Laser capture microdissection coupled mass spectrometry (LCM-MS) for spatially resolved analysis of formalin-fixed and stained human lung tissues

• Published in: Clinical proteomics Vol. 17; no. 1; p. 24
• Main Authors: Herrera, Jeremy A., Mallikarjun, Venkatesh, Rosini, Silvia, Montero, Maria Angeles, Lawless, Craig, Warwood, Stacey, O’Cualain, Ronan, Knight, David, Schwartz, Martin A., Swift, Joe
• Format: Journal Article
• Language: English
• Published: London BioMed Central 17.06.2020; BioMed Central Ltd
• Subjects: Biomedical and Life Sciences; Biotechnology; Cell Biology; Formaldehyde; Life Sciences; Mass spectrometry; Pathology; Proteins; Proteomics; Respiratory tract diseases; Scientific equipment industry; United States
• ISSN: 1542-6416; 1559-0275
• DOI: 10.1186/s12014-020-09287-6

Background Haematoxylin and eosin (H&E)—which respectively stain nuclei blue and other cellular and stromal material pink—are routinely used for clinical diagnosis based on the identification of morphological features. A richer characterization can be achieved by laser capture microdissection coupled to mass spectrometry (LCM-MS), giving an unbiased assay of the proteins that make up the tissue. However, the process of fixing and H&E staining of tissues provides challenges with standard sample preparation methods for mass spectrometry, resulting in low protein yield. Here we describe a microproteomics technique to analyse H&E-stained, formalin-fixed paraffin-embedded (FFPE) tissues. Methods Herein, we utilize heat extraction, physical disruption, and in column digestion for the analysis of H&E stained FFPE tissues. Micro-dissected morphologically normal human lung alveoli (0.082 mm 3 ) and human lung blood vessels (0.094 mm 3 ) from FFPE-fixed H&E-stained sections from Idiopathic Pulmonary Fibrosis (IPF) specimens (n = 3 IPF specimens) were then subject to a qualitative and then quantitative proteomics approach using BayesENproteomics. In addition, we tested the sensitivity of this method by processing and analysing a range of micro-dissected human lung blood vessel tissue volumes. Results This approach yields 1252 uniquely expressed proteins (at a protein identification threshold of 3 unique peptides) with 892 differentially expressed proteins between these regions. In accord with prior knowledge, our methodology approach confirms that human lung blood vessels are enriched with smoothelin, CNN1, ITGA7, MYH11, TAGLN, and PTGIS; whereas morphologically normal human lung alveoli are enriched with cytokeratin-7, -8, -18, -19, 14, and -17. In addition, we identify a total of 137 extracellular matrix (ECM) proteins and immunohistologically validate that laminin subunit beta-1 localizes to morphologically normal human lung alveoli and tenascin localizes to human lung blood vessels. Lastly, we show that this micro-proteomics technique can be applied to tissue volumes as low as 0.0125 mm 3 . Conclusion Herein we show that our multistep sample preparation methodology of LCM-MS can identify distinct, characteristic proteomic compositions of anatomical features within complex fixed and stained tissues.