Introduction: Optimal human islet isolation requires digestion of key islet basement membrane (BM) proteins including Collagen-IV (C-IV), Laminin-521 (L-521), Nidogen and Perlecan (PLC) by collagenolytic enzymes. Previous studies have demonstrated that culturing isolated human islets with a combination of some islet BM proteins can improve islet survival when exposed to hypoxia. In this study, we assessed whether PLC, used alone or combined with C-IV and L-521, has the ability to protect human islets from hypoxia-induced damage.
Methods: Cultured human islets (n = 7) were characterized regarding islet equivalent (IEQ) yield; viability (FDA-PI); apoptosis (Annexin-V) and reactive oxygen species (ROS) production (DCFH-DA). Islets were cultured for 4 – 5 days in (A) supplemented CMRL (sham-treated controls); (B) CMRL + 40 µg/mL C-IV; (C) CMRL + 10 µg/mL L-521; (D) CMRL + 10 µg/mL PLC or (E) CMRL + combined BM proteins at 2% oxygen. Glucose-stimulated insulin secretion (stimulation index [SI]: 2 vs 20 vs 2 mM) and islet chemokine sampling were performed postculture. Data were corrected for IEQ and normalized to hypoxic controls (mean ± SEM).
Results: Compared with controls (100%), significantly more islets could be recovered post culture using C-IV (172 ± 9.2%, p<0.001), L-521 (159 ± 6.1%, p<0.001), PLC (146 ± 8.7%, p< 0.05) or the protein combination (142 ± 5.7%, p<0.05). Islet fragmentation was also substantially reduced compared with controls when islets were treated by C-IV (63 ± 9.0%, p<0.001) or L-521 (64 ± 8.1%, p<0.01) but not when PLC (73 ± 9.7%, NS) or the combination (85 ± 19.3%, NS) was added. Viability was significantly increased in islets treated by C-IV (130 ± 4.5%, p<0.001), L-521 (119 ± 4.4%, p<0.05), PLC (127 ± 6.4%, p<0.01) or the combination (118 ± 2.9%, p<0.05). In contrast to controls (0.94 ± 0.07), the SI indicated intact metabolic function after treatment with C-IV (1.47 ± 0.12, p<0.01), L-521 (1.49 ± 0.07, p<0.001), PLC (1.50 ± 0.07, p<0.001) or the combination (1.43 ± 0.09, p<0.05).

ROS production during hypoxia could be massively reduced using C-IV (55 ± 8.5%, p<0.05), L-521 (51 ± 6.3%, p<0.01), PLC (42 ± 4.7, p<0.001) or the combination (49 ± 6.0%, p<0.01). ROS reduction was associated with decreased TNF-a release observed after adding C-IV (54 ± 7.8%, p<0.05), L-521 (46 ± 3.6, p< 0.05), PLC (29.5 ± 2.8%, p<0.001) or the combination (40 ± 3.5, p<0.01). In accordance, apoptosis was also substantially reduced using C-IV (68 ± 10.4%, p<0.01), L-521 (73 ± 5.5%, p<0.05) or PLC (66 ± 6.9%, p<0.01), but not when adding the combination (87 ± 7.2%, NS).
Conclusions: Our findings indicate that PLC has a similar protective capacity to C-IV or L-521. Importantly, our study suggests that PLC has distinct anti-inflammatory qualities compared with the other BM proteins. As PLC expresses receptors specific for growth factors, future studies will target the use of growth factors relevant for optimizing human islets.