Sylwia Stopka (Presenter)
The George Washington University
Bio: Currently I am a chemistry graduate student at the George Washington University in the research group of Professor Akos Vertes. My research entails developing new mass spectrometry (MS)-based technologies for biomedical analysis, ranging from atmospheric pressure ionization methods to laser desorption ionization (LDI) from nanostructure substrates called silicon nanopost arrays (NAPA). This new matrix-free LDI technique allows for ultra-trace analysis of volume-limited samples (i.e., single biological cells) with extreme sensitivity. Here I am expanding this technology to MS imaging of tissue sections and small cell population analysis on these NAPA substrates. This work shows significant promise for drug distribution studies in tissue sections, and ultimately for exploring cellular heterogeneity by single cell analysis.
Authorship: Sylwia A Stopka (1), Charles Rong (1), Andrew R. Korte (1),Trust T. Razunguzwa (2), Nicholas J. Morris (2), Javad Nazarian (3) and Akos Vertes (1)
(1) The George Washington University, Washington, DC (2) Protea Biosciences Inc., 1311 Pineview Drive, Suite 501, Morgantown, WV (3) Children’s National Medical Center, Washington, DC
| Short Abstract Mass spectrometry imaging (MSI)-based techniques provide versatile spatially-resolved chemical information directly from tissues. Matrix-assisted laser desorption ionization is the dominant technique in MSI, but due the narrow dynamic range of quantitation and the spectral interferences, it is poorly suited for the MSI of small molecules. Here we present the first evidence for a new matrix-free MSI technique using nanophotonic laser desorption ionization (LDI) from a silicon nanopost array (NAPA). In mouse brain and kidney, over ions were mapped and correlated to anatomical features. Additionally, human and microalgae cells were cultured and analyzed on these silicon substrates, allowing correlation of cell number within a given pixel to the corresponding ion intensities. These results demonstrate the utility of LDI-MSI from NAPA for tissue sections and small cell population analysis. |
Long Abstract
Conventional analytical techniques, including autoradiography and histochemical staining, have been widely accepted for the spatial mapping of predefined biomolecules or xenobiotics in biological tissue sections. These targeted analyses, however, require radioactive labels or reporter molecules, ultimately limiting the number of analytes detectable within a single experiment.[1] In recent years, mass spectrometry imaging (MSI)-based techniques have gained ground as a comprehensive method for the detection, localization, and relative quantitation for hundreds of compounds simultaneously. The mainstream MSI method, matrix-assisted laser desorption ionization (MALDI), has primarily been used to explore the spatial distribution of large molecules (e.g., proteins and peptides) in tissue sections.[2] Small-molecule MSI for drugs and metabolites is receiving increasing interest, however, imaging these molecules with MALDI remains a challenge due to the narrow dynamic range and matrix-related spectral interferences present at < 500 Da.
To overcome these limitations, several matrix-free LDI techniques have been developed, but only a handful have been applied to MSI.[3] [4] In previous contributions we have introduced silicon nanopost arrays (NAPA) that exhibited a wide dynamic range spanning > 3 orders of magnitude and extreme sensitivity for volume-limited samples, making this technique amenable for single cell analysis and tissue imaging.[5] Using this technique we uncovered metabolic changes due to oxidative stress perturbations in single yeast cells using LDI from NAPA. [6] Precise nanofabrication techniques (e.g., deep UV projection lithography) produce NAPA substrates that are extremely uniform, in principle enabling sub-micrometer spatial resolution. [7]
In this contribution, we present the first application of LDI-MSI from silicon NAPA substrates for the spatial mapping of metabolites and lipids directly from tissue sections. We also show how NAPA-MSI can be applied to cells cultured directly on the silicon substrate. This analysis can ultimately bridge the gap between MSI and single cell analysis.
To explore the potential of LDI-MSI using NAPA, coronal brain and sagittal kidney sections from mice were imaged on NAPA platforms and several metabolite and lipid distributions were linked to anatomical features. Limited sample preparation was required for analysis. Brain and kidney tissues were sectioned to 10 µm thickness, directly mounted on the NAPA substrate, and dried. The LDI-MSI was performed using an Orbitrap LTQ-XL. Due to the thickness of the sample, a laser fluence of ~100 mJ/cm2 was needed, with 10 laser shots per pixel. The tissue sections were oversampled, with a step size of 50 μm and a laser spot size of ~80 µm x 100 µm. For on-chip growth, human hepatocytes (Hep G2/C3A) were cultured directly on the NAPA substrate, and were maintained in supplemented Eagle’s Minimum Essential Medium. Wild type Chlamydomonas reinhardtii were also cultured in tris acetate phosphate medium and the cells were directly loaded onto the NAPA substrate and allowed to air dry. Due to the thinner nature of both the human and microalgae cells, a lower laser fluence of ~60 mJ/cm2 was used for imaging experiments.
Ion distributions in mouse brain LDI-MSI from NAPA showed over 80 ions in both positive and negative ion mode to be localized to specific regions of the brain. For example, the ion signals assigned as ST(24:1) at m/z 888.6258 and gondoic acid at m/z 309.2795 were mainly detected from the corpus callosum (CC) and the anterior commissure (aco) while the ion PE(34:0) at m/z 718.5453 was only present in the caudoputamen (CP) and the cortex. To further expand the molecular coverage, LDI-MSI experiments were also performed at increased mass resolution; increasing the set mass resolution from 30,000 to 100,000 resulted in an additional ~100 spectral components. This allowed for the mapping of lipid species with Δm of 20 mDa, some of which presented complementary spatial distributions. In sagittal mouse kidney sections, the ion signal at m/z 616.1771 assigned as heme b was only detected within vascular features. In contrast, cholesterol at m/z 369.3516 was detected across the whole tissue section, whereas glucose at m/z 203.0525 was only detected in the renal medulla. Neutral lipids, e.g., TAG(54:6) at m/z 897.7445, were mainly localized to the renal cortex and hilum.
To probe heterogeneity within small cell populations, we detected metabolites from Hep G2/C3A cells cultured on NAPA substrates for six days. Imaging by SEM indicated that the cells attached to the nanoposts and developed extensive lamellipodia that were significantly thinner than the bodies of the cells. Upon LDI from NAPA at a relatively low laser fluence of ~24 mJ/cm2, only 13 sample related ions were observed in the spectra. Subsequent observation by SEM revealed that desorption of material only occurred from the lamellipodia. At an elevated laser fluence of ~48 mJ/cm2, the number of detected components increased to 110, and SEM images revealed that material was desorbed from both the lamellipodia and the cell bodies. Within the lamellipodia, phosphocholine, cholesterol, glutathione, TAG(52:2), and PE(38:4) were identified.
To explore the potential of MSI for high throughput single cell analysis, C. reinhardtii cells were deposited onto the NAPA substrate and imaged by LDI. The pixels of 50 µm in diameter were correlated to specific numbers of cells, by overlaying the optical and chemical images. This way for every MSI pixel a cell number could be rendered. The spatial distribution of DAG(40:4), DGTS(32:0), chlorophyll a, and chlorophyll b signal intensities were correlated with the sizes of cell populations residing in each MSI pixel. A proportionality between population size and signal intensity was observed in the 20 to 100 cell range for all four compounds.
These results demonstrate the first applications of molecular imaging using nanophotonic LDI-MS from silicon NAPA substrates for biological tissue sections and small cell populations. The highly uniform NAPA structures enable the LDI imaging of small molecules and lipids with minimum interference from the substrate.
References & Acknowledgements:
References
[1] A. McEwen, C. Henson, Bioanalysis 2015, 7, 557-568.
[2] D. S. Cornett, M. L. Reyzer, P. Chaurand, R. M. Caprioli, Nature Methods 2007, 4, 828-833.
[3] T. R. Northen, O. Yanes, M. T. Northen, D. Marrinucci, W. Uritboonthai, J. Apon, S. L. Golledge, A. Nordstrom, G. Siuzdak, Nature 2007, 449, 1033-U1033.
[4] B. N. Walker, J. A. Stolee, D. L. Pickel, S. T. Retterer, A. Vertes, Journal of Physical Chemistry C 2010, 114, 4835-4840.
[5] B. N. Walker, J. A. Stolee, A. Vertes, Analytical Chemistry 2012, 84, 7756-7762.
[6] B. N. Walker, C. Antonakos, S. T. Retterer, A. Vertes, Angewandte Chemie-International Edition 2013, 52, 3650-3653.
[7] N. J. Morris, H. Anderson, B. Thibeault, A. Vertes, M. J. Powell, T. T. Razunguzwa, RSC Advances 2015, 5, 72051-72057.
| Description | Y/N | Source |
| Grants | yes | DE-FG02-01ER15129 |
| Salary | no | |
| Board Member | no | |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |