INFLUENCE OF DONOR AGE ON BOVINE PANCREATIC ISLET ISOLATION¹

Abstract

Background. 

Pancreatic islets from pigs are largely used for experimental studies. However, pancreas harvesting requires modification of conventional slaughtering to reduce ischemia time. It has been shown that bovine pancreatic islets can be more easily obtained and they show satisfactory in vitro and in vivo function. To improve the isolation procedure we compared the effect of bovine donor age on islet isolation.

Methods. 

Islets were isolated by collagenase digestion and sequential sieving from calves (6 months of age) and from adult bovine (> 16 months of age). After isolation the number of islet equivalents was calculated and histological and immunohistochemical studies performed. The purity and viability of islet for each preparation was also estimated. In vitro function of islets was evaluated by static insulin secretion assay, and alginate encapsulated islets were transplanted in streptozotocin-induced diabetic rats for in vivo functional evaluation.

Results. 

A significantly higher number of islets were obtained from calf pancreas, compared with adult bovine pancreas. Hystological examination showed intact morphologic features of islets. The purity of islet preparations was higher from calf pancreas than from adult pancreas. Cell viability, and insulin production in presence of high glucose concentration, were not affected by donor age. All animals receiving microencapsulated islets from calves showed normoglycemia for prolonged periods (17–40 days).

Conclusions. 

These results indicate that pancreatic islet isolation is more efficient from juvenile bovine than from adult. Calf pancreas is a good and convenient source of tissue for massive islet isolation for experimental studies.

Pancreatic islet transplantation is believed to represent a potential therapy for insulin-dependent diabetic patients. That islet transplantation has the potential to normalize glycemia has been demonstrated in diabetic patients receiving concomitant transplantation of another organ, most frequently the kidney (^{1, 2}). However, several problems remain to be solved to extend islet transplantation alone to insulin-dependent diabetic patients. A major obstacle to this transplantation is the need to avoid anti-rejection drugs for chronic immunosuppressive therapy. Another difficult task in this area is use of nonhuman islets to overcome donor organ insufficiency. To find a solution to both these problems, research is now ongoing with the aims of developing immunoisolation devices and xenotransplantation. To these purposes, several studies are in progress on isolation and experimental transplantation of islets from pig and from bovine pancreas (^3–6).

Despite porcine islets have been considered the best source for xenogenic transplantation, they have been shown to be fragile and their isolation is rather inconsistent because of loss of viability and islet fragmentation (⁷). Very recently, it has been shown that pancreas from the adult pig are more suitable for islet isolation and show higher degree of viability than islets from juvenile pigs (³). It has also been shown previously that bovine islets can be isolated efficiently from pancreas tissue and they demonstrate satisfactory in vitro and in vivo function (⁵). The main advantage in using bovine pancreas is related to an easier access to the organ. Thus, for porcine pancreas one must avoid conventional animal processing, at the slaughterhouse, that implies too long ischemia time. On the contrary, harvesting the bovine pancreas is feasible during conventional slaughtering processing with an acceptable ischemia time.

With the aim to improve the outcome of experimental procedures for isolation of bovine islets, we compared the effect of donor age on islet viability and function. To this purpose, we isolated islets from bovine pancreas harvested from calves (6 months of age) and from adult bovine (>16 months of age). We showed that islet yielding and purity are higher in calf than in bovine adult pancreas and that islets from calf pancreas can efficiently normalize glycemia when encapsulated in alginate-gel microcapsules and transplanted into rats in which diabetes has been induced with streptozotocin.

MATERIALS AND METHODS

Islet preparation.

Bovine pancreases were obtained from a local slaughterhouse and transported to the laboratory in cooled (4°C) Euro-Collins solution (Salf S.p.A., Bergamo, Italy). Warm and cold ischemia times ranged from 11 to 18 min and from 70 to 100 min, respectively. Bovine donors were either calves (6 months) or adults (16–36 months old). Islet isolation was performed as previously described (⁸). Briefly, the splenic lobe was intraductally injected with Hanks’ balanced salt solution (HBSS) containing 0.8 mg/ml collagenase P (Boerhinger-Mannheim, Mannheim, Germany). Distended pancreatic tissue (40–70 g) was loaded in a shaking water bath at 35°C and 120 revolutions per minute. After 8–9 min incubation the tissue was filtered through 400- and 90-μm mesh stainless steel filters in sequence and digestion was stopped by adding cold (4°C) HBSS containing 4% bovine serum. The same procedures of filtration and washing were repeated for approximately 30 min from beginning of incubation. The preparation was purified by centrifugation on Histopaque 1.077 g/ml (Sigma, St. Louis, MO) as previously described (⁸). After isolation, the islets were resuspended in M199 culture medium supplemented with 10% bovine serum, 25 mM Hepes, and antibiotics (penicillin 100 U/ml, streptomycin 100 μg/ml, gentamicin 50 μg/ml, and amphotericin B 1 μg/ml). To determine the islet yield of the isolation, we counted the total number of islets using an inverted microscope equipped with a calibrated grid, counting only islets >50 μm in diameter. We also evaluated the number of islet equivalents ([IEQ] 150 μm-diameter islets) in each isolation, dividing islet population in four diameter classes, as previously described (⁹). Approximately 3,000–4,000 IEQ were loaded in 75 cm (²) suspension culture flasks and cultured in a standard cell incubator with 5% CO₂. We have previously observed that culture of bovine islets under standard temperature conditions (37°C) results in extensive adhesion of islets to the bottom of the culture flasks, with islet fragmentation and islet cells migration. Islet adhesion could not be prevented even using suspension culture flasks. To overcome this problem, we lowered the culture temperature to 27°C.

Morphologic and immunohistochemical analysis.

Samples of islet preparations were fixed in Bouin’s solution. After storage in 70% ethanol, samples were embedded in paraffin and cut in 3-μm sections (LKB, 2088 Ultrotome V, LKB-Produkter AB, Bromma, Sweden). For morphological evaluations, sections were stained with hematoxylin and eosin.

To localize the insulin producing β cells, the sections were processed for light microscopic immunohistochemical analysis using an avidin-biotin horseradish peroxidase complex technique (ABC method, ABC-Elite; Vector Laboratories, Burlingame, CA, USA). A monoclonal antihuman insulin antibody (Sigma) was used. Briefly, the sections were dewaxed, rehydrated, and incubated for 1 hr with 0.3% H₂O₂ in methanol to quench endogenous peroxidase. A specific binding was blocked by 30 min incubation with nonimmune sera (horse serum). Slides were then incubated for 2 hr in a moist chamber with the primary antibody (anti-insulin 1:6900) in PBS with 1% bovine serum albumin (Miles, Bayer, Milan, Italy), followed by the secondary antibody (biotinylated horse anti-mouse IgG), ABC solution, and finally they were developed with diaminobenzidine. The purity of islet preparation was estimated by standard morphometric technique using light microscopic method to derive volume fraction of tissue occupied by insulin producing cells. To this purpose, a microscope connected to a computer-based image analysis system was used. Images were systematically acquired and digitally overlaid with a hortogonal point grid (9×11 lines). The purity was calculated as the percentage of grid points contained in the insulin-producing cells to the total number of intersections contained in the tissue. An average of 107 grid points were counted on tissue samples.

Islet viability.

Viability of isolated islets was assessed using a fluorometric viability assay kit (Molecular Probes, Eugene, OR, USA), with calcein and ethidium homodimer used for identification of viable and dead islet cells, respectively. Islets were incubated for 30 min. in 4 μM ethidium homodimer and 1.7 μM calcein. The islets were analyzed with a laser scanning confocal microscope (InSight puls, Meridian Instruments, Inc., Okemos, Michigan). Calcein and ethidium were excited separately at 488 nm and at 568 nm, respectively. The fluorescence emission was also acquired separately, for calcein at 530 nm and for ethidium at 645 nm. Digital images were pseudo-colored in green and red for the two dyes, respectively. The percentage of green fluorescent cells was calculated for each islet investigated. On the average 7 (^4–13), islets were evaluated in each isolation.

Glucose stimulated insulin release.

To evaluate in vitro function of bovine islets from calves and bovine adults, we performed static insulin secretion assay. Islets were evaluated after 7 days of culture at 27°C. Groups of 50 IEQ were incubated in 500 μl of Krebs-Ringer bicarbonate solution supplemented with 1.6 mM glucose for 45 min at 37°C. Islets were then washed in Krebs solution and incubated for 1 hr with high-glucose concentration (25 mM glucose plus 5 mM theophylline in Krebs solution at 37°C) and followed by 1 hr incubation in low-glucose concentration (1.6 mM glucose in Krebs solution at 37°C). Supernatants were removed after each hour for insulin measurement and stored at −20°C. Insulin concentration was measured using radioimmunoassay (Sorin Biomedica, Saluggia, Italy).

Encapsulation and transplantation of islets.

For in vivo evaluation, we transplanted alginate encapsulated islets into female MWF rats, a colony of rats selected from the Wistar strain (¹⁰). Diabetes was induced by a single injection of streptozotocin (60 mg/kg body weight; Sigma) into the tail vein of rats weighing 180–200 g, 7–14 days before islet implantation. Only rats with blood glucose levels >400 mg/dl were used. Body weight and blood glucose concentrations were determined three times weekly, before and after islet transplantation, by tail bleedings (Glucocard Memory 2, A. Menarini, Firenze, Italy).

After 7 days of culture, islets were suspended in a solution of 1.7% sodium alginate (Manugel DMB, Monsanto plc, Surray, United Kingdom) in calcium-free Krebs-Ringer solution at a concentration of 3 IEQ/μl. The mixture was extruded through an air jet droplet generator into a solution of 100 mM CaCl₂. Microcapsule diameter ranged from 800 to 950 μm. After gelation, the microcapsules were washed in calcium-free Krebs-Ringer, then in Krebs Ringer-Hepes 25 mM and cultured overnight in complete medium at 37°C. Empty microcapsules or encapsulated islets (1.6×104 IEQ) were introduced into the rat peritoneal cavity through a small (1–2 cm) midline incision under ether anesthesia. Low-dose CsA treatment was instituted in all animals at the dose of 10 mg/kg for 7 days after transplantation and 5 mg/kg for 10 additional days.

Statistics.

All results are expressed as mean±standard error (SE). Data on islet purity and cell viability were analyzed by nonparametric test (Mann-Whitney test). All other data were analyzed by Student’s t-test for unpaired or paired data as appropriate. Statistical significance level was defined as P <0.05.

RESULTS

The results of islets isolation from calves and adult bovine are presented in Figures 1, 2, 3. A significantly higher number of islets was obtained from calf pancreas (on average 76,968±12,120 islets/isolation) compared with adult bovine pancreas (29,656±5,890 islets/isolation, P <0.01). Also the number of IEQ was higher in calves than in bovine adults (58,818±6,676 vs. 21,318±4,851 IEQ/isolation, P <0.01). Despite the difference in number of islet yields, the size of islet populations was comparable in the two experimental conditions. As shown in Figure 2, the average size distribution histograms for calves and bovine adults was comparable for each dimensional class. The morphologic appearance of islets from juvenile or bovine adults is reported in Figure 3 (A,B). Both preparations showed good preservation of islet structure at hematoxylin-eosin staining (Fig. 3 C, D); however, larger amounts of exocrine tissue remained attached to islets from bovine adult isolations compared with calf pancreas. Immunostaining of isolated islets with an anti-insulin antibody is shown in Figure 4. Either preparations from calves (A) or bovine adult (B) pancreas demonstrated the presence of well differentiated islet cell producing insulin. As shown in Figure 4, the amount of acinar tissue is more extensive in the adult (B) than juvenile bovine (A). The percentage of islet tissue measured for both types of preparations is reported in Table 1. In islets isolated from pancreas of young donors, purity was importantly higher (91±2%) than those isolated from pancreas of adult donors (39±7%, P <0.05). We also determined cell viability within the islets using fluorescent microscopic method of calcein and ethidium staining. As shown in Figure 4 (C, D) images acquired using confocal microscopic examination were used to count calcein or ethidium homodimer staining of individual cells within the same islet. The mean percentage of viable cells in both types of preparations, at 1 and 7 day after isolation, is reported in Table 1. On average, the number of alive cells was comparable in islets isolated from calves and from bovine adults at both 1 and 7 days after isolation. Islet viability remained constant with time even at a culture temperature of 27°C. In parallel culture we also verified that viability of calf islets was comparable after 7 days at 27°C and 37°C (data not shown).

Figure 1: Number of islets and islet equivalents isolated from calves (n=7) and bovine adults (n=8) after 1 day of culture. * P<0.01 vs. calf.

Figure 2: Size distribution histograms for islet obtained from calves and bovine adults. Data are the mean of 6 isolations from calf pancreas and 5 isolations from adult bovine pancreas after 1 day of culture.

Figure 3: Phase contrast (A, B, magnification ×125) and hematoxylin-eosin light microscopic analysis (C, magnification ×360), of islets from calves (A, C) and bovine adults (B, D) after 1 day of culture.

Figure 4: Immunohistochemical staining for insulin of islets from calves (A) and bovine adults (B, magnification ×160) after 1 day of culture. Laser scanning confocal microscopic analysis of a representative islet with alive (C), calcein staining, and dead cells (D, magnification ×125), ethidium homodimer staining, after 1 day of culture.

Table 1: Percentage of purity and viability of preparations from calf and adult bovine

We evaluated in vitro function of bovine islets from calves and bovine adults using static incubation, with high and low glucose concentration, and measuring insulin production rate during the incubation periods. As reported in Figure 5, both from islets calves and bovine adults 7 days after isolation showed higher production rate of insulin (1.51±0.39 and 2.06±0.72 pg/min/islet, respectively), when incubated with high (25 mM) glucose concentration than with low (1.6 mM) glucose concentration (0.28±0.11 and 0.46±0.18 pg/min/islet, respectively). Of note, no statistically significant difference was observed between production rates of calves vs. bovine adults.

Figure 5: Insulin production after 7 days of culture by islets from calves and bovine adults during incubation with high (25 mM glucose plus theophylline 5 mM) and low (1.6 mM) glucose concentration in culture medium. *P <0.05 vs. high glucose.

To verify whether islets from calves, beside being vital and responding in vitro to glucose stimulation, also provide in vivo function, we performed islet xenotransplantation in MWF rats made diabetic with streptozotocin. To this purpose we encapsulated calf islets in alginate gel and transplanted them into the peritoneal cavity of four rats. As shown in Figure 6, after islet transplantation blood glucose levels promptly normalized in all animals. On the contrary, glucose levels remained high in control rats receiving empty capsules. The normoglycemic state was maintained in animals for different time periods, ranging from 17 to 40 days. As shown in Figure 6 all animals receiving islets gained weight after islet transplantation, suggesting amelioration of metabolic state, whereas control animals receiving empty microcapsules did not gain weight.

Figure 6: Nonfasting blood glucose level (A) and body weight (B) changes in rats in which diabetes was induced with stroptozotocin. At day 0, animals received 15,000–16,000 islets in alginate microcapsules or received empty microcapsules. Mean values at different time points are calculated for animals with blood glucose levels <200 mg/dl.

DISCUSSION

The results of our present investigation indicate that pancreatic islet isolation is more efficient from juvenile bovine than from adult pancreas. Thus, the number of islet obtained from calves was higher than that obtained from adult bovine, whereas islet dimensions were comparable from both sources. One could expect that difference in yielding may derive from different islet size, that can be retained differently between the two filters. However, the size distributions measured for calves and bovine adults are almost identical suggesting that size was not the explanation of different yields. Islet purity was much higher in calf than adult animal pancreas processing. Because pancreatic tissue of both sources was processed in the same way, the reason for the difference in islet yielding and purity of isolation must be different organization of extracellular matrix that characterizes pancreas tissue in young and adult bovine. Beside the reasons of these differences, it is important to know that better islet preparations can be obtained from young than from old animals and that this result is apparently the opposite of what has been shown for pig pancreas (³).

It has been shown previously that bovine islet recovered secretory functions a few days after isolation (⁵). For this reason, we evaluated in vitro function of islet from juvenile and adult bovine after 7 day of culture. Despite important difference in yielding and purity, we observed that islet viability was comparable between the two tissue sources and was satisfactorily maintained during in vitro culture. As mentioned previously, we observed preliminarily that bovine islets tend to attach to the culture flask with time when cultured at 37°C, resulting in loss of a large amount of islets. To avoid islet adhesion to culture plastics, we lowered temperature to 27°C in the incubator without loss of islet viability. Of interest, we did not observed difference in cell viability between culture temperature of 37°C and 27°C. After 1 week in culture, static incubation of islets with high glucose concentration induced a significant increase in insulin production in both islet populations, compared with insulin production under low glucose concentration. Islet response to glucose challenge was comparable in islets from calves and from bovine adults, in line with the observation that islet viability was comparable in the two conditions.

The isolation of a large number of islets from calves that we have obtained and their purity, prompted us to test islet function in an in vivo setting. To this purpose, we encapsulated bovine islets from calves in alginate gel and transplanted them in streptozotocin-treated rats with stable hyperglycemia before transplantation. All animals reached normoglycemia promptly after islet transplantation and gained weight. Function of transplanted islets was maintained for periods ranging from 16 to 40 days. These results demonstrate that the isolation of islets from young bovine allowed to obtain viable material that retain, at least initially, adequate functional properties. It has been shown previously that bovine islets controlled hyperglycemia in vivo, but this observation was obtained in the nude mice (⁵). Thus, this is the first demonstration that bovine islets successfully control glycemia in diabetic rats when encapsulated in alginate gel. Similar results have been obtained by Lanza and co-workers (¹¹) using bovine islets, but in those investigations pancreas were obtained from newborn calves (0–2 weeks).

We do not specifically investigate the reasons responsible for later loss of secretary function by encapsulated islets with time. Morphologic studies (data not shown) we have performed on microcapsules containing islets after failure of transplanted material did not show evident deposition of cells around the microcapsules. However, the morphologic features of islets appeared markedly altered with development of necrotic areas and loss of β cell structure. Loss of islet viability may depend from nonefficient immunoisolation provided by alginate microcapsule or from lack of oxygen diffusion within the alginate microcapsule (¹²). One possible cause of loss of islet viability with tissue is that we used uncoated alginate to immunoprotect islets. This could expose the islet population to toxic effect of cellular or humoral response of the host to islet transplantation. Despite some reports show that uncoated alginate microcapsule can be used satisfactorily to obtain long-term islet survival and function in xenotransplantation (¹³), later investigations by the same authors (¹⁴) show that polylysine coating is essential to maintain long-term islet function in vivo. Independently from the causes of long-term failure of xenotransplantation, our results show that bovine islets from young animals maintain their ability to function in a “in vivo” setting. This is useful for planning experimental studies aimed at discovering new strategies to prevent islet rejection, either by immunoisolation or by induction of transplant tolerance. Studies are in progress in our laboratory using synthetic membranes for immunoisolation.

In conclusion, our present data indicate that calf pancreas is a good and convenient source of tissue for massive islet isolation for experimental studies. The advantage of using bovine islets over pig islets is related to a easier procedure for harvesting the organ reducing the ischemia time period.

Acknowledgments.

The authors thank Dr. Giuseppe Remuzzi for helpful discussion and Dr. Stefano Cortinovis for assistance during pancreas harvesting and processing. The authors also thank Annamaria Varbaro and Massimiliano Bernardoni for help in morphologic evaluation and Maria Serena Lepre for technical assistance. Sara Gualandris helped in the preparation of this manuscript.