Spatial differences in an integral membrane proteome detected in laser capture microdissected samples

Abstract

The combination of laser capture microdissection and mass spectrometry represents a powerful technology for studying spatially resolved proteomes. Moreover, the compositions of integral membrane proteomes have rarely been studied in a spatially resolved manner. In this study, ocular lens tissue was carefully dissected by laser capture microdissection and conditions for membrane protein enrichment, trypsin digestion, and mass spectrometry analysis were optimized. Proteomic analysis allowed the identification of 170 proteins, 136 of which were identified with more than one peptide match. Spatial differences in protein expression were observed between cortical and nuclear samples. In addition, the spatial distribution of post-translational modifications to lens membrane proteins, such as the lens major intrinsic protein AQP0, were investigated and regional differences were measured for AQP0 C-terminal phosphorylation and truncation.

Figure 1

A diagram of a lens that indicates the anterior, equatorial, and posterior sections used for laser capture microdissection. Four regions of the lens, shaded in black, were the captured and subjected to proteomic analysis.

Figure 2

Images of the laser capture microdissected bovine lens anterior cortical tissue. A) H&E stained tissue before LCM, B) optical image of tissue section after laser capture but before tissue removal, C) optical image of tissue section after LCM and tissue removal, and D) optical image of captured tissue on the LCM cap.

Figure 3

Summary of the cellular localization of the identified proteins. 136 proteins identified in the 12 samples were sorted based on the known cellular localization listed in the Swiss-Prot database annotation.

Figure 4

Base peak chromatograms from LC-MS/MS analysis. Base peak chromatogram indicating major AQP0 tryptic peptide signals from: A) anterior cortex laser capture microdissected sample and B) lens homogenate from manually dissected anterior cortex. Major AQP0 and connexin 50 peptides are highlighted in bold font.

Figure 5

The spatial distribution of AQP0 C-terminal phosphorylation. The relative percentage of phosphorylation in the different regions of the lens are expressed by the ratio of phosphorylation. The ratio of S235 phosphorylation = peak area of phosphorylated 229–238/ peak area of unphosphorylated 234–238; the ratio of phosphorylation on S243 or S245= peak area of mono-phosphorylated 239–259/ peak area of unphosphorylated 239–259. The ratios of phosphorylation are plotted: (A) S235 phosphorylation and (B) S243 or S245 phosphorylation. A single asterisk (*) indicates significance compared to the anterior outer cortex. A double asterisk (**) indicates significance compared to the equator outer cortex (p< 0.01). AC, anterior cortex; EC, equatorial cortex; PC, posterior cortex; N, nucleus.

Figure 6

C-terminal truncation of AQP0 in bovine lens. C-terminal truncation of AQP0 was rarely detected in the outer cortex samples (not shown), but was detected in all of the nucleus samples (gray bars). The relative level of truncation at each site was calculated as the ratio of the peak area of each truncated peptide to the peak area of peptide 239–259.