SIMS2015 Session BI1-TuA: Biological Imaging

ToF-SIMS Imaging is widely applied to the analysis of cells and tissues due to its unique capability of generating chemical high spatial resolution images ^[1-3]. Along with these capabilities there are challenges ToF-SIMS imaging still needs to overcome. One of the most significant challenges is the low secondary ion yield generated from biological samples. This low secondary ion yield could potentially be increased by the application of novel primary ion beams which enhance the protonation of molecular species. In the light of successful previous studies performed with (H₂O) _n⁺ clusters on biomolecular standards ^[4-6] we have explored the application of giant water clusters and water containing Argon clusters as analysis beams for tissue imaging.

A series of experiments performed with 20 keV C₆₀⁺_, Ar₂₀₀₀⁺, (H₂O) ₆₀₀₀⁺ and water-doped Ar₂₀₀₀⁺ were compared. We studied the secondary ion yields obtained from lipid standard samples such as DPPC and brain extract. Results showed an enhancement of secondary ion yield when the analysis beam is water-doped argon or (H₂O) _n⁺ clusters. There was specific enhancement of protonated molecular ions [M+H] ⁺ when the analysis was carried out with (H₂O) _n⁺ clusters.

The same four polyatomic ion beams were then applied to mouse brain tissue imaging. Images of the brain cerebellum were acquired on serial sections. More intact lipid ions of different classes were detected when the analysis beam contained water. Cholesterol was detected in both white and grey matter under (H₂O) ₆₀₀₀⁺ and water-doped Ar₂₀₀₀⁺ bombardment. This suggests a possible suppression effect on cholesterol across the grey matter area, which shows reduced signal when analysed with C₆₀⁺or Ar₂₀₀₀⁺. Moreover, certain protonated lipid species were only detected with (H₂O) _n⁺ clusters. These results suggest the potential benefits when (H₂O) _n⁺ clusters and water containing argon clusters are applied to tissue imaging.

[1] J. S. Fletcher, N. P. Lockyer, J. C. Vickerman, Mass Spectrom. Rev.2011,30, 142–174.

[2] B. C., Touboul, D. and A. Brunelle, (2014), Mass Spectrom. Rev., 33: 442–451. doi: 10.1002/mas.21399

[3] J. C. Vickerman, D. Briggs. TOF-SIMS: Materials Analysis by Mass Spectrometry. IM Publications and Surface Spectra Ltd., Manchester, UK, 2013.

[4] S. Sheraz (née Rabbani), A. Barber, J. S. Fletcher, N. P. Lockyer, J. C. Vickerman, Anal. Chem. 2013 , 85 , 5654 – 5658.

[5] S. Sheraz (née Rabbani)., A. Barber,, I . Berrueta Razo, J.S. Fletcher, N.P. Lockyer and J.C. Vickerman. ( 2014 ) Surf. Interface Anal., 46, 51 – 53 .

[6] I. Berrueta Razo, S. Sheraz, A. Henderson, N.P. Lockyer, and J.C. Vickerman ( 2014 ) Surf. Interface Anal., 46, 136 – 139 .

Breast cancer, the most common cancer among women, is known to vary in responsiveness to chemotherapy. Therefore, the role of changes in tumor metabolism affecting the response to chemotherapy is under scrutiny. Treatment for later-stage breast cancer includes pre-surgical administration of chemotherapeutics. Since treatment occurs with the tumor in place, analysis of biopsies taken pre- and post-treatment allows assessment of tumor response to treatment. The goal of this research is to define the molecular and cellular mechanisms underlying the tumor response to chemotherapy. The sub-micron spatial resolution imaging capability of ToF-SIMS provides a powerful approach to attain spatially-resolved molecular and cellular data from cancerous tissues not available with conventional imaging techniques.

In this work, we use imaging ToF-SIMS and principal component analysis (PCA) of tumor tissue sections from patients both pre- and post-treatment to characterize lipid pathways associated with tumor metabolic flexibility and response to chemotherapy. Analysis will aid in clarifying links between fatty acid composition within a tumor and potential response of that tumor to treatment.

Data were acquired on an IONTOF TOF.SIMS V using a Bi₃⁺ analysis beam in both high mass (HMR) and high spatial resolution (HSR) modes on two sets of pre- and post- therapy tissues. Three 1mm² areas per tissue section were analyzed by stitching together 200μm² raster area scans in HMR. Pre- and post-treatment tissues did not exhibit distinct differences in PCA when the entire raster area was included. We therefore developed a method to isolate regions by utilizing PCA of ToF-SIMS images to analyze specific tissue areas. The application of PCA to ion images separated cellularized and stromal regions as distinct chemistries. These PCA-generated images were then used as masks to reconstruct the representative spectra of that specific region. Stromal and cellular regions of different tissue samples were then spectrally compared with PCA resulting in chemical separation between pre- and post-therapy tissues using negatively-charged ion species. Higher variability remained present using positively charged species. Chemical differences between untreated and treated tissues were correlated to key fatty acids (such as palmitic, oleic, and stearic), monoacylglycerols, diacylglycerols and cholesterol. These results provide a new unsupervised perspective to isolate and interpret metabolic features of tumor regions within tissues. Identification of biomarkers by ToF-SIMS could be developed to monitor response to therapy and potentially improve treatment efficacy.

During the last decades, with the advent of novel and more targeted cancer therapeutic agents there is a need to better understand the mechanism of transport and delivery of drugs at the single cell level. Given that most drugs must reach an intracellular target in order to inhibit tumor growth or proliferation, the question arises as to how effectively they may diffuse into a cell and reach a subcellular organelle. With the recent development of high spatial resolution TOF-SIMS analytical tools (~300 nm with 25 keV Bi₃⁺) combined with soft sputtering beams ( 20keV Ar₁₅₀₀+), the tridimensional analysis of subcellular biological structures has become more attainable. Here, we present a novel experimental workflow and results of the use of confocal microscopy complemented with 3D TOF – SIMS analysis of single cells for the localization of the fate of two therapeutic agents for cancer treatment. PC9GR lung cancer cells were treated with CGM 097 and MLN2480 anti-tumor agents and incubate over time on a Au/Si substrate. After incubation, cell surfaces were washed and 3D TOF-SIMS analysis was performed. Preliminary results showed that CGM 097 incorporates preferentially to the cytoplasm (shown by the molecular fragment C₉H₉N₂O⁺) while MLN2480 incorporates preferentially to the nuclei region (shown by the molecular fragments C₆H₂NClF₃⁺ and C₇H₈N₄O Cl⁺). A labeling strategy was used for the localization of subcellular domains using fluorescent agents (e.g., DAPI, DiI, and DiO) during cell growth and their localization using 3D TOF-SIMS. Distinct secondary ion signals for the cell nuclei (C₄H₅N₂⁺ fragment) and for the cytoplasm (C₁₀H₁₀N⁺ and C₁₁H₁₂N⁺ fragments) were used to confirm the localization of DAPI (DNA stain) and DiI stains.

In the last few decades the pharmaceutical industry has transformed people’s lives, creating medicines that either cure diseases or in the worst case scenario transform death sentences into chronic diseases. However, the development of new drugs is becoming increasingly challenging as a result of several variables including increased difficulty of the biological targets, market demands, diminishing returns (medicines per capita invested) and strict safety requirements. A paradigm shift in the drug discovery workflow is required to reduce attrition and transform typical drug screening assays into translatable analytical techniques for the analysis of drugs in complex environments, both in-vitro and ex-vivo. As the majority of drug targets are intracellular several promising analytical techniques are being developed to study drug uptake and target engagement within cells. The ability to visualize unlabelled compounds inside the cell at physiological dosages can offer valuable insight on the compound behaviour both on and off-target. In this study, we show evidence of unlabelled pharmaceuticals inside both mammalian and bacterial cells using ToF-SIMS. Two model systems were investigated; amiodarone in mammalian cells and a bactericide in a tuberculosis model.

Amiodarone is an anti-arrhythmic drug with well-known toxicity owing to accumulation in the lysosomes of macrophages causing phospholipidosis. For this reason, it was used as a model compound for the mammalian part of the study. We show that it can be detected in 4 distinct mammalian cell lines (NR8383, Hek293, HeLa and HepG2) at therapeutic levels.

Antimicrobial resistance is of great societal concern and there is an urgent need to develop new drugs. A major issue is achieving sufficient uptake of the drug through the bacterial membrane and also overcoming membrane proteins pumping out the drug. Therefore there is a growing need to image the drug within a single bacterial cell. This poses both sensitivity and spatial resolution challenges due to the size of the microorganism, about 2 mm by 4 mm. Here, we study a Mycobacterium tuberculosis (TB) model (mycobacterium bovis BCG) seeing that, according to WHO (World Health Organization) surveys, one third of the world population has latent TB and antibiotic resistant strains are rapidly emerging. This study shows, for the first time, label-free ToF-SIMS images of the drug molecule inside bacteria. This understanding combined with optical microscopy is being used to investigate the mode of action of these bactericides.

The potential and future challenges for ToF-SIMS imaging of drug uptake into cells will be discussed.

Understanding the distribution of a dosed drug within the body is a key part of the drug development process, both in terms of drug efficacy (does the drug reach the target area and at sufficient concentration to have a pharmacological effect?) and potential side effects (where else does the drug go and what does it do there?). Oral administered drug delivery is the largest fraction (38%) of the pharmaceutical market and permeability through the intestine plays a key role in determining the fraction of a drug absorbed. The intestine is an organ with a complex 3D structure at the microscale. In the small intestine, finger-like villi structures protrude into the central intestinal lumen. The villi are lined by an epithelial membrane mainly consisting of adsorptive enterocytes. Below the epithelial layer is the lamina propria containing white blood cells, tissue macropharges and a rich vascular and lymphatic network into which digestive products are absorbed. Mass spectrometry imaging (MSI) provides a powerful route for label-free observation of the spatial distribution (including depth) of key chemical entities in tissue. Imaging via Matrix Assisted Laser Desorption Ionisation (MALDI) MSI is proving to be an influential tool for profiling unlabelled compounds directly in biosamples. However, using MALDI it is not possible to ascertain that an unlabelled drug is within a specific cell type. We therefore explored the use of ToF-SIMS and show, for the first time, the 3D distribution of a drug candidate at better than cellular resolution in the small intestine of mouse and rat following oral administration in comprehensive animal studies.

SIMS imaging demonstrated the highest compound signal was observed in the lamina propria of the rat as areas of compound aggregation, in contrast to the mouse where the candidate drug was observed at higher levels in the enterocytes. Compound distribution at 24 hours post-dose following repeat administration showed that drug was still present as focal aggregates in the lamina propria of the rat, while in mouse no drug was detected. In addition, light microscopy of adjacent haematoxylin and eosin stained tissue sections indicated the presence of foamy macrophages in the lamina propria of rats but not of mice. The high, localised concentration of drug in the rat correlated well with the location of the foamy macrophages. The SIMS data have provided unparalleled information on drug distribution in the small intestine of rodents, indicating uptake into enterocytes and cells within the lamina propria. The new insights gained and the future prospects for the application of SIMS in drug discovery and development will be discussed.

All animal studies were ethically reviewed and carried out in accordance with Animals (Scientific Procedures) Act 1986 and the GSK Policy on the Care, Welfare and Treatment of Animals.

Biological SIMS imaging has been extensively developed for last several years. Major breakthroughs in biological SIMS imaging are based on cluster ion beams such as metal clusters, C60, Ar gas clusters. However, the secondary molecular ion yields are not sufficiently high enough for routine biologcial SIMS imaging applications in biomedical applications. Improvement of secondary ionization probability by orders of magnitudes is mandatory for generating breakthroughs in biological SIMS imaging, which is expected in the near future.

An additional major problem of biological SIMS imaging is the lack of protein and peptide imaging. Most of biological story tellings are mianly based on proteins. The biological implication of lipids and metabolites SIMS imaging would be much higher if protein imaging is provided toghether. Utilizing high secondary ion yields of metals, proteins can be SIMS imaged with metal tagged proteins. Rare earth metals such as Yb, Sm, Pr, Tm, Eu were used for imaing NeuN and HUC proteins in neuron cells in hypothalamus mouse tissues with a spatial resolution of ~2 μm using a TOF-SIMS. Lipids and neurotransmitters images obtained simultaneously with protein images were overlayed for more deeper understanding of neurobiology, which is not allowed by any other bioimaging technqiues. The protein images from TOF-SIMS were compared with confocal fluorescence miscroscopy and nanoSIMS images.

In biological SIMS imaging, it is quite often disregarded that SIMS is extremely surface sensitive. To image single cell membranes in a tissue, we adopted the vibrotome technique to prepare a tissue slice without any fixation and freeze- drying. After appropriate tissue culture, a cell membrane terminated tissue surface was obtained, 3 different types of neuronal cells from a hypothalamus were identified.

In –vivo imaging is of paramount importance in biologcial imaging. To investigate the intrinsic biology and dynamic responses of live cells and tissues in the cellular level, development of an ambient imaging mass spectrometry system with ~5 μm spatial resolution will be reported with preliminary imaging for colonies of HCT-8 cells and hypothalamus tissues.

If SIMS can provide molecular omic imaging including proteins of tissues with single cell identification, I believe we can claim much stronger bioimplications on biomedical applications on screening drugs and understanding fundamental causes of various diseases.

Cluster SIMS has proven to be a useful tool for biological studies involving 2D & 3D imaging of endogenous and exogenous markers in biosamples. However, low ionization probability and beam induced damage still hinder further applications ^[1]. The introduction of the Ar gas cluster ion beam has greatly reduced the build-up of chemical damage induced by the impact of the ion beam to biosamples ^[2]. Moreover, our recent work shows that chemical tuning of the Ar cluster projectile yields a higher intensity of secondary ions when mixed clusters, created by doping with small gas molecules such as CH₄, O₂, CO₂ and HCl, are used ^[3]. Interestingly, HCl doped Ar clusters show ionization enhancement for [M+Cl]^- but not for [M+H]⁺. A more favorable environment is created by depositing a thin layer of ice on the surface of the sample to aid in the disassociation of HCl, resulting in the enhancement of [M+H]⁺, a phenomenon we call dynamic reactive ionization (DRI) ^[4].

Here, this methodology is applied to the imaging of several biomolecules and biological tissue. Varying enhancement is shown for [M+H]⁺ and [M+Cl]^-, especially for lipids. Depth profiling is also explored by dynamically depositing water onto the sample surface to create a more favorable environment for DRI. It is promising that optimized conditions (e.g., the percentage HCl dopant, ice layer thickness) can be applied to a wide range of biosamples to maximize chemical sensitivity.

[1] N. Winograd, Analytical Chemistry 2005, 77, 142a.

[2] H. Gnaser, M. Fujii, S. Nakagawa, T. Seki, T. Aoki, J. Matsuo, Rapid Communications in Mass Spectrometry 2013, 27, 1490.

[3] A. Wucher, H. Tian, N. Winograd, Rapid Commun Mass Spectrom 2014, 28, 396.

[4] H. Tian, A. Wucher, N. Winograd, SIMS XX abstract 2015, submitted.

Detecting drug compounds and metabolites within a cell type is now a priority for pharmaceutical development. In this context, three-dimensional secondary ion mass spectrometry (SIMS) imaging was used to investigate the cellular uptake of amiodarone, a phospholipidosis-inducing pharmaceutical compound. The high lateral resolution and 3D imaging capabilities of SIMS combined with the multiplex capabilities of ToF mass spectrometric detection allows for the visualization of multiple compounds (e.g. pharmaceuticals, metabolites, endogenous molecules) on the single cell level. The intact, unlabeled drug compound was successfully detected at therapeutic dosages in macrophages (cell line: NR8383). Chemical information from endogenous biomolecules was used to correlate drug distributions with morphological features. From this spatial analysis, amiodarone was detected throughout the cell with the majority of the compound found in the membrane and subsurface regions of the cell and absent in the nuclear regions. Similar results were obtained when the macrophages were doped with amiodarone metabolite, desethylamiodarone. The FWHM lateral resolution measured across an intracellular interface in a high lateral resolution ion images was approximately 550 nm. Overall, this approach provides the basis for studying cellular uptake of pharmaceutical compounds and their metabolites on the single cell level.

NanoSIMS imaging makes it possible to visualize stable isotope-labelled lipids in cells and tissues at a scale below 100 nm. In this presentation, we will discuss the use of NanoSIMS imaging to visualize lipids in mouse cells and tissues.

After administering stable isotope-labelled fatty acids to mice, various tissues were processed and 500 nm sections were mounted on conductive substrates for analysis. NanoSIMS imaging allowed us to visualize neutral lipids in cytosolic lipid droplets in intestinal enterocytes, chylomicrons at the basolateral surface of enterocytes, and lipid droplets in cardiomyocytes and adipocytes. After an injection of stable isotope-labelled triglyceride-rich lipoproteins (TRLs), NanoSIMS imaging revealed that lipids were delivered to cytosolic lipid droplets in parenchymal cells in less than an hour. Using a combination of backscattered electron (BSE) and NanoSIMS imaging, it was possible to correlate high-resolution cell morphology information from BSE images with chemical information provided by the NanoSIMS on the surface of the same section. This combined imaging approach allowed us to visualize stable isotope-enriched TRLs along the luminal face of heart capillaries and the lipids within heart capillary endothelial cells. We also observed examples of TRLs within the subendothelial spaces of heart capillaries.

NanoSIMS imaging provided evidence of defective transport of lipids from the plasma LPs to adipocytes and cardiomyocytes in mice deficient in glycosylphosphatidylinositol-anchored HDL binding protein 1. Other potential applications of NanoSIMS analysis on lipid metabolism will also be discussed.