Examination of lipid distributions in hydrogel-expanded mouse brain tissue using imaging mass spectrometry

Imaging is essential in biological research, and imaging mass spectrometry uniquely provides a label-free approach with high molecular specificity. However, imaging mass spectrometry is limited in spatial resolution, which in turn limits the biological structures that can be studied. Custom lens setups and altered optical paths can shrink the diameter of the incident laser beam probe in matrix-assisted laser desorption/ionization (MALDI) imaging mass spectrometry to achieve high spatial resolutions (<5 μm). However, these research-grade instruments are complex and expensive, making high spatial resolution imaging experiments unrealistic for the broader community. There is a need for inexpensive and widely-accessible means for high spatial resolution imaging.

An alternative method for improving spatial resolution is through physical magnification of the substrate, which has been well established in the subfield of expansion microscopy (ExM). ExM leverages superabsorbent hydrogels for isotropic expansion of tissues and retention of fluorescently labeled analytes. While typical ExM involves covalently anchoring analytes to the hydrogel network, lipid retention without anchoring has been recently demonstrated for imaging mass spectrometry. Herein, we demonstrate expansion imaging mass spectrometry (ExIMS) of expanded brain tissue and examine lipid distributions in both positive and negative ion modes across multiple tissue structures. A linear expansion factor of 4.5-fold is achieved and used to obtain high spatial resolution images of mouse brain cerebellum. Approximately 75 % of lipids in both positive and negative ion modes are detected in expanded tissue compared to unexpanded tissue. Additionally, the majority of lipid distributions across the brain are maintained post-expansion, though lipid delocalization can occur for some lipids in the hippocampus and the granular layer of the cerebellum. Alterations to the hydrogel formulation can significantly affect the ability of ExIMS to maintain accurate lipid distributions in expanded tissue. For example, stronger fixation can better maintain lipid distributions, while gentler digests can better maintain lipid distributions at higher tissue expansion factors.

This work provides a methodical analysis of the effect of hydrogel embedding and expansion on lipid distributions across multiple tissue structures, ion polarities, and lipid classes. Computational tools are used to assess the spatial fidelity of lipid distributions prior to and following tissue expansion, which begins to assess the interaction between lipid delocalization and expansion parameters.