Distribution of cell types of the islets of langerhans throughout the pancreas of the Chacma baboon.код для вставкиСкачать
THE ANATOMICAL RECORD 217:172-177 (1987) Distribution of Cell Types of the Islets of Langerhans Throughout the Pancreas of the Chacma Baboon SONIA A. WOLFE-COOTE AND D.F. DU TOIT Research Institute for Medical Biophysics, South African Medical Research Council (S.A.W.-C.), and Department of Surgery, Medical School, University of Stellenbosch (D.I?D.), Tygerberg, 7505, Republic of South Africa ABSTRACT Biopsies of the pancreas head, tail, and uncinate regions of four baboons were processed for immunocytochemical (ICC) studies by using avidinbiotin-peroxidase label for light microscopy (LM). Toluidine-blue- or methylene-bluestained 0.5-pm sections of nonosmicated resin-embedded tissue were viewed to locate areas of suitable islets. For ICC investigations, batches 10 pm apart of ten consecutive 1-pm sections throughout ten islets from each of the three regions were immunolabelled for LM. Four slides in each batch were immunolabelled consecutively for insulin, glucagon, somatostatin, and pancreatic polypeptide, the fifth acting as one of the range of controls in each batch. The number of each of the four cell types was counted in at least ten immunolabelled islets from each of the pancreas heads, uncinate portions, and tails. The uncinate region and not the head, as in most mammals, was found to contain significantly higher numbers of pancreatic polypeptide (PP)cells and lower numbers of A (glucagon) and D (somatostatin) cells (P< .001).The PP cells occurred in clumps and not as described in other mammals as part of the mantle of A, D, and PP cells. PP and A cell numbers were significantly different for each region (P< .001), being lowest in the head for PP and in the uncinate process for A cells. D cell distribution was similar to that of the A cells whilst a significantly smaller number of B (insulin) cells was found in the tail compared with other regions (P< .001). The varying distribution of cells in islets in different parts of the pancreas was reported in 1976 (Orci et al., 1976) in the rat, with glucagon-rich islets occurring in the tail, body, and superior head (dorsal) whilst pancreatic polypeptide (PP)-rich islets were found in the middle and inferior head (ventral). In the same year it was reported that PP cells in the dog were more abundant in the portion of the pancreas adjacent to the duodenum (Larsson et al., 1976)and that they were identical to the previously described F (or X) cells found mainly in the uncinate process Wunger et al., 1965). In 1978 O’Rahilly and Muller reported a heterogeneity in cellular composition of the human endocrine pancreas depending on the anatomic location of the islets and linked it to the dual origin of pancreas observed during embryogenesis in mammals. A 3D reconstruction of rat pancreas (Baetens et al., 1979) confirmed earlier findings (Orci et al., 1976) of PP-rich, glucagon-poor islets in the areas developed from the ventral primordium and glucagon-rich, PP-poor islets in the area developed from the dorsal primordium. The study was repeated on human pancreas (Orci et al., 1978) with similar results. It was suggested from work on the dog and human that the concentration of PP is inversely proportional to that of insulin and glucagon and that PP concentration is significantly higher in the uncinate process and head of the pancreas than in the body and tail (Gersell et al., 0 1987 ALAN R. LISS, INC. 1979). These results seem to suggest a nonrandom arrangement of cells within a n islet or islets within the pancreas and this raises the question of whether there is a functional significance to this arrangement. The distribution of cell types of the islets of Langerhans throughout the pancreas of the Chacma baboon was investigated in this study. MATERIALS AND METHODS Tissue Processing Biopsies from the head, uncinate, and tail regions of the pancreas of four Chacma baboons (Pupio ursinus) of both sexes were taken under anaesthesia induced by sodium pentobarbitone (Sagatal, Maybaker) and maintained with ethrane. The tail biopsy was taken from the horizontal portion of the pancreas, lateral to the superior mesenteric vessels. The uncinate process biopsy was taken from the pancreas, inferior to the superior mesenteric vessels in the portion of the pancreas folding behind these vessels. The head biopsy was taken from the portion lying between the inner curve of the second part of the duodenum and the superior mesenteric vessels (Fig. 1).The tissue was processed for ultrastructural immunolabelling studies in the following way. The biopReceived June 16,1986; accepted August 1, 1986. 173 ISLET CELL DISTRIBUTION IN BABOON PANCREAS lrnrnunocytochernical Procedures A /H ! I I Fig. 1. Diagrammatic representation of the areas of biopsy in the baboon pancreas: head (H),uncinate (U)and tail (T). D = duodenum, a = superior mesenteric artery; v = superior mesenteric vein. Line AB in a represents the section plane viewed in b. sies were placed in a drop of 1% glutaraldehyde (GA) (Sabatini et al., 1963) in 0.1 M sodium cacodylate buffer, pH 7.4, and cut into slices no more than 1mm thick and roughly 2 mm broad by 4 mm long. Tissue slices of each biopsy were placed, per region of the pancreas, into 1% GA in the 0.1 M sodium cacodylate buffer (300 mOsmol) for 2 hours at pH 7.4. At each stage of the tissue processing the bottles were rotated a t room temperature (RT) (Hayat, 1970) in a Taab rotator type N. After two 10-minute washes in 0.1 M sodium cacodylate buffer, pH 7.4, the tissue was dehydrated in the following way: 50%, 70%, and 90% ethyl alcohol (EtOH) for 15 minutes each; 100% EtOH for two changes of 15 minutes each followed by 100% acetone for 30 minutes. The tissue was then placed in a 1:l mixture of acetone and ERL-4206 resin (vinyl cyclohexene dioxide) (Spurr, 1969)for 30 minutes followed by a n addition of a n equal volume of resin for a further 30 minutes and finally into pure resin overnight. After a further 4-6 hours impregnation with fresh resin, each tissue slice was embedded in resin in the flat “lid” end of a n inverted Beem capsule from which the narrow-shaped end had been cut. Thick (0.5-1.0 pm) survey sections of tissue were cut, stained with toluidine or methylene blue and viewed under a Zeiss light microscope until a n area containing suitable large islets was located. Five consecutive slides were prepared of the next ten 1-pm sections. Ten 1-pm sections were discarded before another five slides, each containing two 1-pm sections were prepared. A further ten 1-pm sections were discarded and the procedure was repeated until survey sections, stained at regular intervals through the cutting process, revealed that one or more islets had been cut through. One slide in each batch was immunostained for glucagon, one for insulin, one for somatostatin, one for PP, and the remaining one was used as a different control slide in each batch. In this way the whole range of absorption and method controls was performed over the area of tissue under scrutiny. This procedure was repeated in the head, the uncinate region, and the tail of the pancreas of four baboons. Details of the ultrastructural studies have been submitted separately for publication. Before the immunolabelling procedure for light microscopy (LM) the resin of the 0.5-1-pm sections was removed by adding a drop of sodium methoxide to the warmed slide for 5-10 seconds before washing in methyl alcohol (Mayor et al., 1961). The sections were then quenched of endogenous peroxidase by incubating for 30 minutes in 0.3% hydrogen peroxide in methanol. Immunolabelling was then performed by using the vectastain ABC kit (Vector Laboratories) according to the instructions provided by the manufacturers. Details of the primary antisera used are given in Table 1. The peroxide marker was revealed by incubating the sections for 2-5 minutes in a 0.05% enzyme substrate solution of diaminobenzidine tetrahydrochloride (DAB) containing 0.01% hydrogen peroxide. This resulted in a brown stain deposit at the areas of positive immunoreactivity. The sections were counterstained with methylene or toluidine blue or on occasions with methylene blue-basic fuchsin stain and finally mounted in DPX. Both first- and second-level controls were used (Larsson, 1981).The first-level controls involved replacement of primary antiserum with antiserum absorbed with antigen. The second-level controls involved omission of either the primary antiserum on the second-layer antibody or the avidin D-biotinylated horseradish peroxidase H complex. TABLE 1. Antisera details Antigen to which antiserum raised Porcine PP midpart of the C terminal Pancreatic glucagon Insulin Somatostatin Synthetic ( S ) or Natural (N) S Code No. Antiserum source Gift from Dr. T. Paquette Working dilution (light microscopy) 1:600 N (porcine) 013-P Dako 1:400 S S Dako 043-P Dako 1:lOO 1:400 174 SA. WOLFE-COOTE AND D.F. DU TOIT Statistical Analysis of Cell Distributions Light micrographs of avidin-biotin-peroxidaselabelled islet tissue printed at a final magnification of between 400 and 600 times were used for the following analysis. The number of insulin-, glucagon-, somatostatin-, and PP-containing cells was counted in as many islets as possible from the four different baboon head, uncinate, and tail regions of the pancreas. The counts were then expressed as a percentage of the whole for each islet. The percentage of each cell type in each region was compared in all four baboons for any significant differences which would preclude a pooling of data for a final analysis. The subgroups of each of the cell type percentage data in the three regions of the pancreas were analysed by using paired t comparisons (assuming unequal variances) with Bonferroni probabilities. P values of < .001 were considered significant. Ideally ratios should not be used for analysis because the cell for comparison is involved in both the numerator and denominator so that a dependency is introduced. The ratio will then not follow any statistical distribution and should not be analysed by using a statistical test. Percentage values were analysed however, because the number of cells in an islet is dependent on the size of the islet. Absolute values are just not comparable without considering some method of standardization between islets. Preparation of islet cell distribution maps Adjacent sections from each islet were immunolabelled with ABC for insulin, glucagon, somatostatin, and PP. The immunolabelled cells for each hormone were traced onto a single acetate sheet with a Stabilo superfine overhead projection pen and the resulting maps were photographed. RESULTS Figure 2 provides an illustration of the type of labelling achieved for glucogen in one islet from the baboon pancreas tail. Figures 3-5 were obtained by tracing cells labelled for each hormone on adjacent sections from each islet onto a single acetate sheet which was then photographed to display the distribution of islet cells in sections of islets from each region of the pancreas. A comparison of islet cell counts from four different baboons for the head, uncinate, and tail regions of the pancreas revealed no significant difference between the corresponding counts for each baboon. Data from all of the baboons were thus able to be pooled to compare the variations in distribution of cell types in islets from different regions of the pancreas. Mean percentage content of each individual cell type in each of the three regions of the pancreas is illustrated in Table 2. The standard deviations illustrate the considerable variation within each region but separate and pooled variance analysis revealed a significantly lower proportion of insulin (B) cells ( P < .001) in the tail compared with head and uncinate regions. The numbers of glucagon (A) cells for each of the three regions were significantly different ( P < .001). The highest proportion of A cells was in the tail, with significantly lower values in the head and minimal counts in the uncinate region. The proportion of somatostatin (D) cells was 176 slightly higher in the tail than in the head but both values were significantly higher than in the uncinate region ( P < .001). The values for PP cells were significantly different for all three regions ( P < .001). The highest number occurred in the uncinate region with a gradation through tail to the lowest count in the head. DISCUSSION It was assumed in the early 1970s that all islets of a pancreas were similarly constructed, and when, in the rat, “atypical” islets were observed with a low glucagon cell content but high PP cell content (Orci et al., 1976), islets were systematically sampled from various regions of the pancreas. Glucagon-rich islets were found in the tail, body, or superior part of the head of the pancreas (the dorsal region) which was vascularized by the gastroduodenal and splenic arteries of the coeliac trunk, whilst PP-rich islets always appeared in the middle and inferior part of the head (the ventral region), vascularized by a branch of the superior mesenteric artery (Baetens et al., 1979; Orci, 1982). These results represented another nonrandom situation further evidence for which was obtained from investigation into the cell distribution in the human pancreas (Orci and Perrelet, 1978). Results presented here for the baboon suggest a similar nonrandom situation and although the details of cell distribution for each species are slightly different, there emerges a definite pattern which has been suggested to be related to the embryonic development of the pancreas. In the embryo the pancreas originates as a ventral and a dorsal bud evaginating from the primitive gut (O’Rahilly and Muller, 1978).Eventually the ventral bud moves to a dorsal position to merge with the dorsal bud where it forms the posterior part of the pancreas head. The most striking difference between islets from the ventral and dorsal regions of the pancreas is their content of A and PP cells. In general, there appears to be a complementary arrangement of A and PP cells so that the part of the pancreas developed from the ventral bud, which has been found to be rich in PP cells, is very low in A cell content whilst the dorsal area of the pancreas is rich in A cells and low in PP cells. In the baboons studied, the location of the high PP cell content and low A cell content was even more specifically found in the uncinate process. Throughout the rest of the head (ventral) region even fewer PP cells than in the tail were observed and the A cell content was midway between that of the tail and the uncinate region. In the mammal generally and the rat and human in particular, the PP cells are described as forming part of Fig. 2. Light micrograph of section of an islet from the pancreas tail immunolabelled with avidin-biotin-peroxidasecomplex (ABC) for PP. The section was counterstained with methylene blue-basic fuchsin stain. Final magnification x576. Figs. 3-5. Maps displaying the distribution of islet cells in sections of islets from the head (Fig. 3), uncinate (Fig. 4), and tail (Fig. 5) regions of the Chacma baboon pancreas. Black represents insulin containing B cells; green represents glucagon containing A cells; orange represents somatostatin containing D cells and n1.r-l- - cells. ISLET CELL DISTRIBUTION IN BABOON PANCREAS 175 176 S.A. WOLFE-COOTE AND D.F. DU TOIT TABLE 2. Distribution of cell types in islets from the head, uncinate, and tail regions of the baboon pancreas' Mean percentage distribution of cell types Region of pancreas Ventral head Ventral uncinate Dorsal tail B D A Mean SD 75 62.9 40.3* 20.3 26.5 20.5 Mean 12.7* O* 25.2* SD 12.6 0 13 Mean 10.9 0.1* 17.0 PP SD Mean SD 9.38 0.69 10.07 1.3* 36.9* 17.5* 3.63 27.08 20.04 'B, insulin; A, glucagon; D, somatostatin; PP, pancreatic polypeptide. *Signifies for each cell type, where the distribution differs significantly (P < .001). the mantle of A, D, and PP cells around the central core of B cells. In contrast the PP cells in the baboon seemed to occur in a clump both in the uncinate region and, when they were present, in large numbers in the tail. Only occasionally were they seen singly in the islets of the tail (dorsal) region. In the rat and the human, the distribution of D and B cells had been found to be relatively constant. In the baboon the D cell content was significantly lower in the uncinate region and appeared to mirror closely the A cell distribution, being highest in the tail (dorsal region). The observation, at times, of A cells in the absence of D cells and vice versa in the baboon head or uncinate region has only been reported before in the canine uncinate process (cited in Orci, 1975). The number of B cells was significantly lower in the tail (dorsal) region compared with the head and uncinate (ventral). It is interesting that in the rat, where dorsal and ventral islets were found to have similar numbers of insulin-producing B cells, dorsal islets actually secreted significantly more insulin than did ventral islets in the presence of 16.7 mM glucose and 20 mM D-glyceraldehyde which inhibits somatostatin but augments insulin release (Trimble and Renold, 1981). It is possible that the larger number of glucagon cells in the dorsal islets may play some contributory role in greater insulin secretion since glucagon is known to stimulate insulin release. The authors detected more glucagon secretion in the dorsal islets than the ventral islets in the presence of 16.7 mM glucose. Perhaps, in the baboon, the significantly higher number of B cells in the uncinate region is a n internal compensation for the almost total lack of A cells observed. It would be interesting to investigate in the baboon pancreas the distinct areas which have developed from the dorsal and ventral primordia. It is tempting to speculate that the region developed from the ventral primordium in the baboon is indeed the uncinate region and not the whole posterior portion of the head as described in other mammals. If this is not true then the explanation that the variation in islet composition is a function of the embryological development of the pancreas loses a lot of its credibility. It had been observed 20 years ago that the uncinate process of the canine pancreas contained few, if any, glucagon-producing A cells (Munger et al., 1965) although it has been suggested that these authors incorrectly identified the right lobe of the dog pancreas a s the uncinate lobe. These authors noted another cell type abundant in islets in this region and at that time they called them F cells. An abundance of PP cells was confirmed in the uncinate region of the dog by using im- munocytochemical techniques (Orci, 1975; Floyd et al., 1977). After a personal communication of similar findings from Larsson (1974), Chance et al. (1976) found, using RIA on acid-alcohol extracts of the uncinate and tail, that the uncinate process could have as much as ten times the content of PP as the tail region of the dog. Confirmatory results followed (Baetens et al., 1976, 1979). The finding of most PP cells in the uncinate region of both the dog and the baboon illustrates yet another similarity between the baboon and canine endocrine pancreas. ACKNOWLEDGMENTS We are indebted to the Medical Research Council of South Africa and the University of Stellenbosch and we thank Dr. R.S. Day for his assistance in analysing the data, Messrs. C. Bruintjies, A. Tomboer and J. Witbooi for their technical assistance, and Mrs. W. van Eyssen for typing the manuscript. Antiporcine PP was a generous gift from Dr. T. Paquette, Diabetes Research Centre, University of Washington. The work presented here has been taken from a Ph.D. thesis submitted to the University of Stellenbosch in 1985 under the promotion of Dr. D.F. du Toit. LITERATURE CITED Baetens, D., F. Malaise-Lagae, A. Perrelet, and L. Orci (1979) Endocrine pancreas: Three dimensional reconstruction shows two types of islets of Langerhans. Science, 206:1323-1325. Baetens, D., C. Rufener, and L. Orci (1976) Localization of bovine pancreatic polypeptide-like immunoreactivity in rat pancreatic monolayer culture. Experientia. 32:919-920. Chance, R.E., R.L. Gingerich, and M.H. Greider (1977) Cited in discussion in Floyd et al. (1977) A newly recognized PP. Rec. Prog. Horm. Res., 33.519-570. Floyd, J.C., Jr., S.S. Fajans, and S. Pek (1977) Regulation in healthy subjects of the secretion of human pancreatic polypeptide, a newly recognised pancreatic islet peptide. Trans. Assoc. Am. Phys., 89:146-158. Gersell, D.J., R.L. Gingerich, and M.H. Greider (1979) Regional distribution and concentration of pancreatic polypeptide in the human and canine pancreas. Diabetes, 28:ll-15. Hayat, M.A. (1970) Principles and techniques of electron microscopy: biological applications. Van Nostrand-Rheinhold,New York, Vol. 1. Larsson, L-I. (1981) Peptide immunocytochemistry. Prog. Histochem. Cytochem., 13:(no. 4)l-85. Larsson, L-I., F. Sundler, and R. Hakanson (1976)Pancreatic polypeptide-a postulated new hormone: Identification of its cellular storage site by light and electron microscopic immunocytochemistry. Diabetologia, 12:211-226. Mayor, H.D., J.C. Hampton, and B. Rosario (1961) A simple method for removing the resin from epoxy-embedded tissue. J. Biophys. Biochem. Cytol., 9:909-910. Munger, B.L., I. Caramia, and P. Lacy (1965)The ultrastructural basis for the identification of cell types in the pancreatic islets 11. Rabbit, dog and opossum. Z. Zellforsch Mikrosk. Anat., 67:776-798. ISLET CELL DISTRIBUTION IN BABOON PANCREAS 177 O’Rahilly, R., and F. Muller (1978)A model of the pancreas to illustrate Orci, L., and A. Perrelet (1978) Cell Surface Reviews, Membrane Fuits development. Acta Anat., 100:380-385. sion Vol. 5. G. Poste, and G.L. Nicholson, eds. NY, Elsevier Bumed Press, Ch. 12, p. 629. Orci, L. (1975) Functional subdivision of Islets of Langerhans and Sabatini, D.D., K. Bensch, and R. Barrnett (1963) Cytochemistry and possible role of D cells. Lancet Dec. 20, pp. 1243-44. electron microscopy. The preservation of cellular ultrastructure Orci, L. (1982) Macro- and micro-domains in the endocrine pancreas. and enzymatic activity by aldehyde fixation. J. Cell Biol., 17:19Diabetes, 31:538-564. 58. Orci, L., D. Baetens, M. Ravazzola, Y. Stefan, and F. Malaise-Lagae (1976) Pancreatic polypeptide and glucagon: Non random distribu- Spurr, A.R. (1969)A low viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastructure Res. 26r31-43. tion in pancreatic islets. Life Sci., 19:1811-1815. Orci, L., F. Malaise-Lagae, D. Baetens, and A. Perrelet (1978) Pan- Trimble, E.R., and A.E. Renold (1981) Ventral and dorsal areas of rat creatic polypeptide rich regions in the human pancreas. Lancet pancreas: Islet hormone content and secretion. Am. J. Physiol., Dec. 2,1978; pp. 1200-1201. 240:E422-E427.