Quantification of basal and stimulated ROS levels as predictors of islet potency and function

Abstract

We have developed a luminol-based assay using intact islets, which allows for quantification of reactive oxygen species (ROS). In addition, an index capable of characterizing metabolic and mitochondrial integrity prior to transplantation was created based on the capacity of islets to respond to high glucose and rotenone (mitochondrial respiratory chain complex I inhibitor) by production of ROS. To validate this assay, lipid peroxidation and antioxidative defense capacity were evaluated by detection of malondialdehyde (MDA) levels and glutathione peroxidase activity (GPx), respectively. Also, flow cytometric analyses of ROS (dihydroethidine), apoptosis (Annexin V, active caspases), necrosis (Topro3), and mitochondrial membrane potential (JC-1) were done in parallel to correlate with changes in luminol-measured ROS. ATP/ADP ratios were quantified by HPLC and the predictive value of ROS measurement on islet functional potency was correlated with capacity to reverse diabetes in a streptozotocin-induced diabetic NOD.scid mouse model as well as in human transplant recipients. Our data demonstrate that levels of ROS in islets correlate with the percentage of apoptotic cells and their functional potency in vivo. The ROS indices following glucose and rotenone exposure are indicative of metabolic potency and mitochondrial integrity and can be used as surrogate markers to evaluate the quality of islets prior to transplantation.

Figure 1. Quantification of ROS in rat islets by luminol-based chemiluminescence

After 24 h culture rat islets were either untreated or pre-treated with 16.7 mM, 28 mM glucose or 1 µM rotenone. SOD and catalase were addedas controls. Results were expressed as counts per minute (CPM), normalized to DNA and represent the mean ± SD of three separate experiments. *p < 0.05 vs. baseline.

Figure 2. Function of rat islet grafts

Streptozotocin-induced diabetic NOD.scid mice were transplanted with untreated (■) (n = 8) and cytokine treated (▲) (n = 8) rat islets (500 IEQ). IPGTT was performed by application of 3 g/kg glucose. After 25 days, islet grafts were removed by nephrectomy. (A) Non-fasting BG levels; *p < 0.05 vs. untreated. (B) IPGTT: BG levels (left); *p < 0.05 vs. untreated. Serum-insulin (right); *p < 0.05 vs. cytokine treated. Data are shown as mean ± SD. Shaded area indicates the physiologic range of BG (80–120 mg/dL).

Figure 3. Function of human islet grafts

Streptozotocin-induced diabetic NOD.scid mice were transplanted with high (■) (n = 5) and low (▲) (n = 5) responding human islets from four different islet preparations. IPGTT was performed by application of 3 g/kg glucose. After 25 days, islet grafts were removed by nephrectomy. (A) Non-fasting BG levels of recipients received 1000 (left) and 500 IEQ (right). *p < 0.05 vs. low responding islet grafts. (B) IPGTT, 1000 IEQ: BG levels (left); *p < 0.05 vs. high responding islet grafts. Seruminsulin (right); * p < 0.05 vs. low responding islet grafts. (C) IPGTT, 500 IEQ: BG levels (left); *p < 0.05 vs. high responding islet grafts. Seruminsulin (right); no significant difference between groups. Data are shown as mean ± SD. Shaded area indicates the physiologic range of BG (80–120 mg/dL). (D) Representative images of islet grafts (magnification 10×). Left: HE; grafts from high (top) and low (bottom) responding islets. Right: Immunohistochemistry for insulin (brown) from high (top) and low (bottom) responding islets.