Isolated pancreatic islets of the ratAn immunohistochemical and morphometric study.код для вставкиСкачать
THE ANATOMICAL RECORD 237:489-497 (1993) Isolated Pancreatic Islets of the Rat: An lmmunohistochemical and Morphometric Study MOSTAFA M. EL-NAGGAR, AHMED A. ELAYAT, M. SALLEH M. ARDAWI, AND MOHAMMAD TAHIR Departments of Anatomy (M.M.E.-N., A.A.E., M.T.) and Clinical Biochemistry (M.S.M.A.), Faculty of Medicine and Allied Sciences, King Abdulaziz University, Jeddah, Kingdom of Saudi Arabia ABSTRACT Although there is a recent increase in the use of the isolated pancreatic islets of the rat in the transplantation and functional studies, there has been no detailed quantitative assessment on the size and cellular constituents of islets after the isolation procedure. The present work was undertaken to study the size classes of the isolated islets and the morphometry of their cellular populations. Islets of the rat pancreas were isolated by using the intraductal collagenase digestion technique, the most commonly used procedure for the isolation of pancreatic islets. Different endocrine cells of the isolated islets were stained by immunoperoxidase staining techniques. The distribution of the cellular constituents of the isolated islets was similar to that of the intact islets of the normal pancreas; A, D, and PP cells were peripherally arranged around the centrally located B cells. However, morphometric quantitative study showed that the percent volume and percent number of A, D, and PP cells of the isolated islets were lower than those of the corresponding intact ones. Further, the mean true diameter of the isolated islets was lower than that of the intact ones. These data indicate loss of islet cells during the process of isolation. Most of the lost cells were from the periphery of islets. This may provide an explanation for the incomplete metabolic control and recurrence of hyperglycemia encountered after isolated islet transplantation in the treatment of diabetes mellitus. It seems that further refinements of the isolation techniques are necessary to obtain islet tissue with total cellular integrity, before a complete success in transplantation could be achieved. o 1993 Wiley-Liss, Inc. Key words: Immunohistochemistry, Isolated islets, Morphometry, Pancreas Moskalewski (1965) was the first to isolate islets from the pancreas of the guinea pig using a n enzymatic digestion technique. Lacy and Kostianovsky (1967) modified the technique and used the density gradients to purify isolated rat islets. Further improvement of the technique was made by Sutton et al. (1986) who used the method of intraductal injection of collagenase. This technique facilitated harvesting large number of islets for use in transplantation procedures to treat diabetes (Sutherland et al., 1980; McEvoy and Leung, 1983; Toledo-Pereyra et al., 1984; Alejandro et al., 1986; Dibelius et al., 1986; Hesse et al., 1986; Lake et al., 1988; Tze and Tai, 1988; Huxlin et al., 1990). The isolated islets were also used to evaluate their ability to secrete insulin in vitro in response to glucose load (Lacy et al., 1968; Slavin e t al., 1977; Halban et al., 1986; Eizirik et al., 1989; Grodsky, 1989). The process of isolation inevitably leads to mechanical and enzymatic trauma to the islets themselves (Moskalewski, 1965;Vance et al., 1968; Buchanan and Mawhinney, 1973; Slavin et al., 1977) especially to 0 1993 WILEY-LISS, INC. their peripheral part, which is rich in A, D, and pancreatic polypeptide (PP)cells. The central core of islets, which is formed of insulin-producing B cells, was kept almost intact. These changes in the peripheral part of isolated islets presumably affect the hormonal secretory function of B cells. There is growing evidence that glucose-induced insulin release is dependent not only on the integrity and number of the insulin-containing B cells, but also on their interactions with the neighboring B and non-B cells (Pipeleers et al., 1982). In these studies, however, the effect of the isolation procedure on the different cellular populations and the Received March 9, 1993; accepted July 26, 1993. Address reprint requests to Dr. Mostafa M. El-Naggar, Department of Anatomy, Faculty of Medicine and Allied Sciences, King Abdulaziz University, P.O. Box 9029, Jeddah 21413, Kingdom of Saudi Arabia. 490 M.M. EL-NAGGAR ET AL. size of the pancreatic islets, was not quantitatively documented. In the present work the main objective was to study immunohistochemically the distribution of the cellular constituents of the functionally viable isolated pancreatic islets of the rat together with morphometric evaluation of the different cell types and the size classes of the isolated islets. Intact islets of normal pancreases were similarly studied and were used for comparison. MATERIALS AND METHODS Isolation of the Pancreatic lslets The pancreatic islets were isolated from adult male Lewis rats using the ductal perfusion method of Sutton et al. (1986) which is based on the original collagenase digestion technique of Lacy and Kostianovsky (1967). The rats were anaesthetized with ether inhalation, and the common bile duct was cannulated with a fine polyethylene tube (PE-50) immediately after incision of the abdomen. The duct was then ligated near the duodenum, and the rat was bled to death. The pancreas was infused with 7 ml of cold, freshly prepared, collagenase [Sigma type V (lot 100H6851)I in medium M199. The pancreas was quickly excised and incubated a t 37°C. Different concentrations of collagenase (1,2, and 3 mg/ ml) and different digestion times (16,18,20,22,24,26, and 28 min) were used to obtain the optimal digestion conditions for this batch of collagenase. The action of collagenase was stopped by adding cold medium M199. The digested pancreatic tissue was disrupted by aspiration through a 14-gauge cannula. The aspirate was filtered through a Nytex filter (mesh size 1,000 pm), collected into a centrifuge tube and the volume was adjusted with cold medium M199 and centrifuged a t 400g for 10 min at 4°C. The supernatant was decanted and the pellet was resuspended in the bottom layer of a discontinuous Ficoll gradient with a density of 1.095. Other Ficoll gradient layers with densities of 1.085, 1.075, and 1.045 were carefully layered on top of the bottom layer respectively, and the tube was then centrifuged at 8OOg for 20 rnin a t 4°C. The islets usually collect at the first and the second interface (between the layers with densities of 1.045, 1.075, and 1.085, respectively). The islets were aspirated and placed into another centrifuge tube and washed twice by using cold medium M199; each time these were centrifuged a t 400g for 10 min a t 4°C. The pellet of the islets was resuspended in the medium and transferred to a petri dish. The islets were then further purified by hand picking of the non-islet tissue with the help of a binocular dissecting microscope with a black background, illuminated by two horizontal beams of fiberoptic light. The whole process was carried out under a laminar flow hood to avoid contamination of the isolated islets. The islets isolated from each rat pancreas were counted under the dissecting microscope immediately after purification and the counting was confirmed under phase contrast microscope by a second observer. Using a calibrated grid attached to the phase contrast microscope the purity of the islet preparation was evaluated. The purity was calculated as the percent number of the intersections which overlie islets to the number of the intersections which overlie islet and non-islet tissues. Functional Evaluation Aliquots from the islet preparations which were isolated with different concentrations of collagenase and with different digestion times were functionally evaluated by measuring their insulin and C-peptide contents. Insulin was extracted from the isolated islets by acidified ethanol and quantitated by radioimmunoassay procedure using Coat-A-Count kit. C-peptide concentrations were measured by double antibody radioimmunoassay technique using commercially available kit (Diagnostic Products Corporation, Los Angeles, CA). Radioactivity was determined using Beckman gamma counter Model 5500 for 1min. The results were expressed as units of insulin or C-peptide per g of protein. The protein content of islet preparations was determined by the procedure of Lowry et al. (1951). The islets isolated with the optimal digestion conditions (22 min incubation time and collagenase concentration of 2 mg/ml) were further functionally evaluated by determining their adenine nucleotide contents. ATP concentration was measured by the method of Lamprecht and Trautschold (1974). ADP and AMP concentrations were measured by the method of Jaworek et al. (1974). Samples of freshly prepared isolated islets were incubated in 1 ml of medium M199 containing 5.5 mmol/liter glucose. Incubations were terminated by the addition of 200 p1 of HC10, (25%, w/v) to the incubation medium and cooling the mixture to 0°C a t 0,30, and 60 min of incubation time. The precipitated proteins were removed by centrifugation a t 13,500g for 2 min. The supernatant was neutralized with KOH containing 0.5 M-triethanolamine, and the KClO, was removed by centrifugation at 13,500g for 3 min. Adenine nucleotide concentrations were measured in neutralized extracts of the islets with a Beckman DU-6 recording spectrophotometer. The Concentrations of the nucleotides were expressed as nmol/mg of protein. Adenine nucleotide concentrations were also measured in freeze-clamped tissues of intact pancreatic islets using the procedure described by Ardawi and Newsholme (1985). The ability of the isolated islets to secrete insulin was evaluated at the optimal digestion conditions. The isolated islets were incubated in a 0.5 ml syringe barrel containing medium M199 with 1%(w/v) bovine serum albumin (fraction v). The incubation chambers were gassed with 02/C02(1911) and incubated a t 37°C in a shaking water bath a t 50-60 oscillations/min. An equilibration period of 15 min was allowed and then the islets were challenged twice for 30 min each with medium M199 supplemented with 15 and then 30 mmoYliter glucose, separated by a 60 min rest phase with medium M199 only. After a further rest phase of 60 min, a final challenge of 15 mmol/liter glucose together with 15 mmol/liter theophylline were used to elicit a maximal response from the islets (see Henquin and Meissner, 1984).The insulin release in response to challenge was determined and results were expressed as units of insulin released per unit time per mg of protein. lrnrnunohistochernical Staining Islet specimens isolated from eight rats, with the optimal digestion conditions described in the present 491 ISOLATED PANCREATIC ISLETS OF THE RAT work, were fixed in buffered neutral formalin and embedded in paraffin with 57°C melting point. Five micron serial sections were cut from each specimen, and groups containing 6-8 sections were selected a t 70 p-m intervals from each other. Sections from each group were stained immunohistochemically for insulin, glucagon, somatostatin and PP-containing cells. Paraffin blocks were also prepared from different areas of the pancreases of 6 rats, matched for age, sex, and weight to those of the donor rats used for isolation of the islets. Sections were cut and stained in a similar way to that used for the isolated islets, and were used for comparison. To localize the insulin producing B cells, the indirect immunohistochemical technique was used (Sternberger, 1979). The primary antibody used was guinea pig anti-swine insulin serum (lot 030), diluted in phosphate-buffered saline (PBS) with 1%normal rabbit serum. Different dilutions were used to obtain the optimal one; this was found to be 1500 which gave the best staining with the least background and was used for staining the sections thereafter. The secondary antibody used was rabbit anti-guinea pig immunoglobulin conjugated with peroxidase (dilution 1:200). The avidin-biotin complex (ABC)technique was used to localize glucagon, somatostatin and PP-producing cells as described previously (Hsu et al., 1981). The primary antibodies used were rabbit anti-porcine glucagon serum (lot O l l ) , rabbit anti-human somatostatin serum (lot 050e), and rabbit anti-human PP serum (lot 109e). They were tried in dilutions ranging from 1:10,000 to 1:lOO in PBS with 1%normal swine serum. It was found that the optimal dilutions of the antiglucagon, anti-somatostatin and anti-PP sera were 15,000, 1500, and 15,000, respectively. These dilutions were used for all the staining procedures thereafter. The secondary antibody used was biotinylated swine anti-rabbit immunoglobulin (dilution 1:200). Sera and antisera were obtained from Dako Corporation (Carpenteria, CA). Sections were incubated in the primary antibodies for overnight in a humidity chamber, a t 4°C. The chromogen substrate used was 3,3-Diaminobenzidine tetrahydrochloride (Sigma, St. Louis, MO) and the sections were counterstained with Harris’ hematoxylin to facilitate the nuclear identification. Specificity control for the immunohistochemical staining included omitting the antibody sera, which were replaced by non-immune sera, and absorption of the primary antisera with the purified antigens. method of Weibel (1963). With a calibrated grid inserted into the ocular lens of the microscope, the total number of the intersections over the cytoplasm and the nucleus of the specified cell type (Pc) as well as over the whole islet (Pi)were determined. The percent ratio of the number of intersections of the specific cell type per total islet intersections Ppc = (Pc/Pi) x 100 was expressed as the volume density of this cell type for the islet quantitated (Vvc = Ppc). The numerical percent of each specific cell type (NNc) per total islet cells was calculated using the nucleus as the counting base. The nuclei of the stained and unstained cells per islet profile were counted. The ratio of specific cell nuclei to the total islet nuclei, was expressed as the percent number of that specific cell per islet cells NNC= (NcINi) x 100. Calculations for B, A, D, and PP cells were done in adjacent or semi-adjacent sections stained specifically for these types of cells. The profile diameters (d) of the islets were calculated where a and b are the from the equation d = 2 major and major a t right angle semi-axis, respectively (Williams, 1977).The size-frequencydistribution of the profile diameters of the isolated and the intact islets were plotted. The mean axial ratio of the profiles was calculated. Assuming that the islets are spheroid structures, the formula of Fullman (Williams, 1977), for the ungrouped profile range ofsizes, was used to calculate the mean true diameter (D) for the isolated and the intact pancreatic islets. a, N 7F D=X 2 l/dl + lid2 . . . . 1/& where N represents the total profiles measured and d l , d2, . . . . dN represent the profile diameters. The results of the morphometric study were presented as the arithmetic mean 2 standard error of mean. Student’s t-test for non-paired observations was used for statistical evaluation of the data. RESULTS The freshly prepared isolated islets are recognized under the dissecting microscope by their characteristic well-defined, rounded or oval shape, showing high opacity and milky white color. They are also recognized under the phase-contrast microscope, as well-defined rounded or oval bodies with a faint green tinge, in contrast to the acinar tissue which appeared irregular in shape and showed a lighter tinge. The series of experiments which were performed to obtain the optimal Morphometric Analysis conditions for isolation of the islets, showed that the Morphometric study was carried out on the immuno- highest yield of islets, (342.7 & 15.8 islets/rat pancreas) histochemically stained sections. Twelve to twenty is- was obtained by the use of collagenase at concentration let profiles were chosen a t random from each slide, of 2 mg/ml (Table 1) and incubation time of 22 min with a total of 130-180 from each specimen. They were (Fig. 1).Poor yields of isolated islets were obtained morphometrically analyzed at a magnification of with incubation periods less or greater than 22 min and 1,000 x . Sections of intact islets, taken from different with collagenase concentration of 1or 3 mg/ml. Higher parts of 6 normal pancreases, were similarly analyzed. concentrations of insulin (318.45 t 11.3 mUlg protein) Standard morphometric methods were used to estimate and C-peptide (176.43 2 16.44 pmol/g protein) were the volume density (Vv) and the percent number (NN) also obtained from samples of islets isolated with colfor each cell type (WeJbel, 1963) and to calculate the lagenase a t concentration of 2 mg/ml and incubation mean true diameter (D) of the isolated and the intact for 22 min (Fig. 2); the latter were selected as the optimal digestion conditions. The purity of the isolated islets (Williams, 1977). The volume densities (Vv)of B, A, D, and PP cells per islet preparations was 20-40% after separation with islet tissue were calculated by the point counting the discontinuous Ficoll gradient. Most of the impuri- 492 M.M. EL-NAGGAR ET AL. TABLE 1. The total number of isolated islets per rat pancreas and the % purity of the islets obtained with the use of 22 min incubation time and 1,2,and 3 mg/ml collagenase Collagenase concentration (mglrnl) 1 2 3 No. of islets per rat (mean 2 SEMI No. of rats 6 14 6 140.7 2 11.7 342.7 2 15.8* 132.3 ? 8.6 % purity of the islets (mean 2 SEMI 73 2 1.8 82.6 2 1.9** 70 ? 5 *No. of the isolated islets at 2 mg vs. number at 1 or 3 mg/ml, P < 0.001. **% uuritv a t 2 mg vs. % purity a t 1 or 3 mg, P < 0.01 and P < 0.05, respectivdy . 400 -.s a, 0 L a M \ - 350 4 ld i, 3; d a, 300 T 250 II) ..J c 200 0 L W T 150 0 h 16 18 20 7 8I 22 24 300 300 250 250 200 150 r: . A 100 50 16 7 26 : a, 3 -E -2 -5 4 0 p < o o 1 p<ooo1 350 4 5 400 350 400 Insulin 0C-peptide 1 11 18 20 I 22 200 a l- ; a v 150 o 2 t' 24 26 1 Digestion t i m e ( m i n ) 100 $. 50 I 0 a 0 28 Fig. 2. Insulin and C-peptide concentrations in aliquots of islets isolated at different incubation times with 2 mgiml collagenase. 28 Digestion t i m e ( m i n ) Fig. 1. The number of isolated islets per rat pancreas at different incubation times with 2 mg/ml collagenase. Values at the bottom of each column represent the number of the rats used for isolation of the islets. ties consisted of pancreatic acini, in addition to small ducts, blood vessels and lymph nodes. Handpicking of the non-islet tissues could raise the final purity up to 82.6 t 1.9%(Table 1). Evaluation of the functional integrity of the islets, isolated using the optimal digestion conditions described in the present work showed that the concentrations of ATP and the total nucleotides were not significantly different from freeze-clamped intact pancreas and the isolated islets which were either freshly isolated or incubated for 30 and 60 min (Table 2). However, ATP/AMP concentration ratio was decreased in isolated incubated islets by 16.67% and 35.86% when the islets were incubated for 30 to 60 min, respectively. Repeated challenge of the isolated incubated islets with solutions of 15 followed by 30 mmol/liter glucose resulted in highly reproducible responses of insulin release above the baseline secretion which was 76.67 ? 16.67 pU/hr/mg protein. This response to glucose challenge was independent of the downward drifting baseline. The system functioned for 5 hr without compromising islet performance as witnessed by the maximal response of the islets to theophylline, which led to the release of 318.3 2 60 p.U insulin/hr/mg protein (Fig. 3). The immunohistochemical staining of the isolated islets showed four populations of cells which were distinct from each other not only in their specificity for each primary antibody, but also in their respective number and morphological distribution. The majority of the islet cells were B cells and they reacted positively to anti-insulin serum. The reaction of the isolated islets to the anti-insulin serum was generally weaker than that of the intact islets of the normal pancreas (Fig. 4, compared with Fig. 5). In many isolated islets, particularly the large ones, the cells differed greatly in their degree of reaction; the cells a t the periphery of the islets were almost well granulated, whereas the cells in the center were somewhat degranulated (Fig. 4). The A cells reacted positively with anti-glucagon serum. They were distributed a t the periphery of the isolated islets (Fig. 6), similar to their distribution in the intact islets (Fig. 7). A cells of the isolated islets appeared, however, less well organized than in the intact islets. They did not form a complete mantle around the B cells, but were rather arranged in clusters scattered a t the periphery of the isolated islets. It was also noticed that some of the peripheral cells seemed injured and the blood vessels on the periphery of the islets were torn. The D cells stained positively with anti-somatostatin serum. They were somewhat variable in number and were dispersed a t the periphery of the isolated islets (Fig. 8), whereas in the intact islets they were located either at the periphery of the islet or dispersed in an intermediate position between the A and B cells (Fig. 9). The immunohistochemical reactivity of the D cells of the isolated islets was weaker than that of the intact pancreatic islets. The PP cells stained positively with antiserum to pancreatic polypeptide. The cells were arranged singly or in clusters a t the periphery of the isolated islets (Fig. 10) similar to their arrangement in the intact pancre- 493 ISOLATED PANCREATIC ISLETS OF THE RAT TABLE 2. The concentration of adenine nucleotides in the isolated pancreatic islets compared with that of the freeze-clampedintact pancreatic tissue' Concentration (nmol/me of Drotein) Samnle ~ Isolated islets Isolated islets Isolated islets Freeze-clamped pancreatic tissue Incubation time ( m i d 0 30 60 1.34 f 0.07 1.34 t 0.09 1.28 f 0.19 0.331 f 0.064 0.315 10.029 0.379 0.027 * 0.289 I 0.043 0.341 r+ 0.037 0.422 f 0.040 1.903 I 0.231 1.986 0.099 2.083 f 0.244 * 4.74 3.95 3.04 0 1.67 t 0.25 I 0.17 0.232 t 0.057 * 0.32 7.18 ? 0.39 2.30 SEM, for eight specimens of isolated islets and six specimens of freeze-clamped intact pancreatic tissue. 5 5 mmol glucose/l 15 rnmol glucose/l 30 mmol glucose/l a A \ ATP/AMP ratio 015 mmol Qi 4 z -2 Total adenine nucleotides 300 h 2 AMP 350 400 G ADP - 'Values are presented as means .3 ATP glucose/l a n d 15 mmol theophylline/l 250 200 of the intact islets of the normal pancreas (Fig. 13). This indicates loss of islet mass during the isolation procedure which affected all sizes of the islets. The cumulative frequency distribution of the profile diameters of the isolated islets showed that their population contain a range of sizes, justifying the use of the formula of Fullman (WiJliams, 1977) to calculate the mean true diameters (D) for the isolated and the intact pancreatic islets, which were found to be 104.547 and 196.319 pm, respectively. 150 v DISCUSSION -,s 100 - 50 3 rA G 0 Challenge Fig. 3.Insulin release from islets isolated with 2 mgiml collagenase and 22 min of incubation, in response to a challenge twice with 15and 30 mmol/liter glucose, then a final challenge with 15 mmoliliter glucose together with 15 mmolfliter theophylline to elicit a maximal response. atic islets (Fig. 11). Whereas few isolated islets contained a large number of PP cells, most of the isolated islets contained 1-3 PP cells or were devoid of them (Fig. 10). The morphometric study showed that the cellular populations of the isolated islets were quantitatively different from those of the intact islets. The results of the morphometric study are summarized in Table 3. Whereas the volume density and the percent number of the B cells of the isolated islets were significantly higher than those of the intact islets, the volume densities and the percent numbers of the peripherally arranged A, D, and PP cells of the isolated islets were significantly lower than those of the intact islets (Fig. 12). The mean profile diameter of the isolated islets was significantly lower (P < 0.001) than that of the intact islets (Table 3). The profile sections of the islets showed a mean axial ratio of 1.468 in the isolated islets and of 1.345 in the intact islets, indicating that the islets could be treated as spheroids. The size-frequency distribution of the profile diameters of the isolated islets showed an overall shifting of all the size classes toward smaller sizes as compared to the corresponding classes Isolation of the islets has become a commonly used procedure to obtain a relatively pure endocrine tissue for transplantation experiments or for functional metabolic studies. In the present work, the islets were isolated from the pancreas of the rat by the intraductal collagenase digestion technique of Sutton et al. (1986). This was selected since it is the most commonly used technique for the isolation of pancreatic islets. Our preparations of the isolated islets proved to be functionally intact and viable. This was assessed by measuring the adenine nucleotide concentrations in the isolated islets, which were not significantly different from those of the intact islets of the freeze-clampedpancreatic tissue. In addition, isolated islets which were incubated in vitro for 30 to 60 min maintained their adenine nucleotide concentrations. Repeated challenge of the isolated incubated islets with 15 and 30 mmol/liter glucose resulted in highly reproducible responses of insulin release which were maintained for up to 5 hr. Maximal response was elicited by theophylline, a standard test for measuring the functional integrity of the islets (Lacy et al., 1972; Malaisse, 1973; Henquin and Meissner, 1984). Paraffin sections stained with immunohistochemical techniques showed that the distribution of B, A, D, and PP cells in the isolated islets is not different from that of the intact islets; where A, D, and PP cells are located at the periphery of the Fentral mass of B cells. Similar immunohistochemical localization of B cells of the isolated islets was observed by Groth et al. (1980), Gray et al. (19841, Warnock and Rajotte (19881, Delaby et al. (1989), Kneteman et al. (1989), and Warnock et al. (1989). There has been, to date, no detailed immunohistochemical study showing critical evaluation of other endocrine non-B cells of the isolated islets. The few isolated islets which were detected in the present work containing a relatively large number of PP cells were most probably derived from the lower part of the 494 M.M. EL-NAGGAR ET AL. Fig. 4. An isolated pancreatic islet stained for insulin. Most of the islet cells are positively stained. The peripheral cells show a strong positive reaction, whereas the centrally located cells show a relatively wcnk positivc rcnction. Immunopcroxidnsc stnin for insulin. X 660. Fig. 6. A cells in an isolated pancreatic islet showing positive reaction (brown staining) to the anti-glucagon serum. These cells are located at the periphery of the islet. Immunoperoxidase stain for glucngon. X 660. Fig. 5. Intact islet from normal rat pancreas showing positively stained B cells in the central core of the islet surrounded by a mantle of unstained non-B cells. Immunoperoxidase stain for insulin. X 650. Fig. 7. Intact islet from normal rat pancreas stained for glucagon. A cells show positive reaction (brown staining) and are arranged at the periphery of the islet. Immunoperoxidase stain for glucagon. x 650. juxtaduodenal pancreas, the latter was reported to be rich in PP cells (see Baetens et al., 1979). The results of the morphometric study indicated that the cellular populations of the isolated islets are quantitatively different from those of the intact islets. The B cells formed 92.74% of the volume and 90.75% of the number of cells in the isolated islets. These were significantly higher than the percent volume and number of the B cells of the intact islets, which were 66.04 and 70.32, respectively. On the contrary, the peripherally arranged A, D, and PP cellular populations of the isolated islets showed significantly lower percent volumes and numbers than those of the intact islets. Our quantitative data of the cellular populations of the intact pancreatic islets are in agreement with those described previously. According to Baetens et al. (19791, B, A, D, and PP cells comprised 82.5, 15, 2, and 0.5% of the volume of the intact dorsal pancreatic islets, whereas, in the ventral pancreas, their percent volumes were 82, 1.3, 2.4, and 14.3, respectively. Hellman (1959), Carpenter and Lazarow (19671, and Dean (1973) have shown that the centrally placed B cells comprise 6585% of the total volume of the intact pancreatic islets. However, we could not locate in the available literature any quantitative study on the cellular populations of the isolated islets. The quantitative changes in the cellular populations of the isolated islets were concomitantly accompanied by a dramatic loss of the islet mass. Whereas the mean true diameter of the isolated islets was 104.547 Fm, that of the intact islets was 196.319 Fm. The loss of the islet mass was uniform and affected all the size classes of the isolated islets, as evidenced by shifting of their size-frequency distribution towards the smaller diameters. The decrease in the percent volumes and numbers of A, D, and PP cells, which are located at the periphery of the islets, indicate that the loss of the islet mass during the isolation procedure had affected prin- ISOLATED PANCREATIC ISLETS OF THE RAT 495 Fig. 8. An isolated islet stained for somatostatin. Positively stained D cells are scattered a t the periphery of the islet. The section shows a contaminating pancreatic acinus. Immunoperoxidase stain for somatostatin. x 650. ery of the islet. This islet is most probably from the lower part of the juxtadudenal pancreas (head) which is usually rich in PP cells. The other islet, does not contain any PP cell. Immunoperoxidase for PP. Fig. 9. An intact islet showing D cells with positive reaction (brown staining) to the anti-somatostatin serum. The cells are scattered at the peripheral part of the islet. Immunoperoxidase stain for somatostatin. x 650. Fig. 11. Intact islet from the splenic part of normal rat pancreas. PP-secreting cells appear positively stained with anti-PP serum. They are few in number and scattered a t the peripheral part of the islet. Immunoperoxidase stain for PP. x 650. x 650. Fig. 10. Two isolated islets stained for PP cells. One islet appears to contain a good number of PP cells, which are scattered at the periph- cipally the peripheral parts of the islets. This finding is supported by the presence of irregular peripheral region which could be detected in most of the isolated islets. Injury to the peripheral cells of the isolated islets was also reported by Moskalewski (19651, Vance et al. (1968), Buchanan and Mawhinney (1973), and Slavin et al. (1977). Their observations were made, however, without the support of morphometric quantitative data. Our findings agree also with the observations of Trimble et al. (1980), which described fewer A cells in the intraportally transplanted islets as compared to the islets within the normal pancreas. Vance et al. (19681, and Buchanan and Mawhinney (1973) found that the glucagon which is released from the isolated islets of the rat was more variable than insulin. The variable release of glucagon in the isolated islets was attributed to damage of the A cells, which are situated more peripherally than the B cells. Slavin et al. (1977) attributed the injury of the peripheral parts of the islets to the metabolic changes occurring in the islet cells during the isolation procedure. However, this injury could also be due to enzymatic digestion of the peripheral parts of the islets by the collagenase enzyme and mechanical trauma during the process of isolation. Whether the changes in the normal percentage of the endocrine cells of the peripheral parts of the isolated islets can lead to changes in the hormonal secretory function of the B cells, is not yet known. However, there are various reports indicating the presence of functional cooperation between various endocrine cells 496 M.M. EL-NAGGAR ET AL. TABLE 3. The endocrine cell types of the isolated islets compared with those of the intact islets' Isolated islets Volume density (percent volume) B cells A cells D cells PP cells Percent No. B cells A cells D cells PP cells Profile diameter of the islets (pm) Intact islets * 0.67 (240) 0.52 (331) * 0.4 (205) 92.74 4.46 2 2.18 0.62 k Remarks * 66.04 0.39 (337) 19.38 k 0.71 (232) 6.09 & 0.21 (175) 8.49 0.49 (292) P < 0.001 P < 0.001 P < 0.001 P < 0.001 70.32 2 0.25 (337) 18.77 t 0.71 (232) 4.23 0.13 (175) 6.68 t 0.58 (292) 133.15 1.37 (1036) P < 0.001 P < 0.001 P < 0.001 P < 0.001 P < 0.001 * 0.18 (331) * * * 90.75 0.95 (240) 5.82 0.68 (331) 2.45 t 0.42 (205) 0.98 0.23 (331) 75.41 t 1.53 (1107) * * 'Data are presented as means ? SEM. Values between brackets represent the No. of the islets examined. 100 2 90 e E 3 .Z volume of isolated islets - 20 - 15 - Z n u m b e r of isolated islets % volume of i n t a c t islets 80 25 Z n u m b e r of i n t a c t islets Isolated islets b0 Intact v 70 L 60 ffl 4 # I? 50 k 5 0 0 > 40 k G ; 30 3 - a) e E 3 20 5 - z 0 - e, a 10 10 I I I I I I I I I 0 0 of the islet tissue, which plays an important role in regulating the release of insulin in response to glucose. Communication of the cells of the pancreatic islets occur through gap junctions. These junctions were reported to occur not only between B and B cells and between A and A cells, but also between A, D, PP cells, and adjacent B cells (Orci et al., 1973,1975; Unger and Orci, 1981; Meda et al., 1983). Pipeleers et a1 (1982) found that B cells in the isolated islets of the rat released 30 times more insulin in response to glucose as compared with that released from purified single B cells. In conclusion, the results of the present work show that the islets isolated by the enzymatic digestion technique have suffered considerable loss of their volume during the isolation procedure. Most of the lost cells were from the peripherally arranged A, D, and PP cells, which led to quantitative changes of the cellular populations of the islets. This might offer an explanation for the incomplete metabolic control which has been a consistent finding in the short-term studies of 30 60 90 l l l l l , l j l l l , I 120 150 180 210 240 pancreatic islet transplantation, and the recurrence of hyperglycemia in long-term studies (Steffes et al., 1979; Trimble et al., 1980; Orloff et al., 1987). Evidently, to control diabetes mellitus by transplantation techniques, isolated islet tissue with total cellular integrity seems to be essential. It, therefore, appears that further refinements of the isolation techniques are necessary before a complete success could be achieved. ACKNOWLEDGMENTS We thank Prof. Ali Naji, Department of Surgery, Hospital of the University of Pennsylvania (Philadelphia, PA), for his valuable help in developing the techniques for this work. The authors gratefully acknowledge Mr. Kabuye Sebastian and Mr. Mohamed Abdelrazak Saleeh for their excellent technical assistance and Mrs. Kim Ishgi for her excellent secretarial help. This work was supported by grant No. 171410 from King Abdulaziz University, Jeddah, Kingdom of Saudi Arabia. ISOLATED PANCREATIC ISLETS OF THE RAT LITERATURE CITED Alejandro, R., R.G. Cutfield, F.L. Shienvold, K.S. Polonsky, J . Noel, L. Olson, J. Dillberger, J. Miller, and D.H. Mintz 1986 Natural history of intrahepatic canine islet cell autografts. J . Clin. Invest., 78:1339-1348. Ardawi, M.S.M., and E.A. Newsholme 1985 Fuel utilization in colonocytes of the rat. Biochem. J., 231:713-719. Baetens, D., F. Malaise-Lagae, A. Perrelet, and L. Orci 1979 Endocrine pancreas: three-dimensional reconstruction shows two types of islets of Langerhans. Science, 206t1323-1325. Buchanan, K.D., and W.A.A. Mawhinney 1973 Glucagon release from isolated pancreas in streptozotocin-treated rats. Diabetes, 22: 797-800. Carpenter, A.M., and A. Lazarow 1967 Effect of hyper- and hypoglycemia on beta cell degranulation and glycogen infiltration in normal, subdiabetic and diabetic rats. Diabetes, 16:493-501. Dean, P.M. 1973 Ultrastructural morphometry of the pancreatic B-cell. Diabetologia 9t115-119. Delaby, J., M. Campana, C. Laborie, and R. Assan 1989 Reproducible high yields of rat islets of Langerhans. Diabetes Metab. Rev., 15(3):123-127. Dibelius, A,, H. Konigsberger, P. Walter, W. Permanetter, W. Brendel, and B.U. Von-Specht 1986 Prolonged reversal of diabetes in the rat by transplantation of allogeneic islets from a single donor and cyclosporine treatment. Transplantation, 41(4):426-431. Eizirik, D.L., S. Sandler, 0. Korsgren, L. Jansson, and A. Anderson 1989 Persistent impairment of the insulin response to glucose both in vivo and in vitro after streptozotocin exposure: studies with grafted pancreatic islets and islets maintained in culture. Acta Endocrinol. (Copenh.), 121:849-856. Gray, D.W.R., P. McShane, A. Grant, and P.J. Morris 1984 A method for isolation of islets of Langerhans from the human pancreas. Diabetes, 33: 1055-1061. Grodsky, G.M. 1989 Perspectives in diabetes. A new phase of insulin secretion. How will it contribute to our understanding of B-cell function? Diabetes, 38t673-678. Groth, C.G., A. Anderson, C. Bjorken, R. Gunnarsson, C. Hellerstrom, G. Lundgren, B. Petersson, I. Swenne, and J. Ostman 1980 Transplantation of fetal pancreatic microfragments via the portal vein to a diabetic patient. Diabetes, 29fSuppl. 1):80-83. Halban, P.A., M. Amherdt, L. Orci, and A.E. Renold 1986 Tris (hydroxymethyl) amino-methane inhibits the synthesis and processing of proinsulin in isolated rat pancreatic islets without affecting release of insulin stores. Diabetes, 35:433-439. Hellman, B. 1959 The total volume of the pancreatic islet tissue at different ages of the rat. Acta Pathol. Microbiol. Immunol. Scand., 47t35-50. Henquin, J.C., and H.P. Meissner 1984 Effects of theophylline and dibutyryl cyclic adenosine monophosphate on the membrane potential of mouse pancreatic B-cells. J . Physiol. (Lond.), 351 :595612. Hesse, U., D.E.R. Sutherland, G. Florack, J . Field, S. Ward, and J.S. Najarian 1986 Comparison of two methods of islet preparation and transplantation in dogs. Diabetes, 35r1109-1118. Hsu, S.M., L. Raine, and H. Fanger 1981 Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabelled antibody (PAP) procedures. J. Histochem. Cytochem., 29(4):577-580. Huxlin, K.R., B.E. Tuch, A.J. Sefton, and G. Dixon 1990 Grafting of fetal pancreata into neonatal rat brain. Transplantation, 49t857861. Jaworek, D., W. Gruber, and H.U. Bergmeyer 1974 AMP and ADP determination. In: Methods of Enzymatic Analysis. H.U. Bergmeyer, ed. Academic Press, London, pp. 2127-2131. Kneteman, N.M., D. Alderson, D.W. Scharp, and P.E. Lacy 1989 Longterm cryogenic storage of purified adult human islets of Langerhans. Diabetes, 38:386-396. Lacy, P.E., and M. Kostianovsky 1967 Method for the isolation of intact islets of Langerhans from the rat pancreas. Diabetes, 16: 35-39. Lacy, P.E., M.M. Walker, and C.J. Fink 1972 Perifusion of isolated rat islets in vitro. The participation of the microtubular system in the biphasic release of insulin. Diabetes, 21 r987-998. Lacy, P.E., D.A. Young, and C.J. Fink 1968 Studies on insulin secretion in vitro from isolated islets of the rat pancreas. Endocrinology, 83:1155-1161. Lake, S.P., J . Chamberlain, P. Husken, P.R. Bell, and R.F. James 1988 In vivo assessment of isolated pancreatic islet viability using the 497 streptozotocin-induced diabetic nude rat. Diabetologia, 31(6): 390-394. Lamprecht, W., and I. Trautschold 1974 ATP and glucose 6-phosphate determination. In: Methods of Enzymatic Analysis. H.U. Bergmeyer, ed. Academic Press, London, pp. 2101-2110. Lowry, O.H., N.J. Rosebrough, L.A. Farr, and R.J. Randall 1951 Protein measurement with the Folin-phenol reagent. J. Biol. Chem., 193:265-275. Malaisse, W.J. 1973 Insulin secretion: multifactorial regulation for a single process of release. Diabetologia, 9:167-173. McEvoy, R.C., and P.E. Leung 1983 Transplantation of fetal rat islets into the cerebral ventricles of alloxan-diabetic rats. Amelioration of diabetes by syngeneic but not allogeneic islets. Diabetes, 32: 852-857. Meda, P., R.L. Michaels, P.A. Halhan, L. Orci, and J.D. Sheridan 1983 In vivo modulation of gap junctions and dye coupling between B cells of the intact pancreatic islet. Diabetes, 32t858-868. Moskalewski, S. 1965 Isolation and culture of the islets of Langerhans of the guinea pig. Gen. Comp. Endocrinol., 5r342-353. Orci, L., F. Malaisse-Lagae, M. Ravazzola, D. Rouiller, A.E. Renold, A. Perrelet, and R. Unger 1975 A morphological basis for intercellular communication between A- and B- cells in the endocrine pancreas. J. Clin. Invest., 56:1066-1070. Orci, L., R.H. Unger, and A.E. Renold 1973 Structural coupling between pancreatic islet cells. Experientia, 29(8):1015-1018. Orloff, M.J., C. Macedo, A. Macedo, and G.E. Greenleaf 1987 Comparison of whole pancreas and pancreatic islet transplantation in controlling nephropathy and metabolic disorders of diabetes. Ann. Surg., 206(3):324-334. Pipeleers, D., P. in’t Veld, E. Maes, and M. Van De Winkel 1982 Glucose-induced insulin release depends on functional cooperation between islet cells. Proc. Natl. Acad. Sci. U.S.A., 79:73227325. Slavin, B.G., P.M. Beigelman, and S.P. Bessman 1977 Cytophysiological studies on isolated pancreatic islets in vitro. Anat. Rec., 188: 445-452. Steffes, M.W., D.M. Brown, J.M. Basgen, A.J. Matas, and S.M. Mauer 1979 Glomerular basement membrane thickness following islet transplantation in the diabetic rat. Lab. Invest., 41:116-118. Sternberger, L.A. 1979 The unlabelled antibody peroxidase-antiperoxidase (PAP) method. In: Immunocytochemistry, 2nd ed. S. Cohen and R.T. McCluskey, eds. John Wiley and Sons, New York, pp. 104-169. Sutherland, D.E.R., A.J. Matas, F.C. Goetz, and J.S. Najarian 1980 Transplantation of dispersed pancreatic islet tissue in humans. Autografts and allografts. Diabetes, 29fSuppl. 1):31-44. Sutton, R., M. Peters, P. McShane, D.W.R. Gray, andP.J. Morris 1986 Isolation of rat pancreatic islets by ductal injection of collagenase. Transplantation, 42(6):689-691. Toledo-Pereyra, L.H., K.O. Bandlien, D.A. Gordon, G.H. MacKenzie, and T.A. Reyman 1984 Renal subcapsular islet cell transplantation. Diabetes, 33t910-914. nimble, E.R., C. Karakash, F. Malaise-Lagae, I. Vassutine, L. Orci, and A.E. Renold 1980 Effects of intraportal islet transplantation on the transplanted tissue and the recipient pancreas. I. Functional studies. Diabetes, 29:341-347. Tze, W.J., and J. Tai 1988 Allotransplantation of dispersed single pancreatic endocrine cells in diabetic rats. Diabetes, 37r383-392. Unger, R.H., and L. Orci 1981 Glucagon and the A cell. Physiology and pathophysiology. N. Engl. J . Med., 304t1518-1524. Vance, J.E., K.D. Buchanan, D.R. Challoner, and R.H. Williams 1968 Effect of glucose concentration on insulin and glucagon release from isolated islets of Langerhans of the rat. Diabetes, 17t187193. Warnock, G.L., D.K. Ellis, M. Cattral, D. Untch, N.M. Kneteman, and R.V. Rajotte 1989 Viable purified islets of Langerhans from collagenase-perfused human pancreas. Diabetes, 38fSuppl. l):136139. Warnock, G.L., and R.V. Rajotte 1988 Critical mass of purified islets that induce normoglycemia after implantation into dogs. Diabetes, 37t467-470. Weibel, E.R. 1963 Principles and methods for the morphometric study of the lung and other organs. Lab. Invest., 12(2):131-155. Williams, M.A. 1977 Quantitative methods in biology. In: Practical Methods in Electron Microscopy. A.M. Glauert, ed. North-Holland Publishing Company, Amsterdam, Val. 6, pp. 48-62.