Aim: Characterisation of the pancreatic extracellular matrix (ECM) is important for optimising the digestion phase of human islet isolation. Raman spectroscopy is a laser based technology which exploits the phenomenon of inelastic light scattering. This allows components of complex heterogeneous samples to be identified based on a specified wavelength shift of incident light, which is proportional to the vibrational frequency of specific molecular bonds. Raman has previously been used to characterise ECM structures in a variety of tissues. Here we aim to determine if it is feasible to utilise Raman microspectroscopy to study the ECM at the islet-exocrine interface, by discerning if it spectrally distinct from the rest of the pancreatic tissue.
Methods: With appropriate consent and ethical approval, 0.5cm³ biopsies were taken from the anatomical head of donor pancreases received for clinical islet isolation (CIT <10hrs) and snap frozen. The tissue was then thawed and fixed overnight in 4% PFA, prior to sectioning onto gelatine coated coverslips. Sections underwent immunofluorescent (IF) staining for insulin and collagen I. A confocal microscope, coupled to the Raman spectrometer, was used to direct the Raman laser to three morphologically distinct areas of tissue within each biopsy, as determined by the IF staining: a) the intra-insulin stained region, b) the islet-exocrine interface, and c) the general exocrine pancreas. Raman measurements (n=10 for each of the three regions) were taken with an integration time of 100 seconds, with a reference background spectra (cover glass alone) subtracted after every tenth measurement. Principle component analysis (PCA) was used to determine statistical differences and similarities between the data sets for the three regions measured. ANOVA was used to determine statistical differences between specific peaks generated for each of the three areas of measured tissue.
Results: PCA demonstrated that there was distinct cluster separation for the spectral measurements generated from the islet-exocrine interface. This allowed the islet-exocrine interface to be significantly distinguished from the other two measured tissue regions. The significant differences were accounted to the 1200-1400^cm-1 fingerprint region of the spectra, with specific reference to a peak present at 1271^cm-1, which was significantly more prominent within the islet-exocrine interface (p<0.05, as compared to the other two tissue regions). By using a purified protein reference spectra, this area was determined to be a region specific for collagens.
Conclusion: Raman microspectroscopy is successfully able to discriminate the region encompassing the islet-exocrine interface from the rest of the pancreatic tissue. Providing that the correct purified protein reference standards are used, this study provides preliminary data demonstrating the feasibility of using Raman microspectroscopy as a novel, non-invasive strategy to characterise the islet-exocrine interface.