Comparing C60+ and (H2O)n+ clusters for mouse brain tissue analysis

Time‐of‐flight SIMS is applied to the analysis of single cells and different types of biological tissue samples enabling the generation of images with high spatial resolution and chemical specificity. However, the low yield of secondary ions from this type of sample still remains a challenge. This low yield could potentially be increased by enhancing the protonation of ions with the presence of water. Here, we have explored the application of a prototype water cluster ion beam for the analysis of mouse brain tissue samples. A series of experiments acquired with 20 keV (H2O)3000+ and 20 keV (H2O)4500+ were compared with 20 keV C60+, showing ion yield enhancement when a (H2O)n+ cluster ion is employed in the analysis. The results have demonstrated the potential benefits provided by the use of (H2O)n+ clusters for the analysis of mouse brain tissue samples. © 2014 The Authors. Surface and Interface Analysis published by John Wiley & Sons Ltd.


Introduction
SIMS imaging is a desorption-ionisation technique capable of generating images of biological surfaces with chemical specificity and high spatial resolution without the need for sample staining or labelling. [1][2][3] SIMS has been used in a variety of biological applications including the analysis of single cells and tissue samples [3] to study the distribution of different compounds in situ. Images from endogenous and exogenous molecules from tissue samples have proved to be important in the study of biochemical mechanisms, the development of a disease, targeted drug delivery, and so on.
With the introduction of liquid metal ion guns, it was possible to image biological samples with good spatial resolution, but significant bombardment-induced damage to the sample can be a downside of the approach. [4] The use of polyatomic ion beams such as SF 5 + and C 60 + can significantly reduce this damage. Using C 60 + , it has been possible to image a biological sample without extensive sample preparation and with micrometer spatial resolution without inducing major sample damage. [3,4] Argon cluster beams offer imaging with even less damage but with lower spatial resolution. These polyatomic beams make analysis beyond the static limit possible such that increasing the ion dose results in more signal from each pixel volume or voxel. This also makes possible the generation of 3D images by depth profiling the sample. [4] 3D imaging of single cells and mouse brain tissue has been reported using 20 keV and 40 keV C 60 + clusters. [3] When a tissue section is analysed with time-of-flight SIMS, most of the ions recorded in the mass spectrum correspond to lipids and their fragments. [3] Most of the peaks from the mass spectrum above m/z 200 can be assigned to fragments or molecular ions from diverse lipids species. [3] Nowadays, it is possible to detect intact molecular ions up to 1000 Da although the low secondary ion yield still remains a challenge. [3] There have been efforts to enhance secondary ion yields by adding metallic nanoparticles or matrix-assisted laser desorption/ionization type matrices to the sample. [5,12] Successful attempts to increase secondary ion yields involved the enhancement of the protonation of ions with the presence of water during the analysis. [6][7][8] There have been attempts to analyse the sample in a frozen hydrated state (instead of freeze-drying the sample) [9][10][11]13] or to inject water vapour at the sample's surface. [14] Another approach proposed the use of a water vapour electrospray beam as the primary ion beam to explore its effect on secondary ion yields. [15] In our lab in Manchester, a prototype water cluster ion beam developed by Ionoptika Ltd has been investigated and was successfully applied to the analysis of standard compounds as shown by Sheraz et al. [16] Herein, we discuss the application of this newly developed water cluster ion beam prototype for mouse brain tissue analysis and its effect on ionisation efficiency. The principles of its operation can be found elsewhere. [16] Experimental A whole mouse brain was obtained using ethically approved procedures from the Faculty of Life Sciences, the University of Manchester. It was sectioned in a Cryo-Microtome (Wolfson Molecular Imaging Centre, The University of Manchester, UK) at À20°C. Each section was 5 μm in thickness with sagittal orientation, and these were thaw-mounted on silicon wafers. The substrates were sonicated in methanol before mounting the tissue sections. The sections were stored at À80°C until the day of analysis. They were desiccated for 1 h at room temperature prior to SIMS analysis.
Time-of-flight SIMS analysis was performed on a J105-3D Chemical Imager described previously by Fletcher et al. [8] Data were collected using 20 keV C 60 + , 20 keV (H 2 O) 3000 + and 20 keV (H 2 O) 4500 + as primary ion beams (Ionoptika Ltd). An independent variable sample bias voltage was optimised for each experiment to compensate for charging during the experiments. Positive ion images were acquired using 650 × 650 μm field of view with 64 × 64 pixels. The pixel sizes were chosen to match the measured beam diameter. The structures studied are the caudate and putamen (striatum) of the brain, which are easy to identify because of their striated appearance.

Results and discussion
Comparing 20 keV C 60 + and 20 keV (H 2 O) n + for mouse brain tissue analysis Images of the striatum of the brain were acquired using 20 keV C 60 + and 20 keV (H 2 O) 3000 + . We present an image acquired with 20 keV C 60 + and an accumulated ion dose of 2.5 × 10 12 ions cm À2 . The spectrum is mainly dominated by characteristic ions from the grey matter and white matter of the brain, i.e. phosphocholine head group m/z 184 and cholesterol m/z 369, respectively (Fig. 1A). Ion peaks from a higher mass range (500-1000 Da, where intact lipids species can be found [17] ) are also visible but are low in signal.
After an accumulated ion dose of 2.5 × 10 12 ions cm À2 with C 60 + , analysis with 20 keV (H 2 O) 3000 + was carried out on the same area with a primary ion fluence of 5 × 10 11 ions cm À2 . The result from this experiment (Fig. 1B) shows a clear increase in total signal intensity even after the analysis with C 60 + . The total ion signal intensity is almost three times higher than the signal for the initial experiment with C 60 + . The signal is particularly enhanced for peaks in a higher mass range (500-1000 Da), where one can observe seven times more intensity compared with the initial analysis with C 60 + . Two more images with 20 keV (H 2 O) 3000 + and 20 keV (H 2 O) 4500 + were acquired on two new areas of the striatum of the brain (Fig. 2). The ion dose for each experiment was 5 × 10 11 ions cm À2 . The spectra of these images show the same trend for the distribution of the peaks dominated by the phosphocholine head group peak. Cholesterol and glycerophospholipids are visible with both water cluster primary ions. Again, there is a clear enhancement of the signal intensity in comparison with C 60 + (Fig. 1) wileyonlinelibrary.com/journal/sia especially for ion species in a mass range 500-1000 Da. Figure 2 shows a particular enhancement (two times more intensity) of the secondary ion yield for peaks in this mass range when using (H 2 O) 4500 + clusters compared with (H 2 O) 3000 + although we have to consider a possible variation in the sample's composition even within the same structure of the mouse brain.
Simultaneous beam experiment: analysis with 20 keV C 60 + with a 5 keV (H 2 O) 3000 + in DC mode A dual beam approach was carried out by analysing a new area of the striatum with 20 keV C 60 + (semi-continuous analysis beam) with a 5 keV (H 2 O) 3000 + beam in continuous mode Direct Current (DC) simultaneously. The purpose of this experiment was to exploit the spatial resolution capabilities of the C 60 + beam while using a low energy (H 2 O) 3000 + DC beam to increase the ionisation efficiency. After analysis with 20 keV C 60 + with an accumulated ion dose of 5 × 10 12 ions cm À2 and a spot size of 7 μm, an additional DC 5 keV (H 2 O) 3000 + defocused beam was directed at the region of interest and further images acquired using C 60 + sputtering from the same area with both beams impacting the surface at the same time. The results from this experiment (Fig. 3) show a continuous loss of secondary ion yield while analysing with C 60 + . However, there is a significant increase of secondary ion yield when the DC 5 keV (H 2 O) 3000 + is activated. The intensity of ions such as the cholesterol fragment (m/z 369), phosphocholine head group (m/z 184), and a glycerophosphocholine lipid (m/z 798) [18] show a decay in intensity after accumulation of dose from C 60 + analysis before the signal intensity is recovered by using the DC 5 keV (H 2 O) 3000 + cluster beam. The recovery of the signal intensity can be found in all the peaks in the mass spectrum and therefore cannot be attributed to changes in the composition of the sample as a function of depth.

Conclusions
We have shown that (H 2 O) n + clusters can be successfully applied for the analysis of mouse brain tissue samples. The results from experiments with C 60 + sputtering were compared with (H 2 O) n + analyses, and an increase in secondary ion yields was observed when using 20 keV (H 2 O) 3000 + clusters and 20 keV (H 2 O) 4500 + . We also investigated a dual beam approach using 20 keV C 60 + and 5 keV (H 2 O) 3000 + . The objective of this approach is to fully exploit the spatial resolution capabilities from the C 60 + as a primary ion beam with the presence of a low energy (H 2 O) n + beam wileyonlinelibrary.com/journal/sia operating in DC mode, obtaining an increase in the ionisation efficiency even after accumulating damage from the C 60 + beam. There are clear benefits for the use of (H 2 O) n + clusters for the analysis of mouse brain tissue, although further investigation is required to set the optimal conditions.