MALDI MS Tissue Imaging of Crystallins using an original metyhod to direct protein identification on lens slices

The lens is a transparent, biconvex structure in the eye that, along with the cornea, helps to refract light to be focused on the retina. Crystallins, α, β and γ, are the predominant structural proteins in lens. They constitute 90% of water soluble proteins and contribute to its transparency and refractive properties by a uniform concentration gradient in the lens. Nevertheless, if these crystallins undergo post translational modifications, they become less soluble and the opacity of eye lens increases. This phenomenon defines cataract. Yet, the nature and the mechanism of occurring of these modifications and how they happen are not fully understood.
MALDI mass spectrometry imaging is a recent technique allowing examining proteins in their native location without the need for traditional processing methods such as extraction, homogenization, and separation. Nevertheless, one main difficulty lies in the identification of the detected species, especially proteins. MALDI-In Source Decay (MALDI-ISD) is a fragmentation process occurring in the mass spectrometer ion source. When the analyzed sample is a protein, ISD fragmentation leads to b-, c- and z-ions series, which allows for some sequencing of the protein. One great advantage of ISD is its fastness and easiness to be implemented since there is no need for a special treatment of the sample. The only requirement is the use of “ISD-favourable MALDI matrix” such as 2,5-dihydroxybenzoic acid or 1,5-diaminonaphtalene.
18 µm-thick equatorial sections of frozen porcine eye lenses were realized with a cryostat. 1,5-DAN matrix was either manually deposited or sprayed with an ImagePrep automated device (Bruker Daltonics). Data were acquired with an UltraFlex II MALDI-TOF/TOF mass spectrometer (BD) in positive reflector mode. For imaging experiments, the surface of the sample was divided into 100-µm-wide pixels and 500 shots were averaged on each. Based on calculated mass differences between consecutive ISD fragments peaks, tags of amino acids were established and submitted to a search in protein databases using a BLAST algorithm (search by sequence homology).
Imaging experiments showed that the localization information may be very useful to associate fragments which exhibit close distributions, suggesting they are originating from the same protein. It is thus possible to arrange fragments in groups of probable origin and to extract the mass spectrum of a high-intensity pixel. This allows to work with a “purified” ISD mass spectrum where fragments of only one protein are present and potentially exhibiting a higher number of peaks, leading to a longer tag and to an easier identification. With this imaging strategy, we were able to identify (by homology) the Beta-Crystallins S and B2, the Gamma-Crystallin B, the Alpha-Crystallin A.