Beta secretase inhibition in aging pancreatic beta cells: A possible role for Rooibos

A thesis submitted in fulfillment of the academic requirements for the degree of Doctor of Philosophy in the Department of Biochemistry and Microbiology in the Faculty of Science, Agriculture and Engineering, University of Zululand, 2020

In type 2 diabetes (T2D), hyperglycemia induces pancreatic β-cell glucotoxicity, characterised by β-cell functional mass decline. T2D patients are thus both insulin resistance and insulin deficient. In addition, to changes in insulin production and secretion, the hormone amylin is also dysregulated. Amylin is a pancreatic β-cell peptide hormone with a key role in glucose homeostasis and is co-secreted with insulin. Beta-secretase (BACE)-regulated cleavage of pre-amylin results in the aggregation and deposition of toxic amyloid plaques, characteristic of long-term diabetic patient pancreata. Thus, BACE inhibition is a promising therapeutic target in managing the progression of T2D. Research has shown that unfermented Rooibos has numerous health promoting benefits, including anti-inflammatory, anti-oxidative and anti-diabetic effects. This extract is unique for containing relatively high levels of aspalathin, a glycosyl dihydrochalcone with robust anti-oxidant properties and ameliorative effects against several metabolic complications. In this study, we aimed to assess the effects of chronic glucotoxicity in a novel 3D β-cell spheroid model and the potential therapeutic effects of GRT™ and BACE inhibition, thereof. Additionally, we determined the effect of BACE inhibition and GRT™ in the pancreata of aged C57BLKS db/db mice. The in vitro toxicity model was established using INS-1 β-cell spheroids created using the 3D BioArray Matrix (BAM) culture system, based on T2D-associated glucotoxicity. Spheroids were exposed to the following conditions: Normal glucose control (NG-11 mM glucose), high glucose (glucotoxicity) control (HG-33 mM glucose), high glucose with GRT™ (HG-GRT™) and high glucose with BACE inhibitor LY2886721 (HGBACE). Spheroids ultra-structural features were assessed using transmission electron microscope (TEM) and toluidine blue stain (TB). Spheroid viability and survival were assessed by measuring cellular ATP content, glucose utilization and size profiling. Functional parameters (i.e. glucose stimulated insulin & amylin secretion) and proinflammatory cytokines were assessed. The in vitro findings were validated in vivo using lean non-diabetic db+ and obese diabetic db/db mice. There were three phases : 10, 16 and 32 weeks treatment, divided into five treatment groups (n=8 per group): Control (vehicle equivalent/day), pioglitazone (15 mg/kg/day), BACE inhibitor (30 µM/kg/day), GRT™-1(74 mg/kg/day) and GRT™-2 (740 mg/kg/day). Physiological parameters, glucose metabolism and pancreatic islet morphology were assessed by iv measuring body weight and food intake; performing an oral glucose tolerance test, measuring serum insulin, C-peptide and amylin levels; and immunohistochemical labelling of pancreata with insulin and glucagon. In the spheroids, glucotoxicity (HG) significantly reduced cellular ATP content (73.63% ± 2.13 vs 100.0% ± 0.36; p< 0.001), while GRT™ ameliorated this effect (82.49% ± 1.07 vs 73.63% ± 2.13; p<0.05). Increased insulin secretion after glucose stimulation was observed in the NG compared to basal glucose levels (41.08 ng/mL ± 5.23 vs 22.6 ng/mL ± 3.49; p=0.001) and this effect was negated in the HG group; this was restored in the HG-GRT™ group (30.8 ng/mL ± 2.49 vs 17.9 ng/mL ± 2.54; p=0.05). HG and HG-BACE groups increased amylin secretion following glucose stimulation (41.7 pg/mL ± 7.18 vs 20.3 pg/mL ± 4.39; p=0.01; 43.9 pg/mL ± 5.38 vs 21.5 pg/mL ± 3.57; p=0.007). HG increased TNF-α compared to the NG group (2.4 pg/mL ± 0.24 vs 1.1 pg/mL ± 0.23; p= 0.001), while HG-BACE group had significantly lower TNF-α levels compared to HG group (2.4 pg/mL ± 0.24 vs 0.67 pg/mL ± 0.15; p< 0.0001). Additionally, after 2 weeks of treatment, HG-GRT™ reduced TNF-α levels (1.6 pg/mL ± 0.16 vs 2.6 pg/mL ± 0.33; p= 0.02). Interestingly, HG-BACE had a short-lived inhibitory effect as TNF-α levels increased after 2 weeks (2.4 pg/mL ± 0.24 vs 0.9 pg/mL ± 0.20; p= 0.0001). In vivo, changes were associated with normal animal aging in the db+ mice. In db/db mice, BACE inhibitor and GRT™-1 (phase I) improved glucose tolerance without concomitant BW gain (2412 ± 121.0; p=0.01 and 2293 ± 105.6, p= 0.02 vs 2970 ± 95.62, respectively), had higher HOMA-β and improved islet architecture of older animals (phase II). Pioglitazone significantly improved glucose tolerance (1884 ± 296.0 vs 3238 ± 260.5; p= 0.004), had lower HOMA-IR in phase II. Amylin levels were significantly reduced in phase II compared to phase I in all treatment groups (p<0.001). In phase I, GRT™-2 effectively improved glucose tolerance, whereas GRT™-2 remained ineffective. Additionally, GRT™-2 tended to have lower HOMA-IR and HOMA-β compared to GRT™-1. In conclusion, glucose tolerance was improved by BACE inhibition and GRT™ by regulating both insulin resistance and β-cell survival and function in both an in vitro and in vivo model. Thus, both treatments have the potential to supplement current anti-diabetic therapy

South African Medical Research Council