Daniel Brandhorst,1,2 Heide Brandhorst,1,2 Samuel Acreman,1,2 Yukari Kimura,1,2 Shannon Layland,3 Katja Schenke-Layland,3 Paul R.V. Johnson1,2
1 Research Group for Islet Transplantation, Nuffield Department of Surgical Sciences, John Radcliffe Hospital, University of Oxford, Oxford, United Kingdom
2 Oxford Consortium for Islet Transplantation, Oxford Centre for Diabetes, Endocrinology, and Metabolism (OCDEM), Churchill Hospital, University of Oxford, Oxford, United Kingdom
3 Natural and Medical Sciences Institute, Eberhard Karls University, Tübingen, Germany
Aim
Islet isolation essentially requires dissociation of the islet basement membrane by collagenolytic enzymes. Basement membrane loss is associated with reduced islet function and viability. Previous studies demonstrated that individual extracellular matrix proteins (ECMPs) can increase islet survival pre- and post-transplant. In the present study, we tested our hypothesis, that the combination of different ECMPs, particularly those forming suprastructures, are more efficient than individual ECMPs to protect human islets from hypoxia-induced damage. In contrast to previous studies, we dissolved ECMPs in the media rather than to coat culture surfaces.
Methods
Islets, isolated from pancreases of 11 human DBD donors (50±2 years, 29.3±1.2 BMI, 5.7±0.3 hours CIT), were cultured for 3–4 days in 2% oxygen and suspended in CMRL 1066 (2% FCS) supplemented with either 40 µg/mL of dissolved collagen-IV, 10 µg/mL laminin-521 or 12.5 µg/mL nidogen-1 used individually or as combination. Sham-treated islets (STIs) cultured without ECMPs served as controls. Post-culture characterisation included IEQ yield or islet number (IN), viability (FDA-PI), early plus late apoptosis (annexin V-PI), glucose stimulation index (SI: 2 vs 20 mM) and reactive oxygen species production. Parameters were normalised to IEQ, related to pre-culture data if appropriate and presented as mean ± SEM. Statistical analysis was performed by Friedman test followed by Dunn’s multiple comparison.
Results
Compared with STI (41±7%), post-culture recovery was higher when hypoxic islets were treated with collagen-IV (64±7%, p<0.001), laminin-521 (57±6%, p<0.01) or nidogen-1 (65±6%, p<0.001) used individually or combined (61±7% p<0.001). This correlated with islet fragmentation (IN/IEQ ratio) that was lower when collagen-IV (116±13%, p<0.001), laminin-521 (114±12%, p<0.01), nidogen-1 (121±12%, p<0.01) or combined ECMPs (119±13%, p<0.001) were compared with STIs (155±16%). Reactive oxygen species production in STIs was substantially reduced by 71±6% (NS), 73±6% (p<0.05), 90±2% (p<0.001), and 87±4% (p<0.001) in presence of collagen-IV, laminin-521, nidogen-1 or combined ECMPs, respectively. This resulted in improved viability (83±7% [p<0.01], 79±9% [p<0.01], 84±7% [p<0.001], 83±8% [p<0.001]) compared with STIs (63±7%). While individual ECMPs stabilised or reduced pre-culture apoptosis (94±17% [p<0.05], 117±16% [p<0.05], 68±13% [p<0.001]), combined ECMPs (171±18%, NS) were equal to STIs (196±28%). STIs did not adequately secrete insulin after glucose challenge (SI 0.97±0.13) in contrast to the physiological insulin response after treatment with collagen-IV (1.76±1.18 [p<0.01]), laminin-521 (1.53±0.25 [NS]), nidogen-1 (2.27±0.67 [p<0.01]) or combined ECMPs (1.95±0.25 [p<0.05]).
Conclusion
Among the three individual ECMPs tested, nidogen-1 appears to be most effective to protect human islets from hypoxia-induced damage. As its protective efficiency partially exceeds that of combined ECMPs, we have to reject our hypothesis. Further studies are required to clarify whether collagen-IV, laminin-521 and nidogen-1 spontaneously assemble to suprastructures in vitro.
RECOMBINANT NIDOGEN-1 SIGNIFICANTLY PROTECTS HUMAN ISLETS FROM HYPOXIA-INDUCED DAMAGE