A Vast Amount of Enzyme Activity Fails to Be Absorbed Within the Human Pancreas

Islet isolation success for clinical islet transplantation critically depends on digestive enzyme variables such as enzyme activities (1–3), quantity (2), and ratio of specific components (1, 2, 4–11). Furthermore, enzyme localization in native pancreas tissue may play a role (12, 13). Commercial enzymes represent the greatest reagent costs for human islet isolation, yet little is known about their uptake into the pancreas. Absorption of commercially available digestive enzymes within a donor pancreas is postulated to have a major impact on the outcome of islet isolation. To assess enzyme dynamics on samples obtained during the initial 20 min of enzyme retrograde ductal perfusion into the human pancreas, measurements of total protein as well as the degradation of N-3-([2-furyl]-acryloyl)-Leu-Gly-Pro-Ala (FALGPA) (14), described herein as class II collagenolytic-like activity (C2LA), and Nα-benzoyl-L-arginine-ethyl ester hydrochloride (BAEE) as a measure of tryptic-like activity (TLA), were performed. All pancreases were intact with no abnormal morphology: age (yr), body mass index (kg/m2), islet equivalents per gram pancreas, percent islet purity, and transplanted yes/no (Y/N) were 45, 33.6, 1953, 41%, and Y for pancreas 1; 46, N/A, 3975, 38%, and Y for pancreas 2; 62, 34, 718, 17%, and N for pancreas 3; 56, 23.7, 2446, 68%, and Y for pancreas 4; and 63, 22.7, 1537, 29%, and N for pancreas 5. Collagenase and thermolysin enzymes (Vitacyte, Indianapolis, IN; lot numbers BHA-110919 and TH-101020 for pancreas 1, BHA-120511 and TH-120607 for pancreases 2–4, and BHA-120511 and TH-110130 for pancreas 5) were dissolved in standard enzyme buffer (15). Supplemental enzyme buffer was added as required during infusion into the pancreatic duct (total volumes for pancreases 1–5 were 120, 140, 160, 156, and 135 mL, respectively), and the Clinical Islet Transplantation Consortium recommendations for perfusion pressure were followed. Before the perfusion start and at 5-min intervals (20 min total, the 5-min sample was not collected from pancreases 4 and 5), 500 μL samples were collected and filtrated through a cellulose filter (0.45 μm; Steriltech, Kent, WA) to avoid tissue and cell contamination on subsequent measurements of enzyme activities and protein concentrations. The C2LA and TLA measurements were determined in quadruplicate at 25°C at 0.1 mM FALGPA (Sigma-Aldrich, Stockholm, Sweden) and 0.5 mM BAEE (Sigma), respectively, with a Hitachi U-2910 spectrophotometer. For C2LA and TLA, 1 U was defined as the conversion of 1 μmol FALGPA or BAEE substrate, respectively, to product per minute. Protein concentrations were determined in quadruplicate by Bradford assay with bovine serum albumin standards (Sigma) (16, 17). Values are presented as mean±standard deviation. The changes in total circulating levels of C2LA and TLA correspond to the amount of enzymes absorbed by the pancreas during the 20 min of sustained ductal infusion. The dynamics of C2LA and TLA had diverging patterns, with decreases in C2LA (Fig. 1A) and stable or increasing TLA (Fig. 1B). TLA increased by 56±20 U after 20 min, likely indicating release of TLA enzymes from the organ. The C2LA (Fig. 1A) from the start of the perfusion phase was 1077±143 U. After 10 min, C2LA activities decreased by 35%±12%, indicating it as the major period of enzyme absorption, with little benefit from increased time of perfusion. Alarmingly, even after the completion of the 20 min of recirculation of enzymes into the pancreatic duct, no additional C2LA activity was absorbed within the pancreas. Hence, a vast amount of collagenase remains in the liquid phase with little effect on the digestion of the pancreas, that is, a pancreatic biopsy taken before ductal perfusion of enzymes and subsequently placed in the digestion chamber remains macroscopically intact, whereas the pancreas containing the enzyme blends is efficiently digested. By the end of perfusion, circulating protein levels were similar or increased (Fig. 1C), reflecting low enzyme digestion activity at 8°C to 14°C.FIGURE 1: Box plots showing the range and distribution of collagenolytic-like activities (FALGPA and C2LA), tryptic-like activities (BAEE and TLA), and protein levels measured during the enzyme perfusion phase of islet isolation. Values are means of quadruplicate measurements from five isolations at different time points, except at 5 min, where no samples from two of the isolations were collected. A and B, amounts of C2LA and TLA, respectively, as expressed in total units. C, total protein over time. D, total C2LA/total protein (U/mg). BAEE, Nα-benzoyl-L-arginine-ethyl ester hydrochloride; C2LA, class II collagenolytic-like activity; FALGPA, N-3-([2-furyl]-acryloyl)-Leu-Gly-Pro-Ala; TLA, tryptic-like activity.Previous publications have focused on characterizing analytic enzyme data, for example, protein composition (1, 2, 11), integrity (3, 4), and initial activities added to the pancreas (6, 10). The approach applied here provides new insights into the absorption of enzyme activities in the human pancreas during islet isolation. Surprisingly, an ordinary human pancreas seems to be saturated already after 10 min of low-temperature enzyme perfusion and absorbs only about a third of total C2LA units of collagenase. The presented findings imply that the amount and administration of enzymes utilized for human islet isolation can be dramatically optimized. Enzyme activities can be monitored to identify specific, dynamic variables, potentially leading to optimal application of costly commercial enzymes to increase the yield and quality of isolated human islets. Andrew S.