Ion Mobility Spectrometry Coupled to Laser-Induced Fluorescence

Nov 30, 2012 - simple and robust. It allows one to detect fluorescence from specified ions, including clusters, which would not survive in a mass spec...
0 downloads 0 Views 1MB Size
Download PDF
Technical Note pubs.acs.org/ac

Ion Mobility Spectrometry Coupled to Laser-Induced Fluorescence Vladimir Frankevich, Pablo Martinez-Lozano Sinues, Konstantin Barylyuk, and Renato Zenobi* Department of Chemistry and Applied Biosciences, ETH Zurich, Zurich, Switzerland ABSTRACT: We report on interfacing a differential mobility analyzer (DMA) with laser-induced fluorescence (LIF) to simultaneously retrieve two-dimensional information on the electrical mobility and fluorescence spectroscopy of gas-phase ions. The fact that the separation of ions within DMA takes place in space rather than in time allows for the continuous selection of ion beams within a narrow range of mobilities that are further analyzed by LIF. Combination of DMA with LIF is simple and robust. It allows one to detect fluorescence from specified ions, including clusters, which would not survive in a mass spectrometer. Complex mixtures of fluorescent compounds can be separated by the DMA and studied by LIF. LIF is a sensitive technique and useful in the study of molecular interactions. DMA with LIF detection can be used for studies of gas-phase fluorescence of small molecules such as different dyes and their conjugates. This unique instrument combination may also provide a powerful platform for probing fluorescent proteins in the gas phase, which is of great fundamental interest for better understanding of their physical and chemical properties. In the present work, we have studied the gas-phase laser-induced fluorescence of mobility-selected rhodamine 6G ions. aser-induced fluorescence (LIF) is a well-known spectroscopic method which is widely applied for sensitive probing of the structure of molecules due to its high specificity to the microenvironment.1 One important feature is that it allows one to obtain structural details of biomolecules. Fluorescence spectroscopy permits one to determine binding sites of a protein, as well as its conformational transitions or intramolecular distances. In order to enable analysis by optical spectroscopy, nonfluorescent biomolecules can be tagged with extrinsic fluorescent tags, which are small molecules with specified optical properties.2 Recently, the combination of LIF with mass spectrometry (MS) has been developed.3 This combination opens up new possibilities both for detection purposes and for structural studies of trapped biomolecular ions in the gas phase. LIF data from the gas phase can provide interesting information that is complementary to liquid phase data. Nonvolatile molecules are usually brought into the gas phase in ionic form by soft ionization techniques, such as electrospray ionization (ESI)4 or matrix-assisted laser desorption/ionization (MALDI).5 Ions can be stored in a quadrupole ion trap3f,6 or a Fourier transform ion cyclotron resonance (FTICR) cell7 for subsequent LIF analysis. For example, in their pioneering work, Parks and co-workers monitored dissociation of double-stranded anions in the gas phase by observing the change in the efficiency of fluorescence resonance energy transfer (FRET) between a pair of fluorophores, BODIPY-TMR (donor) and BODIPY-TR (acceptor), labeling two strands of the duplex.3c LIF FTICR experiments have been performed in our lab with an internal MALDI ionization source described in detail elsewhere3e and recently with ESI source.7b Clear evidence of FRET occurring

L

© 2012 American Chemical Society

in the gas phase was demonstrated for the system rhodamine 6G (R6G; donor) and sulforhodamine B (acceptor) covalently bound through a phenyl ring linker.8 It was shown recently that ESI is capable of generating high numbers of gaseous analytes at ambient conditions.9 This was demonstrated by correlating fluorescence spectra recorded in the ESI plume with those of ions isolated in the high vacuum of an FTICR mass spectrometer. Therefore, isolated ions are directly accessible for spectroscopic investigation at ambient conditions, without the need for ion trapping and transport. However, only the total fluorescence signal from all ions can be acquired in an in-plume experiment. There are no possibilities for selective analysis of ions by this method, compared with others based on trapping of ions. Mass selective trapping techniques such as FTICR MS or ion traps require complicated hardware for LIF detection. Here, we propose an alternative technique, which is selective to the ion electrical mobility instead of the mass to charge ratio and does not require sophisticated instrumentation for LIF detection of gas-phase ions. Both LIF spectroscopy and differential mobility analyzers (DMAs) have been extensively used for characterization of different samples but usually separately. An instrument combining DMA and LIF has been developed to simultaneously explore fluorescence and ion mobility. Obviously, when this tandem instrument is coupled with a mass spectrometric detector, it will become even more powerful. Received: October 27, 2012 Accepted: November 30, 2012 Published: November 30, 2012 39

dx.doi.org/10.1021/ac303137e | Anal. Chem. 2013, 85, 39−43

Analytical Chemistry

Technical Note

and detected orthogonally by the cleaved edge of a 3 mm core diameter plastic fiber with a high NA (NA = 0.51, PGR FB 3000, SEDI, France). The other end of this fiber was coupled to a holographic imaging spectrograph (HoloSpec f/1.8i, Kaiser Optical Instruments Inc., Ann Arbor, USA) with CCD detection (LN/CCD-2500-PB/VISAR, Princeton Instruments, Trenton, USA). A 488 nm laser filter (Thorlabs, USA) was used in order to block most of the scattered laser light. Complementary electrospray ionization-mass spectrometry coupled to traveling wave ion mobility spectrometry (ESITWIMS-MS) measurements were carried out on a hybrid quadrupole−ion mobility−time-of-flight mass spectrometer (Synapt G2-S HDMS, Waters, Manchester, UK) equipped with a commercial LockSpray electrospray ion source (Waters). The sample solution of rhodamine 6G and tetraheptylammonium bromide (THBr) in methanol was infused at the flow rate of 5 μL min−1 using an external syringe pump (Harvard Apparatus, USA). Very gentle conditions were used to ensure minimum in-source fragmentation of analytes. The capillary voltage was set to 2.8 kV, and the source temperature was set to 40 °C. Nebulizer gas with a pressure of 6 bar and a desolvation gas at the flow of 500 L h−1 at a temperature of 250 °C were applied to assist the electrospray. The cone voltage and the source offset were both kept at 10 V. TWIMS-MS spectra were acquired in the “resolution mode” (Rfwhm ≥ 20 000) in the m/z range of 50−600 using automatic parameters for the pusher cycle and traveling-wave ion guides including IMS cell. Argon at a flow rate of 2 mL min−1 was used as a trap/collision gas, and the collision energy offset was automatically set to 4 V. Nitrogen at a flow rate of 90 mL min−1 was used as a buffer gas in the ion mobility stage. The helium cell prior to the IMS stage was operated at a helium flow of 180 mL min−1 to provide collisional cooling of ions before they enter the TWIMS guide. A typical spectrum was recorded during 2 min using a 1 s single scan time. The MassLynx v. 4.1 software (Waters, Manchester, UK) was used to manage and process the data. 2D IMS-MS maps were visualized in DriftScope v. 2.1 (Waters, Manchester, UK). The DMA was also coupled to a Fourier transform ion cyclotron resonance (FTICR) mass spectrometer (Ion Spec, CA, USA) for calibration purposes. The DMA was mounted on the FTICR MS instrument instead of the commercial ESI source for these experiments. The exact classification voltage to transmit selected tetraheptylammonium ions (THA) were identified in this fashion.

Ion mobility spectrometry (IMS) allows for the characterization of gas-phase ions according to their electrical mobility, which ultimately renders information on their collisional cross section.10 Traditional time-of-flight IMS admits ion packets into a drift tube, resulting in relatively poor sensitivity (duty cycle of ∼1%). DMAs have also a limited duty cycle because the classification voltage needs to be scanned to cover a given range of mobilities. However, for a given mobility, planar DMAs have shown to have transmission efficiencies >50% and, therefore, are far more efficient than gated IMS producing steady ion beams.11 This high transmission efficiency has enabled the interfacing of DMAs with commercial mass spectrometers with minimal modifications of the latter.11b Similarly, we take advantage here of this feature to integrate for the first time IMS and LIF. In this way, because the DMA acts as a mobility filter, ions of known electrical mobility are continuously selected, and as soon as they exit the DMA, they are excited by a laser to retrieve spectroscopic information. Additionally, the current is collected with an electrometer to provide a mobility spectrum. We provide here a proof of principle of this novel hybrid instrument by characterizing R6G ions.

■

EXPERIMENTAL SECTION The experimental setup of the DMA-LIF combination is shown schematically in Figure 1. A home-built nanoelectrospray

Figure 1. Scheme of the DMA-LIF experimental setup for the study of fluorescent analytes. A home-built nanoelectrospray source produces gas-phase ions (chamber 1) which are separated according to their electrical mobility in the DMA. A small capillary at the output of the DMA guides ions toward the current detector. A 488 nm laser beam crosses the ion path, and fluorescence could be detected by an optical fiber coupled to an optical spectrograph and a CCD camera (chamber 2).

■

RESULTS AND DISCUSSION It was recently discovered that ESI is capable of generating substantial amounts of isolated gas-phase species at ambient pressure a few millimeters downstream from the ESI tip.9 Because the original droplets generated in nanoelectrospray (nano-ESI)12 are substantially smaller than in conventional high-flow ESI,12b nano-ESI is very efficient for conversion of ions present in solution to ions in the gas phase. Preliminary characterization of the nano-ESI source itself was conducted by mapping the fluorescence properties of the plume. Consistent with our previous work,9 we found different fluorescence regimes during the excitation of the nano-ESI plume along its axial direction. Figure 2 shows LIF spectra of R6G at different distances from the electrospray tip. Similar to the experiment with conventional ESI source, the fluorescence maximum of R6G in the plume close to the emitter shows almost no shift relative to that in solution (λ ≈ 555 nm). As the laser beam was

source (nano-ESI) was used to produce gas-phase ions (chamber 1, Figure 1). The inner diameter of the capillary for sample introduction was 50 μm (PicoTip emitter), the electric potential was in the range of 3.2 kV relative to the upper electrode of the DMA. The distance between the tip of the capillary and the DMA inlet slit was ∼5 mm. A methanol solution containing rhoadmine 6G (99% pure, laser grade, from Acros Organics, USA) was infused through the capillary with a backing pressure of 70 mbar (∼200 nL/min). At the outlet of the commercial DMA (Model P4, SEADM, Spain), we integrated a vacuum cross, holding 2 windows and an electrometer (Lazcano Inc., Spain). In addition, we drilled an orifice at the center of the cross to insert an optical fiber for collecting the fluorescence emission. Fluorescence was excited by a 488 nm, 20 mW CW Laser (OBIS LS, Coherent, USA), 40

dx.doi.org/10.1021/ac303137e | Anal. Chem. 2013, 85, 39−43

Analytical Chemistry

Technical Note

Figure 2. LIF spectra recorded directly in the nanoESI plume (chamber 1 in Figure 1) at different distances from the electrospray capillary tip: red line, 2 mm; green line, 5 mm; blue line, 10 mm. Accumulation time 60 s; background subtracted. Spikes in the spectra are due to cosmic rays.

Figure 4. DMA spectrum of a mixture of rhodamine 6G (△) and tetraheptylammonium (THA) (star). The inset shows a peak identified as THA by a DMA-FTICR MS experiment. High mobility peaks were identified in a complementary TWIMS MS experiment (Figure 5).

shifted down 5 mm from the tip, an additional shoulder corresponding to gas-phase R6G ions appears in the spectrum (Figure 2, green line). Only a gas-phase R6G peak with a maximum around 505 nm is visible in the spectrum at a distance of 10 mm. During subsequent experiments, the actual position of the nano-ESI tip from the DMA entrance was around 5 mm to improve ion transmission. To assist ion desolvation, the ions entered the DMA against a counterflow of 0.3 L/min. The first proof of the potential to simultaneously retrieve information on the electrical mobility and fluorescence properties of gas-phase ions is given in Figure 3. It shows the characteristic mobility spectrum of R6G (left) and its corresponding fluorescence spectrum (right). The latter was obtained by fixing the DMA voltage at 1250 V during 10 min, whereby pure gas-phase R6G ions were transmitted through the DMA. As expected, the maximum of the LIF spectrum appears at 505 nm, matching the typical gas-phase R6G LIF spectrum shown in Figure 2 (blue trace). An additional advantageous feature of DMAs is the fact that they usually operate in the so-called linear regime (i.e., low electric field), and therefore, they measure true mobility.11a For this reason, a single standard compound suffices to calibrate the whole mobility range. Therefore, to transform the DMA scanning voltage scale into electrical mobility, we prepared a solution of R6G containing THA as internal standard.13 Figure 4 shows the mobility spectrum of the mixture containing R6G and THA recorded on the DMA. The x axis shows the mobility scale based on the mobility of the THA monomer (0.97 cm2 V−1 s−1).13 The voltage at which this ion is transmitted through

the DMA under the same experimental conditions, as in Figure 4, was determined during preliminary experiments in which the DMA was interfaced with a Fourier transform mass spectrometer (inset in Figure 4 shows a DMA selected FTICR mass spectrum of the THA monomer). Thus, we conclude that the measured mobility for R6G was 1.04 cm2 V−1 s−1. We are not aware of any previous measurement of R6G’s electrical mobility; thus, we cannot compare this value against the literature. Additionally, this mixture was characterized in a commercial nonlinear TWIMS-MS system (Synapt G2-S; Figure 5). Consistent with the DMA measurements, the mobility of THA was found to be higher than that of R6G. However, due to nonlinear effects of the TWIMS cell of the Synapt, the relative difference in arrival time between both species was 16.6%, while the actual difference in inverse mobility was found to be 7.3%. This highlights the advantage of using a linear IMS if further collisional cross section estimations want to be pursued in combination with LIF. In addition, note that the DMA transmits low mobility ions above an inverse mobility of 1.2 cm−2 V s. In contrast, the Synapt mobility spectrum does not show any additional feature beyond the THA monomer. As shown below, these low mobility peaks must correspond to R6G clusters, which are stable enough to survive the transition through the DMA, but decompose within Synapt. Presumably,

Figure 3. Raw DMA spectrum of R6G (left). LIF spectrum at DMA voltage of 1250 V, 600 s accumulation time; background subtracted (right). 41

dx.doi.org/10.1021/ac303137e | Anal. Chem. 2013, 85, 39−43

Analytical Chemistry

Technical Note

R6G peak (Figure 3). Further LIF measurements of the THA monomer peak (1.03 cm−2 V s) did not produce any fluorescence signal. This is as expected since THA is not a fluorescent compound. Interestingly, the LIF analysis of the low mobility region (1.2−1.5 cm−2 V s) showed a florescence signal in the 600 nm wave range (Figure 6A). As rhodamine was the only fluorescent compound in our mixture, this observation suggests that this peak corresponds to metastable rhodamine containing species. The shift in wavelength as compared to pure gas-phase R6G is consistent with previous work: gas-phase fluorescence of rhodamine complexes exhibit a red shift compared with pure rhodamine.9 It is also known that the fluorescence emission of R6G dimers in solution is in the range of 650 nm.10 This is thus a probable explanation for the presence of the ≈600 nm LIF peak. The fluorescence of gasphase species is usually blue-shifted compared to that in solution. We observe a shift of 50 nm for R6G, for example. The exact identity of the R6G cluster(s) remains unknown since it was not observable by mass spectrometry. However, this observation underlines the importance of conducting mobility measurements of labile ions at conditions as close as possible to the ion production, since otherwise they may go undetected. Additionally, it shows the usefulness of performing LIF downstream from the DMA, because otherwise it would be impossible to link these peaks to R6G by DMA (or IMS) alone.

Figure 5. (Top) Mass spectrum of a mixture of R6G (red) and THBr (blue) and (bottom) ion mobility spectrum. The sample was the same that was used to obtain the data shown in Figure 4, which shows similar peak positions in the DMA spectrum.

this occurs due to radio frequency heating in the ion guide region of the mass spectrometer.14 The mobility spectrum shown in Figure 4 was mapped in tandem by LIF. As shown in Figure 5, the high mobility region corresponds to unidentified contaminants unrelated to R6G. As expected, the high inverse mobility region covering 0.4−0.9 cm−2 V s showed no fluorescence signal (Figure 6C). In contrast, LIF measurements of the peak transmitted at 0.96 cm−2 Vs (R6G monomer) showed a LIF peak with maximum at 505 nm (Figure 6B). This is in agreement with the experiment with pure rhodamine solution when we observed the gas-phase

■

CONCLUSIONS DMAs are IMS instruments that (i) measure true mobility, (ii) act as mobility filters with high transmission, and (iii) operate at atmospheric pressure, whereby fragile ions produced during electrospray are preserved. These three main features make them ideal to be interfaced with LIF to investigate simultaneously mobility and fluorescence properties of gasphase ions. Here, we presented a proof of concept of this hybrid instrument by characterizing R6G. Given that both conformation and fluorescence are key properties to investigate relevant biological molecules in the gas phase, we expect that this new hybrid instrument will have potential to address open questions, for example whether or not gas-phase proteins retain their liquid-phase native structure.

■

AUTHOR INFORMATION

Corresponding Author

*Address: ETH Zurich, Department of Chemistry and Applied Biosciences, HCI E 329, 8093 Zurich, Switzerland. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are most grateful to SEADM (Gonzalo Fernandez de la Mora) for the loan of the DMA. We thank Myriam Macia and Guillermo Vidal for their support during the setting up of the DMA. We also thank Pavel Sagulenko for help in FTMS measurements and René Dreier for machining the parts of the experimental setup. This research was supported by a Marie Curie European Reintegration Grant (PMLS) within the seventh European Community Framework Programme (276860).

Figure 6. LIF spectra of the R6G/THBr mixture at different DMA electrical mobilities. 42

dx.doi.org/10.1021/ac303137e | Anal. Chem. 2013, 85, 39−43

Analytical Chemistry

■

Technical Note

REFERENCES

(1) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Springer: Berlin/Heidelberg, 2006. (2) Giepmans, B. N. G.; Adams, S. R.; Ellisman, M. H.; Tsien, R. Y. Science 2006, 312 (5771), 217−224. (3) (a) Li, G. Z.; Vining, V. A.; Guan, S. H.; Marshall, A. G. Rapid Commun. Mass Spectrom. 1996, 10 (14), 1850−1854. (b) Friedrich, J.; Fu, J. M.; Hendrickson, C. L.; Marshall, A. G.; Wang, Y. S. Rev. Sci. Instrum. 2004, 75 (11), 4511−4515. (c) Danell, A. S.; Parks, J. H. Int. J. Mass Spectrom. 2003, 229 (1−2), 35−45. (d) Khoury, J. T.; Rodriguez-Cruz, S. E.; Parks, J. H. J. Am. Soc. Mass Spectrom. 2002, 13 (6), 696−708. (e) Frankevich, V.; Guan, X. W.; Dashtiev, M.; Zenobi, R. Eur. J. Mass Spectrom. 2005, 11 (5), 475−482. (f) Sagoo, S. K.; Jockusch, R. A. J. Photochem. Photobiol., A: Chem. 2011, 220 (2−3), 173−178. (g) Qunzhou, B.; Forbes, M. W.; Talbot, F. O.; Jockusch, R. A. Phys. Chem. Chem. Phys. 2010, 12 (11), 2590−2598. (4) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science 1989, 246 (4926), 64−71. (5) (a) Hillenkamp, F.; Karas, M.; Beavis, R. C.; Chait, B. T. Anal. Chem. 1991, 63 (24), A1193−A1202. (b) Zenobi, R.; Knochenmuss, R. Mass Spectrom. Rev. 1998, 17 (5), 337−366. (6) Louris, J. N.; Cooks, R. G.; Syka, J. E. P.; Kelley, P. E.; Stafford, G. C.; Todd, J. F. J. Anal. Chem. 1987, 59 (13), 1677−1685. (7) (a) Marshall, A. G.; Hendrickson, C. L.; Jackson, G. S. Mass Spectrom. Rev. 1998, 17 (1), 1−35. (b) Chingin, K.; Chen, H. W.; Gamez, G.; Zenobi, R. J. Am. Soc. Mass Spectrom. 2009, 20 (9), 1731− 1738. (8) Dashtiev, M.; Azov, V.; Frankevich, V.; Scharfenberg, L.; Zenobi, R. J. Am. Soc. Mass Spectrom. 2005, 16 (9), 1481−1487. (9) Chingin, K.; Frankevich, V.; Balabin, R. M.; Barylyuk, K.; Chen, H. W.; Wang, R.; Zenobi, R. Angew. Chem., Int. Ed. 2010, 49 (13), 2358−2361. (10) Borsdorf, H.; Eiceman, G. A. Appl. Spectrosc. Rev. 2006, 41 (4), 323−375. (11) (a) Fernandez de la Mora, J. The differential mobility analyzer (DMA): Adding a true mobility dimension to a preexisting API-MS. In Ion Mobility Spectrometry - Mass Spectrometry: Theory and Applications; Wilkins, C. L.; Trimpin, S., Eds.; CRC Press: Boca Raton, FL, 2010; pp 105−136; (b) Rus, J.; Moro, D.; Sillero, J. A.; Royuela, J.; Casado, A.; Estevez-Molinero, F.; Fernández de la Mora, J. Int. J. Mass Spectrom. 2010, 298 (1−3), 30−40. (12) (a) Schmidt, A.; Karas, M.; Dulcks, T. J. Am. Soc. Mass Spectrom. 2003, 14 (5), 492−500. (b) Kebarle, P.; Verkerk, U. H. Mass Spectrom. Rev. 2009, 28 (6), 898−917. (13) Ude, S.; de la Mora, J. F. J. Aerosol Sci. 2005, 36 (10), 1224− 1237. (14) (a) Hogan, C.; Fernandez de la Mora, J. J. Am. Soc. Mass Spectrom. 2010, 21 (8), 1382−1386. (b) Hogan, C. J., Jr.; Fernandez de la Mora, J. Phys. Chem. Chem. Phys. 2009, 11 (36), 8079−8090.

43

dx.doi.org/10.1021/ac303137e | Anal. Chem. 2013, 85, 39−43