Advertisement

Identification and Partial Structural Characterization of Mass Isolated Valsartan and Its Metabolite with Messenger Tagging Vibrational Spectroscopy

  • Olga Gorlova
  • Sean M. Colvin
  • Antonio Brathwaite
  • Fabian S. Menges
  • Stephanie M. Craig
  • Scott J. Miller
  • Mark A. Johnson
Research Article

Abstract

Recent advances in the coupling of vibrational spectroscopy with mass spectrometry create new opportunities for the structural characterization of metabolites with great sensitivity. Previous studies have demonstrated this scheme on 300 K ions using very high power free electron lasers in the fingerprint region of the infrared. Here we extend the scope of this approach to a single investigator scale as well as extend the spectral range to include the OH stretching fundamentals. This is accomplished by detecting the IR absorptions in a linear action regime by photodissociation of weakly bound N2 molecules, which are attached to the target ions in a cryogenically cooled, rf ion trap. We consider the specific case of the widely used drug Valsartan and two isomeric forms of its metabolite. Advantages and challenges of the cold ion approach are discussed, including disentangling the role of conformers and the strategic choices involved in the selection of the charging mechanism that optimize spectral differentiation among candidate structural isomers. In this case, the Na+ complexes are observed to yield sharp resonances in the high frequency NH and OH stretching regions, which can be used to easily differentiate between two isomers of the metabolite.

Graphical Abstract

Keywords

Vibrational spectroscopy Mass spectrometry IR-spectroscopy Metabolomics Metabolite Drug discovery Conformer differentiation 

Introduction

Metabolomics is a rapidly expanding field that focuses on the identification of the biological degradation products of small molecules [1]. Ultrasensitive detection of and differentiation between these metabolites is, in fact, an expensive bottleneck in the development of new drugs, and to this end mass spectrometry has played an important role [2]. Typically, such analyses are carried out after an initial separation of crude extracts with liquid chromatography (LC), followed by two-stage mass spectrometry (MS/MS) in order to identify the metabolites by their fragmentation pattern upon collision-induced dissociation (CID) [3, 4, 5]. Because the latter step requires an extensive library of standards [6, 7, 8], it would be useful to develop complementary strategies that can directly provide structural information about the ions [7, 9]. A promising method on the horizon that is capable of providing such information characterizes mass-selected ions with vibrational spectroscopy, carried out in an action mode by resonant IR photodissociation [9, 10]. This provides a spectral axis for secondary analysis of ions [10] that encodes the local interactions of various functional groups in the target species. The first report of this application in the context of metabolomics appeared in the past year [10], where vibrational spectra were obtained for several prototypical ions over the fingerprint region (200–1800 cm−1) using a very powerful free electron laser (FELIX in this case, located in Nijmegen, The Netherlands) to dissociate covalently bound, room temperature ions by sequential absorption of many (n = 10–20) IR photons using IRMPD (infrared multiple photon dissociation):
$$ \mathrm{A}{\mathrm{B}}^{+}+\mathrm{n} hv\to \mathrm{A}+{\mathrm{B}}^{+} $$
(1)
While this clearly represents a breakthrough in the field, here we describe the results obtained with a variation of this approach achieved by adding a cryogenic ion processing stage in the low kinetic energy region of the ion transport optics to allow attachment of weakly bound, relatively inert atoms or molecules (typically rare gases, H2 or N2). This method has the advantages that the ions are cooled close to their vibrational zero-point levels and that the recorded vibrational spectra are obtained in a linear action mode (i.e., dissociation with a single IR photon). Together, these features are the so-called “messenger tagging” approach [11]:
$$ {\mathrm{AB}}^{+}+\mathrm{RG}\kern0.3em \to \kern0.3em {\mathrm{AB}}^{+}\cdot \mathrm{RG}\kern0.3em \hspace{0ex}\left(\mathrm{RG}=\mathrm{rare}\kern0.3em \mathrm{gas}\kern0.3em \mathrm{or}\kern0.3em \mathrm{molecules}\kern0.3em \left({\mathrm{H}}_2,\kern0.3em {\mathrm{N}}_2\right)\right) $$
(2)
$$ {\mathrm{AB}}^{+}\cdotp \mathrm{RG}+\mathrm{h}\upnu \left(600\hbox{-} 4000{\mathrm{cm}}^{\hbox{-} 1}\right)\to {\mathrm{AB}}^{+}+\mathrm{RG} $$
(3)

The practical consequence of this procedure is that spectra obtained in the laboratory can be directly compared with theoretical predictions for candidate structures, which are routinely available with electronic structure calculations of the linear absorption spectra of their equilibrium geometries [12].

To illustrate how this approach can be applied to drug discovery, we consider the case of the well-known pharmaceutical Valsartan. Valsartan (I in Figure 1) is metabolized by P450 CYP2C9 hydroxylation to produce 4-hydroxy valsartan (II in Figure 1), which is the dominant metabolite [13]. Its structure has been established [13] via standard analytical techniques, and features oxidation of the C4 atom of the alkyl chain on the valeroyl group. Our goal here is to establish the inherent performance of vibrational spectroscopy to differentiate between the structures that occur in metabolic degradation, not to survey the actual limits of its sensitivity in a practical application. We note that there is a very recent, similar investigation by the FELIX group on the application of IRMPD spectroscopy to atorvastatin metabolites [14].
Figure 1

Valsartan (I) and its naturally occurring metabolite (II) by hydroxylation of the valeroyl tail at the C4 position. The structural isomer III was synthetically prepared for comparison with the spectroscopic behavior of II

We first evaluate the performance of messenger tagging IR spectroscopy by characterizing three commonly available modes for electrical charging of nominally neutral molecules in electrospray ionization (ESI) mass spectrometry: protonation, deprotonation, and complexation with Na+. We then demonstrate how structural details can be inferred from cold ion vibrational spectra when combined with conformer-selective, IR-IR double resonance spectroscopy as well as with site-specific isotopic substitution. We conclude by highlighting the ability of messenger tagging vibrational spectroscopy to reveal variations in the NH and OH stretching regions of the key hydroxyl groups in the natural metabolite II, as well as the synthetic, isomeric compound III, to test the efficacy of the method in the differentiation of structurally similar isomers.

Experimental

The samples of interest were purchased or synthesized [see Supplementary Information (SI) for details], and then dissolved in 50:50 methanol:water solutions, with traces of sodium originating from the glassware. Neat ~1 mM solutions were then electrosprayed into the photofragmentation mass spectrometer at Yale (Figure 2), which has been described in detail previously [15, 16]. The generated ions were guided through four differentially pumped vacuum stages using two rf-only quadrupole guides and an octopole guide before they entered the 90° turning quadrupole to isolate the charged species from the remaining neutral molecules. The turned ions then entered another octopole guide before arriving into the quadrupole ion trap (Jordan TOF Products, Inc., Grass Valley, CA, USA) where they were collisionally cooled by a He buffer gas containing 10% N2. The buffer gas mixture was passed through a liquid N2 trap to reduce impurities in the gas manifold. The cold ion trap was attached to the second stage of a closed-cycle helium cryostat (Sumitomo, Woburn, MA, USA, 1.5 W at 4.2 K) which, using a 100 W resistive heater, allowed for temperature control down to ~40 ± 5 K and N2 adduct condensation onto the ions. The cooled ions were then ejected into the extraction region of a Wiley-McLaren time-of-flight (TOF) mass spectrometer, after which the mass-selected and N2-tagged ions of interest were intersected with the output of a pulsed, tunable OPO/OPA infrared laser (LaserVision, Bellevue, WA, USA) pumped by a Nd:YAG source (Continuum Surelite EX, San Jose, CA, USA, 10 Hz, 7 ns). At a resonant transition of the tagged-ion of interest, the absorbed energy was redistributed and resulted in evaporation of the weakly bound N2 tag molecule. The vibrational predissociation spectra reported here were obtained by monitoring the generation of the photofragments as a function of laser frequency. All of the spectra were recorded in a single photon absorption regime, which was verified by ensuring that the energy of the laser relates linearly to the photo-evaporation yield. Spectra were then normalized by laser energy to account for fluctuations in laser power over the scan range.
Figure 2

Present layout of the Yale instrument. This custom-built, mass-selective apparatus provides a variety of spectroscopic capabilities for structural determination of cold, isomer-selected gas-phase ions

For comparison, mass spectra (MS1) and collision induced dissociation (CID) mass spectra (MS2) were obtained using the linear ion trap of a Thermo Scientific LTQ Orbitrap XL Hybrid FT Mass Spectrometer (Waltham, MA, USA). Sample solutions of II and III in 50:50 methanol:water (same as those used for the spectroscopic measurements) were electrosprayed with the needle voltage held at ~4 kV. Ions were then guided to the ion trap, mass selected, and fragmented using He as a collision gas. The mass spectrometer was calibrated with Pierce LTQ ESI Positive Ion Calibration Solution, Thermo Fisher Scientific Inc.

Results and Discussion

Dependence of Vibrational Patterns on the Charging Method

One issue that immediately arises in the application of vibrational spectroscopy for structure elucidation is the choice of a charging scheme. In particular, messenger tagging vibrational spectroscopy is most effective if the key distinguishing features appear as sharp, well separated vibrational bands. Since the primary ring scaffolds are identical for oxidation products II and III, it would be valuable if the OH stretches of the hydroxyl groups in the two positions (C4 versus C5 of the valeroyl alkyl tail) could be used directly as reporters for compound identification. Such bands are often very diffuse in the condensed phase because of local perturbations arising from thermal fluctuations in the surrounding environment. In the analysis of the vibrational predissociation spectra, on the other hand, these features are often sharp and appear at frequencies that reflect the minimum energy conformations of the cold, isolated ions [12, 17]. In particular, the isomer-dependence of folded configurations with intra-molecular H-bonds promises to provide a natural mechanism to enhance differences in otherwise similar OH groups [18]. These conformations will, in turn, be driven by the nature of the charging mechanism essential for the application of messenger tagging vibrational spectroscopy. As such, we first survey the spectra of parent molecule I when charged by the typical methods: protonation, deprotonation, and sodiation.

Figure 3 presents the vibrational predissociation spectra of four easily prepared ions derived from the N2-tagged Valsartan molecule I. Note that the sharp bands in the free OH and NH regions above 3400 cm−1 (denoted gold) indeed depend strongly on the charging motif. For example, the neutral molecule I should yield one NH and one OH stretching fundamental, but only the I + Na+ spectrum (Figure 3b) displays two sharp bands in the region above 3400 cm−1 where these fundamentals are expected to occur (see Figure S1). The I + H+ spectrum (Figure 3c), on the other hand, exhibits only a single feature in this range despite the fact that there should be an additional OH or NH group in this cation (two are expected for the neutral), depending on the protonation site [17]. A similar situation occurs in the spectrum of N2-tagged, deprotonated I (I – H+, Figure 3d), which should still retain either the NH or OH functionality, but now displays no sharp bands above 3100 cm−1. Rather, diffuse bands appear near 3000 cm−1 in the region nominally associated with the CH stretching fundamentals. In fact, the I – H+ spectrum is similar to the that of the N2-tagged anion formed by removal of two protons and complexation with Na+ to yield the singly charged anion (I – 2H+ + Na+, Figure 3e). Note that the I + H+, I – H+, and I – 2H+ + Na+ complexes all display strongly red-shifted features in the lower energy region near the C=O stretching fundamentals, suggesting the formation of cyclic, intramolecular H-bonds, which are known to dramatically red-shift the NH and OH stretching fundamentals [15, 19, 20, 21, 22]. In the case of I + Na+, however, we observe two sharp features at the expected locations along with a third broad band appearing between these two stronger transitions, indicating that the Na+ ion acts to break up these intramolecular hydrogen bonding interactions.
Figure 3

Vibrational predissociation spectra of N2-tagged I with various charging schemes: (a) FTIR of I in methanol, (b) I + Na+ (sodiated), (c) I + H+ (protonated), (d) I –H+ (deprotonated), and (e) I – 2H+ + Na+ (doubly-deprotonated and sodiated). Dagger in (b) denotes unexpected band described in detail in the text, v CO COOH and v CO amide in (d) and (e) indicate the locations of amide I and carboxylic carbonyl stretches, determined using site-specific isotopic labeling (see Figure S4 in Supplementary Material)

Remarks on Strategies for Specific Conformer and Band Assignments

We emphasize that our purpose here is not to determine the detailed ion-induced conformations from their spectra. Indeed, the conformationally flexible scaffold of this system, combined with the spectroscopic complexities that arise from ion-driven, intramolecular H-bonds (vide infra) often present profound challenges for assignments based on harmonic predictions. Even within this more limited scope, however, it is useful to determine the spectral pattern of all conformers that are generated in the ion source. We then demonstrate how we can experimentally identify features arising from the various functional groups, and then use this information to characterize the intramolecular interactions driving the observed structures.

To determine whether multiple conformers contribute to the I + Na+ spectrum in Figure 3b, we exploit the photochemical hole burning method [18, 23, 24, 25, 26]. This involves an IR-IR double resonance scheme carried out with two tunable IR lasers and three stages of mass selection (IR2MS3). The first MS stage isolates a parent mass for interaction with a scanning IR laser (the pump), and the second stage of MS separates the undissociated, tagged parent mass from the fragments generated by resonant excitation at the first laser crossing. A second IR laser (the probe) then interrogates the undissociated, tagged parent ion, and photofragments from this excitation are isolated with the third MS stage. To acquire a conformer-specific spectrum, the probe laser is fixed at a particular resonance, and the pump laser is scanned through the entire spectrum. In this way, whenever the pump laser excites a resonance that is also associated with the transition probed by the fixed probe laser, the probe fragment ion signal is decreased, and the complete spectrum associated with that species appears as dips in the probe laser signal because of the depletion in the population of this conformer by the scanning pump laser.

Figure 4 presents the results of an IR2MS3 study of the N2-tagged I + Na+ ion. The top trace is the linear action spectrum from this mass-selected ion packet, whereas the three lower traces show the dip spectra associated with fixing the probe on three transitions in the 1600–1800 cm−1 region. The fact that the dip spectra are different immediately confirms that at least two conformers are present. Interestingly, probing at the typical strong bands (blue and green) near the positions of non-interacting amide and carboxylic acid carbonyls (blue and green arrows Figure 4a) yields dip features that are quite close to non-interacting NH and OH stretches (arrows in Figure 4a). Probing the interloper (red band (†) at ~1725 cm−1 in Figure 4a), however, yields the dip pattern displayed in Figure 4c. The free OH expected for a non-interacting acid is completely missing, whereas the band near the free NH stretch remains intact. This suggests the presence of a conformer with an intramolecular H-bonding interaction that is compromising the behavior of the acid functionality.
Figure 4

IR–IR double resonance vibrational predissociation spectra of I + Na+·N2. Spectral regions of key bond-specific transitions of I + Na+ are reproduced in trace a. Dip spectra (b-d) arising from the probed transitions correlate absorptions in the fingerprint region to bands in the OH/NH region originating from two different isomers; ‡ indicates an isomer specific feature in the NH/OH region. Regions assigned above the spectra are from SarSarH+·D2 and ImH+·Ar [18, 22]

Because the interloper (†) falls between the previously observed amide I and free acid C=O stretches, it is useful to establish which functionality is responsible for this feature. This can be accomplished by synthetically incorporating a heavy isotope at key positions on the scaffold likely to be involved in H-bonding (see SI for synthetic routes). Then, by taking the difference between the normalized spectrum of the labeled species and the unlabeled one, bands associated with the labeled site appear as derivative-like line shapes while common features are suppressed [24]. Application of this method to the I + Na+ complex is presented in Figure 5. Labeling the amide carbonyl (blue) with 13C yields the spectrum displayed in Figure 5b, with the difference spectrum shown in the box to the right (trace 5d). Only one transition is affected by this substitution, establishing that the band at 1627 cm−1 is uniquely assigned to the amide C=O stretch.
Figure 5

Isotopic substitution for the assignment of C=O related bands. N2-predissociation spectra of I + Na+ (a) and its isotopomers with 13C-labeled amide carbonyl (b) in blue and carboxylic carbonyl (c) in green per colored schematic on the left. Subtraction spectra (traces d and e) for assignment clarification are shown in the inset to the right

The acid group is more complicated, however, as is evident in trace 5c. 13C labeling at the central carbon atom of the carboxylate group displaces two bands: the sharp 1777 cm−1 feature at the expected location for a free acid functionality (v CO COOH , green), and the interloper (†, red) that was identified in the IR-IR hole burning study as arising from a second conformer. Thus, the acid group is unambiguously identified as the carrier of the † feature, which is red-shifted from that arising from a free acid by about 50 cm−1 and thus denoted v CO COOH bound . We note that a similar situation was observed earlier in the case of the SarSarH+ dipeptide spectrum [18], where an intramolecular H-bond to the acid carbonyl group in one of its conformers displayed a significant (24 cm−1) red-shift in the C=O stretching frequency.

Although the conformer with the strong intramolecular H-bond (Ib + Na+) presents a formidable challenge for structural assignment by comparison with spectra computed at the harmonic level [27, 28], we have found a compelling match for the more open Ia conformer, with the structure and calculated spectrum reported in Figure S3. The general spectral trends (free NH and red-shifted carboxylic acid carbonyl) of the conformer Ib + Na+ are also recovered by the calculated structure of another isomer reported in Figure S3. The two intramolecular binding motifs at play are highlighted in Figure 6. In particular, structure Ia + Na+ on the left of Figure 6 features the Na+ ion binding the amide C=O group to the tetrazole ring, leaving the acid group free, thus accounting for the free OH stretch in its spectrum (Figure 4b). On the other hand, the spectrum of the Ib + Na+ (Figure 4c) does not display a free OH stretching band and has a strongly red-shifted C=O group on the acid. These features are consistent with the expected behavior of the minimum energy structure Ib + Na+ at the right of Figure 6. In this arrangement, the Na+ ion is bound in a bidentate fashion to the carboxylate group such that the acidic proton is strongly H-bonded to the nearby amide carbonyl. This arrangement is based on a partially formed, metal ion-induced zwitterion, a scenario commonly encountered in sodiated peptides [29]. The diffuse band near 2450 cm−1 (highlighted red in Figure 6) is then a likely candidate for the acidic OH stretch, which is often strongly anharmonic and coupled with background states as discussed extensively in the case of protonated peptides [20]. Because only one transition is associated with the amide I C=O group, we conclude that this feature is common to both isomers, presumably because in both cases it is proximal to a positive charge center (H-bond to an OH group and complexation with Na+). It is precisely because of such ambiguities that we are not attempting a detailed structural analysis based on comparison with harmonic calculations.
Figure 6

The vibrational predissociation spectrum of I + Na+·N2 (2300 to 2700 cm−1) labeled with bond-specific spectral signatures. This range displays the N2 tag stretch (pink) and a broad feature around 2450 cm−1, which is often attributed to an intramolecular H-bond. Conformer-specific spectral signatures are labeled according to the two plausible binding motifs Ia and Ib, whose molecular structures are optimized at the B97D3/6-31++g(d) level of theory. Ib + Na+ exhibits a wire-like binding interaction for the bands circled in red. Previously observed experimental assignments are represented as dotted lines from Refs [18, 22]

Metabolite Identification through the OH Stretching Bands in the Vibrational Spectra of the Cold Na+ Complexes

To address the hypothetical issue where it is important to differentiate between structural isomers (in addition to the natural metabolite), we synthesized III with the hydroxyl group at the terminal C5 position (structure indicated on the lower right in Figure 1, with synthetic pathway detailed in SI). We note that these two compounds would present a challenging case for the widely used MS2 approach for metabolite identification [30], since they yield identical fragmentation patterns as illustrated in Figure S2.

Figure 7 presents the vibrational spectra of the II + Na+ and III + Na+ complexes in traces 7b and 7c, respectively, along with that of the I + Na+ reported above in trace 7d. Note that the signature bands of the parent scaffold (the C=O stretches in blue at low energy and the sharp NH and OH stretches in gold and green respectively at higher energy) are maintained in the spectra of both II + Na+ and III + Na+. Interestingly, the red shifted acidic carbonyl (†) is completely absent in both hydroxylated derivatives. In fact, the bands associated with conformer Ia + Na+ (lower left in Figure 6, green) of the parent are essentially intact, suggesting that both derivatives feature a similar complexation motif with the Na+ ion. This allows assignment of the remaining sharp bands (purple) in the II + Na+ and III + Na+ to the new OH groups on the alkyl chain, which are indeed well differentiable, and thus provide definitive spectral signatures of the two structural isomers. It is important to note that this distinction results from the ability to measure the vibrational spectra in isolation at low temperature. To emphasize this point, trace 7a includes the FTIR spectrum of II in the condensed phase at 300 K, where the NH and OH stretches blend together in a diffuse band centered at 3500 cm−1.
Figure 7

(a) FTIR of II; and N2 predissociation spectra of (b) II + Na+, (c) III + Na+, and (d) I + Na+. Band-specific spectral regions are labeled above trace (a) with acid C=O referring to carboxylic acid, amide I to carbonyl excitations of the amide group, and amide II to primarily to the C–N stretches of the amide group. The purple colored peaks [* in (c) and dotted to (b)] are spectral signatures of the respective hydroxyl moieties of the natural metabolite and its synthetic analogue, respectively; § in (b) and (c) indicates bands that are specific to the metabolite and its structural isomer

The preservation of the type Ia conformer motif in the hydroxylated derivatives is interesting in light of its postulated structure. Retention of the free carboxylic OH and associated carbonyl stretch indicates that the conformer Ia + Na+ binding motif of the acid to the metal remains intact. On the other hand, there is a small (15 cm−1) blue shift in the amide I band (blue). A plausible assembly motif that accounts for these features is presented in Figure 8. The proximity of the OH on the alkyl chain to that carbonyl would then account for the ~160 cm−1 red shift of the OH stretch in II versus III.
Figure 8

Plausible assemblies of Ia + Na+, Ib + Na+, II + Na+, and III + Na+. Functional groups are highlighted according to spectral signatures in Figure 7. Molecular structures were geometrically optimized at the B97D3/6-31++g(d) level of theory. The harmonic spectra corresponding to these structures are presented in Figure S3

The spectroscopic behavior of the hydroxylated derivatives emphasizes the fact that the differentiation arises from the local folded forms of conformers at low temperature. This also creates interesting possibilities for enhanced discrimination techniques where a strategic guest molecule is docked to the target species in a manner that optimizes host–guest interactions for a specific isomer. This yields an IR variant of molecular recognition methods that have been demonstrated in a mass spectrometric venue for chiral recognition [31, 32, 33].

Summary

We illustrated how messenger tagging vibrational spectroscopy can be used to differentiate metabolites in small molecule drug discovery within the context of structural mass spectrometry. The importance of the charging mechanism was explored by recording the vibrational spectra of protonated, deprotonated, and sodiated ions, with the result that only the sodiated ions yield clear bands associated with the NH, OH, and CO functionalities, whereas the other charging schemes yield intermolecular H-bonds, which complicate and obscure these features. The sodiated parent ion of Valsartan was observed in two conformations, which yield very different characteristic absorptions for the functional groups. These patterns were identified using conformer-selective IR-IR double resonance spectroscopy combined with site-specific isotopic substitution, and assigned to different classes of metal coordination and intramolecular H-bonding. We then obtained the vibrational spectra of the sodiated Valsartan metabolite and a synthetic isomer, which differ only in the addition of a hydroxyl group at either the C4 or C5 position of the valeroyl alkyl chain. The spectra of both sodiated derivatives were simpler than that of the parent as they exhibit only one of the conformers adopted by the sodiated parent. The sharp, free OH transitions of the hydroxyl groups are separated by over 160 cm−1, thus enabling a clear differentiation of the derivatives by coupling vibrational spectroscopy with the sensitivity of mass spectrometry. The large shift between the two fundamentals is discussed in terms of their likely intramolecular H-bonding interactions, thus highlighting the importance of quenching the internal energy of these conformationally flexible systems prior to analysis with vibrational spectroscopy. From these results, we suggest the potential role of neutral complex formation with strategically chosen partners based on molecular recognition as a way to further refine the selectivity afforded by cryogenic ion chemistry and spectroscopy.

Notes

Acknowledgements

M.A.J. thanks the National Science Foundation for support under grant number CHE-1465100. This work was supported in part by the Yale University Faculty of Arts and Sciences High Performance Computing Facility (and staff). A.B. acknowledges financial support from the National Science Foundation through the HBCU-UP award no. 1505095. We also thank Nan Yang and Chinh Duong for their work on the updated instrumental capabilities utilized in this experiment (Figure 2).

Supplementary material

13361_2017_1767_MOESM1_ESM.pdf (10.2 mb)
ESM 1 (PDF 10398 kb)

References

  1. 1.
    Fessenden, M.: Metabolomics: small molecules, single cells. Nature 540, 153–155 (2016)CrossRefGoogle Scholar
  2. 2.
    Zhang, J., Gonzalez, E., Hestilow, T., Haskins, W., Huang, Y.: Review of peak detection algorithms in liquid-chromatography-mass spectrometry. Curr. Genom. 10, 388–401 (2009)CrossRefGoogle Scholar
  3. 3.
    Wang, Y., Liu, S., Hu, Y., Li, P., Wan, J.-B.: Current state of the art of mass spectrometry-based metabolomics studies - a review focusing on wide coverage, high throughput, and easy identification. RSC Adv. 5, 78728–78737 (2015)CrossRefGoogle Scholar
  4. 4.
    Halket, J.M., Waterman, D., Przyborowska, A.M., Patel, R.K.P., Fraser, P.D., Bramley, P.M.: Chemical derivatization and mass spectral libraries in metabolic profiling by GC/MS and LC/MS/MS. J. Exp. Bot. 56, 219–243 (2005)CrossRefGoogle Scholar
  5. 5.
    Xu, Y., Heilier, J.-F., Madalinski, G., Genin, E., Ezan, E., Tabet, J.-C., Junot, C.: Evaluation of accurate mass and relative isotopic abundance measurements in the LTQ-Orbitrap mass spectrometer for further metabolomics database building. Anal. Chem. 82, 5490–5501 (2010)CrossRefGoogle Scholar
  6. 6.
    Vaniya, A., Fiehn, O.: Using fragmentation trees and mass spectral trees for identifying unknown compounds in metabolomics. Trend Anal. Chem. : TRAC. 69, 52–61 (2015)CrossRefGoogle Scholar
  7. 7.
    Neumann, S., Böcker, S.: Computational mass spectrometry for metabolomics: Identification of metabolites and small molecules. Anal. Bioanal. Chem. 398, 2779–2788 (2010)CrossRefGoogle Scholar
  8. 8.
    Scheubert, K., Hufsky, F., Böcker, S.: Computational mass spectrometry for small molecules. J. Cheminf. 5, 12–12 (2013)CrossRefGoogle Scholar
  9. 9.
    Cismesia, A.P., Bailey, L.S., Bell, M.R., Tesler, L.F., Polfer, N.C.: Making mass spectrometry see the light: the promises and challenges of cryogenic infrared ion spectroscopy as a bioanalytical technique. J. Am. Soc. Mass Spectrom. 27, 757–766 (2016)CrossRefGoogle Scholar
  10. 10.
    Martens, J., Grzetic, J., Berden, G., Oomens, J.: Structural identification of electron transfer dissociation products in mass spectrometry using infrared ion spectroscopy. Nat. Commun. 7, 11754 (2016)CrossRefGoogle Scholar
  11. 11.
    Okumura, M., Yeh, L.I., Lee, Y.T.: The vibrational predissociation spectroscopy of hydrogen cluster ions. J. Chem. Phys. 83, 3705–3706 (1985)CrossRefGoogle Scholar
  12. 12.
    Wolk, A.B., Leavitt, C.M., Garand, E., Johnson, M.A.: Cryogenic ion chemistry and spectroscopy. Acc. Chem. Res. 47, 202–210 (2014)CrossRefGoogle Scholar
  13. 13.
    Nakashima, A., Kawashita, H., Masuda, N., Saxer, C., Niina, M., Nagae, Y., Iwasaki, K.: Identification of cytochrome P450 forms involved in the 4-hydroxylation of valsartan, a potent and specific angiotensin II receptor antagonist, in human liver microsomes. Xenobiotica 35, 589–602 (2005)CrossRefGoogle Scholar
  14. 14.
    Martens, J., Koppen, V., Berden, G., Cuyckens, F., Oomens, J.: Combined liquid chromatography-infrared ion spectroscopy for identification of regioisomeric drug metabolites. Anal. Chem. 89, 4359–4362 (2017)CrossRefGoogle Scholar
  15. 15.
    Kamrath, M.Z., Relph, R.A., Guasco, T.L., Leavitt, C.M., Johnson, M.A.: Vibrational predissociation spectroscopy of the H2-tagged mono- and dicarboxylate anions of dodecanedioic acid. Int. J. Mass Spectrom. 300, 91–98 (2011)CrossRefGoogle Scholar
  16. 16.
    Robertson, W.H., Kelley, J.A., Johnson, M.A.: A pulsed supersonic entrainment reactor for the rational preparation of cold ionic complexes. Rev. Sci. Instrum. 71, 4431–4434 (2000)CrossRefGoogle Scholar
  17. 17.
    Fournier, J.A., Wolk, A.B., Johnson, M.A.: Integration of cryogenic ion vibrational predissociation spectroscopy with a mass spectrometric interface to an electrochemical cell. Anal. Chem. 85, 7339–7344 (2013)CrossRefGoogle Scholar
  18. 18.
    Leavitt, C.M., Wolk, A.B., Fournier, J.A., Kamrath, M.Z., Garand, E., Van Stipdonk, M.J., Johnson, M.A.: Isomer-specific IR-IR double resonance spectroscopy of D2-tagged protonated dipeptides prepared in a cryogenic ion trap. J. Phys. Chem. Lett. 3, 1099–1105 (2012)Google Scholar
  19. 19.
    DeBlase, A.F., Scerba, M.T., Lectka, T., Johnson, M.A.: Vibrational predissociation spectroscopy of Ar-tagged, trisubstituted silyl cations. Chem. Phys. Lett. 568, 9–13 (2013)CrossRefGoogle Scholar
  20. 20.
    Leavitt, C.M., DeBlase, A.F., van Stipdonk, M., McCoy, A.B., Johnson, M.A.: Hiding in plain sight: unmasking the diffuse spectral signatures of the protonated N-terminus in simple peptides. J. Phys. Chem. Lett. 4, 3450–3457 (2013)CrossRefGoogle Scholar
  21. 21.
    DeBlase, A.F., Kass, S.R., Johnson, M.A.: On the character of the cyclic ionic H-bond in cryogenically cooled deprotonated cysteine. Phys. Chem. Chem. Phys. 16, 4569–4575 (2014)CrossRefGoogle Scholar
  22. 22.
    Gerardi, H.K., Gardenier, G.H., Viswanathan, U., Auerbach, S.M., Johnson, M.A.: Vibrational predissociation spectroscopy and theory of Ar-tagged, protonated Imidazole (Im) Im1-3H+·Ar clusters. Chem. Phys. Lett. 501, 172–178 (2011)CrossRefGoogle Scholar
  23. 23.
    Myshakin, E.M., Jordan, K.D., Sibert, E.L., Johnson, M.A.: Large anharmonic effects in the infrared spectra of the symmetrical CH3NO2 -·(H2O) and CH3CO2 -·(H2O) complexes. J. Chem. Phys. 119, 10138–10145 (2003)Google Scholar
  24. 24.
    Garand, E., Kamrath, M.Z., Jordan, P.A., Wolk, A.B., Leavitt, C.M., McCoy, A.B., Miller, S.J., Johnson, M.A.: Determination of noncovalent docking by infrared spectroscopy of cold gas-phase complexes. Science 335, 694–698 (2012)CrossRefGoogle Scholar
  25. 25.
    Guasco, T.L., Elliott, B.M., Johnson, M.A., Ding, J., Jordan, K.D.: Isolating the spectral signatures of individual sites in water networks using vibrational double-resonance spectroscopy of cluster isotopomers. J. Phys. Chem. Lett. 1, 2396–2401 (2010)CrossRefGoogle Scholar
  26. 26.
    Guasco, T.L., Johnson, M.A., McCoy, A.B.: Unraveling anharmonic effects in the vibrational predissociation spectra of H5O2 + and its deuterated analogues. J. Phys. Chem. A 115, 5847–5858 (2011)CrossRefGoogle Scholar
  27. 27.
    DeBlase, A.F., Dziekonski, E.T., Hopkins, J.R., Burke, N.L., Sheng, H., Kenttämaa, H.I., McLuckey, S.A., Zwier, T.S.: Alkali cation chelation in cold β-O-4 tetralignol complexes. J. Phys. Chem. A 120, 7152–7166 (2016)CrossRefGoogle Scholar
  28. 28.
    Gord, J.R., Hewett, D.M., Hernandez-Castillo, A.O., Blodgett, K.N., Rotondaro, M.C., Varuolo, A., Kubasik, M.A., Zwier, T.S.: Conformation-specific spectroscopy of capped, gas-phase Aib oligomers: tests of the Aib residue as a 310-helix former. Phys. Chem. Chem. Phys. 18, 25512–25527 (2016)CrossRefGoogle Scholar
  29. 29.
    Bush, M.F., Forbes, M.W., Jockusch, R.A., Oomens, J., Polfer, N.C., Saykally, R.J., Williams, E.R.: Infrared spectroscopy of cationized lysine and ε-N-methyllysine in the gas phase: effects of alkali-metal ion size and proton affinity on zwitterion stability. J. Phys. Chem. A 111, 7753–7760 (2007)CrossRefGoogle Scholar
  30. 30.
    Huan, T., Tang, C., Li, R., Shi, Y., Lin, G., Li, L.: MyCompoundID MS/MS search: metabolite identification using a library of predicted fragment-ion-spectra of 383,830 possible human metabolites. Anal. Chem. 87, 10619–10626 (2015)Google Scholar
  31. 31.
    Awad, H., El-Aneed, A.: Enantioselectivity of mass spectrometry: challenges and promises. Mass Spectrom. Rev. 32, 466–483 (2013)Google Scholar
  32. 32.
    Fujihara, A., Maeda, N., Doan, T.N., Hayakawa, S.: Enantiomeric excess determination for monosaccharides using chiral transmission to cold gas-phase tryptophan in ultraviolet photodissociation. J. Am. Soc. Mass Spectrom. 28, 224–228 (2017)CrossRefGoogle Scholar
  33. 33.
    Gaye, M.M., Nagy, G., Clemmer, D.E., Pohl, N.L.B.: Multidimensional analysis of 16 glucose isomers by ion mobility spectrometry. Anal. Chem. 88, 2335–2344 (2016)CrossRefGoogle Scholar

Copyright information

© American Society for Mass Spectrometry 2017

Authors and Affiliations

  • Olga Gorlova
    • 1
  • Sean M. Colvin
    • 1
  • Antonio Brathwaite
    • 2
  • Fabian S. Menges
    • 1
  • Stephanie M. Craig
    • 1
  • Scott J. Miller
    • 1
  • Mark A. Johnson
    • 1
  1. 1.Department of ChemistryYale UniversityNew HavenUSA
  2. 2.College of Science and MathematicsUniversity of the Virgin IslandsSt. ThomasVirgin Islands (U.S.)

Personalised recommendations