Using imaging secondary ion mass spectrometry to determine mammalian plasma membrane component distribution and cell differentiation state

Frisz, Jessica

Permalink

https://hdl.handle.net/2142/32025 Copy

Description

Title

Using imaging secondary ion mass spectrometry to determine mammalian plasma membrane component distribution and cell differentiation state

Author(s)

Frisz, Jessica

Issue Date

2012-06-27T21:28:45Z

Director of Research (if dissertation) or Advisor (if thesis)

Kraft, Mary L.

Doctoral Committee Chair(s)

Kraft, Mary L.

Committee Member(s)

• Sweedler, Jonathan V.
• van der Donk, Wilfred A.
• Schroeder, Charles M.

Department of Study

Chemistry

Discipline

Chemistry

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Secondary ion mass spectrometry (SIMS)
• Plasma membrane
• Sphingolipids
• Cholesterol
• Time-of-flight SIMS
• Multivariate analysis
• Differentiation

Abstract

The organization of the plasma membranes of mammalian cells has been studied for over forty years, with our understanding evolving from the fluid mosaic model of Singer and Nicholson to a more nuanced model acknowledging lateral heterogeneity in protein and lipid distributions. Within this area of research, lipid rafts have received intense focus, perhaps because they have defied easy characterization. They are thought to be sphingolipid- and cholesterol-enriched features tens to hundreds of nanometers in size dependent upon cholesterol, although their composition and size are still debatable as are other characteristics such as their lifetime, location, percent coverage of the membrane, and their mechanisms of formation and maintenance. The capability of chemically assaying the lipid composition at relevant size regimes has been unattainable by any single technique until the recent application of high resolution imaging mass spectrometry to biological samples. Using a high resolution imaging secondary ion mass spectrometry instrument (NanoSIMS 50, Cameca), we have chemically mapped the distributions of labeled sphingolipids and cholesterol within the plasma membrane of murine fibroblasts with sub-100 nm resolution. This required the incorporation of non-perturbing heavy isotope labels into lipid molecules of interest and sample preparation that preserved native lipid organization. These experiments have revealed sphingolipid-enriched domains with an average size of 200 nm that adopt a non-random clustering into micron-scale patches in the membrane of fibroblast cells. Analysis of the cholesterol distribution reveals a relatively homogeneous distribution throughout the membrane that does not colocalize with sphingolipid-enriched domains. Depletion of cholesterol alters the surface coverage of domains and their long-range organization, but does not affect average microdomain size or short-range organization. These results are contrary to the lipid raft hypothesis, which states that lipid rafts are biologically relevant to cell function, they are defined by enrichment of both cholesterol and sphingolipids, and their formation is driven by the interaction of these two molecules. These results contribute significantly to our understanding of membrane organization and may reshape the conception of lipid distributions within mammalian plasma membranes. Further studies using this methodology to chemically image mammalian cell membranes may also elucidate the mechanisms that produce lipid heterogeneity and the purposes these domains serve in the cell. Imaging mass spectrometry, in conjunction with multivariate analysis, has also been applied to the analysis of cell membranes for the purpose of classifying cell type or differentiation state. Cells display a characteristic combination of proteins, lipids, and glycans based on their origin or developmental state. Time-of-flight (TOF) SIMS samples the full mass spectrum, to which multivariate analysis such as principle component analysis (PCA) or partial least squares discriminant analysis (PLS-DA) can be applied. Such analysis reveals characteristic patterns of peak intensities rather than single ions, which can be used to classify or predict cell type. Using this approach, we were able to predict the differentiation state of primary hematopoietic cells harvested from mice with better than 88% accuracy when the cells in the training set and the test set were from mice of similar age and a restricted peak set including only ions known to originate from biomolecules was utilized to build the PLS-DA model. We were also able to distinguish, with some success, murine fibroblast NIH-3T3 cells that differed only in the amount of sialic acid expressed in the membrane. This technique will be of use for characterizing cells grown on substrates that exhibit spatial variations intended to guide particular stem cell fates, where bulk measurements are inadequate and location-specific information about cell identity is needed.

Graduation Semester

2012-05

Permalink

http://hdl.handle.net/2142/32025

Copyright and License Information

Copyright 2012 Jessica Faith Frisz

Owning Collections