Rodgers, M. T.; Armentrout, P. B. Collision-induced dissociation measurements on Li+(H2O)n, n=1−6: The first direct measurement of the Li+-OH2 bond energy. J. Phys. Chem. A 1997, 101, 1238−1249.

6064

DOI: 10.1021/jp511975f J. Phys. Chem. A 2015, 119, 6057−6064