Obliterative Segmental Sclerosis of Pancreatic Islets - Europe PMC

Obliterative Segmental Sclerosis A Possible Consequence

of

Pancreatic Islets

of Hypotensive

T. A. SEEMAYER, J.-F. YALE, J.-P. de CHADAREVIAN, M. GROSE, and E. B. MARLISS

Shock in Young BB Rats

From the Department of Pathology, The Montreal Children's Hospital, and McGill Nutrition and Food Science Centre, Royal Victoria Hospital, McGill University Faculty of Medicine, Montreal, Canada

During the conduct of a time-course study of peripheral blood T lymphocytes in BB rats, a hitherto unreported islet lesion was recognized in 8 of 17 glucose-tolerant rats which had been subjected to repetitive cardiac puncture in early life and sacrificed at 186-190 days of age. Some of the large islets contiguous to fibrous septa demonstrated variable degrees of obliterative sclerosis and reduction of islet cell mass, associated with evidence of remote periand intrainsular hemorrhage. In 3 rats, additional features included an active, proliferative replacement of islet substance by fibroblasts and histiocytes. These islet alterations are not characteristic of BB rats, nor have

identical lesions been described in studies of experimental or human diabetes meflitus. Affected rats had sustained repetitive episodes of blood loss (from 15% to 28% of total blood volume each time) during the early phase of the study. It is possible, but not proved, that hypovolemic shock was induced in some of these animals. Ischemic necrosis of islets has been reported in premature and young infants succumbing to shock. It is posited that hypovolemic shock in the young animal may, as in the infant, result in islet cell ischemic necrosis, sufficient to produce, with time, the described islet morphologic features. (Am J Pathol 1986, 125:327-331)

THE PANCREATIC ISLETS of Langerhans are subject to a diverse assortment of pathologic alterations. Indeed, numerous viruses, chemicals, and endocrine alterations may, under appropriate conditions, visit cellular injury on islets, sufficient to induce glucose intolerance and diabetes mellitus. Islets are also liable to structural and/or functional modification with aging, as well as with systemic diseases which affect the pancreas. Type 1 (insulin-dependent) diabetes in the BB rat and man classically shows "insulitis" (mononuclear cell infiltration of the islets with selective beta-cell necrosis) in the active phase of the pathologic process. In the course of examining the pancreas after a study of T-lymphocyte subsets in the BB rat, unusual islet lesions were recognized in the absence of the classical insulitis or any demonstrable abnormality of glucose tolerance. These lesions were seen in 2 of 21 rats sacrificed at 120-125 days of age and in 8 of 17 animals sacrificed at 186-190 days of age. The islet changes, present in both diabetes-prone (BBdp) and non-diabetesprone (BBn) rats, consisted of variable degrees of obliterative sclerosis, often segmental, associated with intra- and periinsular aggregates of hemosiderin-laden

macrophages. The description of this peculiar islet alteration forms the substance of this report.

Materials and Methods Experimental Animals Pregnant BB (BBdp and BBn) rats were obtained from Dr. Pierre Thibert, (Animal Resources Division, Health and Welfare Canada, Ottawa, Ontario, Canada), and pregnant Wistar rats unrelated to the BB strain from Charles River Laboratories (St-Constant, Quebec, CanSupported in part by grants from The Medical Research Council of Canada (MA 7636-TAS; MA 6540-EBM). Dr. J.F. Yale was the recipient of a Medical Research Council of Canada Postdoctoral Fellowship. Presented in part at the XII Congress of the International Diabetes Federation, Madrid, Spain, 1985 (Diabetes Res Clin Pract 1985, Suppl 1:S507). Accepted for publication June 23, 1986. Address reprint requests to Thomas A. Seemayer, MD, Department of Pathology, Montreal Children's Hospital, 2300 Tupper Street, Montreal, Quebec, Canada H3H 1P3.

327

328

SEEMAYER ET AL

ada), for time-course studies of peripheral blood Tlymphocyte subsets. Results from this immunologic study are presented elsewhere. ' The dams and their litters were maintained in cages in laminar flow hoods in rooms controlled for temperature (20 C) and humidity (70%). All animals had free access to food (Purina Rat Chow, Ralston Purina, St. Louis, Mo) and water.

AJP * November 1986

Table 1-Blood Volumes Removed During Sampling Estimated Blood volume % Total Body removed blood volume total blood weight Age removed (ml) volume (ml)* (g) (days) 27 ± 1 13-18 34.3 + 0.8 2.06 ± 0.05 0.54 ± 0.02 5.22 ± 0.16 26-32 87.1 ± 2.7 42-54 215.8 ± 6.6 12.95 ± 0.40 68-78 313.5 ± 11.5 18.81 + 0.69 *

Experimental Design Blood was sampled regularly between 13 and 72 days of age (Table 1), sequentially in all rats. Rats received light ether anesthesia, and blood was withdrawn by cardiac puncture into a heparinized syringe with a 25gauge needle. The amount of blood removed at the youngest age was the maximum compatible with 10007 survival, as established in previous experiments. After cardiac puncture, the animals were kept in a warmed box until awake and walking and then were returned to their dams. The latter were exposed briefly to ether to prevent rejection of the pups. Body weight, urine volume, and urine glucose content were recorded daily throughout the study period. A glucose tolerance test (plasma glucose measured fasting and 60 minutes after a 2.5-g/kg glucose load by gavage) was performed at 120 days in BBdp rats not yet diabetic. Thirty (7 BBdp, 14 BBn, 9 Wistar) glucosetolerant rats were then sacrificed. Thirty-seven other rats (23 BBdp, 14 BBn) that developed neither diabetes nor glucose intolerance were followed until 186-190 days of age, at which time they were sacrificed. At sacrifice, the pancreas was dissected, weighed, and either frozen for measurement of insulin content (n = 14 BBdp, 6 BBn) or fixed in Bouin's solution for histologic examination (n= 16 BBdp, 22 BBn, 9 Wistar). Rats in which diabetes or glucose intolerance developed showed the previously described morphologic changes2 and are not included in the present description. Assays Plasma glucose was measured by the glucose oxidase method on a Beckman II glucose analyzer (Beckman Instruments, Fullerton, CA, USA). The pancreatic extracts for total insulin content measurement were prepared by the use of methods previously described.3 The pancreases were homogenized by ultrasonic disintegration in 8 ml of an acid-ethanol solution (75'%o vol/vol ethanol, 1.5%o vol/vol 12 mol/l HCI, 23.5%o vol/vol distilled water); the extracts (supernatant after a single centrifugation at 6000g for 15 minutes) were kept at - 20 C until assay. Insulin was measured on appropriate dilutions of pancreatic extracts with the use of antibeef insulin antibody (Dr. Peter Wright), '251-porcine insulin, and rat insulin standard (24.5 IU/mg) (Novo

1.47 ± 0.08 2.35 ± 0.05 2.64 ± 0.05

28 ± 1 19 + 1 15 + 1

Theoretical blood volume according to weight 41

Research Laboratory, Copenhagen, Denmark) with dextran-coated charcoal separation of bound from free insulin. Histology The fixed pancreases were embedded in paraffin. Step sections were prepared with hematoxylin-phloxinesaffron (HPS), Congo-red, and Prussian blue stains. Serial sections were prepared for immunocytochemical study employing the peroxidase-anti-peroxidase (PAP) method to detect insulin- and glucagon-reactive cells, as previously described.2

Results The blood volumes removed at 13-18, 26-32, and 42-54 days of age were estimated to be 19-2807o of total blood volume4 (Table 1). The volumes removed at 68-78 days were about 15%7o. All rats survived, although many under 30 days of age were cold to the touch and unable to walk normally until 30-60 minutes after the blood sampling. Their growth, as estimated by body weight gain, was normal (data not shown). The mean (± SEM) pancreatic insulin content was similar for BBdp (167.5 ± 23.7 gg/g) and BBn rats (161.9 + 28.2 ig/g). Insulitis was not present in either group of rats. Foci of periductular and/or acinar mononuclear cell infiltration' 6 were identified in 11 of 16 BBdp rats and 4 of 22 BBn rats. As a rule, these infiltrates were more numerous and intense in BBdp rats. Distinct alterations were recognized in approximately 1-10%7o (morphometric studies were not performed) of islets in 5 BBdp and 3 BBn rats at 186-190 days of age, and in 2 BBn rats at 120-125 days of age. No abnormalities were found in any of the Wistar rats examined at 120-125 days of age. Involved islets, generally large islets contiguous to fibrous septae, were the site of segmental or multifocal obliterative sclerosis, with replacement by dense, often hyalinized, collagen (Figures 1 and 2). In some, the glossy character of collagen resembled amyloid, although Congo-red-stained sections were devoid of apple-green birefringence. While most islet lesions were sclerotic, cellular (proliferative) alterations

ISLET SCLEROSIS IN BB RATS

Vol. 125 * No. 2

329

Figure 1 -Photomicrograph of sclerotic islet contiguous to ductules and vessels. (HPS, x 100)

Figure 3-Photomicrograph of islet illustrating percentric proliferation of fibroblasts. (HPS, x 200)

were recognized in 3 (2 BBn, 1 BBdp) rats (Figures 3 and 4). Such islets were infiltrated by fibroblasts, histiocytes, and scant numbers of lymphocytes. In some of these islets, mitoses and extracellular deposits of fibrin were noted. Immunocytochemical studies revealed glucagon- and insulin-containing cells in both islet lesions, although the intensity of staining was reduced considerably, particularly in the cellular lesions. Hemosiderin-laden macrophages were noted within and, more especially, around sclerotic islets in contiguous fibrous septa (Figure 5). Apart from discrete foci of periductular and acinar lymphocytic infiltrates,5 no alterations were recognized in the exocrine or vascular pancreas.

This report describes an unusual pancreatic islet alteration in 5 (of 16) diabetes-prone glucose-tolerant and 5 (of 22) non-diabetes-prone BB rats. Approximately

1-1007e of large islets contiguous to fibrous septa were the site of cellular and/or sclerotic obliterative lesions. The effects seen would be expected to be insufficient to produce compromise of insulin-secretory capacity. This was confirmed by normal glucose tolerance and by normal pancreatic insulin contents. The cellular variant, seen in 3 rats, was represented by extensive islet infiltration with fibroblasts, histiocytes, and lesser numbers of lymphocytes. The sclerotic variant, present in all 10 affected rats, consisted of variable degrees of fibrous replacement of islet substance by dense collagen. The latter process could be likened to glomerular changes associated with sclerosing glomerulopathies. The sclerotic islets were surrounded by large numbers of hemosiderin-laden macrophages. Both insulin- and glucagon-immunoreactive cells were present in affected islets, although the intensity of staining was less than normal, especially in the cellular lesions. Intrainsular fibrosis has been described in both Type 2 and Type 1 human diabetes mellitus. In the former,

Figure 2-Photomicrograph of sclerotic islet illustrating segmental nature x 200)

Figure 4-Photomicrograph of islet illustrating cellular (proliferative) architectural effacement. (HPS, x 200)

Discussion

of desmoplasia. (HPS,

330

SEEMAYER ET AL

i .'j,1 ~i'

AJP

'"t>;!',9.EI~ ~. :Ii . .'% -!

... !,

....

;I

Ii:. :;'

U:~~~~~~~~~~~:

:;l1,

..:i

*:'R..

' . :',;*r; .

.,;,. . . .

d.1.

,

; g;...

)

...

., jilL S5

Figure 5-Photomicrograph of islet illustrating periinsular deposits of iron. (Prussian blue, x 200)

islet fibrosis is associated with vascular sclerosis and acinar fibrosis.' In Type 1 diabetes, islet fibrosis is seen infrequently, most often in pseudoatrophic islets devoid of B cells.8 Gepts suggests that such fibrosis probably represents a response to antecedent insulitis.8 Hyalinization of islets (amyloid alteration) is a relatively common finding associated with long-standing diabetes mellitus and aging.9 Some have proposed that a peculiar host response to endogenous insulin may be the basis for the formation of islet amyloid.10 Islet fibrosis in association with acinar fibrosis has been described in acromegalics I and patients with hemochromatosis.12 In the latter, the pancreas is the site of an extensive deposition of iron which presumably contributes to the process of fibrosis. Selective intrainsular fibrosis devoid of inflammation has been noted in high birth weight infants of diabetic mothers. 13 Experimental intrainsular fibrosis in association with insulitis has been induced after the administration of heterologous or homologous insulin14 and islets." Under appropriate conditions, inoculation of mice with the M variant of encephalomyocarditis virus may lead to islet necrosis with variable degrees of fibrosis within and around islets. 16 The islet alterations herein described are not those previously associated with the autoimmune attack directed to islets in BB rats.217 Review of our protocol disclosed that significant amounts of blood were withdrawn repetitively during this study (Table 1). Indeed. several rats (less than 30 days of age) became cold to the touch and weak after blood removal. It is possible, though not proven, that rats which manifested islet lesions sustained hypovolemic shock as a consequence of repeated blood drawing. Variable degrees of islet cell coagulative and/or hemorrhagic necrosis have been reported in stillborns and infants who succumbed early in life to shock and asphyxia."182' Had these infants sur-

*

November 1986

vived longer, it is possible their islets might have demonstrated fibrous replacement and the associated deposits of hemosiderin. The specificity of the effects of the postulated shock toward a certain population of islets could be related to the development and/or vascular perfusion of the islets. In the human, at this early stage, numerous single or grouped islets of Langerhans are situated in the fibrous stroma, and many of them reach considerable size.22 These are followed later by the development of islets in the lobules, a process which continues well into infancy. It has been shown in the bovine pancreas that there is a gradual growth of small islets throughout the last fetal trimester and during the early neonatal period, while there appears to be no additional significant increase in the diameter of large, earlier formed islets.23 While an analogous developmental sequence may not prevail in the rat pancreas, the vascular perfusion of islets in this species is a function of their size and volume: larger islets have decreased perfusion (relative to size and volume) than smaller islets.24 Thus, in the rat, the larger islets would be most vulnerable to ischemic injury stemming from shock. In this context we note that infusion in vivo of epinephrine induces a marked selective reduction in islet blood flow,25 a situation possibly relevant to the rats of the present study. Whether the islets of BB rats are more susceptible to development of this lesion than other rats that do not develop spontaneous diabetes remains to be established.

References 1. Yale JF, Grose M, Marliss EB: Time-course of the lymphopenia in BB rats: Relation to the onset of diabetes. Diabetes 1985, 34:955-959 2. Seemayer TA, Tannenbaum GS, Goldman H, Colle E: Dynamic time course studies of the spontaneously diabetic BB Wistar rat: III. Light-microscopic and ultrastructural observations of pancreatic islets of Langerhans. Am J Pathol 1982, 106:237-249 3. Junod A, Lambert AE, Stauffacher W, Renold AE: Di-

abetogenic action of Streptozotocin: Relationship of dose

to metabolic responses. J Clin Invest 1969, 48:2129-2139

4. Yale CE, Torhorst JB: Critical bleeding and plasma volumes of the adult germfree rat. Lab Anim Sci 1972, 22:497-502 5. Seemayer TA, Colle A, Tannenbaum GS, Oligny LL, Guttmann RD, Goldman H: Spontaneous diabetes mellitus syndrome in the rat: III. Pancreatic alterations in aglycosuric and untreated diabetic BB-Wistar-derived rats. Metabolism 1983, 32(Suppl 1):26-32 6. Guttmann RD, Colle E, Seemayer TA, Michel F: Spontaneous diabetes mellitus in the rat: A powerful probe of MHC and non-MHC immunogenetic interactions. Trans Proc 1982, 14:547-549 7. LeCompte PM, Gepts W: The pathology of juvenile diabetes, The Diabetic Pancreas. Edited by BW Volk, KF Wellman. New York, Plenum Press, 1979, pp 325-363 8. Gepts W: The pancreatic islets in diabetes. Am J Med 1981, 70:105-115

Vol. 125 * No. 2

9. Ehrlich J, Ratner IM: Amyloidosis of the islets of Langerhans: A restudy of islet hyaline in diabetic and nondiabetic individuals. Am J Pathol 1961, 38:49-59 10. Maloy AL, Longnecker DS, Greenberg ER: The relation of islet amyloid to the clinical type of diabetes. Hum Pathol 1981, 12:917-922 11. Volk BW, Wellmann KF: Hormonal diabetes,' pp 271-290 12. Volk BW, Wellmann KF: Hemochromatosis and diabetes,' pp 317-324 13. Hultquist GT, Olding LB: Pancreatic-islet fibrosis of young infants of diabetic mothers. Lancet 1975, 2:10151016 14. LeCompte PM, Steinke J, Soeldner JS, Renold AE: Changes in the islets of Langerhans in cows injected with heterologous and homologous insulin. Diabetes 1966, 15:586-596 15. Heydinger DK, Lacy PE: Islet cell changes in the rat following injection of homogenized islets. Diabetes 1974, 23:579-582 16. Craighead JE: Pathogenicity of the M and E variants of the encephalomyocarditis (EMC) virus. Am J Pathol 1966, 48:375-386 17. Nakhooda AF, Like AA, Chappel CI, Murray FT, Marliss EB: The spontaneously diabetic Wistar Rat: Metabolic and morphologic studies. Diabetes 1976, 26:100-112 18. Bernstein J: Renotubular and pancreatic islet necrosis in newly born infants. Am J Dis Child 1958, 96:705-710 19. Emery JL, Bury HP: Involutionary changes in the islets of Langerhans in the fetus and newborn. Biol Neonate 1964, 6:16-25

ISLET SCLEROSIS IN BB RATS

331

20. Coen R, McAdams AJ: Visceral manifestation of shock in congenital heart disease. Am J Dis Child 1970, 119: 383-389 21. Seemayer TA, Osborne C, de Chadarevian J-P: Shockrelated injury of pancreatic islets of Langerhans in newborn and young infants. Hum Pathol 1985, 16:1231-1234 22. Falin IL: The development and cytodifferentiation of the islets of Langerhans in human embryos and foetuses. Acta Anat 1967, 68:147-168 23. Bonner-Weir S, Like AA: A dual population of islets of Langerhans in bovine pancreas. Cell Tissue Res 1980, 206:157-170 24. Lifson N, Lassa CV, Dixit PK: Relation between blood flow and morphology in islet organ of rat pancreas. Am J Physiol 1985, 249 (Endocrinol Metab 12):E43-E48 25. Rooth P, Taljedal IB: In vivo fluorescence microscopy of islet blood flow. Diabetes Research and Clinical Practice 1985, Suppl 1: S477

Acknowledgments Dr. Pierre Thibert, of Ottawa, Canada, provided the experimental animals for this study. The authors are indebted to Marie Montambault and Leti Pegorari for expert technical assistance, to Dr. Susan Bonner-Weir for very helpful discussion, and to Jessie Blais and Laura Mennitto for secretarial assistance.