Ims By Nu Bioprobes[1]

Nuclear Instruments and Methods in Physics Research B 267 (2009) 2144–2148

Contents lists available at ScienceDirect

Nuclear Instruments and Methods in Physics Research B journal homepage: www.elsevier.com/locate/nimb

Imaging mass spectrometry with nuclear microprobes for biological applications Y. Nakata a,*, H. Yamada a, Y. Honda a, S. Ninomiya b, T. Seki a,d, T. Aoki c,d, J. Matsuo b,d a

Department of Nuclear Engineering, Kyoto University, Sakyo, Kyoto 606-8501, Japan Quantum Science and Engineering Center, Kyoto University, Uji, Kyoto 611-0011, Japan Department of Electronic Science and Engineering, Kyoto University, Nishikyo, Kyoto 615-8510, Japan d CREST, Japan Science and Technology Agency, Chiyoda, Tokyo 102-0075, Japan b c

a r t i c l e

i n f o

Available online 24 March 2009 PACS: 79.20.Rf 82.80.Ms 82.80.Rt Keywords: Mass spectrometry Molecular imaging Secondary ion Pollen Mammalian cell

a b s t r a c t A mass spectrometric technique using nuclear microprobes is presented in this paper for biological applications. In recent years, imaging mass spectrometry has become an increasingly important technique for visualizing the spatial distribution of molecular species in biological tissues and cells. However, due to low yields of large molecular ions, the conventional secondary ion mass spectrometry (SIMS), that uses keV primary ion beams, is typically applied for imaging of either elements or low mass compounds. In this study, we performed imaging mass spectrometry using MeV ion beams collimated to about 10 lm, and successfully obtained molecular ion images from plant and animal cell sections. The molecular ion imaging of the pollen section showed high intensities of PO3 ions in the pollen cytoplasm, compared to the pollen wall, and indicated the heterogeneous distribution in the cytoplasm. The 3T3-L1 cell image revealed the high intensity of PO3 ions, in particular from the cell nucleus. The result showed that not only the individual cell, but also the cell nucleus could be identified with the present imaging technique. Ó 2009 Published by Elsevier B.V.

1. Introduction Two modern techniques for elemental analysis, namely particle-induced X-ray emission (PIXE) and Rutherford backscattering (RBS) spectrometry, involve irradiation of the analyte with energetic ion beams, resulting in the generation of characteristic X-rays and backscattered ions with energies that depend on the constituent elements of the surface. Ion beam irradiation also leads to emission of atomic and molecular ions originating from the solid surface, and the ions represent the surface composition of the analyte. The mass spectrometric technique using keV primary ion beam is known as secondary ion mass spectrometry (SIMS) and has been employed in surface characterization and depth profiling of inorganic and organic materials [1,2]. During the last decade, the SIMS technique has been applied to a new mass spectrometric analysis named imaging mass spectrometry, which reveals the spatial distribution of molecular species in biological tissues and cells [3–5]. In analyzing such biological samples, the molecular ion emission is of particular importance since these samples consist of various molecular species, such as amino acids, peptides or proteins. However, due to the low yields of large molecular ions, SIMS * Corresponding author. Tel.: +81 774 38 3977; fax: +81 774 38 3978. E-mail addresses: [email protected], [email protected] (Y. Nakata). 0168-583X/$ - see front matter Ó 2009 Published by Elsevier B.V. doi:10.1016/j.nimb.2009.03.094

imaging generally provides information on elements or molecules with low mass (<500 u). In contrast to the keV ion beam irradiation, high-energy ions (> MeV) have the ability to desorb large molecules. The first report on desorption of biomolecular ions by irradiation with MeV-energy ions was in 1974 [6]. The mass spectrometric technique that uses fission fragments from a 252Cf source is called plasma desorption mass spectrometry (PDMS) and has been applied to the analysis of antibiotics, toxins, and proteins with masses up to 20,000 u [7–9]. The desorption mechanism of PDMS has been studied in detail by, for example, the Uppsala group, who suggested the mechanism governed by high-energy density of electronic excitation around the ion track [10,11]. During 1970s and 1980s, the PDMS analysis was one of the most useful methods for detecting large biomolecular ions, and has generated much interest in its physics and use for biological mass spectrometry. However, the number of PDMS studies gradually decreased after the advent of other promising techniques of fast atom bombardment (FAB) and matrix-assisted laser desorption/ionization (MALDI) [7,8]. In recent years, we have developed a new system for imaging mass spectrometry using MeV ion beams, termed MeV-SIMS here, and demonstrated the capability of MeV-SIMS to provide images of large molecules (>1000 u) by using pure films of angiotensin II, bradykinin, and substrate peptide [12]. However, these pure films are different from ‘‘real-world” samples such as biological tissues

Y. Nakata et al. / Nuclear Instruments and Methods in Physics Research B 267 (2009) 2144–2148

and cells, where the targeted compounds are diluted in very complex matrices. In this paper we present our first results on MeV-SIMS imaging of plant and animal cells, and suggest that the current technique is useful for mass spectrometric imaging of biological tissues and cells. 2. Experimental 2.1. Mass spectrometry Mass spectrometry analysis was performed using a 6 MeV Cu ion beam from Kyoto University’s tandem accelerator. The experimental setup is shown in Fig. 1. The primary ion beam was incident on a sample at an angle of 20° to the surface normal, after being collimated by a 5 lm aperture (National Aperture Inc., USA) installed upstream, 40 mm from the sample. The beam spot on the sample surface increased to about 10 lm due to the beam divergence of 0.14 mrad, as determined by an upstream X–Y slit. The irradiated location could be checked with the CCD camera by fluorescence from the screen. The primary ion beam current was several tens pA when with a 1 mm aperture, and decreased to about 2000 primary ions/s with the 5 lm aperture. The sample holder was mounted on a piezoelectric stage (M-2689, MESS-TEK, Japan) with a nominal resolution of several nm, and thus the sample surface was scanned over a square area of 150 lm  150 lm with any step size. In this experiment, the sample holder was biased at 6 kV, and ejected electrons and negatively charged particles were extracted into a linear-type time-of-flight (TOF) mass spectrometer. Due to the high electron velocity, we could use the electron signal as a start signal, and the signal from the secondary ions as a stop signal for the TOF measurement. The secondary ion intensity was integrated every relevant m/z and the selected ion image could be displayed on the PC monitor. Typically, the working pressure in the analyzing chamber was 2  10 5 Pa. 2.2. Materials and sample preparation The emphasis of this paper is on the capability of MeV-SIMS to obtain secondary ion images from a variety of samples, ranging from peptide sample to plant and animal cells. Firstly, the micropatterned surface was investigated for evaluation of the performance of the MeV-SIMS technique, and the results could give us general information about image contrast, acquisition time, and lateral resolution. As a next step, we analyzed the ‘‘real-world”

samples such as plant and animal cells. The aim of this study is to demonstrate that the spatial distributions of molecular species in the cells can be revealed by the MeV-SIMS technique. 2.2.1. Test sample A micropatterned peptide surface was prepared as a target for testing the MeV-SIMS imaging technique. A peptide sample, glycylglycylglycine (triglycine, molecular weight 189.17) was purchased from the Peptide Institute (Japan). The triglycine thin film was prepared by spin-coating a 10 lL of aqueous solution onto a clean Si wafer, after which the film was introduced into a vacuum evaporation chamber, where gold was evaporated onto the film covered with a Ni fine mesh (70 wire/inch, bar 30 lm). The micropatterned peptide sample was then introduced into the analytical chamber and analyzed with MeV-SIMS. 2.2.2. Plant cell Pollen grains were collected from marvel-of-Peru (Mirabilis jalapa) and let to dry overnight. On average, the pollen size was 150 lm, which is a little larger than usual, and thus its cross section could be imaged by using the 10 lm diameter primary ion beam. Impurities such as lipids present on the pollen wall were rinsed with acetone, after which the pollen grains were dissolved in acetone and stored at room temperature until used. The acetone solution of pollen was dripped onto a Si wafer and allowed to dry. Then, it was embedded in araldite and cut using a microtome knife mounted on a rotary microtome (PR-50, Yamato Kohki Industrial Co., Japan). 2.2.3. Animal cell The 3T3-L1 cell line was purchased from Dainippon Sumitomo Pharma Co. (Japan). The 3T3 cells were grown in a nutritive media (DMEM) onto 10 mm  10 mm indium tin oxide (ITO) thin films. After the cells reached about 80% confluency on the ITO substrate, the culture medium was exchanged with one containing about 500 ppm 5-bromo-2´-deoxyuridine (BrdU) (Nacalai Tesque, Japan). BrdU (C9H11BrN2O5) is a thymidine analog that is incorporated into nuclear deoxyribonucleic acid (DNA). After the cells were exposed to the BrdU-containing medium for 1 h, the substrates were removed from the medium and quickly rinsed with water, and then allowed to dry in a desiccator. To obtain a section of biological tissues and cells, several techniques are used, for example, a freezefracture technique has been successfully used to prepare the cell sections by fracturing the frozen cells sandwiched between two

Detector (Microchannel plate)

Flight tube

Seco

nd a

ry io

ns

2145

Deflectors Einzel lens

Piezoelectric Stage

Primary ion beam Aperture

Faraday cup Sample holder

Fig. 1. The experimental setup for MeV-SIMS imaging.

View port

Y. Nakata et al. / Nuclear Instruments and Methods in Physics Research B 267 (2009) 2144–2148

Si wafers [13]. However, this technique seems to require great care and technical training, and is now under development in our laboratory. In this experiment, we performed Ar gas cluster irradiation on the 3T3 cell surface to obtain the cell section. An Ar gas cluster ion consists of a number of Ar atoms, ranging from 100 to 10,000 Ar atoms, and is characterized by high sputtering yield, low surface damage and efficient surface smoothing [14–16]. To date, the gas cluster ion irradiation has been employed for the smoothing and etching of semiconductors: lately, a new application to the cell surface processing using this method has been studied [17]. In this study, the 3T3 cells on the ITO substrate were irradiated with an 8 keV Ar gas cluster ion beam at a fluence of 5  1015 ions/cm2. The irradiated cells were observed with an optical microprobe and the surface profile was measured using a laser microprobe (VK8500, KEYENCE, Japan). 3. Results and discussion

250

150

100

50

0 0

3.1. Test sample

10

20

30

40

50

60

70

80

Displacement [µm]

Mass spectrometric imaging of the micropatterned triglycine surface was carried out to evaluate the performance of the MeV-SIMS imaging prior to analyzing plant and animal cells. A collimated 6 MeV Cu ion beam was used for scanning the 150 lm  150 lm surface area of the triglycine sample to produce a 30  30 pixel digital image. Fig. 2(a) displays the negative ion mass spectrum from the triglycine surface. The deprotonated triglycine was observed at m/z 188 with the high yield of 0.03 molecular ions/primary ion. Fig. 2(b) shows the deprotonated triglycine image obtained with high contrast, and is very similar to the optical image in Fig. 2(c). This image was acquired during about 110 min with 6000 primary ions/pixel. To evaluate the beam spot size on the sample surface, a line-scan of the mesh-covered triglycine surface was performed, and showed the beam diameter to be about 10 lm (Fig. 3). 3.1.1. Plant cell Next, we conducted MeV-SIMS imaging of the pollen collected from the marvel-of-Peru. The imaging analysis of pollen is important for the study of pollen allergens and, for example, the CN and O2P distributions were investigated with conventional SIMS [18]. In this study, pollen grains embedded in araldite were cut using the microtome knife and pollen sections were successfully obtained (Fig. 4(a)). From the pollen cytoplasm, PO3 ; H2 PO4 and deprotonated palmitic acids (C15H31COO ) were observed (Fig. 4(b)). The ion images of total ions (Fig. 4(c)), PO3 (Fig. 4(d)) and deprotonated palmitic acids (Fig. 4(e)) showed high concentrations of the molecular ions in the cytoplasm compared to the pollen wall. It has been found that phytic acid (C6H18O24P6) was a major constituent of the

Fig. 3. Linescan of a triglycine surface covered with a Ni fine mesh. The vertical axis represents the deprotonated triglycine intensity and the variation shows the beam diameter to be about 10 lm.

pollen of many plant species [19]. Thus the pollen grains contain large amount of phosphoric acids. The high intensity of PO3 ions in the cytoplasm agrees well with the abundance of phytic acids. On the other hand, the outer pollen wall (exine) is made of sporopollenin, which is a mixture of biopolymers and characterized by its rigid structure. The inner pollen wall (intine) is composed of cellulose (C6H10O5)n, which is a polysaccharide and chemically stable. Therefore, it is natural that no PO3 ions were detected from the pollen wall. Fig. 4(d) indicated the heterogeneous distribution of PO3 ions in the cytoplasm, providing qualitative agreement with the pollen imaging obtained with conventional SIMS [18]. In addition, Fig. 4(e) reveals that the deprotonated palmitic acids were localized at the cytoplasm. The peak at m/z 255 showed bad signal-to-noise ratio, and hence the isolated red spots are probably of statistical nature. 3.1.2. Animal cell The 3T3-L1 cells cultured on the ITO substrate were analyzed with MeV-SIMS. The cells were subjected to Ar gas cluster irradiation, and the irradiated cell surface was observed with optical and laser microscopes. Fig. 5(a) and (b) shows that the cell nucleus and membrane still remain on the substrate without visible degradation of cell structure. Main component of the cell membrane is phospholipids, which are also included in the cytoplasm. Fig. 5(c) showed that the PO3 ions were detected from the whole cell

50

(a)

4+

6MeV Cu => triglycine 40 Counts

10 µm

200

Intensity [arb. uni]

2146

triglycine-H

(b)

(c)

-

30

30

m

0

ly t r ig

20 10 0

255

20 0

50

100

150 m/z

200

250

cine

m

300

Fig. 2. (a) Negative ion mass spectrum of a triglycine sample bombarded with 6 MeV Cu4+ ions. (b) Secondary ion image of the deprotonated triglycine. The number of primary ions was about 6000 ions/pixel. (c) Optical image of the micropatterned triglycine surface.

2147

Y. Nakata et al. / Nuclear Instruments and Methods in Physics Research B 267 (2009) 2144–2148 4+

6 MeV Cu => pollen cytoplasm

(a) Counts

2000

-

PO3

(b)

Negative-Ion Mode 30

1500 255

1000 -

HPO4

500

0

0

100 200 300 400 500 600 700 800 900 1000 m/z

(c)

5080

(d)

(e)

36

0

80

Total ion

172

0

m/z 79

m/z 255

Fig. 4. (a) An optical microscope image of the pollen section; (b) negative ion mass spectrum of the pollen cytoplasm bombarded with 6 MeV Cu4+ ions; secondary ion images of: (c) total ions, (d) PO3 ions (m/z 79) and (e) deprotonated palmitic acids (m/z 255). The scale bars represent 30 lm.

Fig. 5. (a) An optical microscope image of the 3T3-L1 cell section. (b) A laser microscope image of the cell section. (c) An MeV-SIMS image (50  50 pixels) of PO3 ions (m/z 79) over a 100 lm  100 lm field of view with a pixel size of 2 lm. This image was obtained with 5000 primary ions/pixel; the maximum count was 252; the scale bars represent 20 lm.

section, indicating that the cell membrane, nucleus, and cytoplasm are present on the substrate. The high PO3 intensity was obtained in particular from the cell nucleus. It indicates that the nuclear membrane was etched by the Ar gas cluster ion beam irradiation and the PO3 ions were produced from deoxyribonucleic acid (DNA) phosphate groups in the cell nucleus bombarded with MeV ions. Fig. 5(c) shows that not only the individual cell but also the cell nucleus could be identified with the MeV-SIMS technique. The 3T3 cells were treated with 500 ppm BrdU, a thymidine analog that is incorporated into DNA, but no signals from BrdU were observed in the nucleus region. The natural isotopes of bromine are 79Br and 81Br, with a relative abundance of 50.7% and 49.3%, respectively. In the obtained spectra, 79Br ion peak is considered to interfere with the PO3 ion peak at m/z 79, which was the most prominent peak. On the other hand, no intense peak at m/z 81 appeared in the mass spectra. Assuming that the yield of Br ions ejected from BrdU was 5  10 5 Br /primary ion, the detected Br ion counts were calculated from 5  10 5 Br /primary ion multi-

plied by 5  103 primary ions, that is, 0.25 Br ions. Therefore, if the number of incident ions increases to 5  105 Cu ions, the Br ion signals would reach 25 counts, a value sufficiently high for high-contrast image. However, increasing the number of incident ions per pixel would result in longer acquisition time, and hence, in order to obtain the trace compounds, such as an administered drug or BrdU in a short time, the beam current density should be improved by installing a focusing lens after the beam collimator. Additionally, in this study, large molecular ions (>m/z 500) were not identified due to poor signal-to-noise ratio. We need to increase the mass resolution to assign those peaks and improve signal-tonoise ratio, resulting in high-contrast images of large molecular ions. We have employed the linear TOF mass spectrometer (m/ Dm  200) to perform the MeV-SIMS imaging, but in the future, a reflectron TOF mass spectrometer (m/Dm  6000) will be introduced to the system, and the MeV-SIMS imaging of large molecular ions will be performed using focused MeV ion beam. In pharmaceutical research field, a high-resolution molecular imaging technique

2148

Y. Nakata et al. / Nuclear Instruments and Methods in Physics Research B 267 (2009) 2144–2148

for localizing drugs in biological cells is required. Also in biology, revealing the spatial distributions of molecular ions of lipids in the cell membrane is a challenging issue. It is suggested that the MeV-SIMS technique would offer new opportunities in pharmaceutical and biological research fields by visualizing the distributions of lipids, drugs, and metabolites in tissues and cells.

Science and Technology Agency (JST). It is also supported in part by the Research Fellowships of the Japan Society for the Promotion of Science (JSPS) for Young Scientists. This paper was edited for publication by Edit Associates (http://www.editassociates.com), and the authors gratefully acknowledge its editor-in-chief, Dr. Rafael Manory.

4. Conclusions

References

We performed mass spectrometric imaging of plant and animal cell sections using an MeV ion beam collimated to 10 lm in diameter. The MeV-SIMS imaging of the pollen section showed high intensities of negative ions in the cytoplasm and indicated the heterogeneous distributions of the PO3 ions. From the 3T3-L1 cell section, the PO3 ion image was successfully obtained, revealing the abundance of this ion in the cell nucleus. This result showed that not only the individual cell, but also the cell nucleus could be identified with the current imaging technique. It is suggested that the MeV-SIMS technique could be applied to pharmaceutical and biological studies, such as the investigation of drug and metabolite distributions in biological tissues and cells, or elucidation of lipid distributions within a cell membrane. Acknowledgements The authors gratefully acknowledge the valuable discussions with Prof. A. Itoh, Prof. H. Shibata and Dr. H. Tsuchida. We also would like to thank K. Yoshida, K. Norizawa, M. Naito and M. Hada for their technical support. This work is supported by the Core Research for Evolutional Science and Technology (CREST) of the Japan

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]

S. Hofmann, Progress Surf. Sci. 36 (1991) 35. A. Benninghoven, Surf. Sci. 299–300 (1994) 246. S.G. Ostrowski, C.T. Van Bell, N. Winograd, A.G. Ewing, Science 305 (2004) 71. J.L. Guerquin-Kern, T.D. Wu, C. Quintana, A.P. Croisy, Biochim. Biophys. Acta. 1724 (2005) 228. M.L. Kraft, P.K. Weber, M.L. Long, I.D. Hutcheon, S.G. Boxer, Science 313 (2006) 1948. D.F. Torgerson, R.P. Skowronski, R.D. Macfarlane, Biochem. Biophys. Res. Commun. 60 (1974) 616. R.D. Macfarlane, Braz. J. Phys. 29 (1999) 415. R.D. Macfarlane, Biol. Mass Spectrom. 22 (1993) 677. B.U.R. Sundqvist, P. Roepstorff, J. Fohlman, A. Hedin, P. Håkansson, I. Kamensky, M. Lindberg, M. Salehpour, G. Säwe, Science 226 (1984) 696. A. Hedin, P. Häkansson, M. Salehpour, B.U.R. Sundqvist, Phys. Rev. B 35 (1987) 7377. R.E. Johnson, B.U.R. Sundqvist, Phys. Today 45 (1992) 28. Y. Nakata, Y. Honda, S. Ninomiya, T. Seki, T. Aoki, J. Matsuo, J. Mass Spectrom. 44 (2009) 128. S. Chandra, D.R. Lorey II, Int. J. Mass Spectrom. 260 (2007) 90. I. Yamada, J. Matsuo, Z. Insepov, M. Akizuki, Nucl. Instr. and Meth. B 106 (1995) 165. J. Matsuo, T. Seki, T. Aoki, I. Yamada, Nucl. Instr. and Meth. B 206 (2003) 838. I. Yamada, N. Toyoda, J. Vac. Sci. Technol. A 23 (2005) 1090. H. Yamada, K. Ichiki, Y. Nakata, S. Ninomiya, T. Seki, T. Aoki, J. Matsuo, Trans. Mater. Res. Soc. Jpn., accepted for publication. F. Lhuissier, F. Lefebvre, D. Gibouin, M. Demarty, M. Thellier, C. Ripoll, J. Microsc. 198 (2000) 108. J.F. Jackson, G. Jones, H.F. Linskens, Phytochemistry 21 (1982) 1255.