Time-resolved ICP-MS Measurement: a New Method for Elemental and Multiparametric Analysis of Single Cells

Transcription

1 ANALYTICAL SCIENCES FEBRUARY 2014, VOL The Japan Society for Analytical Chemistry Time-resolved ICP-MS Measurement: a New Method for Elemental and Multiparametric Analysis of Single Cells Special Reviews Shin-ichi MIYASHITA,* Alexander S. GROOMBRIDGE,** Shin-ichiro FUJII,* Akiko TAKATSU,* Koichi CHIBA,* and Kazumi INAGAKI* * Environmental Standards Section, National Metrology Institute of Japan (NMIJ), National Institute of Advanced Industrial Science and Technology, Tsukuba Central 3-10, Umezono, Tsukuba, Ibaraki , Japan ** Nano Doctoral Training Centre, The Cavendish Laboratory, The University of Cambridge, CB3 0HE, England Time-resolved inductively coupled plasma mass spectrometry (ICP-MS) has attracted much attention for elemental and multiparametric analysis of single cells, instead of a classical bulk analysis of large amount of cells after a dissolution. In the time-resolved measurement, cells are directly introduced into the plasma via nebulizing or micro drop dispensing, and then ion plumes corresponding to single cells are individually detected with a high time resolution. The sensitivity and cell throughput in the measurement strongly depend on the time resolution. A high cell introduction efficiency into the plasma supports for a reduction of cell consumption. Biomolecules can also be measured through the attachment of elemental tags, and then the amount distribution of elements and biomolecules in single cells can be evaluated, while providing information concerning cell-to-cell variations. By applying ICP time-of-flight mass spectrometry (ICP-TOFMS), multiparametric analysis of elements and biomolecules can be achieved similar to that by a flow cytometer. This article highlights the technical aspects of the time-resolved ICP-MS measurement technique for elemental and multiparametric analysis of single cells. Keywords Single cells, ICP-MS, ICP-TOFMS, time resolved measurement (Received October 7, 2013; Accepted November 19, 2013; Published February 10, 2014) 1 Introduction Overview of Time-resolved ICP-MS for Single-cell Analysis Technical Issues for Single-cell Analysis by Time-resolved ICP-MS Time resolution 3 2 Cell introduction efficiency 4 Future Prospects: Improving Reproducibility and Quantitative Capability Concluding Remarks Acknowledgements References Introduction In biochemical research fields, the utilization of inductively-coupled plasma mass spectrometry (ICP-MS) has been extended from trace elemental and speciation analysis to biomolecule and cell analysis ICP-MS has been the state-of-the-art technique for the detection of elemental and Shin-ichi MIYASHITA received his Ph.D. in 2011 from Tokyo University of Pharmacy and Life Sciences. He works at National Metrology Institute of Japan in National Institute of Advanced Industrial Science and Technology (NMIJ/AIST) as a Researcher (2011 ). His current research interest is the development of a high-efficiency cell introduction system for inductively coupled plasma (ICP) spectrometries and its application to the trace elemental analysis of individual microbial cells with time-resolved ICP mass spectrometry. Alexander S. GROOMBRIDGE received his MChem in Chemistry with Study in Japan from The University of Sheffield in 2013, and is currently a Ph.D. student at the Nano Doctoral Training Centre at the University of Cambridge. In 2011 and 2012 he worked at the National Metrology Institute of Japan under Dr. Kazumi Inagaki, conducting research into sample introduction systems in inductivelycoupled plasma optical emission and mass spectrometers. To whom correspondence should be addressed. k-inagaki@aist.go.jp

2 220 ANALYTICAL SCIENCES FEBRUARY 2014, VOL. 30 isotopic composition of matter since its first implementation. 11 In the early stages of its utilization, trace elements were major targets of measurements for investigating their biological effects and functions. Since around 2002, the birth year of the concept of Metallomics, 12,13 the application of ICP-MS has been spreading to quantify biomolecules with direct relation to microbial cells, based on its excellent elemental detecting capability. Not only metal-containing biomolecules, but also other biomolecules, such as amino acids, peptides, proteins, nucleic acids and microbial cells, can be successfully quantified by measuring heteroatoms, such as phosphorus and sulfur, and/or by supporting elemental tags, such as metal-labeled antigens and nanoparticles In recent years, cytometric analysis of single cells with a time-resolved ICP-MS measurement is receiving much attention for elemental and multiparametric analyses of single cells. Due to cellular heterogeneity within an isogenic cell population, individual analysis of cells will lead to a more sensitive representation of cell-to-cell variations, instead of a stochastic average analysis by bulk measurements Classical ICP-MS analysis of cells is the bulk analysis of large amounts of cells with a lysis, extraction or digestion; 30 this is an approach that is hardly adapted for evaluating any cell-to-cell variance. Time-resolved measurement via a direct cell introduction method is a relatively recent method for individual analysis of cells by ICP-MS, an approach that can detect the elements in individual cells with a high sensitivity, often called single particle mode. When cells are directly introduced into the plasma, ion plumes of the cells are generated after their vaporization. If one cell is introduced over the duration of the ion plume generation, and the acquisition time per data point is shorter than the duration of the ion plume generation, the ion plumes corresponding to individual cells can be detected as high spikes or short transient signals without overlapping with each other and averaging with background signals. The transient signal intensity of an element can be directly related to the amount of the element present in the individual cells. Therefore, the amount distribution of the elements in single cells, including the cell-to-cell variations, can be evaluated. 17,18 Through combination with elemental tagging techniques, the amount distributions of biomolecules in individual cells can also be evaluated similar to a flow cytometer Time-resolved measurement is not a new technique for ICP optical emission spectrometry (ICP-OES) and ICP-MS, but it has been mainly applied to the elemental analysis of airborne particle matter and observation studies of particle decomposition inside the plasma In the past decade or so, time-resolved ICP-MS has attracted much attention for highly sensitive elemental and multiparametric analyses of single biological cells. The elemental analysis of individual mammalian cells by time-resolved ICP-OES was first reported in 1994 by Nomizu et al. 38 Calcium signals were successfully measured corresponding to individual cells; however, other elements, such as Mg, K and Na, could not be measured due to their insufficient detection limits. In the past decade, Li et al. applied time-resolved ICP-MS to investigate the behavior of uranium compound-exposed bacteria in the ICP. 14 They concluded that the quantification of uranium in intact bacteria is possible using an inorganic uranium standard, providing a correction factor of atomization/ionization ratio. 14 Tanner and Bandura et al. developed a specialized ICP time-of-flight mass spectrometer (ICP-TOFMS) system for the multiparametric analysis of single human cells in Cells are labeled with multiplex antibodies tagged with enriched isotopes, and then the labels are quasi-simultaneously measured for individual cells by ICP-TOFMS. This method takes advantage of its multiparametric analysis capability, similar to flow cytometric analysis using multiparametric fluorescence detection. The instrument has been launched by DVS Science (CyTOF TM ). 39 Ho et al. successfully performed single-cell analysis of algal cells (1 6 μm) using quadrupole ICP-MS (ICP-QMS) at its lowest integration time (10 ms), focussing on the 25 Mg isotope. 15 They applied this method for tracking bismuth antiulcer drug uptake in Helicobacter pylori cells. 16 Shigeta et al. applied a micro droplet dispensing unit to introduce selenized yeast cells into the plasma. Uniformly-sized droplets (23 μm diameter) were successfully dispensed at a rate of 50 Hz, and were completely introduced into the plasma after passing through a desolvation unit, used for deducing the droplet size. 18 They obtained the frequency distribution of Cu, Zn and Se in individual selenized yeast cells. Groombridge et al. developed a high-efficiency cell introduction unit, which was based on a total consumption sample introduction device. 17 A cell introduction efficiency of over 70% into the plasma was achieved for the analysis of single yeast cells. They also conducted a preliminary investigation into the potential for multielement correlation analysis of individual yeast cells by time-resolved ICP-TOFMS. The time-resolved profiles of six elements (Mg, P, Mn, Fe, Cu and Zn) were quasi-simultaneously obtained. A relatively strong correlation was observed for the spectra between P and Zn (correlation factor: 0.69), between P and Mg (0.63), and between Mg and Zn (0.63). The application of time-resolved ICP-MS has been spreading in recent years, especially for the mass cytometric bioassay of single cells using the CyTOF TM instrument This article highlights the technical aspects of the time-resolved ICP-MS measurement for elemental and multiparametric analyses of single cells. 2 Overview of Time-resolved ICP-MS for Single-cell Analysis Schematic images of the time-resolved ICP-MS measurement for single-cell analysis are shown in Fig. 1. A cell suspension is Shin-ichiro FUJII received his Ph.D. in 2004 from the University of Tokyo. He worked at the University of Tokyo as Research Fellow of the Japan Society for the Promotion of Science ( ). He has worked at National Metrology Institute of Japan in National Institute of Advanced Industrial Science and Technology (NMIJ/AIST) as a Researcher ( ) and as a Senior Researcher (2013 ). His current research interest is the development of accurate analysis technology for nucleic acids. Kazumi INAGAKI received his Ph.D. from Nagoya University in He has worked at National Metrology Institute of Japan in National Institute of Advanced Industrial Science and Technology (NMIJ/AIST) as a Researcher ( ) and as a Senior Researcher (2008 ). He has developed many certified reference materials for trace element analysis. His current research interest is the development of a novel sample introduction system for plasma spectrometry, such as ICP-OES and ICP-MS.

3 ANALYTICAL SCIENCES FEBRUARY 2014, VOL. 30 Fig Schematic image of time-resolved ICP-MS measurement of single cells. introduced into the plasma via nebulizing14 17,19 23 or micro droplet dispensing,18 and ion plumes corresponding to single cells are measured with a high time resolution. The time resolution is defined here as the acquisition time (mainly integration time) per data point. When the interval of the cell introduction into the plasma is longer than the resolution, spike or short transient signals corresponding to single cells can be measured. ICP-QMS and sector field-type ICP-MS (ICP-SFMS) instruments, most commercial ICP-MS instruments, can acquire only one m/z at a time in the time-resolved measurement, since the settling and hopping time for each m/z is longer than the duration of ion plume generation. On the other hand, the ICP-TOFMS can quasi-simultaneously acquire a wide range of m/z in time-resolved measurements due to its fast scanning speed, and it can be applied for the multiparametric analysis of single cells.19,20 The duration of the ion plume generation depends on the center gas flow rate (the sum of the carrier gas and other support gas flow) and the on-axis sampling position (distance from the load coil).40 A schematic image of the ion sampling from the ion plume of a single cell is shown in Fig. 2. The center gas flow rate, which controls the residence time of the cells inside the plasma, is controlled for completely decomposing the cells to atoms inside the plasma. The on-axis sampling position is set at the position where the cell is completely decomposed to atoms, generated ion plume, and less diffused the ions inside the plasma, since ions in a small region of the plasma can only be sampled into the mass filter via the sampling interface. The duration of the ion plume generation also depends on the size of cells. Under typical optimum conditions of the center gas flow and sampling position, which depend on ICP-MS instruments, the durations of the ion plume generation corresponding to cells having a diameter of 1 to 10 μm are typical in the range of μs.19,20 The time resolution has a large influence on the sensitivity of Fig. 2 Schematic image of ion sampling from the ion plume of a single cell. the measurement. Figure 3 shows time-resolved ICP-QMS measurement profiles of phosphorous in yeast cells (S. cerevisiae) with different time resolution settings (10 and 0.05 ms). A fast scanning pulse counter directly connected to an electron multiplier was used for the measurements. When the time resolution is longer than the duration of the ion plume generation, cell events are measured as current spike signals, which are averaged with the background signals in the integration. On the other hand, cell events are measured as

4 222 ANALYTICAL SCIENCES FEBRUARY 2014, VOL. 30 Fig. 3 Time-resolved ICP-QMS measurement profiles of phosphorous in yeast cells (S. cerevisiae) with different time-resolution settings ((a)10 ms, (b) 0.05 ms). The figures on the right side, (a ) and (b ), are enlarged figures of the ion plume corresponding to a single cell. transient profiles when the time resolution is shorter than the duration. The signal-to-background (S/B) ratio of cell events is improved by increasing the time resolution due to minimize the averaging with the background signals, and the maximum S/B ratio would be obtained at a time resolution of 0.01 to 0.05 ms, which is ca. one-tenth the duration of the ion plume, because each cell event is measured as a transient signal profile. The time resolution also influence the cell throughput in the measurement. The cell throughput can be increased by increasing the time resolution, but the maximum cell throughput is limited by the duration of ion plume generation. This is because spike or short transient signals corresponding to single cells can be measured when the interval of the cell introduction into the plasma is longer than the time resolution. Olesik et al. calculated the probabilities of detecting signals from one particle or more than one within an integration time, when the duration is ca. 400 μs. 40 The probability of detecting signals from more than one particle can be reduced by increasing the time resolution (decreasing the integration time), but the number of particles entering the plasma must be limited to keep the probability of the overlapping the signals acceptably low. The theoretical limitation of cell throughput is ca cells per second when the duration is μs, 23,40 though it would vary with the size of cells and the operating conditions of ICP-MS, since these change the duration of ion plume generation. 3 Technical Issues for Single-cell Analysis by Time-resolved ICP-MS For evaluating the cell-to-cell variance of elements based on their amount distributions, a highly efficient analysis (high sensitive, high sample throughput, and low sample consuming) is desirable. Several technical issues regarding the time resolution (data acquisition parameter setting) and cell introduction efficiency to the plasma should be addressed for realizing a reliable single-cell analysis by time-resolved ICP-MS. The technical issues are summarized as follows. 3 1 Time resolution Although less than 0.10 ms of the time resolution would be desirable for the measurement, commercial ICP-QMS and ICP-SFMS instruments suffer from limitation in their time resolution setting on their operating software, which is typically 1 to 10 ms resolution. 40 In addition, there may be short delays, such as the settling time of mass filter, the detector dead time, and the gate closing time of a detector between the acquisition of the data points. Therefore, the acquisition parameter setting should be verified before the time-resolved measurement when a default operating software is utilized. Several researchers constructed a system using a digital oscilloscope or other digital processor for directly reading ion currents from an electron multiplier on a fast time scale ( μs per point). The reading process has a shorter systematic delay, but it still suffers from the detector dead time when highly intense cell events are measured. 40 ICP-TOFMS can quasi-simultaneously acquire a wide range

5 ANALYTICAL SCIENCES FEBRUARY 2014, VOL of m/z with a high time resolution, 13 μs for CyTOF TM, for example. 19,20 However, the signal intensity measured is one to two orders of magnitude lower than that of ICP-QMS. This is due to the duty cycle of the TOF mass analyser, which is defined as the fraction of extracted ions that make it into the TOF mass analyzer. In the ICP-TOFMS measurement, an ion packet is extracted from the plasma in several micro-seconds, and is then acquired by sweeping a wide range of m/z over 10 μs. Therefore, a substantial amount of ions per m/z integrated in the acquisition is lower than in the case of ICP-QMS and ICP-SFMS Cell introduction efficiency Efficient cell introduction to the plasma is also needed to reduce the sample consumption. The conventional sample introduction systems in ICP-MS, however, have poor sample transport efficiencies to the plasma, typically <1%. 15,43 Several special devices have been developed for improving the cell introduction efficiency. Nomizu et al. implemented a desolvating system that used a ribbon heater in tandem with α-emissions from 241 Am to neutralize the charge on the cells after nebulization. It was noted that the achieved cell introduction efficiency was 0.1%. 33 The cell introduction system of CyTOF TM consisted of a conventional pneumatic nebulizer, a low-volume on-axis spray chamber with a heater, and an aerosol splitter. The cell introduction efficiency quoted was 20 30%. 21,23 Groombridge et al. developed a high-efficiency cell introduction system consisting of a high-performance concentric nebulizer (HPCN) and a small-volume cylinder chamber, with a cell introduction efficiency of over 70% being achieved. 17 The time resolution in this case was 10 ms, and the cell throughput was 13 cells per second; this prevented any overlapping of separate cell events. The cell throughput will be improved by improving the time-resolution by less than 0.1 ms. Shigeta et al. applied a micro droplet dispensing unit in tandem with a desolvation system for cell introduction. 18 The cell introduction efficiency was 100% with the frequency of droplet dispensing being 50 Hz, but it was noted that the number concentration of cells in a suspension was diluted to 1 or 5 cells per 100 droplets so as to avoid the agglomeration of cells into single droplets. Therefore, the cell throughput calculated was 0.5 to 2.5 cells per second, which would be improved by increasing the droplet generation frequency while maintaining stable dispensing. 4 Future Prospects: Improving Reproducibility and Quantitative Capability The reproducibility of the measurement should be ensured for evaluating the cell-to-cell variance, and thus the change and drift in and between measurements should be normalized. This is especially important when comparing cell populations sampled from different specimens. Finck et al. used bead standards for normalizing the data in mass cytometry. 47 Polystyrene bead standards embedded with five enriched isotopes ( 139 La, 141 Pr, 159 Tb, 169 Tm, and 175 Lu) representative of the mass and dynamic ranges of the elements in label antibodies were used as internal standards for correcting the variation of the mass cytometric measurement. For any quantification of the signals obtained in ICP-MS, a form of calibration is necessary due to the widely varying ionization characteristics of elements within the plasma. The use of a standard solution, as employed in our previous work, 17 can only be used semi-quantitatively due to different ionization characteristics in the plasma compared to particles. As noted by Ho et al., the ion plumes resulting from single cells in the plasma exhibit a loss of the analyte through diffusion from the central plasma channel, thereby decreasing the resulting peak height intensities. With a standard solution, the ions are evenly distributed in the central channel of the plasma, and thus exhibiting less loss by diffusion. 15 Furthermore, the slight variation of the trajectories of cells leading to the sampling cone through the plasma can also result in decreasing the resulting signals. 48 Frequently suggested in the literature is the use of particle standards for the quantification of signals related to single cells. 15,33 This method certainly improves upon that of using a standard solution, but still has many disadvantages: for example, the high cost of monodisperse particle standards that can be easily introduced into the plasma in solution, and obtaining polydisperse particles with the ideal average particle size needed. The best option for accurate quantification would be having reference biological cells, such as E. coli or S. cerevisiae, cultured in a standard medium, though this would be very difficult to maintain. 5 Concluding Remarks Technical aspects in using time-resolved ICP-MS measurements for elemental and multiparametric analysis of single cells were reviewed. Technical subjects regarding time resolution and the cell introduction efficiency to the plasma should be addressed for realizing a highly efficient single-cell analysis. Commercial ICP-QMS and ICP-SFMS instruments suffer from limitations in their time-resolution settings on their operating software, but this limitation could be overcome through updating their software. A highly time-resolved measurement of less than 0.1 ms could easily be done without using an external digital processor in near future. An acceptable cell introduction efficiency can be obtained for yeast cells (ca. 3 μm in diameter) by using a highly efficient micro-flow nebulizer 17 or micro droplet dispenser. 18 These approaches would also be applicable for larger cells (over 10 μm), but some modifications, i.e. nozzle being replaced with a larger diameter one, would be required. Normalizing and robust calibrating methods are also required. More convenient normalization methods and standards for calibration would be desirable instead of nanoparticles or polystyrene beads. Time-resolved ICP-MS has attracted much more attention by resolving the above-mentioned technical issues, and its applications are set to spread to various biological research fields. 6 Acknowledgements This work was partly supported by Grant-in-Aid for Scientific Research (C) (No ) from the Japan Society for the Promotion of Science. 7 References 1. J. Szpunar, Analyst, 2005, 130, R. Lobinski, D. Schaumloffel, and J. Szpunar, Mass Spectrom. Rev., 2006, 25, Y. Ogra, Anal. Sci., 2009, 25, A. Prange and D. Proefrock, J. Anal. At. Spectrom., 2008, 23, J. Bettmer, M. Bayon, J. Encinar, M. Sanchez, M. de la Campa, and A. Sanz-Medel, J. Proteomics, 2009, 72, 989.

6 224 ANALYTICAL SCIENCES FEBRUARY 2014, VOL M. Wang, W. Feng, Y. Zhao, and Z. Chai, Mass Spectrom. Rev., 2010, 29, C. Engelhard, Anal. Bioanal. Chem., 2011, 399, D. Profrock and A. Prange, Appl. Spectrosc., 2012, 66, C. Giesen, L. Waentig, U. Panne, and N. Jakubowski, Spectrochim. Acta, Part B, 2012, 76, A. Sanz-Medel, M. Montes-Bayon, J. Bettmer, M. Fernandez-Sanchez, and J. Encinar, TrAC, Trends Anal. Chem., 2012, 40, R. Houk, V. Fassel, G. Flesch, H. Svec, A. Gray, and C. Taylor, Anal. Chem., 1980, 52, J. Szpunar, Anal. Bioanal. Chem., 2004, 378, H. Haraguchi, J. Anal. At. Spectrom., 2004, 19, F. Li, D. W. Armstrong, and R. S. Houk, Anal. Chem., 2005, 77, K. S. Ho and W. T. Chan, J. Anal. At. Spectrom., 2010, 25, C. N. Tsang, K. S. Ho, H. Sun, and W. T. Chang, J. Am. Chem. Soc., 2011, 133, A. Groombridge, S. Miyashita, S. Fujii, K. Nagasawa, T. Okahashi, M. Ohata, T. Umemura, A. Takatsu, K. Inagaki, and K. Chiba, Anal. Sci., 2013, 29, K. Shigeta, H. Traub, U. Panne, A. Okino, L. Rottmann, and N. Jakubowski, J. Anal. At. Spectrom., 2013, 28, S. D. Tanner, D. R. Bandura, O. Ornatsky, V. I. Baranov, M. Nitz, and M. A. Winnik, Pure Appl. Chem., 2008, 80, D. Bandura, V. Baranov, O. Ornatsky, A. Antonov, R. Kinach, X. Lou, S. Pavlov, S. Vorobiev, J. Dick, and S. Tanner, Anal. Chem., 2009, 81, O. Ornatsky, D. Bandura, V. Baranov, M. Nitz, M. A. Winnik, and S. Tanner, J. Immunol. Methods, 2010, 361, S. Bendall, E. Simonds, P. Qiu, E. Amir, P. Krutzik, R. Finck, R. Bruggner, R. Melamed, A. Trejo, O. Ornatsky, R. Balderas, S. Plevritis, K. Sachs, D. Pe er, S. Tanner, and G. Nolan, Science, 2011, 332, S. C. Bendall, G. P. Nolan, M. Roederer, and P. K. Chattopadhyay, Trends Immunol., 2012, 33, R. Trouillon, M. Passarelli, J. Wang, M. Kurczy, and A. Ewing, Anal. Chem., 2013, 85, S. Date, H. Mizuno, N. Tsuyama, T. Harada, and T. Masujima, Anal. Sci., 2012, s, Y. Lin, R. Trouillon, G. Safina, and A. Ewing, Anal. Chem., 2011, 83, T. Masujima, Anal. Sci., 2009, 25, E. Turner, D. Cohen, H. Pugsley, D. Gomez, C. Whitmore, C. Zhu, and N. Dovichi, Anal. Bioanal. Chem., 2008, 390, J. V. Sweedler and E. A. Arriaga, Anal. Bioanal. Chem., 2007, 387, G. Cerchiaro, T. M. Manieri, and F. R. Bertuchi, Metallomics, 2013, 5, H. Kawaguchi, N. Fukasawa, and A. Mizuike, Spectrochim. Acta, Part B, 1986, 41, U. K. Bochert and W. Dannecker, J. Aerosol Sci., 1989, 20, T. Nomizu, S. Kaneco, T. Tanaka, T. Yamamoto, and H. Kawaguchi, Anal. Sci., 1993, 9, T. Myojo, M. Takaya, and M. Ono-Ogasawara, Aerosol Sci. Technol., 2002, 36, S. E. Hobbs and J. W. Olesik, Anal. Chem., 1992, 64, J. W. Olesik, Appl. Spectrosc., 1997, 51, 158A. 37. L. A. Allen, J. J. Leach, and R. S. Houk, Anal. Chem., 1997, 69, T. Nomizu, S. Kaneco, T. Tanaka, D. Ito, H. Kawaguchi, and B. T. Vallee, Anal. Chem., 1994, 66, J. W. Olesick and P. J. Gray, J. Anal. At. Spectrom., 2012, 27, S. D. Tanner, O. Ornatsky, D. R. Bandura, and V. I. Baranov, Spectrochim. Acta, Part B, 2007, 62, R. Thomas, Practical Guide to ICP-MS, 2013, 3rd ed., CRC Press, US. 43. J. L. Todolí and J. M. Mermet, Liquid Sample Introduction in ICP Spectrometry: A Practical Guide, 2008, Elsevier Science Publishers B.V., Amsterdam. 44. K. Inagaki, S. Fujii, A. Takatsu, and K. Chiba, J. Anal. At. Spectrom., 2011, 26, Y. Takasaki, K. Inagaki, A. Sabarudin, S. Fujii, D. Iwahata, A. Takatsu, K. Chiba, and T. Umemura, Talanta, 2011, 87, A. S. Groombridge, K. Inagaki, S. Fujii, K. Nagasawa, T. Okahashi, A. Takatsu, and K. Chiba, J. Anal. At. Spectrom., 2012, 27, R. Finck, E. F. Simonds, A. Jager, S. Krishnaswamy, K. Sachs, W. Fantl. D. Pe er, G. P. Nolan, and S. C. Bendall, Cytometry, Part A, 2013, 83A, J. Tuoriniemi, G. Cornelis, and M. Hassellöv, Anal. Chem., 2012, 84, 3965.