Collagenase penetrates human pancreatic islets following standard intraductal administration.
Cross SE., Hughes SJ., Partridge CJ., Clark A., Gray DW., Johnson PR.
BACKGROUND: : To optimize human islet isolation, it is important to improve our understanding of the collagenase digestion phase. Previous studies of collagenase action were mostly concerned with optimizing its composition, but the delivery and distribution of collagenase at the islet-exocrine interface is likely to be important for liberation of intact islets. The aim of this study was to characterize collagenase distribution in relation to islets in infused human pancreases. METHODS: : Human pancreases were retrieved from multiorgan donors with appropriate consent. Tissue samples were taken from the neck, body, and tail regions before and after collagenase infusion by manual syringe-loading (n=10) or recirculating perfusion (n=8), and snap frozen in liquid nitrogen. Frozen sections were immunolabeled for collagenase, insulin, CK19, collagen VI and CD31, then assessed by confocal microscopy. RESULTS: : Collagenase labeling was widespread throughout the pancreas, associated with collagen VI, and adjacent to CK19-labeled ducts. Collagenase was found within 67%+/-2% of islets ("intraislet"), associated with capillaries (CD31-positive). Intraislet collagenase was observed in 70%+/-3% of islets in the pancreatic tail, compared with 58%+/-2% and 53%+/-2% of islets in the body and neck, respectively (P<0.05 tail vs. neck), and was more prevalent in islets with diameters more than 150 microm (98%+/-1% of islets >150 microm vs. 52%+/-2% of islets <150 microm, P<0.05). There was no difference in intraislet collagenase labeling between perfused and syringe-loaded pancreases. CONCLUSIONS: : Using current infusion techniques, collagenase penetrates the islet interior. This could cause islet fragmentation, and consequently, low islet yields. This study underlies the need to optimize collagenase delivery to preserve intact islets.