INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

Hie quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

A Bell & Howell information Company 300 North Zeeb Road. Ann Arbor. Ml 48106-1346 USA 313/761-4700 800/521-0600 CHARACTERIZATION OF THE SPONTANEOUSLY DIABETIC BB WISTAR RAT

DISSERTATION

Presented in Partial fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

By

James Roscoe Wright, Jr., B.S., M.A., M.D,

*****

The Ohio State University 1995

Dissertation Committee: Approved by C.C. Capen D.A. Senhauser

H.M. Sharma riser R.E. Stephens Ddpa^ftment of Pathology A. J . Yates UMI Number: 9526106

UMI Microform 9526106 Copyright 1995, by UMI Company. All rights reserved.

This microform edition is protected against unauthorized copying under Title 17, United States Code.

UMI 300 North Zeeb Road Ann Arbor, MI 48103 To Allan J. Yates, M.D., Ph.D.

He chased me around the world until I agreed to finish this 1 ACKNOWLEDGEMENTS

My sincere appreciation and thanks are extended to the following people:

To Allan J. Yates for his supervision in all aspects of this Ph.D. program. He has been a tremendous role model as both a clinician and a scientist. To Charles C. Capen, Donald A. Senhauser, Hari M. Sharma, Ralph E. Stephens, and Allan J. Yates for their help throughout this Ph.D. program and for their suggestions, criticisms, and editing of this dissertation. To the Ohio State University Department of Neuropathology for wonderful histologic, photographic, and secretarial support; without this help, none of the many papers included in the appendix of this dissertation would have been possible. To Hari M. Sharma and Chull Shim who patiently taught me histopathology. To Ralph E. Stephens and Kenneth C. Ford who taught me tissue culture and autoradiography. To John C. Neff and Ken Smith who taught me immunofluorescence. To David W. Covey for maintaining our BB Wistar rat colony and to Pierre Thibert for providing the rats for these studies. To Rachel L. (Tigner) Wright for support and statistical assistance. To Mary Larlham for typing this dissertation. VITA

September 21, 1955 ..... Born - Huntsville, Alabama 1977 . . . ;...... B.S., Ohio State University, Columbus, OH 1977-1979...... Graduate Research Associate Depts. of Surgery & Pathology, Ohio State University 1984 ...... M.A., Ohio State University 1984 ...... M.D., Ohio State University 1984-1988 ...... Resident, Dept, of Pathology Barnes Hospital, Washington University, St.Louis, MO 1985-87 ...... NIADDKD Trainee (Diabetes) Washington University 1988-1990 ...... Fellow, Pediatric/Perinatal Pathology, IWK Children's Hospital & Grace Maternity Hospital, Dalhousie University, Halifax, Nova Scotia, Canada 1989-1993 ...... Clinical Assistant, IWK Children's Hospital 1989-1994 ...... Assistant Professor of Pathology Dalhousie University 1990-present ...... Assistant Professor of Surgery Dalhousie University 1993 ...... Pathologist, IWK Children’s Hospital & Grace Maternity Hospital 1994-present ...... Associate Professor of Pathology Dalhousie University

iv PUBLICATIONS Wright JR Jr and Kearns H: Long-term culture, low temperature culture, and hyperoxic culture do not prolong fish-to-mouse islet xenograft survival. Xenotransplantation - in press. Nguyen T, Wright JR Jr, Nielsen PF, and Conlon JM: Characterization of the pancreatic hormones from the Brockmann body of the tilapia - implications to islet xenograft studies. Comp. Biochem. Physiol. - in press. Wright JR Jr and Kearns H: Fish-to-mouse islet xenograft survival is markedly prolonged by 15- deoxyspergualin (DSG). Transplant. Proc. - in press. Wright JR Jr and Schrezenmeir J: Transplantation of Fish islets. In Methods in Cell Transplantation, edited by C Ricordi, RG Landes Co., Austin - in press. Wright JR Jr: Relationship of surgical oncology and pathology in early 20th century America. In History of Surgery: Proceedings of the 17th International Symposium on the Comparative History of Medicine -- East and West, edited by Y Kawakita, Division of Medical History, The Taniguchi Foundation, Osaka - in press. Wright JR Jr, Thompson DL, McBride JA, and Liston RM: Pulmonary hyperplasia associated with tracheal atresia: pathological and prenatal sonographic findings. Pediatric Pathol. Lab. Med. 15: 81-97, 1995. Wright JR Jr: Procurement of fish islets (Brockmann bodies). In Pancreatic Islet Transplantation Series. Volume 1: Procurement of Pancreatic Islets, edited by RP Lanza and WL Chick, RG Landes Co., Austin, 1994, pp. 123-33. Aiken KD, Yu W, Wright JR Jr, and Roth KA: Adaptation of enteroendocrine cells in response to jejunal-ileal transposition in the rat. Gastroenterol. 106: 1576-83, 1994 . Wright JR Jr, Stinson D, Wade A, Haldane D, and Heifetz SA: Necrotizing funisitis associated with Actinomyces meyeri infection: A case report. Pediatric Pathol. 14: 927-34, 1994.

v Yang H, Yu W, Wan X, Al-Abdullah I, and Wright JR Jr: Recyclable Islet isografts: A biological insulin pump to facilitate surgery in diabetic rats. Cell Transplant. 3: 107-11, 1994. Yang H, Issekutz TB, and Wright JR Jr: Treatment with anti-VLA-4 and LFA-1 monoclonal antibodies prolongs intraportal rat islet allograft survival. Transplant. Proc. 26: 3325-6, 1994.

Wright JR Jr and Kearns H: Immunomodulation and fish- to-mouse islet xenograft survival. Transplant. Proc. 26: 4089, 1994, Wright JR Jr, Kearns H, and MacDonald AS: Leflunomide and cyclosporin-A prolong fish-to-mouse islet xenograft survival in Balb/c mice. Transplant. Proc. 26: 1310, 1994.

Theil M, McAlister VC, He G, Wright J, MacDonald AS, Bitter-Suermann H, Lee TDG: Effect of auxilliary liver transplantation on the simultaneously transplanted small bowel allograft. Transplant. Proc. 26: 1620, 1994. He G, McAlister VC, Lee TDG, Bitter-Suermann H, Theil M, Wright J, and MacDonald AS: Oral leflunomide prevents small bowel allograft rejection in the rat. Transplant. Proc. 26: 1613, 1994. Wan X, Yu W, Yang H, Al-Abdullah I, Coddington D, Bitter-Suermann H, and Wright JR Jr: Long-term survival of Wistar-Furth islet allografts in Lewis rats is associated with spontaneously tolerated orthotopic liver allografts. Transplant. Proc. 26: 723-4, 1994. Wright JR Jr, Kearns H, Polvi S, MacLean H, and Yang H: Experimental xenotransplantation using principal islets of teleost fish (Brockmann bodies): Graft survival in selected strains of inbred mice. Transplant. Proc. 26: 770, 1994. Yu W, Wan X, Wright JR Jr, Coddington D, and Bitter- Suermann H: Heterotopic liver transplantation in rats: the effect of intrahepatic islet isografts and split portal blood flow on liver integrity following auxiliary liver isotransplantation. Surgery 115: 108- 117, 1994. Lawen JG, Yu W, Cook HW, and Wright JR Jr: Failure of donor essential fatty acid deficiency to prevent renal allograft rejection in rats. Transplantation 56: 1269- 70, 1993.

vi Wright JR Jr and Gillis DA: Mediastinal foregut cyst containing an intramural adrenal cortical rest: A case report and review of supradiphragmatic adrenal cortical rests. Pediatric Pathol. 13: 401-407, 1993. Wright JR Jr: review of A.M. Silverstein's A History of Immunology (San Diego: Academic Press, 1989). Bull. Hist. Med. 67: 199-200, 1993.

Wright JR Jr: Experimental transplantation using principal islets of teleost fish (Brockmann bodies). In Pancreatic Islet Cell Transplantation: 1892-1992 ^ One Century of Transplantation for Diabetes, edited by C Ricordi, RG Landes Co., Austin, 1992, pp. 336-51.

Wright JR Jr, Polvi S, and MacLean H: Experimental transplantation using principal islet of teleost fish (Brockmann bodies): Long-term function of tilapia islet tissue in diabetic nude mice. Diabetes 41: 1528-1532, 1992. Yu W, Wan X, Wright JR Jr, Coddington D, and Bitter- Suermann H: The effect of intrahepatic islet isografts on hepatocyte integrity following auxiliary heterotopic liver isotransplantation in rats. Transplant. Proc. 24: 2920-2921, 1992.

Wright JR Jr, Polvi SL, Schrezenmeir J, and Al-Abdullah I: Longterm function of teleost fish principal islets (Brockmann bodies) after transplantation under the renal capsule in diabetic nude mice. Transplant. Proc. 24: 3029-3030, 1992. Lawen JG, Wright JR Jr, Yu W, and Epstein HR: Donor essential fatty acid deficiency does not prolong heterotopic cardiac allograft survival in rats. Transplantation 52: 178-180, 1991. Wright JR Jr, Haliburton B, Russell H, Henry M, Fraser R, and Cook HW: The anti-diabetogenic effect of essential fatty acid deficiency in multiple low-dose streptozotocin-treated mice persists if essential fatty acid repletion occurs outside of a brief window of susceptibility. Diabetologia 34: 709-714, 1991. Wright JR Jr, McIntyre L, Rand JJ, and Hixson EG: Nordic ski jumping injuries: A survey of active American jumpers. Am. J. Sports Med. 19: 615-619, 1991.

vii Wright JR Jr, Hixson EG, and Rand JJ: Ski jumping injuries. In Winter Sports Medicine, edited by MJ Casey, C Foster, and EG Hixson, FA Davis Co., Philadelphia, 1990, pp. 324-30. Wright JR Jr: review of J.S. Olsen's The History of Cancer: An Annotated Bibliography (Westport, CN: Greenwood Press, 1989). Bull. Hist. Med. 64: 353, 1990. Mendola J, Wright JR Jr, and Lacy PE: The effect of oxygen free radical scavengers on immune destruction of murine islets in allograft rejection and in multiple low-dose streptozocin-induced insulitis. Diabetes 38: 379-385, 1989. Wright JR Jr, Hauptfeld V, and Lacy PE: Induction of la antigen expression on murine islet parenchymal cells does not diminish islet allograft survival. Am. J. Pathol. 134: 237-242, 1989.

Wright JR Jr and Lacy PE: Silica prevents the induc­ tion of diabetes with CFA and low-dose streptozotocin in rats. Diabetes Res. 11: 51-54, 1989. Wright JR Jr and Lacy PE: Synergistic effect of adjuvants, endotoxin, and fasting on induction of diabetes with multiple low doses of streptozocin in rats. Diabetes 37: 112-118, 1988. Wright JR Jr, Mendola J, and Lacy PE: Effect of niacin/ nicotinamide deficiency on the diabetogenic effect of streptozotocin. Experientia 44: 38-40, 1988. Wright JR Jr, Kyriokos M, and DeSchryver-Kecskemeti K: Malignant fibrous histiocytoma of the stomach: A report and review of malignant fibrohistiocytic tumors of the alimentary tract. Arch. Pathol. Lab. Med. 112: 251- 258, 1988. Wright JR Jr, Epstein HR, Hauptfeld V, and Lacy PE: Tumor necrosis factor enhances interferon-induced la antigen expression on murine islet parenchymal cells. Am. J. Pathol. 130: 427-430, 1988. Wright JR Jr: Nordic ski jumping fatalities in the United States: a 50 year summary. J. Trauma 28:848- 851, 1988. Wright JR Jr, Lefkowith JB, Schreiner G, and Lacy PE: Essential fatty acid deficiency prevents multiple low- dose streptozotocin-induced diabetes in CD-I mice. Proc. Natl. Acad. Sci. USA 85:6137-6141, 1988. Wright JR and Lacy PE: Spontaneous hydronephrosis in C57BL/KSJ mice. J. Comp. Pathol. 99: 449-454, 1988. Wright JR Jr: The 1917 New York biopsy controversy: A question of surgical incision and the promotion of metastases. Bull. Hist. Med. 62: 546-562, 1988. McAlister WH, Wright JR Jr, and Crane JP: Main stem bronchial atresia: Intrauterine sonographic diagnosis. Am. J. Roentgenol. 148:364-366, 1987. Wright JR Jr: Atrophy, testis, rat. In Genital System: Monographs on Pathology of Laboratory Animals, edited by TC Jones, U Mohr, and RD Hunt, Springer-Verlag, Berlin, 1987, pp. 218-225. Wright JR Jr, Lacy PE, Unanue ER, and Hauptfeld V: Induction of class II MHC antigens on human and rodent islet parenchymal cells in vitro (letter). Diabetologia 30:441, 1987.

Wright JR Jr: review of Bert E. Park's The Impact of Illness on World Leaders (Philadelphia: University of Pennsylvania Press, 1986). Missouri Medicine 84: 225- 226, 232, 1987. Wright JR Jr: review of W.D. Foster's Pathology as a Profession in Great Britain and the Early History of the Royal College of Pathologists (London: Royal College of Pathologists, 1983). Bull. Hist. Med. 61: 679, 1987. Wright JR Jr, Barth RF, Neff JC, Poe ET, Sucheston ME, and Stempel LE: Gastrointestinal malformations associated with prune belly syndrome: Three cases and a review of the literature. Pediatric Pathol. 5:421- 448, 1986. Wright JR Jr, Lacy PE, Unanue ER, Muszynski C, Hauptfeld V: Interferon-mediated induction of la antigen expression on isolated murine whole islets and dispersed islet cells. Diabetes 35:1174-1177, 1986. Wright JR Jr, Hixson EG, and Rand JJ: Injury patterns in nordic ski jumpers: A retrospective analysis of injuries occurring at the Intervale ski jump complex from 1980 to 1985. Am. J. Sports Med. 14:393-397, 1986. Wright JR Jr, Stephens RE, Ford KC, Thibert P, and Yates AJ: DNA excision repair in spontaneously diabetic BB Wistar rats. Mech. Ageing Develop. 36:109- 116, 1986. Wright J, Yates A, Sharma H, and Thibert P: Histopatho- logical lesions in the pancreas of the BB Wistar rat as a function of age and duration of diabetes. J. Comp. Pathol. 95:7-14, 1985.

Wright JR Jr: The development of the frozen section technique, the evolution of surgical biopsy, and the origins of surgical pathology (William Osier Medal Essay). Bull. Hist. Med. 59:295-326, 1985. Wright JR Jr, Yates AJ, Shah NT, Neff JC, Covey DW, and Thibert P: Hematological characteristics of the BB Wistar rat. Vet. Clin. Pathol. 12:9-13, 1983. Wright JR Jr, Sharma HM, Yates AJ, Thompson BM, and Thibert P: Supra-diaphragmatic accessory lobe of the liver in BB Wistar rats. Teratology 27: 117-120, 1983. Wright JR Jr, Yates AJ, Sharma HM, and Thibert P: Central pontine myelinolysis following saline treatment of a diabetic rat for dehydration. J. Comp. Pathol. 93:509-514, 1983.

Wright JR Jr, Senhauser DA, Yates AJ, Sharma HM, and Thibert P: Spontaneous thyroiditis in BB Wistar diabetic rats. Vet. Pathol. 20:522-530, 1983. Wright JR Jr, Yates AJ, Sharma HM, and Thibert P: Pathological lesions in the spontaneously diabetic BB Wistar rat: A comprehensive autopsy study. Metabolism 32 (suppl.1):101-105, 1983. Wright JR Jr, Yates AJ, Sharma HM, Shim C, Tigner RL, and Thibert P: Testicular atrophy in the spontaneously diabetic BB Wistar rat. Am. J. Pathol. 108:72-79, 1982. Wright JR Jr, Yates AJ, Sharma HM, and Thibert P: Spontaneous gastric erosions and ulcerations in BB Wistar rats. Lab. Animal Sci. 31:63-66, 1981.

x Wright JR Jr: review of Jonas Rishel’s The Indian Physician/ containing a new system of practice, founded on medical plants: together with a description of their properties, localities, and method of using and preparing them.... (New Berlin, PA: Joseph Miller, 1828; photostatically reprinted by the Ohio State University Libraries Publication Committee, 1980). Ohio State University College of Medicine Journal 31: 17, 1981. Parrish WB, Bronn DG, Wright JR Jr, and Minton JP: Elemental analysis of biological material in the frozen state. Ohio J. Sci. 79:271-273, 1979.

FIELDS OF STUDY Major Field: Pathology Studies in Experimental Pathology (Allan J. Yates, M.D., Ph.D.) Islet Transplantation (Paul E. Lacy, M.D., Ph.D.) Immunology (Emil R. Unanue, M.D.) Anatomical Pathology (John Kissane, M.D. ) Pediatric Pathology (Blaise E. Favara, M.D.) Placental Pathology (Stephen A. Heifetz, M.D.)

Minor Field: History of Medicine (John C. Burnham, Ph.D. TABLE OF CONTENTS

DEDICATION ...... ii ACKNOWLEDGEMENTS .... iii

VITA ...... iv LIST OF TABLES ...... XV LIST OF FIGURES ...... xvii

CHAPTER PAGE I. INTRODUCTION ...... 1 II. BACKGROUND INFORMATION...... 3 A. History of the BB Wistar r a t ...... 3 B. Other diabetes models ...... 8 III. METHODOLOGY...... 14

A. Colony Husbandry ...... 14 B. Necropsy procedure...... 15 C. Hematology...... 18 D. DNA Repair Studies ...... 19 E. Statistics...... 22 IV. RESULTS AND DISCUSSION...... 23 A. Morphology of OrganSystems ...... 23 1. Pancreas (exocrine and endocrine) ...... 23 2. Endocrine...... 34

A. T h y r o i d ...... 34 B. Parathyroid...... 37 C. A d r e n a l ...... 38 D. Pituitary...... 40

xii 3. Male Reproductive...... 43

A. T e s t e s ...... 43 B. Prostate...... 47

4. Female Reproductive ...... 49 A. U t e r u s ...... 49 B. O v a r y ...... 50 C. Mammary gland ...... 52 5. Digestive...... 54 A. Gastrointestinal...... 54 B. L i v e r ...... 58

6. U r i n a r y ...... 63 A. B l a d d e r ...... 63 B. K i d n e y ...... 64 7. Respiratory (Lungs) ...... 70 8. Circulatory...... 75 A. H e a r t ...... 75 B. Vasculature...... 78 9. Hematopoietic/Lymphatic ...... 83 A. Thymus ...... 83 B. Lymph n o d e ...... 84 C. S p l e e n ...... 88

10. Nervous System ...... 89 A. B r a i n ...... 89 B. Peripheral Nerve ...... 95

11. O t h e r ...... 95 A. C u t i s ...... 95 B. E y e s ...... 96 C. Soft t i s s u e s ...... 97 B. Hematology...... 98 C. DNA Repair and Longevity...... 102 V. GENERAL DISCUSSION (SUMMARY) ...... Ill

xiii REFERENCES ...... 119

APPENDICES A. Histopathological lesions in the pancreas of the BB Wistar rat as a function of age and duration of diabetes ...... 182

B. Spontaneous thyroiditis in BB Wistar diabetic rats ...... 191 C. Testicular atrophy in the spontaneously diabetic BB Wistar rat ...... 201

D. Atrophy, testis, rat ...... 210 E. Spontaneous gastric erosions and ulcerations in BB Wistar rats ...... 219 F. Supradiaphragmatic accessory lobe of the liver in BB Wistar rats ...... 224 G. Central pontine myelinolysis following saline treatment of a diabetic rat for dehydration ...... 229

H. Hematological characteristics of the BB Wistar rat ...... 236 I. DNA excision repair in spontaneously diabetic BB Wistar rats ...... 242 J. Pathological lesions in the spontaneously diabetic BB Wistar Rat: A comprehensive autopsy study ...... 251

x i v LIST OF TABLES

TABLE PAGE 2.1 Comparison of selected spontaneous rodent models for diabetes mellitus - metabolic considerations ...... 11 2.2 Comparison of selected spontaneous rodent models for diabetes mellitus - pancreatic beta cell morphology ...... 12 2.3 Comparison of selected spontaneous rodent models for diabetes mellitus - maintenance information ...... 13 4.1 Incidence of hepatic lesions by age ...... 105 4.2 Incidence of pulmonary lesions by age ...... 106 4.3 Infectious organisms isolated from BBW rat tissues ...... 107

4.4 Incidence of cardiac lesions by age ...... 108

4.5 Incidence of lymphatic lesions by age ...... 109 4.6 Sites of lymphomatous involvement ...... 110 5.1 Criteria for a good animal model ...... 118 A.l Age incidence of pancreatic lesions in BB Wistar rats ...... 185

B.l Character of thyroiditis in BB Wistar rats ...... 193 B.2 Incidence of lymphocytic thyroiditis in BB Wistar rats by age and duration of diabetes ...... 195 C.l Distribution of age and duration of diabetes by degree of testicular atrophy ...... 204

xv C.2 Body and testicular weights ...... 204 C.3 Numbers of rats with different degrees of severity of testicular histologic abnormalities and their relationship to degree of atrophy and age ...... 205

E.l Comparisons of treatment and health status of BBW diabetic rats with and without gastric lesions ...... 222 H.l Complete blood count and platelet values ..... 239 H.2 Differential white cell counts ...... 240 I.1 Analysis of variance parameters ...... 246 J.l Frequency of strain-related lesions (percent) ...... 253

J.2 Frequency of diabetes-related lesions (percent) ...... 253 J.3 Frequency of nonspecific lesions (percent) ... 255

xvi LIST OF FIGURES

FIGURE PAGE

4.1 Pituitary chromophobe adenoma in a 327 day old BBWd rat. The lesion was roughly spherical with a large central hematoma. Each large unit on rule is 1 cm...... 139 4.2 Photomicrograph of the same adenoma. The tumor appeared somewhat pleomorphic with few mitotic figures. A portion of the neurohypophysis is seen on the left slightly above the lesion (Original magnification Reticulin, x25) ...... 140

4.3 Inflamed seminiferous tubules in the periphery of an otherwise atrophic testis from a 397 day old BBWd rat. Tunica albuginea is markedly thickened. Rat had bilateral orchitis (Original magnification HE, x25) ...... 141 4.4 BBWd rat with multifocal granulomatous lesions. Several necrotic foci are present. All seminiferous tubules were totally devoid of germ cells and Sertoli cells. Thickening of tunica albuginea is also present (Original magnification HE, x25) ...... 142 4.5 Granuloma with several multinucleated giant cells (Original magnification HE, x250) ...... 143

4.6 Acute prostatitis in a BBWd rat (HE, original magnification x25) ...... 144 4.7 Normal prostate from 551 day old W rat (HE, x25) ...... 145

4.8 Normal prostate from 551 day old W rat (HE, xlOO) ...... 146

xvii 4.9 Normal prostate from 551 day old W rat. Epithelium is columnar and negative image of the golgi is visible at the apical end of the cell (HE, x630). Lumens contain pale eosinophilic secretions ...... 147 4.10 Inactive prostate from 177 day old BBWd rat. Epithelium is low cuboidal to simple squamous. No secretion in the lumens. (Original magnifications HE, x25) ...... 148 4.11 Inactive prostate from 177 day old BBWd rat. Epithelium is low cuboidal to simple squamous. No secretion in the lumens. (Original magnifications HE, xlOO) ...... 149 4.12 Large uterine stromal polyps in 372 day old BBWd rat. Histology showed squamous metaplasia and acute inflammation in the lumen. No inflammatory infiltrates were seen in the stroma. Each unit on rule is 1 m m ...... 150 4.13 Mammary adenocarcinoma found in milk line near base of tail of a 248 day old BBWnd rat. Tumor measures approximately 3 cm. in diameter. Tumor was firm, white, and lobulated. No metastases were seen. Photograph was taken after the tumor had been shaved to remove the hair ...... 151

4.14 Photomicrograph of the mammary adenocarcinoma. Epithelium has squamous elements. Ductules contain a deeply eosinophilic secretion (Original magnification HE, xlOO) ...... 152 4.15 Idiopathic megacolon in a 327 day old BBWd rat. The rat presented with abdominal distention and anorexia and was treated with enemas. Treatments were temporarily effective. At necropsy, the colon and caecum were observed to be distended (up to 5 cm in diameter) with hard fecal material. Nothing was found constricting the rectum when the pubic symphysis was removed. All tissue in the area of the rectum was examined histologically and was normal. The duration of diabetes in this rat was 205 days .... 153 4.16 412 day old BBWd rat with erosions and ulcerations in both the thick-walled glandular portion and the thin-walled squamous portion of the stomach. The glandular

xviii portion is severely hemorrhagic. The lesions in the squamous forestomach (see arrows) are surrounded by white, raised borders ...... 154 4.17 Small erosion in the thin-walled squamous portion of the stomach (forestomach). Area of coagulation necrosis extends to the muscularis mucosa. Epithelial surface has sloughed. (Original magnification HE, xlOO) ...... 155 4.18 Severe hepatic fatty change in the region of several portal triads in a 150 day old BBWd rat. Duration of diabetes was only 57 days. (Original magnification HE, x250) ...... 156 4.19 Several cysts consisting of dilated tubules filled with eosinophilic proteinacious fluid were seen in the kidney of a 412 day old BBWnd rat. The cyst just to the right of center is associated with focal chronic inflammation of the tubule. This pattern was seen in several rats and is typical of some of the changes seen in chronic renal disease in the rat. Most cysts in rat kidneys were not inflamed. (Original magnification HE, xlOO) ...... 157 4.20 A small focal subcapsular granuloma in the renal cortex of a 200 day old BBWd rat (Original magnification HE, xlOO) ...... 158 4.21 Chronic interstitial nephritis in a 210 day old BBWd rat. Mononuclear infiltrates, interstitial fibrosis, tubular atrophy, and relative sparing of glomeruli are prominent. (Original magnification HE, xlOO) ...... 159 4.22 Gross specimens from different lobes of a 325 day old BBWd rat with severe bilateral bronchopneumonia. The cut surfaces of the specimens on the left and right reveal large abscesses. The two center specimens show pleural adhesions ...... 160

4.23 Severely congested, edematous lung from a 192 day old BBWd rat. Presence of neutrophils in the alveoli indicate bronchopneumonia (Original magnification HE, x250) ...... 161 4.24 Murine chronic respiratory disease in a young W rat. Main features seen in this photomicrograph are peribronchial lymphoid

xix hyperplasia, perivascular lymphoid aggregates, and chronic pneumonitis (Original magnification HE, x25) ...... 162 4.25 551 day old W rat with limited multifocal chronic myocarditis-predominately in the left ventricle. Antischkow cells are prominent in the chronic infiltrate. This picture is typical of myocardial degeneration with fibrosis in aged rats. (Original magnification HE, x250) ...... 163

4.26 Pregnant female BBWnd rat (age unknown) with vasculitis in all organ systems. A necrotizing vasculitis was seen in the pancreas (Original magnification HE, x250) ...... 164 4.27 Several less involved vessels were seen in the submucosa of the stomach (Original magnification HE, xlOO). Other lesions in this rat included a small organizing mural thrombus in the heart, disseminated intravascular coagulation in the kidney (see Figure 4.29), severe extramedullary hematopoiesis in the spleen, and an infarct in a mesenteric node. Russell bodies were very prominent in the lymph nodes. The rat was also pregnant ...... 165 4.28 Focal vasculitis in the colon of a 320 day old BBWd rat. Histiocytes are seen predominantly in the intima and media and eosinophils are in the adventitia. This rat also had systemic eosinophilia. (Original magnification HE, xlOO) ...... 166

4.29 Two glomeruli with large microthrombi in the kidney of a near-term pregnant BBWnd rat (unknown age) which died with disseminated intravascular coagulation (DIC) secondary to the widespread vasculitis. (Original magnification HE, x400) ...... 167

4.30 Hyperplastic mesenteric lymph nodes from 320 day old BBWd rat. Nodes are enlarged and fused. Normal nodes are small and discrete ...... 168 4.31 Typical lymph node from a 300 day old W rat with prominent follicular germinal centers and abundant lymphocytes (Original magnification HE, xlO) ...... 169

xx 4.32 Hyperplastic lymph node from a 412 day old BBWd rat. Sinusoidal dilatation and lymphocytic depletion are prominent histologic features. (Original magnification HE, xlO) ...... 170 4.33 Necrotizing granuloma in a pancreatic lymph node from a 125 day old BBWd rat. Additional sections were stained with acid fast blue, GMS, and gram stain. No organisms were seen. (Original magnification HE, x25) ...... 171 4.34 Lymphoma mass in a 334 day old female BBWd rat found at the junction of the cecum and colon. Tumor is firm, lobulated, and hemorrhagic. Tumor was also present in colon and pancreas ...... 172 4.35 Histologically, the lymphoma is diffuse with a total loss of normal nodal architecture. (Original magnification HE, x25) ...... 173 4.36 The tumor is histiocytic with plasmacytoid features. Several binucleated plasma cells are shown as well as a mitotic figure in the lower left corner. (Original magnification HE, x400) ...... 174 4.37 This lymphoma was found in a 248 day old BBWd rat. The tumor mass was found at the junction of the ileum and caecum. The 3 specimens in the photograph are organized proximal to distal from left to right. The lumen is totally occluded in the specimen on the left. Each large unit on the rule is 1 c m ...... 175

4.38 Photomicrograph of the area on the left. The neoplasm in the submucosa of the ileum is a histiocytic lymphoma with plasmacytoid features. (Original magnification HE, x25) ...... 176 4.39 Single, small, well-circumscribed area of necrosis in the proximal ileum (50 cm from cecum) of a 315 day old BBWd rat. Normal mucosa seen to right of necrotic area. Histological examination revealed much necrosis and a histiocytic lymphoma with plasmacytoid features. The adventitia of the colon was also involved, but this was not grossly visible. Lymph nodes were grossly

xxi large and histologically hyperplastic, but no tumor was seen. Spleen showed severe extramedullary hematopoiesis, but was not markedly enlarged ...... 177

4.40 Lymphoid depletion in the spleen of a 320 day old BBWd rat. Distinction between red and white pulp is obscured. (Original magnification HE, x25) ...... 178 4.41 Spleen from a BBWnd rat with a lymphoma (age unknown). Spleen measures 7 cm long and weighs 7.5 gms. Normal values are 2-3 cm and 0.5-0.75 gm. Splenomegaly is a result of severe extramedullary hematopoiesis and tumor infiltration ...... 179 4.42 Massive subcutaneous spindle cell sarcoma in BBW rat of unknown age ...... 180 4.43 Massive eosinophilic infiltrate in the submucosa of the colon of a 350 day old BBWd rat. The muscularis externa is also involved. Eosinophilic infiltrates were also seen in the other organs of this rat. (HE, original magnification xlOO) ...... 181 A.l Insulitis in a 115-day-old BBWnd rat. The mononuclear infiltrate is predominantly peri-insular with minimal islet invasion. Surrounding acinar tissue is not involved. HEx50 ...... 186

A.2 Islet from a 135-day-old BBWd rat showing marked invasion by mononuclear cells. Onset of diabetes was 55 days earlier. HEX125 ...... 186

A.3 Periductal eosinophilic infiltrate in the pancreas of a 280-day-old BBWd rat. HEX125 ...... 186 A.4 Pancreatic granuloma in a 385-day-old BBWnd rat. An islet in the upper left corner is surrounded by a predominantly lymphocytic infiltrate. Granuloma is also surrounded by lymphocytes. HExl25 ...... 186

xxii A.5 Islet from a 137-day-old BBWd rat showing islet cells with pyknotic nuclei and acidophilic cytoplasm and intact islet cells. No leukocytic infiltrates were seen in any islets in this rat. Onset of diabetes was 47 days earlier. HEx200 ...... 187 A.6 One of many hyperplastic islets in a 178-day -old BBWd rat showing a peri-insular mononuclear infiltrate. Onset of hyperglycaemia was 60 days earlier. HEx50 ...... 187

A.7 Atypical small end-stage islet consisting of only a few cells, and without evidence of necrosis. HEx200 ...... 187 B.l Extensive multifocal mononuclear cell infiltrate in BB Wistar rat thyroid tissue. Infiltrate is predominantly interstitial and perivascular. HE ...... 194 B.2 Large numbers of plasma cells and plasmacytoid lymphocytes composing some of the infiltrates. Infiltrate is in close association with thyroid follicles with some disruption of follicular basement membrane. HE ...... 194

B.3 Mixed chronic inflammatory infiltrate adjacent to several thyroid follicles ...... 194 B.4 Extensive follicular involvement by inflammatory infiltrate with apparent disruption of follicular architecture and dissolution of thyroid epithelial cells in the most severely affected thyroids. HE ...... 194 C.l Bar graph showing percentage of rats with testicular lesions by age in days. Rats appear in the following order for all age groups: BBWd, BBWnd, and Wistar. The number of rats in each group is noted above each bar ...... 203 C.2 Transected rat testes. Normal Wistar rat on the left and BB Wistar diabetic rat with bilateral atrophy on the right. Line is 1 c m ...... 205

xxiii BB Wistar diabetic rat with atrophy. Atrophic seminiferous tubule with several multinucleated giant cells in the lumen. Interstitial tissue shows Leydig cell hyperplasia. Tubules located peripherally to the right, left, and above the central tubule contain only Sertoli cells. (x400) ...... 206 C.4 Testes from 3 rats showing different degrees of atrophy and other variables. A - Normal seminiferous tubules from a 361-day-old Wistar rat. Full complement of germ cells present. B - Testis of a Wistar rat 551 days old with severe atrophy, thickening of interstitial connective tissue (3+), and Leydig cell hyperplasia (3+). C - Severely atrophic testis in a 387-day-old BBW diabetic rat. Most tubules have only Sertoli cells. Tubules retaining some germ cells show sloughed epithelium. There is one relatively normal tubule in the field. (H&ExlOO) (With a photographic reduction of 30%) ...... 207

C.5 A - Normal Wistar rat testis demonstrating normal cellularity of seminiferous tubules and normal thickness of tunica albuginea. (H&ExlO) B - BB Wistar diabetic rat testis with diffuse, severe atrophy of seminiferous tubules and a thickened tunica albuginea. (H&ExlO) ...... 208

D. 263 (upper left). Severe atrophy, testis, rat. Seminiferous tubules are lined by Sertoli's cells; germ cells are absent; tubules are decreased in diameter. The space between the tubules and interstitial tissue is an artifact. Formalin fixation causes tubular shrinkage. H&Ex400 ...... 212 D. 264 (lower left). Atrophy, seminiferous tubules, testis, rat. Note the small distorted tubules without germinal epithelium with marked thickening of the tunica propria. Leydig's cells contain PAS-positive, acid-fast pigment. PASx400 ...... 212 D. 265 (upper right). Focal atrophy, testis, 551-day -old Wistar rat. At the bottom are some normal tubules and interstitial tissue. Severely

xxiv atrophic tubules and markedly increased interestitial tissue (upper). Focal dystrophic calcification (arrow). H&Ex63 ...... 212

D.266 (lower right). Leydig's cell hyperplasia, testis, 2-year-old Wistar rat. Nodular (A) and (B). H&Ex400 ...... 212

D.267 (upper left). Seminiferous tubule, rat. Normal germinal epithelium and tubular diameter ...... 213

D.268 (lower left). Seminiferous tubule, diabetic BB Wistar rat testis. Hypocellular germinal epithelium and decreased tubular diameter. H&EX400 ...... 213 D.269 (upper right). Seminiferous tubules, rat. Changes typical of chronic vitamin A deficiency. Note reduced tubular diameter and preservation of a thin rim of spermatogonia and Sertoli's cells. (Mason 1933,p232) H&EX400 ...... 213

D.270 (lower right). Seminiferous tubule, rat deficient in vitamin E. Note multinucleated giant cells consisting of fused spermatid nuclei with typical cresentic degeneration. (Slide courtesy of Dr. James S. Nelson) H&EX400 ...... 213 E.l Multifocal gastric erosions and ulcerations with poorly defined margins in thick-walled glandular portion of stomach. Many lesions are at base of rugae and are darkly stained. Each unit on rule = 1 m m ...... 221 E.2 Small gastric erosion comprised of well circumscribed area of necrosis involving gastric mucosa. Margins of lesion are shown by lighter arrows. Darker arrows point to region of necrosis containing brown pigment with staining properties of aposiderin. Hematoxylin and eosin stain. Line = 96 p m ...... 222 F.l Thoracic cavity of the second (nondiabetic) rat is opened and the thoracic contents are reflected caudally to reveal a supra­ diaphragmatic accessory liver lobe (arrows outline the margins) in the right hemithorax.

xxv Pooled blood is also present in the thoracic cavity. A, left lung; B, right lung; C, liver; D, heart ...... 226 F.2 Photomicrograph of an accessory liver lobe in the first (diabetic) rat showing extensive areas of fibrosis and abnormal lobular architecture. H&E stain, original magnification x25 ...... 227 F.3 Fibrous septum showing bile duct hyperplasia in the accessory lobe of the first rat. Hemosiderin-laden macrophages are present near the large blood vessel in the center of the septum. Severe congestion is seen in the dilated hepatic sinusoids. H&E stain, original magnification x25 ...... 227 F.4 Fibrous portion of the diaphragm of the first rat containing hepatic elements near the site of attachment of the accessory lobe. Numerous bile ducts are seen in the center and to the right. Several hepatocytes are present in the left portion of the diaphragm. H&E stain, original magnification x400...... 228 G.l Transverse section of pons and cerebellum showing marked pallor of myelin in basal region and tegmentum. Note the margin of normal myelin (arrow). Luxol fast blue xlO ...... 231

G.2 Central pons with macrophages and several unidentifiable necrotic cells showing karyorrhectic and pyknotic nuclei. Luxol fast blue x630 ...... 232 G.3 Central pons with axons traversing the area. Bodian's silver stain x630 ...... 232 1.1 Unscheduled DNA synthesis (average grains/ nucleus) in BB and W rats as a function of fluence at 0, 2, 4, 6, and 8 h of incubation. Repair at 40 J/m2 does not exceed that at 20 J/m2 ...... 246

1.2 Unscheduled DNA synthesis (average grains/ nucleus) in BB and W rats as a function of time at a dosage of 20 J/m2. Pooled standard

xxvi errors of the means at 0, 2, 4, 6, and 8 h are 0.28, 0.67, 0.65, 0.63 and 0.72 grains/ nucleus, respectively ...... 247

xxvii CHAPTER I INTRODUCTION

The validity of studying animal models for human disorders is a longstanding dilemma for the medical scientist. The obstacles for studying factors involved in the pathogenesis of a disorder in a population of human subjects are large. The importance of both genetic predisposition and environmental factors to the manifestation of many disorders is unquestionable. Since it is ethically unacceptable to control the genetic background of human subjects and since it is exceptionally difficult to control chronic environmental conditions, the usefulness of human prospective studies is severely restricted. Furthermore, the cost of prospective studies is immense. On the other hand, retrospective epidemiological studies are relatively inexpensive but, unfortunately, must rely on information from numerous sources of uncontrollable reliability. In addition, selection of a homogenous group of appropriate human subjects is not an easy task. Upon superficial examination, many of these important variables appear more easily controlled by utilizing animal models. Unfortunately, any animal model introduces a new set of uncontrollable variables. Rarely is any model for a

1 2 human disorder a normal animal with a single variable changed. Frequently, the metabolic disorder that precipitates the overt symptoms of the desired disorder may be either the cause or the result of other underlying disorders. Even when the abnormalities in the animal model appear identical to those of the human disease, there is no reason to believe that the underlying defects causing those abnormalities are related. In other words, similar abnormalities do not necessarily imply a similar etiology or pathogenesis. This makes the choice of an appropriate animal model for an experimental study as important as the overall experimental design. One would like to choose the model that most closely approximates the human disease. Unfortunately, the information necessary to make this decision is not always readily available. The goal of my dissertation is to characterize several aspects of the spontaneously diabetic BB Wistar rat, a model for insulin- dependent (juvenile-onset) diabetes mellitus (IDDM). This will permit future investigators to determine whether this model is appropriate for the objectives of their studies. CHAPTER II

BACKGROUND INFORMATION

A. History of the BBW Rat The spontaneously diabetic Bio Breeding Wistar (BBW) rat was discovered by chance in 1974 at the Bio Breeding Laboratories of Canada. It was noted that the high mortality rate within an outbred colony of Wistar rats at the facility was also associated with persistently wet bedding. After the possibility of a defective watering system was ruled out, other possibilities were explored. The presence of polyuria and the discovery of both glucose and ketones in the urine was definitive for the diagnosis of diabetes mellitus. Because of the obvious importance of maintaining and studying the colony, and because of the expenses associated with that task, funding was requested from the Canadian Medical Research Council (MRC) to support the endeavor. After receiving MRC support, studies on the BBW rat were expanded. The pilot studies were performed through the collaborative efforts of Drs. Nakhooda, Wei, and Marliss at the University of Toronto, Dr. Like at the

3 University of Massachusetts, and Dr. Chappel at the Bio

Breeding Laboratories (Nakhooda et al., 1977; Nakhooda et al.t 1978). These pilot studies resulted in several very important observations that suggested that the BBW rat was like no other model of IDDM yet studied. First of all, the diabetic syndrome in the BB Wistar rat was characterized by variable degrees of hyperglycemia, glycosuria, hypoinsulinemia, hyperglucagonemia, and ketoacidosis. These symptoms could be minimized by proper insulin treatment. Untreated BBW rats could be divided into three groups based on level of ketosis: 1) severely ketotic (total blood ketone body levels between 6-13 mM), 2) moderately ketotic (1-5 mM), and 3) stable (<1 mM). Severely ketotic rats showed rapid weight loss and dehydration in one to six days. Moderately ketotic rats gradually lost weight over 15 days but maintained marked polyuria and glycosuria. Stable rats maintained weight, polyuria, and glycosuria in excess of 40 days.

Second, light microscopic examination of the pancreas from stable and early ketotic rats indicated that islets were small, frequently inflamed, and had reduced numbers of beta cells showing variable degranulation. Islets from severely ketotic rats were exceedingly small and rare; beta cells were seldom present. From these findings Nakhooda et al. (1977) concluded that a widespread insulitis was responsible for active beta-cell destruction. A milder form of this 5 insulitis was also observed In a small number of BB Wistar litter mate controls. Further study demonstrated that most of these rats were "chemically diabetic", i.e. they had abnormal glucose tolerance tests but no hyperglycemia or glycosuria. It was also determined that abnormal glucose tolerance preceded the onset of overt diabetes in most of the BBW diabetic rats (Nakhooda et al., 1978). Although complete remission of the diabetic syndrome was occasionally seen, the diabetic state was usually severe and permanent (Nakhooda et al., 1978). Like reported that only 6 spontaneous cures were observed in more than 1000 diabetic animals studied (Like et al., 1982). Continued interest by the Canadian Government, and financial difficulties at the Bio Breeding Laboratories resulted in the transfer of the colony to the Animal Resources Division of Health and Welfare Canada in December of 1977 and January of 1978 (Chappel and Chappel, 1983). New maintenance techniques, such as monogamous sibling mating, implemented following the transfer increased the average incidence of diabetes within the litters of diabetic matings from 20-25% (average onset at 63 days of age) to 70- 80% (average onset at 85 days of age). Furthermore, the number of unproductive matings was decreased from 50% to 15%. Unfortunately, sibling mating also resulted in smaller litters and numerous developmental abnormalities and had to be discontinued after about eight generations. Technically 6 speaking, these BB Wistar rats were still outbred, but skin graft rejection studies indicated that these animals were for all practical purposes inbred. At the time our studies were performed, the overall incidence of hyperglycemia in these animals was about 60% with an average postnatal time of onset of 85 days, but this pattern was quite variable. Furthermore, the number of pups/litter varied from 6-15 depending on whether one or both parents were diabetic. When both parents were diabetic, fecundity was decreased but incidence within the litter was increased (Personal Communication, P. Thibert). Following the transfer to Health and Welfare, the animals became more readily available to the scientific community. By 1979, spontaneously diabetic BB Wistar rats were available from: Pierre Thibert, D.V.M. Chief, Animal Resources Division Health Protection Branch Health and Welfare of Canada Sir Frederick G. Banting Research Centre Tunney's Pasture, Ottawa, Ontario Canada, K1A 0L2 (613) 957-0865

A breeding colony of BB Wistar rats was initiated in the Pathology Department at the Ohio State University in April of 1979. Animals were shipped from Ottawa by commercial airlines to Columbus via Pittsburgh. Many of the animals in our first few shipments from the parent colony in Ottawa died of pneumonia acquired in transit or while housed 7 in a non-barrier facility in Graves Hall. The colony was later moved to a semi-barrier facility in Wiseman Hall in July of 1979 to decrease the incidence of pulmonary infections. A necropsy study was initiated to assure quality control within our colony and was expanded following a series of interesting findings. This study is based in part on those findings. In the early 1980s, the University of Massachusetts Medical School received a contract from the National Institutes of Diabetes and Digestive and Kidney Diseases (NIDDK) to breed and distribute diabetes-prone and diabetes- resistant BB Wistar rats to eligible investigators. In Worcester, several inbred lines were developed and characterized. BB/Wor rats became available from: Dennis L. Gubereski, BB\Wor Project Administrator Department of Pathology University of Massachusetts Medical School 55 Lake Avenue North Worcester, Massachusetts 01605 (508) 856-3366

Since that time, the BB Wistar rat has become one of the most extensively studied animal models for diabetes. We estimate that over a thousand papers have been published using this model since its development. Many of these findings have been described in a series of comprehensive review articles (Like et al., 1982; Marliss et al., 1982; Like and Rossini, 1984; Rossini et al., 1985; Eisenbarth, 1986; Mordes et al., 1987; Kastern et al., 1990; Scot et 8 al., 1990; Crisa et al., 1992; Rossini et al., 1993; Guberski, 1994).

B. Other Models of Diabetes

Numerous spontaneous models for diabetes mellitus have been studied in the past 50 years (Hansen and Hansen, 1990). Rather than reviewing each model separately, I have constructed the tables 2.1, 2.2 and 2.3 which permit comparison of salient features. These tables are the result of an extensive literature review but are still not exhaustive. Since many of these models were characterized 25 or more years ago, mechanized information searches are of limited value in accessing some of this information.

Table 2.1 shows 15 selected spontaneous rodent models for diabetes mellitus. Although other models exist (many are hybrids resulting from cross breeding these models), these 15 are the best known and most frequently utilized. It is useful to describe each rodent as a model for either IDDM or non-IDDM (NIDDM). In this present study we are interested in the models that are similar to human IDDM, but we will briefly examine the models for NIDDM. In general, NIDDM models are obese, hyperglycemic, nonketotic, and have elevated serum insulin (Table 2.1). The pancreatic islet morphology of these models is similar to that seen in human NIDDM. Most of these show beta cell hyperplasia and an absence of definitive beta cell degranulation, necrosis, or insulitis (Table 2.2). Models of NIDDM are reviewed in the National Research Council's Institute of Laboratory Animal

Resources (ILAR) News in the Summer 1990 issue (volume 32, number 3).

Six models are similar to human IDDM. These are the Chinese hamster, the South African hamster, the New Zealand

White Rabbit, the non-obese diabetic (NOD) mouse, the Long-

Evans Tokushima Lean (LETL) rat, and the BB Wistar rat.

They share certain metabolic features including hyperglycemia, the absence of obesity, and decreased plasma insulin. Ketoacidosis, an important feature of human IDDM, is seen only in the Chinese hamster, the South African hamster, the LETL rat, and the BB Wistar rat. The typical human IDDM pattern of islet morphology is B cell degranulation, glycogen deposition, and necrosis. Only the New Zealand White Rabbit differs from this general pattern

(Table 2.2). Insulitis, a transient feature of the human disease, is seen only in the BB Wistar rat, the LETL rat, and the NOD mouse. Beta cell hyperplasia is not seen in either human IDDM or the IDDM models (Table 2.2). Models of IDDM are reviewed in the Winter 1993 issue of ILAR News (volume 35, number 1). There are certain inherent advantages of the BBW rat as a model for IDDM. First, it is one of only three models that develops an insulitis at the onset of the diabetic state. Second, it is a much larger animal than the NOD mouse, the Chinese hamster, or the South African hamster. Animal size is important for many types of experimentation.

Third, it is much more widely available and better characterized than the LETL rat. Finally, the BBW rat has gentle temperament compared to the Chinese hamster, an animal known for its ferocious nature. For these reasons, the BB Wistar rat is one of the most promising model for

IDDM research. Therefore, I have characterized various previously unexamined features of this model. TABLE 2.1 COMPARISON OF SELECTED SPONTANEOUS RODENT MODELS FOR DIABETBS MELLITUS

METABOLIC CHARACTERISTICS

T Y P E O P HYPER­ KETO PLASMA INSULIN MODEL DIABETES GENETICS LETHALITY OBESITY GLYCEMIA ACIDOSIS IRI RESIST­ ANCE NIDDM a u t o s o m a l H o u s e 4 4^ - t 4 d o m i n a n t Obese Hyperglycemia (ob/ob) NIDDM a u t o s o m a l 4 t r a n s i e n t - f / n o r m a l r e c e s s i v e

Diabetic (db/db) House NIDDM a u t o s o m a l 4 4 4 u t r a n s - i e n t r e c e s s i v e New Zealand obese (NZO) Mouse NIDDM p o l y g e n i c " 4 4^ - T 4

Japanese KK House NIDDM p o l y g e n i c - 4 t r a n s i e n t - _4

Spiny Mouse (Aconvs cahirinusi NIDDM polygenic 4_ 4 4 4

Egyptian Sand Rat (Ptaamomvs obesus. NIDDM Desert Rat) p o l y g e n i c 4 t r a n s i e n t 4 t t r a n s - i e n t

spontaneously Diabetic Wistar Rats NIDDM 7 ( J a p a n ) - - 4 - I 7

Zucker "fatty" rat NIDDM a u t o s o m a l - 4 - * T - r e c e s s i v e

NIDDM p o l y g e n i c - 4 4 - t l 4 Chinebe Hamster (Cricetulus ariseusl IDDM p o l y g e n i c - 4 4 J/normal trans-ient

south African Hamster (Mystromvs IDDM polygenic 4_ elbicaudatus, white tailed rat) - 4 4 7 7

New Zealand White Rabbit IDDM ? - _ 7 4 1 7 BB Wistar (BBW) Rat IDDM p o l y g e n i c + - 4 4 i 7 Non-obese diabetic (NOD) mouse IDDM p o l y g e n i c 4 - 4 4 t _ Long-Evans Tokushima Lean (LETL rat IDDM p o l y g e n i c 4 - 4 4 7 H a n 1 IDDM p o l y g e n i c 4 - 4 4 t - TABLE 2.2 COMPARISON OF SELECTED SPONTANEOUS RODENT MODELS FOR DIABETES MELLITUS

PANCREATIC BETA CELL MORPHOLOGY

MODEL B CELL HYPERPLASIA e CE L L degranulation B CELL GLYCOGEN B CELL NECROSIS IHSUUTtS | OOese Yef'j* M a & e 4 ♦ 4

| Obese H/pergTcemc Mouse 4 trann'crt -

D-aPe!-: Mjuse * traiert - 4 Ne* Zaaland Obese U o h s a ?

Japanese KK Mouse 4 t m i e r * -

Sp

1 Egypuan Sand R* 4 ♦ 4 S Sporcaneous Diabdic We t * Rat ? ? ? 1 (Japan) 7

ZucLet ra! * - -

. Ma n (N DDM) 4 4 • •

Chmese Hanster • 4 4 X South Afncan Hamsier X ? 4 4 N e » Zealand h^ t i R a h M - * ------[ NO D M* * e 4 .? ■ 4 B B We : * Ra! 4 ? 4 4 | LETL ra: * 4 ? 4 j Ma n (lOOM) * 4 4 4 4 TABLE 2.3 COMPARISON OF SELECTED SPONTANEOUS RODENT MODELS FOR DIABETES MELLITUS

MAINTENANCE INFORMATION

COMPLETE

INSULIN REOUIRED FOR PATHOLOGY PROFILE MAINTENANCE IMPAIRED INCIDENCE WITHIN AVAILABLE MODEL BREEDING UTTER AGE OF ONSET LIFE SP A N BODY WEIGHT

- 4 50% puberty 4 Obese Yellow Mouse max > 1 8 mo < 80g

Obese Hyperglycemic - 4 25% 4-6 wk. (ob/ob) Mouse - 15 mo. < 120g

Diabetic (db/db) Mouse - 4 25% 5-6 wk. partial 3-6 mo. 45g (12 mo. max) N e w Zealand Obese ? - 4 ? partial ? (NZOJ Mouse < 72g

Japanese KK Mouse - - 50-90% 6-8 wk. ? < 45g Spiny Mouse - - 15% 6-17 mo. ? 60-85g Egyptian Sand Rd - ? diet dependent ? max. > 32 mo. 118-225g Sporeaneousty Diabetic - ? ? ? Wstar Rats (Japan) ? ?

Chinese Kamsiei 4 4 < 100% 1-12 mo. max. > 23 mo 22-35g South African Hamster - ? ? -4 mo. 14 mo. max 75-185g N e w Zealand White 4 ? 19% 1-3 yrs. Rabbit 5yr. ?

N O D Mouse ' ? > 90% female 7 mo. ? <30g < 10% male

B B Wstar (BBW) Rat 4 4 70% 2-4 mo 4 14 mo. max < 400g

L E U R a 4 ? 14-64% 2-5 mo - ? ? 12 mo. max CHAPTER III METHODOLOGY

A. Colony Husbandry A breeding colony of BB Wistar diabetic (BBWd) rats was established in the Pathology Department of the Ohio State University from the parent colony at the Health Protection Branch of the Canadian Government. Animals were shipped by commercial airlines from Ottawa to Columbus, Ohio where they were originally housed in a standard animal housing facility, and later in a semi-barrier housing facility. The barrier facility was constructed with glazed tile walls and epoxy-terrazo floors. The room was maintained under slight positive pressure at 22°C and was supplied with fresh air passed through high efficiency particle fitters at a rate of 16 air changes per hour. Humidity varied between 45 and 55% and a constant photoperiod was maintained with a

12 hour light-dark cycle.

Rats were housed as mating pairs in 27.9 x 48.3 x 17.8 cm. polycarbonate cages with sanitized hardwood chip bedding

(Sani-Chips, FJ Murphy Co., Moonachie, NJ). Cages were sanitized at 82.2°C and changed twice each week. All rats had free access to pellet rat chow (Purina Laboratory Chow,

14 Ralston Purina Co., Richland, Indiana) and water. Water bottles were acid cleaned weekly and fresh tap water was supplied daily. All rats were weighed and monitored daily for ketones (Ketostix, Ames Co., Elkhart, Indiana) and glucose (Testage, Eli Lilly, Indianapolis, Indiana) in the urine with laboratory test strips. Glycosuria was estimated semiquantitatively on a scale from 0 to 4 plus; ketosuria was recorded as zero, small, or large. Protamine zinc (U- 40) insulin (Eli Lilly, Indianapolis, Indiana) was administered subcutaneously each morning to maintain 4+ glucose (without ketosis) in the urine. The initial dosage was determined on the basis of body weight and then adjusted according to individual responses to therapy determined by urinalysis. Daily treatments were recorded on treatment report forms. Non-diabetic siblings, BB Wistar rats that did not develop diabetes (BBWnd), and commercially purchased outbred Wistar (W) rats (Harlan Industries, Indianapolis,

Indiana) were housed under identical conditions but did not require insulin therapy.

B. Necropsy Procedure All rats were necropsied as soon as possible following expiration and were stored in a cold room following the discovery of their death or their sacrifice until the time of their necropsy. Each animal was examined externally and 16 then laid on its back. A midline excision was made from the mandible to the anus and the skin and musculature was reflected to the side. The thorax was then opened and the clavicles were cut. The trachea was ligated above the thyroid and the body organs were removed en bloc. The cranium was opened to remove the brain. All organs were examined individually for gross abnormalities and observations were recorded on Autopsy Report forms.

Interesting findings were frequently photographed. Whole body weights and weights of certain organs (heart, lungs, liver, spleen, pancreas, kidney, testes, brain) were routinely recorded. General sources on rat anatomy (Bivin et al., 1979; Hebeland and Stromberg, 1986) were consulted as necessary. Tissues were fixed in neutral phosphate-buffered formalin, trimmed, placed in labelled cassettes, and then submitted for processing by the histology laboratories in either the Division of Anatomical Pathology or the Division of Neuropathology. Tissue specimens were processed by

Autotechnicon (Technicon Corporation, Tarrytown, NY) as follows. Tissues were dehydrated through a graded ethanol series (70%, 80%, 95%, 100%), cleared with xylene, and infiltrated with paraffin. Then, specimens were embedded in paraffin blocks and sectioned at 4-6 microns with a microtome. Sections were floated in a waterbath, mounted on glass slides, and then routinely stained with hematolylin 17 and eosin (HE). Tissue and fluid specimens suspected of

antemortem bacterial infection were placed in disposable culture collection tubes (Septi-Seal culturette, "Marion Scientific Corporation, Rockford, Illinois) and cultured in the Microbiology Division of the Department of Pathology. Specimens were plated on blood agar, chocolate agar, MacConkey's agar, Sabourand agar, thioglycollate broth,

Schaedlers blood agar, Schaedlers blood agar with Kanamycin

and Vancomycin, Kamanycin Bile esculin agar, chopped meat, and biphasic medium for mycoplasma. Paraffin sections of such tissues were stained with Gram's stain, acid fast blue, or Gomori's methenamine silver for bacteria and fungi.

Histological preparations of the following tissues were routinely prepared and examined (the number in parenthesis indicate the minimal number of tissue sections processed): heart (1), thoracic aorta (1), lungs (2), liver (1), spleen (1), pancreas (2), stomach (glandular and squamous portions) (1), small intestine (1), colon (1), adrenal (2), kidney (2), bladder (1), testes (2), prostate (1), uterus (2 horns,

1 body), ovary (2), mediastinal lymph node (1), mesenteric lymph node (2), brain (2), pituitary (1), and thyroid (3). Parathyroid and thymus were occasionally examined. The skin, mammary gland, muscle, bone, eyes, spinal cord, and bone marrow were histologically examined only when disease was grossly apparent. Histological findings were recorded on Histology Report forms. Photomicrographs were taken with 18 a Leitz Wetzlar Orthoplan large field microscope equipped with a Leitz Wezlar Orthomat camera on Kodak Ectachrome 64 film at 50 ASA.

C. Hematology Rats used in hematological studies were anesthetized with ether, and blood (300 lambda) was collected in capillary whole blood collectors (Microtainers, Becton-

Dickinson, Rutherford, New Jersey) by inserting a heparinized capillary tube (American Hospital Supply Corp., Miami, Florida) into the retro-orbital plexus of vessels at the inner canthus of the eye. This technique was chosen because it offers more consistent results than tail vein bleeding (Hulse, 1965). Complete blood counts (CBC), differential white cell counts, and platelet counts were performed on BBWd (16 male and 11 female), BBWnd (12 male and 10 female), and control (9 male and 10 female rats). (Control rats were a Wistar line obtained from Bio Breeding Laboratories in Ottawa, Canada.) CBC's and platelet counts were accomplished with a Coulter S-plus. Platelet counts in excess of 700,000 were diluted 1:2 until there was agreement in two of three simultaneous measurements. Blood for serum studies was also collected by retro- orbital bleeding. Samples were allowed to clot and then centrifuged in a table top centrifuge (Model HN, International Equipment Co.) with a fixed angle rotor at 19 1500 rpm. The supernatant was pipetted into polypropylene micro sample tubes (Kew Scientific, Columbus, Ohio) and frozen at -40°C until use. Protein electrophoresis was performed on agarose plates at pH 8.6. Total protein was measured by the biuret method. Samples from five rats in each of the six categories were compared by two-way analysis of variance.

D. DNA Repair Studies Because eukaryotic cells undergo semiconservative DNA synthesis only during the S phase of the cell cycle, unscheduled DNA synthesis (UDS), a measure of excision repair, can be quantified in non-S phase cultured fibroblasts using an autoradiographic procedure. If S phase cells are cultured with tritiated thymidine and then examined autoradiographically, the nuclei will show intense staining with silver grains while non-S phase cells will not show significant nuclear labelling. However, non-S phase cells undergoing UDS because of repair will contain nuclear silver grains when examined autoradiographically. The number of silver grains per cell undergoing UDS can then be counted and these counts reflect the amount of DNA repair synthesis. In order to measure UDS, hydroxyurea is used to selectively inhibit S-phase DNA synthesis. Hydroxyurea inhibits ribonucletide diphosphate reductase, an enzyme used to synthesize deoxynucleoside triphosphates from the 20 corresponding ribonucleoside diphosphates. Hydroxyurea appears to have little or no effect on UDS because UDS does not require a large pool of DNA precursors and can proceed using the pre-existing pool (Friedberg, 1985). Primary fibroblast cultures were established from skin samples derived from neonatal offspring of either a diabetic BB Wistar rat mating pair or a control Wistar rat mating pair. The pups were killed and their abdomens cleaned with 70% ethanol. Skin samples (lxl cm) were excised and then washed three times in phosphate buffered saline (PBS). The samples were placed in 0.5 ml of 0.01% trypsin and then minced between two razor blades. The minced dermis, in 25 ml flasks containing 5 ml of 0.01% trypsin, was then digested for 30 minutes in a 37° C water bath/shaker. The contents of the flasks were then poured into 50 cc plastic centrifuge tubes and then inactivated with 40 ml of B medium

(Eagles MEM supplemented with 1.5x essential amino acids, 2x non-essential amino acids, and 1.5x vitamins) containing 10% fetal calf serum and 0.02% gentamycin. The tubes were centrifuged at 1100 RPM for 15 minutes and then the medium was decanted. Pellets were resuspended in 8 ml of B10 2X medium, divided into four T25 flasks (Corning Glass Works, Corning, N.Y.), and incubated at 37°C with 95% air/5% carbon dioxide. The fibroblasts were confluent at 12 days and were harvested with 0.01% trypsin, split 1:2, and plated.

The fibroblasts were split 1:2 for each successive passage. 21 Passage 3 fibroblasts were used for the experiment.

Fibroblasts from each cell line were attached to 22 x 11 mm glass coverslips as described below. The fibroblasts were suspended at a concentration of 9 x 1010 cell/ml into 100 mm diameter culture plates containing ten coverslips. Each coverslip received 0.5 ml cell suspension. The fibroblasts were permitted to attach and grow. After two days, the medium was replaced with medium containing 2 x

10-3M hydroxyurea. After 15 hours, the medium was aspirated and the cells washed twice with PBS. For both BBW and W cell lines, the petri dishes were divided into 3 groups: (a) control plates receiving no radiation, (b) plates treated with 20 J/m2 ultraviolet (UV) radiation, and (c) plates treated with 40 J/m2 UV radiation. Next, both the irradiated and unirradiated plates were then exposed to medium containing 2 x 10"3 M hydroxyurea and 2 jxCi/ml tritiated thymidine (3HdThd, specific activity = 2 0 Ci/mmol, New England Nuclear) for 0, 2, 4, 6, and 8 h. The coverslips were then washed thrice in PBS and then fixed thrice for 5 min in freshly prepared Carnoy's solution. The fibroblast-bearing coverslips were then dehydrated through an ethanol series and mounted on labeled glass slides cell- surface up. The slides were placed in a drying oven. After drying overnight, the glass slides were dipped in a photographic emulsion (0.67% glycerol + 32.67% distilled water + 66.67% Kodak NTB-2) and stored in black slide boxes containing calcium carbonate at 4°C. The black slide boxes were tightly sealed with electrical tape. After 6 days, the emulsion coated coverslips were developed with Kodak D19, washed, fixed, and lightly stained with toludine blue.

Fibroblast nuclei were viewed with a microscope (630X). Nuclear grains were counted electronically using a Docuval camera and an Artex 880 counter in approximately 50 non-S phase cells on each coverslip. All analyses were performed on grain counts corrected for background. Corrections for background were made by subtracting the mean number of grains present in adjacent equivalent areas without cells on each coverslip. Three coverslips per cell line were examined for most group and dose interactions. The magnitude of scheduled DNA synthesis for each group and dose was equivalent to the average number of grains per nucleus for irradiated cells minus the values for unirradiated cells.

E. Statistics

Statistical analyses were performed as described in individual papers listed in the appendices. CHAPTER IV

RESULTS AND DISCUSSION

A. Morphology of Organ Systems 1. Pancreas

A wide variety of exocrine pancreatic lesions have been reported in diabetics, but the significance of many of these, if any, remains obscure. Several sources have reported that the diabetic pancreas has a decreased organ weight (LeCompt and Gepts, 1977), but this is minimal in the absence of other significant pancreatic pathology (e.g.- acinar atrophy) since the islets constitute such a small part of the entire organ (Warren et al., 1966). Fibrosis is the most commonly reported finding in the exocrine pancreas of human diabetics and is sometimes associated with a degree of interstitial chronic inflammation. Very high incidences have been reported in some studies (Warren et al., 1966; Kothare, 1974) but the incidence in the control group of non-dabetics may also be nearly as high (Kothare, 1974). Therefore, this is not a specific change seen in the diabetic pancreas. Other frequent findings include fatty infiltration, arteriosclerosis and pancreatitis. Although the incidence of pancreatitis in some studies has been as high as 20-40% (Kothare, 1974), some autopsy studies have

23 24 reported low (5%) incidences (Goto et a l . , 1974; Parson et a l . , 1968. On the other hand, it is well documented that either acute or chronic pancreatitis may lead to overt diabetes (Johansen and Ornshott; 1972; Warren et a l . , 1966), but this is a rare mechanism for the onset of diabetes. There is presently no compelling evidence suggesting that pancreatitis is necessarily more common in diabetics than in the general population (Warren et al., 1966).

Rare findings that appear to be somewhat more frequent in diabetics than in the general population are pancreatic carcinoma and pancreatic calculi (Warren et al., 1966). Because of their low incidence, neither warrants further discussion, but I will digress momentarily because of the historical significance of one of these. A case report by

Barron in 1920 in which a calculus blocking a pancreatic duct resulted in acinar atrophy with persistence of the islets is credited with suggesting to Banting his classic duct ligation experiment resulting in the discovery of insulin (Bliss, 1982). In summary, exocrine pancreatic lesions may be common in diabetes, but most appear to have limited significance.

On the other hand, the pathology of the endocrine pancreas in diabetes is very important and has been extensively characterized. Furthermore, the islet pathology differs distinctly in IDDM and NIDDM. 25 In IDDM, the islets show severe and pathogomonic changes (Gepts and LeCompte, 1981). The most striking change is that islets are decreased in both size and number.

Frequently, these endocrine cells are scattered as single cells in the exocrine tissue rather than existing as discrete islets. The appearance of pyknotic nuclei and acidophilic cytoplasm in some islets (Warren et al., 1966) lead to the classic description of "atrophic" or "inactive" islets in IDDM. Immunocytochemical studies later showed that essentially only the beta cells were reduced in number and that the diabetic islets were composed predominately of alpha, delta, and PP cells. It has also been determined that beta cells seldom totally disappear in the years following clinical onset of IDDM (Gepts and LeCompte, 1981; Foulis and Stewart, 1984; Rahier et al., 1983; Orci and Perrelet, 1979). In most cases of IDDM, the number of beta cells is greatly reduced at the onset of diabetes, but this is not always the case. Occasionally, hypertrophic islets, composed mainly of beta cells showing signs of functional hyperactivity are seen in the pancreas of young diabetics who die shortly after onset. Another finding that is frequently observed in the islets of IDDM patients that die shortly after clinical onset (usually less than 1 year) is insulitis. Insulitis, a term coined in 1940 by Von

Meyenburg, is lymphocytic infiltration selectively involving the pancreatic islets (Gepts, 1977; Foulis and Stewart, 26 1984). Insulitis usually affects some but not all islets of the pancreas. In most cases, only small lymphocytes are seen, but the presence of macrophages, plasma cells, eosinophils, and neutrophils have also been reported (Wellmann and Volk, 1980). One final lesion that has interesting historical significance is hydropic change, the presence of cells (now known to be beta cells) with clear cytoplasm. In the pre-insulin era, this was a common finding to which great significance was described. It was considered a degenerative change that was forerunner of islet cell atrophy, but this explanation was dismissed in

1951 when Toreson deomonstrated that the hydropic change was due to glycogen deposition (Gepts and LeCompte, 1981). Hydropic change is now fairly rare because few cases of diabetes go untreated. In contrast to the clear pattern in IDDM, the pancreatic islet pathology in NIDDM is extremely variable and is not pathognomonic (Gepts and LeCompte, 1981). Often, there is some decrease in the number of cells in the islets of individuals with NIDDM but this change is not nearly as severe as in IDDM. Others may have beta cell hyperplasia. Probably the most typical lesion in NIDDM is hyalinosis, a change seen in up to 50% of all NIDDM pancreases (Wellmann and Volk, 1980). Hyalinosis is not specific for NIDDM since it is seen on rare occasions in both IDDM and in non­ diabetics (Warren et al., 1966; Wellmann and Volk, 1980). 27 Islet fibrosis is another common finding in NIDDM but it is also occasionally found in IDDM. In summary, none of these changes are specific for (or even usually present in) NIDDM. Islet pathology is even seen in the infants of diabetic mothers. Several characteristic changes have been reported. These include islet hypertrophy and hyperplasia, an increase in the percentage of beta cells, and eosinophilic infiltrates. These changes have been attributed to either maternal hyperglycemia or insulin antibodies crossing the placental barrier (Barressi et al., 1978). According to Nakhooda et al. (1978), BBW rat islets typically show three general patterns: (1) normal islet histology in prediabetic and nondiabetic BBW rats; (2) mononuclear cell (occasionally with eosinophils) insulitis at the onset of diabetes and in the following few weeks; and

(3) "end-stage" islets, which are small, decreased in number, and composed almost entirely of non-beta cells, and a few granulated beta cells, in chronically diabetic rats. This picture is confirmed by our study and by others (Seemayer et al., 1982a). Most of the BBWd rats in our study had typical small "end-stage" islets (Figure A.7) often with numerous cells with pyknotic nuclei (Figure A.5). These were particularly common in the older BBWd rats and were not seen in BBWnd or W rats. In almost all instances, at least a few small islets remain regardless of the severity of the diabetic state. In several instances, 28 hyperplastic islets were seen in BBWd rats after the onset of diabetes (Figure A.6). Islet histology in prediabetic and nondiabetic BBW rats was usually normal. Insulitis was the most common inflammatory pancreatic lesion in both BBWd and BBWnd rats (Table A.l) and is characterized as a mononuclear cell infiltrate (almost always lymphocytic) that is specific for the islets. In most instances, the infiltrate is predominantly peri-insular (Figure A.l) but in the most severe lesions may also include frank invasion of the islets (Figure A.2). Frequently a few eosinophils are present in the periphery of inflamed islets. (Wright et al., 1985; Appendix A).

In BBWd rats, insulitis occurred most frequently in the younger age groups (Table A.l). The ages of BBWd rats with insulitis ranged from 107 days to 317 days of age (mean

+SD=164.1 +61.9 days). Since insulitis is transient, it was seen only in those rats that died shortly after onset of hyperglycemia. The longest period of time following onset of diabetes at which a BBWd rat still had full-blown insulitis was 69 days (the mean duration of diabetes for BBWd rats with insulitis when they died was 49.9 + 16.9 days). In reviewing the literature concerning insulitis in BBW rats, I was struck by the straight forward and oversimplified picture presented in previous studies.

Seemayer et al. (1982a) examined three early diabetic rats 29 (1-three days after detection of glycosuria); three unstable diabetic rats (7-22 days after detection), three stable diabetic rats (41-63 days after detection), eleven prediabetic rats (four at 50 days and seven at 65 days of age), and six nondiabetic control rats (BBW rats over 120 days of age that had not developed glycosuria). Seemayer et al. (1982a) reported no structural or inflammatory changes in the control (BBWnd) rats and variable degrees of periductal and acinar mononuclear infiltration and insulitis in the prediabetic rats (particularly in the 65 day old group). All three early diabetic rats had insulitis usually involving only the peripheral portion of the islets but occasionally permeating the whole islet. Involved islets were enlarged, edematous, and poorly delineated from the acinar tissue; islet cells showed cytoplasmic swelling, vacuolization, intense acidophilia, refractility, and nuclear pyknosis. Periductal infiltrates were also present.

One rat had focal acinar involvement by mononuclear cells. In the three unstable diabetic rats, islets were small, sparse, and showed no evidence of active cellular injury. Rare islets had residual mononculear cells. Two of the three stable diabetic rats had a single islet infiltrated by limited numbers of mononuclear cells. No islets were normal and several showed evidence of cellular injury. One rat had periductal mononuclear cell infiltrates. Since insulitis is a transient manifestation of the onset of diabetes in BBW rats, many questions remain to be answered about the time frame surrounding this event. Seemayer et al. (1982a) observed insulitis in most of their 50 and 65 day old prediabetic rats, and they suggest that the destructive process in the islets starts well in advance of the clinical syndrome. It is not presently known whether glycosuria is always preceded by insulitis. If so, why does

it remain severe in some BBWd rats 60-70 days after onset of glycosuria and is not seen in others only a few days after onset. Furthermore, why have other investigators observed fullblown insulitis only in early diabetes (i.e. a few days after onset). Our study reveals that it may be quite prominent several months after the onset of glycosuria (Figure A.2). A likely explanation is that our rat populations are different. Seemayer's study utilized rats

that were selected because they were a very homogenous

population for each group (i.e. early diabetic, stable,

unstable, and prediabetic). In at least some cases, animals within these groups were littermates. This might in part explain the clear pattern in this study. On the other hand, our rats are a rather heterogenous group with respect to

both ancestry and age of onset of diabetes. In order to examine whether age of onset had any effect on the persistence of insulitis, we examined all rats that died or were sacrificed within 70 days of onset of glycosuria and 31 found that 45 BBW rats were in this category, but there was no statistically significant difference in the age at onset of glycosuria for those rats within this subset that had insulitis and those that did not. Another point of confusion is the 16.3% incidence of insulitis in our BBWnd rats (Table A.l). In the nondiabetic rats, there was no specific age predilection for insulitis as with the diabetic rats. In fact, insulitis was observed in BBWnd rats ranging from 96 to 422 days of age (mean + S.D. = 306.5 + 132.4). Only one of these rats could be considered a typical prediabetic rat (i.e. less than 120 days old). If the insulitis is actually responsible for the destruction of beta cells and the subsequent onset of hyperglycemia, then why does insulitis occur in about one- fifth of the BBWnd rats in our study and others. For instance, Nakhooda et al. (1978) also reported an incidence ranging from 17 to 25% in BBWnd rats, although Seemayer et al. (1982a) did not observe insulitis in his BBWnd series. Either these animals were developing overt diabetes at the time of their death (it is well known that many older BBWnd rats eventually convert into BBWd rats) or insulitis does not always result in clinically apparent diabetes.

In addition to insulitis, other inflammatory lesions were also common (Table A.l). Chronic interstitial inflammation (i.e. inflammation with little acinar involvement) was the most frequent of these occurring in all 32 three groups of rats at incidences between 11% and 14%. Except for chronic interstitial inflammation, no other pancreatic lesions were seen in W rats. Low incidences of chronic pancreatitis (i.e. inf lamination with diffuse, extensive acinar involvement), acute pancreatitis, acute and chronic pancreatitis, acute interstitial inflammation, and acute and chronic interstitial inflammation were present in BBW rats. In the acute lesions, eosinophils were often present either exclusively or with neutrophils (Figure A.3).

No age pattern was discernable for any of these inflammatory lesions. Granulomas were occasionally observed in BBW pancreatic tissue (Table A.l). In no instance could any organisms be demonstrated with gram stain, GMS stain, or acid fast blue stain. In several rats, granulomas and insulitis were present together (Figure A.4). Eosinophilic infiltrates were occasionally present with granulomas.

The most common pancreatic lesion in most strains of aging rats is primary acinar atrophy and fibrosis (Berg, 1967; Anver and Cohen, 1979; Burek, 1978). This change is characterized by shrinkage of acini, absence of cell granules, transformation of acini into duct-like structures lined by cuboidal epithelium, and complete degradation of acini with fibrosis of the empty supporting tissue (Berg,

1967). Fat cells are often interspersed within areas of pancreatic lobular atrophy (Anver & Cohen, 1979). One study 33 has reported asymptomatic "chronic relapsing pancreatitis" to be very common in some inbred rat strains (Kendrey and Roe, 1969). It appears that many of these instances are equivalent to what we refer to as chronic interstitial inflammation rather than actual pancreatitis (i.e. extensive diffuse acinar involvement). They also reported changes consistent with acinar atrophy (always associated with some inflammation) and a few cases of polyarteritis nodosa. Two types of endocrine pancreatic lesions occur frequently in aging rats: islet atrophy and islet hyperplasia. Islet atrophy is not common and is usually secondary to extensive exocrine damage whereas islet hypertrophy is more common and may occur spontaneously. Beta cells in the latter islets become both hypertrophic and hyperplastic with increased granularity. Intra- and peri- insular fibrosis may become marked and the islets may eventually coalesce to form giant conglomerates or even tumors. This change is usually associated with abnormal glucose tolerance curves.

Neoplastic changes in the pancreas of the rat are exceedingly rare (Altman and Goodman, 1979; Rowlatt, 1967). In fact, a review by Altman and Goodman (1979) states that only one spontaneous exocrine adenoma and 7 spontaneous exocrine adenocarcinomas have been reported in the literature. Islet tumors are more common than exocrine tumors but are still rare. The majority of endocrine 34 pancreas tumors have been islet cell adenomas. Only seven islet cell carcinomas have been reported in the literature (Altman and Goodman, 1979).

2. Endocrine a) Thyroid The relationship between diabetes and thyroid disease is presently uncertain. Published reports pertaining to

thyroid function in diabetes are fragmentary and

contradictory. Although very little is known about the

functional state of the thyroid in diabetes, there is a small subgroup of the human IDDM population for which this relationship is better established. Several studies have reported an association between IDDM and autoimmune thyroid disease (Irvine et a l ., 1970 and Nerup and Lernmark, 1981). Relevant to this is the observation that HLA-B8 and Dw3 histocompatability antigens occur with a greater frequency in patients with IDDM, Graves' disease, and Hashimoto's thyroiditis (Christy et a l ., 1977 and Moens and Farid,

1978). Furthermore, a higher incidence of thyrotoxicosis has been reported in diabetics as well as frequent carbohydrate metabolic problems in patients with thyrotoxicosis. In addition, autopsy studies have reported an occasional association of IDDM and chronic thyroiditis. Other autoimmune disorders in endocrine glands also are occasionally present in human IDDM. This has been termed 35 "the syndrome of polyendocrine autoimmunity (Bottazzo et al., 1974; Doniach and Bottazzo, 1981) and is now recognized as a clinical entity. The occurrence of a spontaneous lymphocytic thyroiditis in BBWd rats suggests that BBW rats may closely mimic this syndrome of polyendocrine autoimmunity (Sternthal et al., 1981; Wright et al., 1983a, see Appendix B ). In general, non-neoplastic thyroid disorders are infrequently observed in most strains of laboratory rats. Burek (1978) reported occasional stratified squamous-lined

ultimobranchial duct cysts, "colloid" cysts, and focal

periarteritis of thyroid arteries in several strains of rats. Organized colloid cysts were very frequently observed in all three groups in our series and were occasionally inflamed. Coleman et al. (1977) reported a 0.7% incidence

of diffuse thyroid hypertrophy in Fischer 344 male rats. In

this present study, hypertrophy was not common and was seen

predominantly with severe thyroiditis. Others have reported atrophic changes (Bullock et al., 1968), marked variation in follicular size and colloid content (Anver and Cohen, 1979), and basophilic debris and calcified concretions (Anver and

Cohen, 1979) in the follicles of aged rats. Except for cysts, the only lesion seen in the thyroid of any of the W rats was focal C cell hyperplasia in a 551 day old male W rat. 36 Lymphocytic thyroiditis was observed in 36 BBWd rats (63.2%) and in eight BBWnd rats (42.1%). A complete description of the histology of these lesions, a review of the pertinent literature, and a discussion of the significance of this finding is present in Appendix B

(Wright et al., 1983a). During the past decade, spontaneous thyroiditis in BBW rats has been extensively studied and the BBW rat has become an important animal model for experimental thyroiditis (Boitard et al., 1985; Allen et al., 1986a; Allen et al., 1986b; Allen et al., 1987; Allen et al., 1990; Voorby et al., 1990; Gottlieb et al., 1991; Rajatanavin et al., 1991; Allen, 1992). Although neoplastic lesions of the thyroid are frequently observed in various strains of aged rats, none were seen in our study. Several studies have reported incidences of naturally occurring medullary thyroid carcinomas in 16-40% of several strains of rats including the Long-Evans, Sprague-Dawley, Wistar, WAG/Rij, and wild rat (Rattus norviqicus) (Boorman and Hollander, 1976;

Thompson and Hunt, 1963; Gilbert and Gillman, 1958).

Medullary carcinomas in the rat arise from C cells; tumors of follicular cell origin are rare in rats (Boorman and Hollander, 1976). Thyroid tumors may also be induced in rats by feeding a low iodine diet (Axelrod and Leblond,

1955). On the other hand, several strains of rats have very low incidence of thyroid neoplasms. Coleman et al. (1977) 37 reported a 2.1 and 0.7% incidence of medullary thyroid carcinoma and thyroid adenoma respectively in a study on aging Fischer 344 rats. b) Parathyroid No relationship between the parathyroid and diabetes has been theorized, but a few scattered reports indicate that the gland may be adversely affected by the diabetic state. Then again, it is likely that every organ in the body is adversely affected. Fraley and Totten (1968) reported a significantly higher (P < 0.05) incidence of histological abnormalities in the parathyroids of diabetic subjects at autopsy. Hansson (1964) reported an increased volume and nuclear size in the parathyroids of alloxan diabetic rats. On the other hand, Warren et al. (1966) reported no consistent abnormality in the parathyroids of diabetic patients, but often found secondary chief cell hyperplasia in patients with diabetic nephropathy.

Although one or more parathyroids were examined in approximately one half of the rats in this study, no parathyroid lesions were observed in any of the three groups of rats. Sporadic non-neoplastic parathyroid lesions reported in other strains of senile rats include interstitial fibrosis, cystic changes, and hyperplasia secondary to renal disease (Burek, 1978; Altman and Goodman, 1979). Some strains of aged rats rarely have parathyroid 38 lesions and other strains, e.g. Long Evans rats, have incidences as high as 40% (Russfield, 1967). Parathyroid neoplasms are a very rare finding in most strains of rats (Altman and Goodman, 1979) and these consist almost exclusively of adenomas. Sass et al. (1975) have reported a single parathyroid carcinoma in a 28 month old

Fischer 344 rat. This is possibly the only report of a malignant neoplasm of the parathyroid in a rat. c ) Adrenal The functional state of the adrenal gland in diabetes mellitus is uncertain. Clinical association of diabetes with Cushing's Syndrome (i.e. hyperactivity of the adrenal cortex) is common knowledge. About 25% of patients with Cushing's Syndrome have symptomatic diabetes. Occasional autopsy studies have reported a significantly higher incidence of adrenal abnormalities (Fraley and Totten, 1968) and adenomas (Russi et al., 1945) in diabetics. Irisawa et al. (1966) have reported that the zona fasciculata is characteristically hyperplastic in diabetes. The present knowledge in this area is at best confusing and the true relationship of adrenal pathology and diabetes is difficult to assess.

A few isolated lesions were observed in BBW rats in this study but we have no reason to believe that they are related to diabetes since all are known to occur in 39 nondiabetic rat strains. Ectopic adrenal tissue (outside the glandular capsule) was observed in two BBWd rats, but this is not an uncommon finding in other strains of rats

(Russfield, 1967; Burek, 1978). Another BBWd rat had three distinct adrenal glands. Other isolated findings observed in BBWd rats that have been described in the literature include "foci of cellular alteration" (Burek, 1978), dilatation of sinusoids associated with blood-filled cysts or thrombosis (Burek, 1978; Muraoka et al., 1977; Anver and Cohen, 1979), focal degeneration and necrosis (Burek, 1978; Anver and Cohen, 1979), sudanophilia and pigmentation (Muraoka et al., 1977), hyalinization of the capsule (Anver and Cohen, 1979) and fatty change (Coleman et al., 1977; Cohen et al., 1978). Several types of non-neoplastic medullary lesions have also been described. These changes included "foci of basophilic change" (Burek, 1978) and thrombosis (Cohen et al., 1978). The significance (if any) of non-neoplastic adrenal lesions in rats has not been established; but they may occur frequently in some strains (Burek, 1978). Although no neoplastic adrenal lesions were observed in this study, spontaneous adrenal tumors are common in some strains of aged rats. Both adenomas and carcinomas are observed in older rats with adenomas predominating (Altman and Goodman, 1979). Incidences of adrenal cortical adenomas between 2-40% have been reported in old rats (Sass et al., 40 1975; Cohen et al., 1978; Burek, 1978). Single incidences of myelolipoma and sarcoma have also been reported (Altman and Goodman, 1979). Nothing is known about the endocrine activity of these cortical tumors (Uebeberg and LUtzen, 1979). Medullary tumors in rats consist almost exclusively of pheochromocytomas. The incidence and age of onset of this lesion varies drastically even in strains of rats derived from common ancestors. Burek (1978) reported these lesions in only 8% of the females and 1% of the males in Wistar-derived WAG/Rij rats (mean age 34 months) while Gilbert and Gillman (1958) reported a 50% incidence in female and an 82% incidence in male South African Wistar rats living over a year.

d) Pituitary

A relationship between pituitary disease and diabetes has been suspected for many years (Warren et al., 1966). Marie's original paper on acromegaly in 1889 reported that these two diseases may occur together. Harvey Cushing in

1912 produced strong evidence for an association between glycosuria and excess secretion of the pituitary. Experimental observations by Houssay in 1931 lead to the description of the Houssay phenomenon: amelioration of diabetes following removal of the pituitary (Warren et al., 1966). Observations consistent with this phenomenon have since been reported with Sheehan's Syndrome (postpartum 41 pituitary necrosis), chromophobe adenoma, and late inactive eosinophilic adenoma (Bonar, 1977.) Several investigators have noted the occurrence of the Houssay phenomenon in man due to pituitary infarction, but others have not seen any improvement. Essentially no consistent morphological changes have been reported in the pituitaries of diabetics in the recent literature, although Warren et al., (1966) did report a three-fold increase in the incidence of pituitary infarcts in diabetics. Autopsy studies on IDDM patients that died in the pre-insulin era showed several characteristic pituitary changes including a reduction in the gross weight of the pituitary, a reduction in the number and size of eosinophils (often with pyknosis), and increased numbers of infarcts and fibrotic foci. Kraus believed these changes to be secondary to IDDM and later suggested that the discrepancies between his study and later studies might be due to the moderating effects of insulin treatment and the inclusion of both IDDM and NIDDM patients in later studies (Kraus, 1944). Except for a single chromophobe adenoma in a 327 day old male BBWd rat (Figures 4.1 and 4.2), no pituitary lesions were observed in either BBW or W rats in our study. This is undoubtedly an insignificant lesion since tumors of the pituitary gland are extremely common in many strains of aged rats and is the main factor limiting the lifespan of some strains. In general, female rats have a higher incidence than males (Altman and Goodman, 1979). Incidence is very low in rats under 18 months (Snell, 1963), but adenomas have been reported in rats as young as seven months (Altman and Goodman, 1979). The vast majority of pituitary tumors in rats are chromophobe adenomas. Grossly, most are soft with an irregular surface and have prominent hemorrhagic areas. They are generally well circumscribed and consist of large polygonal cells with prominent vesicular nuclei and abundant non-granulated eosinophilic cytoplasm. Basophilic adenomas, acidophilic adenomas, chromophobe carcinomas, a craniopharyngioma, a granular cell myoblastoma, and a fibrosarcoma have been previously reported in rats but are rare (Altman and Goodman, 1979). Magnusson et al. (1979) has reported a 0.42% incidence of malignant pituitary tumors in outbred Sprague-Dawley rats. These showed infiltration into brain tissue but not metastasized extracranially.

Non-neoplastic lesions of the pituitary are both uncommon and insignificant (Burek, 1978). Colloid cysts are occasionally seen in the anterior pituitary of rats older than 20 months of age. Berg (1967) reports a generalized increase in chromophobe cells and a decrease in acidophils.

Increased reticular connective tissue and melanin pigment have also been reported in aged rats (Burek, 1978; Anver and Cohen, 1979). 43 3. Male Reproductive a) Testes It is well known that impotence affects approximately 50% of diabetic men of reproductive age (Konez and Balodimos, 1970; Wiles, 1992). Schoffling et al. (1963) attempted to relate testicular histological findings to the clinical presence of impotence because none of the testicular biopsies from diabetic patients with impotence was entirely normal. They found variable degrees of tubular atrophy, basement membrane thickening, decreased spermatogenesis, and a relative preponderance of Sertoli cells. Federlin et al. (1965), also reported a significant decrease in the early stages of spermatogenesis in diabetics with "hypogonadism". Neither Schoffling nor Federlin examined testicular histology in diabetics with normal sexual function. Singhal et al. (1969), examined testicular biopsies from diabetics with and without impotence as well from age-matched normal controls. They reported histological findings similar to the previous studies but also observed that these changes were not limited entirely to diabetics with impotence. Although these changes were more frequent in the diabetics than in the controls, there was no significant correlation with loss of sexual potency. Impotence is presently believed to be due to diabetic peripheral neuropathy and to be unrelated to these testicular changes (Konez and Balodimos, 1970; Wiles, 1992). 44 Although testicular histology has been studied extensively in chemically-induced diabetes (Scaffidi and Rotolo, 1974; Schoffling et al., 1967; Rosenman et al.,

1974; Soulairac et al., 1948; Oksanen, 1975; Lukens, 1948; Chesler and Tislowitz, 1945; Hunt and Bailey, 1961), it has

been examined in only two other strains of spontaneously diabetic animals, the obese-hyperglycemic AO mouse (Heilman et al., 1963) and the Chinese hamster (Schoffling et al., 1967). Both of these studies described hypocellularity or maturation arrest as the predominant lesion rather than total absence of germ cells within seminiferous tubules. Histological changes observed in the BB Wistar rat included both mild and severe atrophy, in addition to these milder changes (Wright et al., 1982, Appendix C; Murray et al., 1983; Murray et al, 1985).

Pancreatectomy in the rat also results in degenerative changes of the germinal epithelium (Lema et al., 1965) which were observed only after the onset of diabetes. The degree of change varied with the severity of diabetes. Without

exception, the thickness of the basement membrane of the tubules and blood vessels were normal. Atrophy of secondary sexual organs were occasionally observed in rats with severe testicular damage. These investigators also reported tubular changes consistent with maturation arrest and a decreased ability to impregnate female rats. Pancreatectomy has also been reported to result in testicular atrophy in 45 cats and roosters (Lema et al., 1965). The severity of testicular lesions in diabetic men and

animals varies widely from study to study. Several have reported only a hypocellularity of germinal cells (Heilman et al., 1963; Scaffidi and Rotolo, 1974), others more severe atrophy (Schoffling et al., 1967), and some only Sertoli cells present (Rosenmann et al., 1974; Soulairac et al., 1948). This discrepancy may possibly be explained by the

variable dosages and/or toxicities of the chemicals used to induce diabetes or by severity of the diabetic state in these different studies. Several investigators have suggested that testicular atrophy is more frequent in poorly controlled diabetics (Schoffling et al., 1967; Warren and LeCompte, 1952). Since testicular atrophy is prevalent in most strains of aged rats (Burek, 1978; Coleman et al., 1977; Lutzen and

Ueberberg, 1973), age is certainly a contributing factor to the high incidence of testicular lesions in the BB Wistar rats. Burek (1978) reported that at least limited atrophy was present in all rats over 18 months of age and was usually severe in rats older than 24 months. He found that atrophic testes of old rats frequently had Leydig cell

hyperplasia, interstitial edema, tubules with multinucleated

giant cells or only Sertoli cells. The changes described by Burek were consistent with the histological changes observed

in all three types of rats in our study (Wright et al., 46 1982, appendix C). Several investigators have reported benign or malignant tumors to be frequently associated with testicular atrophy in senile rats (Burek, 1978; Coleman et al., 1977; Lutzen and Ueberberg, 1973), but these were not seen in any rats in our study. This is possibly due to the greater age of their animals. In our study, testicular atrophy was first observed in BBWnd and W rats at 355 and 361 days of age respectively; the earliest age of onset in the BBW diabetic rats was 148 days. Figure C.l shows clearly that the incidence of atrophy increased with age in all rats but occurred at a much younger age in BBWd rats. Saksena et al. (1979) have reported that sperm production in Sprague Dawley rats remains maximal from age 72 days to beyond 450 days of age. This appears consistent with our histological findings in standard Wistar rats but not in the BBWd rats. Age is undoubtedly a very important factor determining the presence of atrophy in all groups of rats but the duration of diabetes is a much better predictor of atrophy in BBWd than is age. Other causes of testicular atrophy in rats are extensively reviewed in Appendix D

(Wright, 1987). Inflammatory testicular lesions were uncommon. Two diabetic rats had orchitis (Figure 4.3) and another had bilateral granulomatous lesions (Figures 4.4 and 4.5). 47 b) Prostate

Reports of rat prostatic pathology are limited because the organ is often not included in necropsy studies, but the available literature indicates that prostatitis is common (Miintzing et al., 1979). Burek (1978) reported that the most common lesion in the prostate was diffuse, multifocal, or focal lymphocytic and plasmacytic infiltration. Abscesses were occasionally observed. Unfortunately, Burek did not report any frequencies of prostatic lesions. Cohen et al. (1978), reported a 22.9% incidence of chronic prostatitis in aging Crl:CD(SD) BR rats. This is consistent with the findings of Coleman et al. (1977) who reported a 19.5% incidence in Fischer 344 rats; but they also reported a 33.1% incidence of focal suppurative prostatitis. In this present study, the prostate was examined in only a small percentage of the rats, but it is still apparent that prostatitis is common in all three groups of rats. It is interesting that chronic prostatitis predominated in W rats and acute prostatitis in BBW rats (Figure 4.6). No prostatitis was observed in W rats under one year of age, but chronic prostatitis was seen in three of seven W rats (42.9%) in the 481-600 day old group. No acute prostatitis was observed in W rats. On the other hand, acute prostatitis was observed in one of 15 BBWnd and three of 24

BBWd rats. No age pattern was discerned. In several of these rats the inflammatory infiltrate included many 48 eosinophils. In one instance, acute prostatitis caused severe bowel obstruction. An isolated case of chronic prostatitis was observed in a 137 day old BBWd rat. Small eosinophilic or basophilic concretion bodies (similar to corpora amylacea in man) are common in the acinar lumens in the prostates of older rats (Anver and Cohen, 1979). The significance of these concretions is unknown.

Acinar atrophy is apparently an infrequent finding in rats regardless of age. The only report that I could find in the literature was a single case out of 144 male Fischer 344 rats (Coleman et al., 1977). This is consistent with my observations in the Wistar rats (Figures 4.7, 4.8, and 4.9), but the high incidence in the BBW rat is difficult to explain (Figures 4.10 and 4.11). Sufrin and Prutkin (1974) reported that insulin deficiency causes testosterone to have less effect on the sex accessory glands in the male rat. This is consistent with the findings of Hunt and Bailey (1961) who reported prostatic atrophy similar to that observed following castration in untreated alloxan diabetic rats. They observed normal prostatic histology in rats treated properly with insulin. Although diabetes may explain the prostatic atrophy in the BBW diabetic rats, the explanation for the similar findings in their non-diabetic siblings is uncertain. Furthermore, similar findings have not been reported in human diabetics. 49 Prostatic hyperplasia, consisting of stratification of acinar epithelium, has been occasionally observed in rats greater than two years of age (Anver and Cohen, 1979). It is uncertain whether this is precancerous but Shain et al. (1975) have reported simultaneous incidence of spontaneous adenocarcinomas and prostatic hyperplasia and hypertrophy in the ventral prostate in aged AXC rats. The incidence and types of prostatic tumors have been reviewed by Altman and Goodman (1979).

3. Female Reproductive a) Uterus

Uterine changes in diabetic women have not been previously reported, except as related to the formation of the placental unit. These changes include excessive placental weight, villous immaturity, villous infarcts, chorangiosis, intervillus thrombi, obliterative endarteritis or thrombosis of fetal stem arteries, thickening of the trophoblastic basement membrane, villus fibrinoid necrosis and fibrosis (Heifetz, 1992; Benirschke and Kaufmann, 1990). Placental morphology in BBW rats has been studied by Brownscheidle and Davis (1981). They reported a slight decrease in mean placental weight and occasional large multilocular cysts in the basal zone, atrophy of the trilaminar trophoblast, and engorgement of the material blood sinuses. 50 Pathological changes were not seen in the vaginas or uteri of any of the three groups of rats in our study, except for large uterine stromal polyps in a 372 day old BBWd rat (Figure 4.12). Three BBWd and 1 BBWnd rats died while pregnant. Pregnancies were bilateral except in one rat.

Non-neoplastic lesions of the uterus in senile rats are infrequently reported. With advancing age, a gradual transformation of interstitial reticulum into collagen is seen in the uterus, cervix, and vagina. Hydrometra, pyometra, and cystic endometrial hyperplasia have also been reported (Anver and Cohen, 1978; Berg, 1967). Neoplastic lesions of the uterus are relatively uncommon in most strains of rats. Most uterine tumors are endometrial stromal polyps (Altman and Goodman, 1978), but a wide variety of other benign or malignant tumors have been reported. These have been reviewed by Baba and Von Haam (1967 and 1976). b ) Ovary

Although decreased fertility and pregnancy complications are frequent in female diabetics, ovarian pathology is unremarkable (Warren et al., 1966). Several studies have reported that ovarian atrophy and fibrosis are more frequent in diabetics than in the general population (Fraley and Totten, 1968; Warren et al., 1966) but this is 51 more likely associated with generalized wasting due to nutritional deficiencies rather than a specific diabetic alteration. This is supported by the observation that the atrophy was more severe prior to the discovery of insulin.

In fact, in about two-thirds of autopsy cases prior to insulin therapy the ovaries were less than half the size of those from non-diabetic women (Warren et al., 1966). No definite lesions were observed in the ovaries of any of the rats in our study. There did appear to be some decrease in the number of corpora lutea in young BBWd rats, but no morphometric studies were done to confirm this. In fact, two BBWd rats under one year of age had no apparent corpora lutea. Ovaries from sexually mature BBWnd and W rats appeared normal. Ovarian pathology in aging rats has been summarized in detail by Russfield (1967). Basic age-related changes include decreased oocytes, follicles, and corpora lutea and an increase in interstitial gland tissue, cyst formation, and degenerative changes in the corpora lutea. Most of these are features of ovarian atrophy. Burek (1978) claims that most, if not all rats over 18 months of age have atrophic ovaries, but the degree of atrophy is variable. Since there is a strong correlation between abnormal estrous cyclicity and morphological changes in the uterus, ovaries, and mammary gland (Anver and Cohen, 1979), a proper morphologic study must also include vaginal smears to 52 determine the cycling patterns of the rats. Since breeding problems are common in BBW rats, more definitive studies on this area might be interesting.

Inflammatory changes in the ovaries of female rats are not uncommmon and ovarian abscesses are probably the most significant of these (Bullock et al., 1968). On the other hand, ovarian tumors are infrequent in most strains of rats. When present, they are usually granulosa-theca cell tumors. Other types of ovarian tumors are very rare (Altman and Goodman, 1979).

c) Mammary Gland There are no changes in the mammary glands of diabetic women that are specific to that disease. Merriam and Sommers (1957) observed a frequent presence of hyaline substance surrounding the mammary ducts in diabetic women.

The only other pertinent study is one showing that breast cancer is less frequent in diabetic women because its onset occurs much later in life (Anderson, 1971). This may, in part, explain why no mammary tumors were observed in any of the BBWd rats in our study. It is possible that female BBWd rats do not develop mammary tumors because they do not live long enough. The only mammary lesion observed in our study was a single mammary adenocarcinoma in a 248 day old BBWnd rat (Figures 4.13 and 4.14). Other histologic abnormalities may have been present but were not diagnosed because only 53 grossly apparent lesions were examined. The mammary tumor is the most frequently occurring neoplasm in the female rat (Ueberberg and Lutzen, 1979; Altman and Goodman, 1979). Sher (1972) has reviewed the incidence of these for a variety of strains. Many strains have frequencies around 50% and a few up to 90%. Reports of 15 to 20% frequencies in males are not uncommon (Sass et al., 1975; Altman and Goodman, 1979). Generally, most murine mammary tumors are fibroadenomas; adenomas are infrequent. Benign and malignant fibromas and other mesenchymal tumors are occasionally seen. Adenocarcinomas occur in less than 10% of all spontaneous mammary tumors. Other carcinomas occasionally seen are papillary carcinomas, comedocarcinomas, and squamous cell carcinomas. Rat mammary carcinomas, while histologically malignant, remain reasonably well circumscribed and only moderately invasive (Altman and Goodman, 1979). The incidence of mammary tumors increases with age.

They are rare before one year of age and the incidence increases rapidly after 18 months (Altman and Goodman,

1979). According to Burek (1978), the incidence of fibroadenomas peaks between 31 and 36 months of age and the risk of occurrence decreases after 37 months. In contrast, he also reported that there is no age where the incidence of adenocarcinomas is higher than any other age. Since the 54 mammary tissue in the rat extends from the axilla to the inguinal region on either side of the ventral midlines, tumors may present anywhere in this area.

4. Digestive A) Gastrointestinal

Morphological changes in the gastrointestinal tract are not a characteristic of diabetes mellitus in man, but gastrointestinal motility problems are not infrequent. Decreased motor activities have been demonstrated in the esophagus, stomach, small intestine, and colon (Battle et al., 1980), but these findings are usually present only in diabetics with severe peripheral neuropathy (Battle et al., 1980; Hoeffel et al., 1980). Although gastric retention is seen in a small percentage of diabetics (Hoeffel et al.,

1980), constipation is a more common complaint (Battle et al., 1980). There is some evidence that the gastro­ intestinal complications of diabetes mellitus can be improved with meticulus metabolic control (White et al., 1981).

We observed a low frequency of diabetic rats with severe constipation, distended abdomens, and anorexia. These rats were treated with warm enemas which provided temporary relief but these rats invariably died of this disorder. Several of these were due to bowel obstruction caused by bowel necrosis or prostatitis, but two of these 55 had no physical obstruction and were termed "idiopathic megacolon" (Figure 4.15). Both of these rats were about one year old and had been diabetic in excess of 200 days. We examined the rectum and colon of these rats with multiple histological sections to rule out aganglionic megacolon. In all instances, the ganglion cells of the myenteric plexus were present. Since the distended abdomen did not appear soon after birth, it was obvious that these ganglion cells were originally functional. Therefore, a possible explanation was that they were involved by autonomic neuropathy (Yagihashi and Simi, 1985a; Yagihashi and Sima, 1985b). Our original finding has since been confirmed by Meehan et al. (1994); they prospectively examined a larger number of cases and found that these lesions also occur in BBWnd rats. This suggests that autonomic neuropathy is not responsible and they have postulated an autoimmune hypothesis (Meehan et al., 1994).

By far the most common lesion in the gastrointestinal tract of BBW rats was gastric erosions and ulcerations (Wright et al., 1981). Gastric erosions were observed in 32.1% of the BBWd and 9.7% of the BBWnd rats. None were seen in W rats. The stomach of the rat has three subdivisions: The forestomach, the fundus and the antrum.

The forestomach is lined by keratinized stratified squamous epithelium and is separated from the fundus by a transverse ridge. Both the antrum and the fundus are composed of 56 glandular epithelium. The former secretes mucus and the latter secretes acid. This anatomical distinction is important when examining gastric erosions and ulcerations. Spontaneous acute ulcerations in the forestomach are a common finding in some strains of rats (Burek, 1978; Berg, 1967; Muraoka et al., 1977), but apparently are not seen in others (Coleman et al., 1977; Berg, 1967). Forestomach 0 lesions may be focal or multifocal and are frequently associated with hemorrhage into the lumen. Histologically, many of these are inflamed and edematous. Some of these may be so severe that the ulcers penetrated the entire stomach wall resulting in peritonitis (Burek, 1978). Although stress ulcerations of the glandular portions of the stomach may be induced in rats experimentally (Glavin, 1980) they occur spontaneously in rats very rarely, if ever. Except for our report in the BB Wistar rat (Wright et al., 1981, appendix E), I was unable to discover a single reported case in routine rat necropsy studies. Although most investigators fail to mention findings that are absent,

Burek (1978) reported that none were observed in his studies. Several of our findings are of interest: (1) glandular erosions (Figures E.l and E.2) predominated; (2) forestomach erosions (Figures 4.16 and 4.17) were very uncommon; (3) forestomach erosions were always associated with glandular lesions. The morphological and etiological features of these lesions in BBW rats have been described 57 previously (see Appendix E, Wright et al., 1981). Interestingly, our study prompted a study by Nishimura et al. (1983) which demonstrated that a high incidence of spontaneous gastric erosions also occur in two other spontaneous models of diabetes, the NOD mouse and the KK-A mouse. In both strains, erosions occurred in both diabetic and non-diabetic mice. No erosions were seen in other mouse strains not predisposed to diabetes that were housed in the same facility. We have no explanation for our findings or those of Nishimura et al. (1983).

Tumors of the gastrointestinal tract are so rare that most reports are single cases (Altman and Goodman, 1979). A total of 22 cases of various types of tumors of the small intestine are in the literature (Altman and Goodman, 1979) and tumors of the colon are equally rare, except for one totally unexplainable short duration epidemic of spontaneous colon adenocarcinomas in AS rats (Heslop, 1969). The near absence of colon cancer in rats is in total contrast to the very high incidence in man (Willis, 1935). The only GI tract neoplasms observed in our study were lymphomatous infiltrates present in the colon and distal small bowel of a few of the BBW rats. These will be discussed in the section on the lymphoid system.

With the exception of the hemorrhaging ulcers, it is very questionable whether any of these GI tract lesions are very significant, life-threatening maintenance problems. 58 B) Liver The relationship between liver disease and diabetes mellitus has been extensively studied, but definite answers are lacking. It is now believed that in some cases liver disease may be induced by diabetes; in others, diabetes by liver disease (i.e. Naunyn's "liver diabetes"); and in still others, both diseases may result from a common disorder (Creutzfeldt et al., 1970). Because of the liver's important role in carbohydrate metabolism, liver diseases would seem a natural consequence of diabetes mellitus, but this has been difficult to prove. In other words, there does not appear to be a specific diabetic "hepatopathy" as in the case with diabetic nephropathy, or retinopathy, or neuropathy. Recently, a lesion called "diabetic hepatitis" has been described in a subset of type II diabetics (Nagore and Scheur, 1988). This lesion resembles alcoholic hepatitis. The constant findings are fatty change, glycogen vacuolization of hepatocyte nuclei, and hepatocyte ballooning. Inflammation, acute or chronic, is variably present as are Mallory bodies and fibrosis. This trend is also seen in diabetic children (Lorenz and Barenwald, 1979).

A high frequency of gallstones has been reported in diabetics at autopsy, but it is now apparent that this is not true (Creutzfeldt et al., 1970). The simultaneous occurrence of diabetes and cholelithiasis can be explained simply on the basis of common obesity. 59 The relationship between diabetes and steatosis is difficult to assess (Creutzfeldt et al., 1970; Stone and van Thiel, 1985). Liver biopsy studies and autopsy studies have reported incidences of 21%-78% and 20%-30% respectively. The true incidence of fatty liver in diabetes is unknown because liver biopsies are performed selectively on those with hepatomegaly or abnormal bromsulphalein retention and also because autopsy statistics are not representative due to the complications of the terminal diseases. Furthermore, the true incidence of fatty liver in the "normal" population is difficult to estimate, since these people are unlikely to be biopsied. These studies are further complicated by the obesity factor, since approximately 50% of obese persons have fatty livers. It is therefore difficult to conclude that steatosis in obese diabetics is a result of diabetes (Creutzfeldt et al., 1970). Still, a significant, positive correlation exists between age of onset and fatty liver. In

IDDM, the incidence of fatty degeneration is low and appears to be controlled with insulin treatment. No such correlation with good blood sugar control was observable in type II diabetics (Creutzfeldt et al., 1970). The fatty change is thought to be due to an imbalance between the level of triglyceride synthesis and the ability of the liver to secrete triglycerides in VLDL particles (Stone and van Thiel, 1985). 60

Lipid overloaded Kupffer cells appear to be common in diabetes, but, once again, this condition is not specific for diabetes. In these instances the whole reticulo­ endothelial system may be likewise involved (Cabarrou et al., 1973). The incidence of liver cirrhosis appears to be significantly elevated in diabetes mellitus (Creutzfeldt et al., 1970; Stone and van Thiel, 1985). Autopsy studies have cited this incidence to be between 5.7% and 21.4% with male diabetics more frequently afflicted than females. Clinical studies indicate that the onset of diabetes is frequently preceded by cirrhosis. It is, therefore, unlikely that diabetes has a causative role in the pathogenesis of cirrhosis, and the nature of this relationship is presently unknown.

The diabetic liver is rich in glycogen, but the amount of hepatocyte glycogen does not normally correlate with fasting blood glucose, the type of diabetes, the degree of ketosis, or the fat content of the cells. Nuclear glycogen deposits are found in 60-75% of the diabetic population, but this change is not specific for diabetes mellitus (Chips and

Duff, 1942; Creutzfeldt et al., 1970). There is some evidence of a reciprocal relationship between cytoplasmic and nuclear glycogen in diabetes, but this is debated.

Although both mode of development and the significance of nuclear glycogen is presently unknown, nuclear glycogen does 61 not occur in normal subjects.

The incidence and types of hepatic lesions in rats vary greatly depending on the age and strain (Anver and Cohen,

1979). Because of the wide-spread utilization of rats in drug and chemical studies, the naturally-occurring hepatic pathology of rats has been extremely well-characterized. In

fact, a workshop was held to standardize the classification of specific hepatocellular lesions in rats (Squire and Levitt, 1975). Cholangiofibrosis, mucopolysaccharide- containing cysts, and telangiectasis are the most frequently observed age-associated non-neoplastic hepatic lesions

(Anver and Cohen, 1979).

Non-neoplastic hepatic lesions were relatively uncommon in our study (Table 4.1) and presented a largely random, non-strain related pattern. Hepatic fatty change, although uncommon, was somewhat more frequent in BBWd rats than in BBWnd and W rats (Figure 4.18). Since the BBWd rat is a non-obese model for IDDM, a low incidence of fatty change is consistent with findings in human diabetics. Mild periportal inflammation (data not shown) and hepatic infarcts were occasionally present in all three groups of rats. The latter finding is surprising in light of the dual blood supply of the liver. Other isolated findings included an abscess, bile duct hyperplasia, and centrilobular necrosis. Similar findings have since been reported by Bernuau et al. (1985). These investigators also performed 62 morphometric studies demonstrating perisinusoidal fibrosis that was strain-specific but not diabetes-specific. Probably the most interesting hepatic lesion in this study was a rare developmental abnormality, a supradiaphragmatic accessory lobe of the liver (Wright et al., 1983b, appendix F), that has been reported in only one other strain, the Gunn rat (Machado and Lozzio, 1972). Supradiaphragmatic accessory livers were present in two very closely related BBW rats, one a diabetic male and the other a nondiabetic female. Since this abnormality has been found in only two strains of inbred rats, the pattern of occurrence suggests a genetic mode of inheritance that is either polygeneic or autosomal recessive with low penetrance. The gross (Figure F.l) and histologic (Figures F.2, F.3, and F.4) appearances of these lesions are shown in appendix F (Wright et al., 1983b).

No hepatic neoplasms were observed in this study.

Evaluation of the literature pertaining to hepatic neoplasms in rats is difficult due to past confusion and ambiguous terminology. The term "hepatoma" does not indicate whether the tumor is benign or malignant. Some sources believe that benign hepatic tumors do not exist in rats (Squire and Levitt, 1975). Another frequently used confusing term is the "hyperplastic nodule", a discrete circumscribed, nodule of proliferating hepatocytes which compress the surrounding parenchyma. These are believed to be preneoplastic. 63 Spontaneous hepatic sarcomas have frequently been reported in the past, but these are now believed to occur only in rat colonies infected with taenia taeniformis, the cat tapeworm (Altman and Goodman, 1979). This parasite is known to induce hepatic sarcomas in rats. Therefore, sarcomas in rats should not be considered spontaneous. On the other hand, hepatocellular carcinomas do occur spontaneously with a low incidence.

6. Urinary a) Bladder Urinary bladder infections and bacteriuria are common in diabetes, particularly in diabetic women. The prevalence of bacteriuria in diabetic women has been reported to be 11-

18%. This is approximately two to three times the rate in nondiabetic women (Schoenbaum, 1979). Although a number of factors are involved, the occurrence of urinary bladder infections appears to correlate best with the degree of microangiopathy. Other risk factors included neurogenic bladder, prior instrumentation of the urinary tract, and duration of diabetes for greater than 20 years (Schoenbaum, 1979). Warren et al. (1966), reported "readily recognized" cystitis in 52 or 351 diabetic autopsies.

Urinary bladder pathology was rarely seen in our study. Only one rat, a female BBWnd rat of unknown age, developed cystitis. Calculi of the bladder were infrequently seen in 64 BBW rats and did not result in infection. Urolithiasis has been reported to be common in several strains of rats and to be rare in others. These may be found in the renal pelvis, ureters, or bladder (Magnusson and Ramsay, 1971; Anver and Cohen, 1979) and can cause stasis, dilation and infection.

Other non-neoplastic lesions of the bladder appear to be uncommon in rats. Neoplastic lesions are also rare and have been reviewed by Altman and Goodman (1979). b) Kidney One of the most severe clinical sequelae of IDDM is diabetic nephropathy. In fact, renal disease accounts for the deaths of approximately one half of all diabetics that die while under the age of 20 (Westberg, 1980). Presently, diabetic glomerulopathy is the leading cause of renal insufficiency in the United States (Stout et al., 1994). A multitude of different forms of renal disease are more frequent in diabetics than in the general population, but most of these are infections that are secondary to involvement of the bladder by autonomic neuropathy. Diabetic glomerulosclerosis is the only specific form of renal disease seen in IDDM. Not only is it specific for diabetes, its incidence increases with the severity of the carbohydrate intolerance. Diabetic glomerulosclerosis is by far the most important form of renal disease seen in diabetic patients. 65 There are essentially two basic changes present in all forms of diabetic glomerulosclerosis, basement membrane (BM) thickening and mesangial thickening (Mauer et al., 1981). There is considerable evidence that both of these changes may be due to non-enzymatic glycosylation (Brownlee, 1992). Osterby (1975) has shown by morphometric analysis of electron micrographs that both BM thickening and mesangial thickening are absent in young patients shortly after onset of IDDM, but are apparent within two years and are prominent within five years of onset. Morphologically, diabetic glomerulosclerosis may take two distinct courses, benign or accelerated. The benign course consists of concurrent thickening of the capillary basement membrane and diffuse glomerulosclerosis (i.e. - mesangial thickening without nodules). This may progress slowly over many years but rarely leads to renal failure. There is also an accelerated course which is superimposed on the changes seen in the benign course. This is a rapid process that frequently leads to renal failure (Bloodworth, 1978). The natural history of the accelerated lesions has been extensively studied by Bloodworth (1978). First, the changes seen in benign diabetic glomerulosclerosis (i.e. capillary BM and mesangial thickening) occur. After 5-10 years of diabetes (regardless of age), renal arterio- and arteriolar sclerosis are present with some resulting ischemia. Glomerular capillary microaneurysms, sometimes 66 occupying as much as one-third of the diameter of a glomerular tuft, develop. The mesangium of the lobule from which the microaneurysm originated proliferates rapidly and pushes laterally into the microaneurysm forming a nodule. These nodules are originally quite cellular and contain fine fibers and osmiophilic granules between cells but they eventually develop into hyalinized masses (Kimmelstiel- Wilson nodules) that are negative to amyloid stains and weakly positive with eosin and PAS. Silver stains frequently show a laminated pattern. Kimmelstiel-Wilson nodules are considered pathognomic for diabetes but are found in only 9-25% of diabetic patients. Large microaneurysms of Kimmelstiel-Wilson nodules are often found adjacent to glomerulo-capsular adhesions, areas of fibrosis binding the peripheral glomerular capillaries to the adjacent parietal layer of Bowman's capsule. Usually both parietal and visceral epithelium are absent in these adhesions. Bloodworth suggests that these are due to large structures (e.g. Kimmelsteil-Wilson nodules) pushing the glomerular tuft against the capsule. Although few diabetic kidneys reach this stage, obliterative diabetic glomerulosclerosis is the endstage lesion in this process and is believed to be a result of ischemia due to a combination of pressure from excessive deposition of BM and due to arterio- and arteriolar sclerosis of renal vessels.

These glomeruli eventually degenerate to a compact tangle of 67 BM material. This has been described as the "large hyaline ball" change.

Diabetes is also characterized by a series of insudative lesions (i.e., intramural accumulations of plasma proteins and lipids) which can be subdivided by location (Stout et al., 1994). Insudative lesions within Bowman's capsule are called capsular drops. Within peripheral glomerular capillaries, insudative lesions are called fibrin caps. Stout et al. also suggest that insudation within afferent and efferent arterioles is responsible for the hyalinization of these structures that is common in diabetics. However, none of these lesions is 100% specific for diabetes (Stout et al., 1994).

Inflammatory lesions are also frequent in the diabetic kidney. In the later stages of diabetes, autonomic peripheral neuropathy often leads to bladder dysfunction and urinary stasis. This may result in urinary tract infections. Acute febrile pyelonephritis and glomerulonephritis are common complications of diabetes. Therefore, when a diabetic develops signs of renal disease earlier than expected, some type of kidney infection should be suspected (Westberg, 1980). Surprisingly, there is little evidence that one form of renal infection, chronic interstitial nephritis, has a higher incidence in diabetics (Westberg, 1980). 68 Renal lesions were very uncommon in our study. Cysts containing proteinaceous fluid were occasionally seen, but these were non-specific findings and were not tabulated. Infrequently, the formation of these cysts from dilated tubules was observed to be associated with focal lymphocytic inflammation (Figure 4.19) and acute pyelonephritis.

According to Burek (1978), this is typical of the chronic progressive glomerulonephropathy characteristic of the rat. Chronic renal disease is a major cause of morbidity and mortality in many strains of rats (Anver and Cohen, 1979). These changes appear to be unrelated to infection because chronic renal disease is also seen in specific pathogen free and germ free rats (Anver and Cohen, 1979). In some strains, this appears to be related to nutritional factors (Bras and Ross, 1964; Snell, 1967; Anver and Cohen, 1979). The morphology of chronic renal disease in rats is well established (Anver and Cohen, 1979; Berg, 1967; Hirokawa, 1975). The initial lesion occurs at 3-6 months of age and consists of an eosinophilic PAS-positive thickening of the glomerular mesangial matrix. At first, only scattered glomeruli are affected. End stage changes may be seen in rats as young as one year old. Glomerular changes include basement membrane thickening in the capillary tufts and

Bowman's capsule, eosinophilic deposits, adhesions between the parietal and visceral layers of Bowman's capsule, and partial or complete hyalinization or sclerosis. Many of the 69 tubules are dilated or cystic with flattened epithelium and proteinaceous casts in the lumens while others are collapsed and atrophic. The cortical interstitium is fibrotic and infiltrated by mononuclear cells. Protein excretion increases from 5 mg/dl in young rats to 20 mg/dl in most rats over 2 years of age.

Immunofluorescence studies have shown that immunoglobulins are present in the glomeruli. Couser and Stilmant (1976) reported that IgM was bound in the mesangium of 50% of glomeruli of 3-6 month old Sprague-Dawley rats. IgG deposits were minimal. Hirokawa (1975) reported a progressive increase with age in IgG deposition in the glomerular basement membrane and mesangium. Circulating antibodies to renal or other tissue antigens are not present in serum from aged rats with extensive glomerulosclerosis (Couser and Stilmant, 1976). Both authors agree that glomerulosclerosis in rats is not an age-related autoimmune phenomenon and that antigens within the immune complex are probably exogenous in origin. Inflammatory lesions were also uncommon in our study.

Small focal granulomas, present in only five BBWd rats, were the most common inflammatory finding in the kidney (Figure 4.20). Incidental findings included chronic interstitial nephritis (Figure 4.21) in one BBWd and focal acute pyelonephritis in two BBWnd rats. 70 Although, nephroblastomas, adenomas, adenocarcinomas, mixed tumors, lipomas, sarcomas and hamartomas have been reported in rats, no renal neoplasms were seen in our study.

Nephroblastomas are the most common, but renal neoplasms are rare in most strains of rats (Altman and Goodman, 1979). Although the kidneys of rat and man are embryologically and histologically very similar, they are physiologically very different (Snell, 1967). For instance, man may rely on sweat glands to remove a significant percentage of his body

fluid; rats have no such glands. Furthermore, the concentrating ability of the rat kidney far exceeds that of man. Man frequently produces renal amyloid, but this is not found in the rat (Snell, 1967). Therefore, it seems a bit tenuous to expect great similarity in the patterns of human and rat renal pathology. This appears to be the case, since no lesions resembling human diabetic nephropathy were seen in BBWd rats. It is also possible that none of the rats in this study lived long enough to develop renal manifestations of diabetes.

7. Respiratory (Lungs)

One of the very early historical observations in the study of diabetes was that diabetics were more susceptible to pulmonary infection. In fact, one older study reported a

50% positive rate for tuberculosis in autopsied diabetics (Root, 1934). Statistical studies of past influenza 71 epidemics with high mortality indicate that diabetics are at a higher risk of complications and death following influenza infection (Schoenbaum, 1979). Other types of pulmonary infections are also frequently seen in diabetics. Pneumonia has been recently reported as the second leading infectious cause for hospitalization of diabetic patients (Whitehouse, 1973). On first glance, this appears to be particularly interesting in light of the high incidence of bronchopneumonia in the BB Wistar diabetic rats, but only if one fails to remember that the BBW nondiabetic siblings also have a similar incidence of bronchopneumonia (Table 4.2). It is also important to consider the absence of bronchopneumonia in any of the standard Wistar rats. Since all animals were housed in the same room under similar conditions, one would expect a similar incidence of bronchopneumonia in all three groups. Since this was not observed, it is obvious that the BBW strain is more susceptible to pulmonary infection and that this is unrelated to the diabetic state. This observation, along with others that are mentioned elsewhere in the text, suggest that some form of immunosuppression may play an etiological role in the many afflictions of the BB Wistar rat. It has been suggested that most laboratory animals have an organ system that is so frequently diseased that it seriously restricts the usefulness of the species for 72 research purposes. In the rat this is the respiratory system and the disorder is murine chronic respiratory disease (CRD) (Lindsey et al., 1971). Murine CRD is a serious, contagious syndrome that is ubiquitous. A British study reported a high incidence of CRD in commercially available specific pathogen free rats which were cesarean derived and maintained under strict isolation (Lamb, 1975). Lamb histologically examined the lungs of 200 rats from nine major breeders and found that only two of the nine sources had rats with consistently normal lungs (Lamb, 1975). The natural history of CRD in the rat is reasonably well defined. Newborn rats are almost always normal until weaning (approximately three weeks). A few of the animals in an affected colony show minimal encrustations around the external nares and eyes by 4-8 weeks. These animals make abnormal "snuffling" sounds which are indicative of catarrhal rhinitus. At the same time, many animals develop a suppurative otitis media that is clinically inapparent.

Some of these animals develop lesions characterized by varying degrees of peribronchial infiltration of lymphoid cells and bronchiectasis with squamous metaplasia of the bronchial epithelium due to mucous plugs. This stage is the most severe seen in W rats in our study. BBW rats often had more severe disease with one or more lobes containing massive multifocal abscesses (Figure 4.22). Mortality rate for CRD is usually low in younger rats, but is much higher 73 in rats over 2 years old. The incidence of CRD may be as high as 50 to 100% (Lindsey et a l ., 1971). Rats from all three groups (BBWd, BBWnd, and W) had some degree of respiratory disease. The diagnosis of bronchopneumonia was made only when neutrophils could be demonstrated within alveoli (Figure 4.23). When neutrophils were seen only in bronchi, acute bronchitis was diagnosed.

Since CRD is ubiquitous and, therefore, essentially "normal" in the rat, this diagnosis was not tabulated, but Figure

4.24 shows one of the most severe cases seen in a W rat in this study. Since CRD may progress to pneumonia (Lindsey et a l ., 1971), it seems likely that the bronchopneumonia in the BBW rats is simply a more severe reaction of an immuno­ compromised strain to microorganism in their environment in spite of the semi-barrier housing facilities. The etiology of CRD is uncertain and the following agents have been suggested: bacteria, conventional viruses, slow virus. Mycoplasma pulmonis, and various combinations of these agents. At present, the case for M;. pulmonis is strongest (Lindsey et a l ., 1971). Lung specimens from rats with grossly apparent pneumonia were sent to the Microbiology Division of the

Pathology Department of University Hospital for examination for bacteria, fungus, and/or mycoplasma. The results of these examinations were most frequently negative or believed to be due to postmortem growth. A complete listing of 74 microbiological findings is included in Table 4.3. Although Mycoplasma was never demonstrated, there is strong circumstantial evidence implicating this is in the pulmonary lesions in the BB Wistar rat. First, tetracycline is effective in suppressing these symptoms in the BBW rat as has been reported in standard rats experimentally infected with Mycoplasma pulmonis (Lindsey et al., 1971). Secondly, the location of pulmonary lesions following gross examination is similar in both of these groups. The anterior lobe of the left lung and the cranial and azygous lobes of the right lung were the most frequent sites of infection. Small granulomatous lesions were seen in two BBW rats (Table 4.2), but both were negative for organisms. In addition to these, multifocal histiocytosis was infrequently observed in the lungs of BBW rats. Grossly, these appeared as small brownish discrete or confluent nodules. Histologically, the lesions consisted of focal accumulations of foamy macrophages in expanded alveolar spaces. These findings are of limited significance and are frequently observed in other strains of aged laboratory rats (Yang et al., 1966). In summary, BBW rats are very susceptible to a variety of pulmonary disorders, the most severe of which is bronchopneumonia. These infections may occur in epidemics and could exterminate whole colonies of rats if extreme care 75 is not exercised.

8. Circulatory a) Heart In affluent societies, almost 50% of diabetics die of heart disease (Asmal et al., 1980; Pyorala, 1989). The risk of myocardial infarction is estimated to be 2 1/2 times greater than that of non-diabetics (Asmal et al., 1980) and to be even higher in obese female diabetics (Sinclair-Smith,

1979). The Framingham study found that diabetic men have twice the frequency of congestive heart disease as non­ diabetic men and that diabetic women have a five-fold increased risk (Kannel et al., 1974). The International Atherosclerosis Project, an autopsy study of coronary arteries from over a dozen countries, also found a marked increase in the extent to atherosclerotic lesions in diabetics compared to control patients (Pyorla, 1989).

Although there was a slight increase in fatty streaks, the increase in raised lesions was much more dramatic. Until relatively recently, the increased incidence of heart disease in diabetics was considered to be entirely a manifestation of accelerated atherogenesis, but this does not seem to be the case. Several studies utilizing non- invasive techniques for studying left ventricular function have shown a characteristic pattern of shortened left ventricular ejection time and a prolonged pre-ejection 76 period in diabetic patients. Some of these patients had normal coronary arteriograms. These studies and others suggest that the heart muscle is abnormally stiff in diabetics (Ledet et al., 1979). Similar results have been obtained using alloxan diabetic dogs (Regan et al., 1974). Furthermore, histological studies have suggested that microangiopathy alone is not responsible for diabetic heart disease (Sinclair-Smith, 1979). One study (Ledet, 1976) reported only minor endothelial proliferation and increased PAS material in arterioles, but no changes in capillaries. Present evidence suggests that diabetic cardiomyopathy is a multifactorial disorder. Vascular disease is probably a major factor, but other biochemical and metabolic abnormalities are also important (Sinclair-Smith, 1979; Ledet et al., 1979).

Most cardiac lesions in rats are age-associated (Wilens and Sproul, 1938). In fact, heart lesions are seldom seen in rats under 12 months of age but are common in rats older than 18 months. The most frequent lesion is myocardial degeneration with fibrosis. This lesion is not usually observable grossly, but may be quite extensive in histological sections. The microscopic appearance is myocardial atrophy, degeneration, and necrosis with condensation fibrosis of the stroma and an inflammatory infiltrate consisting of lymphocytes, Antischkow cells, and macrophages (Figure 4.25). The most frequent site of 77 involvement is the papillary muscles and their attachment sites in the wall of the left ventricle (Burek, 1976; Anver and Cohen, 1979). Maximum incidence may be as high as 60- 80%. Cardiac performance is not impaired (Anver and Cohen,

1979). Table 4.4 shows that low incidences of these lesions were observed in the older age groups in all three groups of rats in our study. Endocardial and subendocardial proliferative lesions have also been reported in several strains of rats including Wistar rats (Anver and Cohen, 1979). This condition is most common in rats between 25 and 30 months of age. Early lesions consist of subendothelial, undifferentiated, mesenchymal cells and scattered lymphocytes. Advanced lesions had subendocardial accumulations of fibroblasts and collagen. The left ventricle is most commonly affected (Boorman et al., 1973). One lesion of this type was observed in a 198 day old female BBWd rat which also had a lymphoma. Other lesions occasionally found in the hearts of aged rats include intracardiac thrombi and valvular endocarditis (Anver and Cohen, 1979).

Acute myocarditis was observed in two BBW rats. One was in response to a focal fungal infection at the site of an indwelling catheter. The etiology of the other acute myocarditis is not known. 78 b) Vasculature

Atherosclerosis is presently responsible for the death of approximately three-quarters of American diabetics, but accounted for only 16-23% of diabetic deaths in the pre­ insulin era (Warren et al., 1966). Although some of this difference can be explained by the former tendency of insulin-dependent diabetics to die of ketoacidosis and emaciation, the vast majority of diabetics, then and now, are non-insulin-dependent and, therefore, not susceptible to ketoacidosis. Therefore, other explanations obviously must be invoked. Approximately 75% of diabetics less than 40 years of age have clinically significant atherosclerosis (<5% in non­ diabetic controls) and essentially all diabetics, those with IDDM or NIDDM, have clinically significant atherosclerosis within ten years of onset (Bierman, 1992).

The mechanism(s) by which diabetes mellitus influences atherogenesis is not fully understood (Bierman, 1992; Schwartz et al., 1992; Ross and Agins, 1992). It is well established that platelet aggregation is enhanced in diabetes. In addition to clotting abnormalities, evidence suggests that diabetic serum contains elevated levels of various factors (e.g. - growth hormone, insulin) that enhance the growth of arterial smooth muscle cells. It is well-known that non-insulin-dependent diabetics may secrete more insulin as a result of their insulin resistance, even 79 in the absence of obesity. This is particularly interesting in light of the evidence suggesting that hyperinsulinemia is a risk factor in diabetic atherogenesis. Other factors tending to increase the incidence of atherosclerosis in type II diabetics include obesity, hypertension and

hyperlipidemia. Finally, non-enzymatic glycosylation of low density lipoproteins (LDL) appears to promote its affinity to LDL receptors while non-enzymatic glycosylation of high density lipoproteins (HDL) promotes its degradation. Therefore, the athero-protective HDL is lost while the effects of athero-promoting LDL is enhanced. All of these factors combine to promote atherogenesis in diabetes (Bierman, 1992; Schwartz et al., 1992). According to Warren et al. (1966), the most frequent sites of fatal atherosclerotic lesions in diabetics are the heart (78.7%), brain (10.6%), kidneys (6.4%), and extremities (3.2%). When these values were compared to their own nondiabetic autopsy series, only the prevalence of lethal cardiac and peripheral artery atheromas are significantly increased. They also noted a tendency in diabetics for more severe involvement of muscular arteries rather than the larger elastic arteries. Several roentgenographic studies have observed that Monckeberg's sclerosis (medial calcification) is more common in diabetics. Ferrier (1964) observed two to three-fold increases in the prevalence of medial arterial calcification 80 in the knees (20.4%) and feet (23.6%) of diabetics when compared to nondiabetics. Medial calcification also occurs at a much younger age in diabetics. In fact, one study reported an 18.3% incidence of medial calcification of leg arteries in diabetic children (Warren et al., 1966). Fortunately, medial calcification is rarely associated with any clinical consequences. Diabetic microangiopathy is another devastating vascular complication of diabetes (Laffel and Krolewski, 1989). It is characterized by diffuse basement membrane (BM) thickening. It involves capillaries in most organs but may be most symptomatic when skin, retina, or renal glomeruli are significantly involved. It may also involve non-vascular structures such as renal tubules, Bowman's capsule, peripheral nerves, seminiferous tubules, and placenta villi. The BM thickening can be best demonstrated at the light microscopic level as a hyaline-like material after periodic acid-Schiff (PAS) staining. Even though the

BM is thickened, the involved capillaries tend to leak plasma proteins. Advanced glycosylation end-products play an important role in the pathogenesis of microangiopathy (Brownlee, 1992). Unfortunately, no atheromatous lesions were observed in the thoracic aorta, coronary arteries, cerebral arteries or any other visceral arteries of any of the rats in our study.

This is not surprising since atherosclerosis has been only 81 infrequently observed in autopsy studies on senile rats. This is presumably, in part, because HDL, which has antiatherogenic properties, is the major lipoprotein fraction in the rat, whereas LDL, which is atherogenic, predominates in man. Several investigators have extensively studied atherosclerosis in the rat. Wilens and Sproul (1938) observed medial calcification and necrosis in the aorta, iliac, coronary, renal, cerebral, and other visceral arteries of Mendel-Sherman rats. Berg (1967) has also occasionally found medial calcification in the aorta and iliac arteries of senile rats. Anver and Cohen (1978) suggest that aortic lesions in rats usually resemble

Monckeberg's medial sclerosis rather than atherosclerosis. On the other hand, coronary atherosclerosis, with both medial and intimal lesions, has been reported in various strains of aged rats that have been repeatedly bred (Wexler, 1964). Medial lesions include smooth muscle edema and hypertrophy (sometimes causing lumen stenosis), basophilic degeneration, and mucopolysaccharide vacuolization; intimal lesions are internal elastic membrane disruption, endothelial hyperplasia, and subintimal mucopolysaccharide accumulations (Wexler, 1964). Carotid and cerebral arteriosclerosis has also been reported in senile, repeatedly bred Sprague-Dawley rats (Wexler and True, 1963).

One interesting observation that resulted from the preceding 82 studies is that islet hyperplasia (Wexler and Fischer, 1963a) and glucose intolerance (Wexler and Fischer, 1963b) frequently accompany arteriosclerosis in breeder rats. The most common vascular pathology seen in senile rats is polyarteritis nodosa (PAN). Frequencies between zero and

60% have been reported in various rat strains (Berg, 1967). The incidence in Wistar rats is reported to be extremely low (Anver and Cohen, 1978). PAN is a vasculitis of unknown etiology that involves muscular arteries. A single artery may have lesions varying from acute to chronic. Prominent histologic features are degeneration of the media and adventitia, fibrinoid necrosis, marked infiltration of PMNs and mononculear cells, and disruption of the internal elastic lamina. Pancreatic, mesenteric, and spermatic arteries are most frequently involved (Anver and Cohen,

1978). Acute vasculitis and/or polyarteritis nodosa were seen infrequently in BBWd and BBWnd rats (Figures 4.26,

4.27, and 4.28), but were never seen in W rats. In one pregnant rat this was also associated with disseminated intravascular coagulation (Figure 4.29). Although morphometric studies were not performed, the walls of the muscular arteries in many of the BBW rats appeared thicker than the walls of arteries with similar lumen size in W rats. This may suggest that BBW rats are somewhat hypertensive. 83 No lesion suggestive of microangiopathy was seen in any of the rats. PAS staining for basement membrane was not routinely performed.

9. Hematopoietic/Lymphatic a ) Thymus Since the insulitis associated with the onset of IDDM is probably a cell-mediated immune response, it is likely that the thymus is involved in the initiation of this process. In fact, Like et al., (1982) has reported that neither insulitis nor diabetes develops in BBW rats that have had neonatal thymectomies. Nonetheless, thymic lesions are not reported in autopsy studies of juvenile onset diabetics. I was able to find only one study specifically examining the thymus. Souadjian et al. (1970) reported a significant increase in the number of Hassalls' corpuscles and in the diameter of the largest Hassall's corpuscle as well as a decrease in the number of epithelial cells per unit area when comparing the thymuses of juvenile-onset diabetics to those of age-matched controls. The significance of this finding is unknown but it might be more rewarding to examine thymic function rather than morphology in such cases.

Atrophy was the only lesion of thymus observed in our study. It is a universal finding in all aged rats (Burek and Meihuizen, 1977). In fact, the atrophy is so severe that the thymus is very difficult to locate. Thymic sections were prepared by embedding the whole superior mediastinum after division into three pieces. With this procedure, we were able to examine the thymus histologically in one-third to one-half of the rats in our study. No other non-neoplastic lesions of thymus are known to occur in rats.

Various strains of aged rats develop a low incidence of thymomas, but tumors of other types are very rare (Altman and Goodman, 1979). No primary thymic neoplasias were observed in our study. b) Lymph Node Although lymph node lesions have not been reported to be more frequent in the human diabetic population than in nondiabetics, a variety of strain specific (but unrelated to diabetes) lymphadenopathies were nearly ubiquitous in our BBW rats. This is in stark contrast to the nearly homogenous, normal appearance of the lymph nodes in our W rat population. Rats have multiple mesenteric lymph nodes at the junction of the cecum and the colon. In W rats these were small (2-4 mm diameter) and distinct, while in BBW rats they were usually fused into one or more longitudinal nodes, some of which were very large (Figure 4.30). Although there was no uniform histological appearance in these longitudinal nodes, most had some form of sinusoidal hyperplasia. Three distinct cell populations predominated in these nodes. In 85 some, sinusoids were massively dilated with plasma cells and differed from plasmacytomas only by the presence of

occasional cortical follicles. Others were consistent with the diagnosis of sinus histiocytosis. Still other rats had nodes with mixed cell types in the dilated sinusoids. Regardless of the type of sinusoidal hyperplasia, germinal centers were rarely seen in the cortical follicles. Lymph nodes from the W rats in our study had distinct follicles, frequently with germinal centers (Figure 4.31), and never showed the sinusoidal hyperplasia typical of BBW rats (Figure 4.32).

Granulomatous lesions were not uncommon (3-5%) in BBW rats (Table 4.5), but were never seen in W rats. Grossly, these nodes were white, non-hemorrhagic, firm, and frequently fused together. In no case could organisms be demonstrated in these granulomas (Figure 4.33).

Although aged rats are less immunocompetent than younger rats, age-related morphological changes within the lymph nodes are usually minimal. Lymph nodes from aged rats are grossly normal (Pollard and Kajima, 1970); the general histological trend with age is toward a "less active" appearance with a decrease in the number of germinal centers per node. Frequently the medullary cords are densely packed with plasma cells that may extend into the paracortical region. Hemosiderin-containing macrophages are often present in the sinuses (Burek, 1978). This appears to be a 86 physiologic change associated with aging rather than a pathologic change. The only common lesions in the nodes of aged rats are cystic changes in the sinusoids (Anver and Cohen, 1979; Burek, 1978). Lymphomas were the predominant neoplastic lesions observed in our study. They were found in a modest percentage (3-4%) of older BBW rats (Table 4.5), but were not seen in W rats. The mesenteric lymph nodes and the colon were the most frequent sites of involvement (Table 4.6). Grossly most were firm, lobulated, white, and severely hemorrhagic (Figure 4.34). Histologically, lymphomas were diffuse (Figure 4.35), and were "histiocytic" with plasmacytoid differentiation (Figure 4.36). Colon involvement occasionally caused severe bowel obstruction

(Figures 4.37 and 4.38). In one rat (Figure 4.39) the gross appearance of the lymphomas was that of a focal ulceration of the intestinal mucosa without any associated nodal involvement. Kalant and Seemayer (1979) have previously reported a 12.6% and 1.2% incidence of mesenteric lymphomas in BBWd and BBWnd rats respectively. These tumors were histologically either plasmacytoid or had features of immunoblastic sarcomas. Cytoplasmic immunoglobulin was demonstrated by immunoperoxidase confirming their B-cell origin (Seemayer et al., 1982b; Seemayer et al., 1983). We are not aware of any studies on lymphoproliferative disease in BBW rats since 1983. Using the "Working Formulation for 87 Clinical Usage" classification scheme, these lesions are now best classified as large cell immunoblastic lymphomas. Interestingly, 50% of B-immunoblastic lymphomas in man are associated with a previous history of autoimmune disorders or systemic immunodeficiency (Cotran et al., 1994).

A wide variety of neoplastic lesions have been reported to occur sporadically in aged rats. These include lymphomas, lymphosarcomas, lymphoreticular tumors, reticulum cell sarcomas, reticulosarcomas, and plasmacytomas (Altman and Goodman, 1979). It is difficult to make generalizations about their incidence because the literature describing many of these predates modern systematic description of hematopoietic system tumors. Therefore, tumor classifications are ambiguous (Snell, 1963); some investigators group all primary lymph node tumors as lymphomas. Thompson et al. (1961) found one lymphoma in a series of 125 Sprague-Dawley rats. Ratcliffe (1940) observed only one lymphoma in a series of 468 Wistar rats. Olcot (1950) and Kim et al. (1960) also reported very low incidences of lymphomas. Most large studies have reported none (Davis et al., 1956; McCoy, 1909; Boorman and Hollander, 1973; Burek, 1978; Snell, 1963). Crain (1958) reported a higher incidence of "malignant lymphomas" in his study but his classification system is ambiguous. It seems likely that the rate of spontaneous lymphomas in aged outbred rats is less than 1%. 88 Lymph node neoplasms are apparently most common in 2-3 year old rats (Burek and Hollander, 1977; Boorman and

Hollander, 1973; Ueberberg and Lutzen, 1979) and have not to my knowledge been previously reported in rats less than 17 months of age (Burek and Hollander, 1977). Since the youngest rat mentioned in the literature with a true lymphoma is a 20 month old Sprague-Dawley rat (Thompson et al., 1961), it is interesting that BBW rats develop lymphomas at such a young age. c ) Spleen The incidence of splenic lesions in human diabetics has not been reported in the literature, but there is little reason to expect any significant change from normal. Some older sources claimed that there is an increased lipid content in the spleens of diabetics. If this is true, it is a minimal increase (Warren et al., 1966). Age-related pathological changes in the spleen of the rat are infrequent as well (Burek, 1978). One study

(Andrew, 1946) did report some minor morphological changes in the spleens of old rats including less discrete

Malpighian corpuscles, absence of germinal centers, and a more sinusoidal arrangement of the red pulp with an increase in hemosiderin-filled macrophages. Megakaryocytes may also be more common. This picture was occasionally seen in some of the older BBW rats (Figure 4.40), but was not observed in 89 our W rats. Mild or moderate extramedullary hematopoiesis was the only common splenic finding and was present in greater than one quarter of the rats in this study regardless of strain. Although mild extramedullary hematopoiesis is a very common finding in many strains of rat, it is most frequently severe in tumor-bearing animals (Anver and Cohen, 1979). In our study, BBW rats with lymphomas frequently had severe splenomegaly secondary to extramedullary hematopoiesis and/or tumor infiltration (Figure 4.41). It is interesting that extramedullary hematopoiesis was very rare at any other site in our study.

Altman and Goodwin (1979), in their review of neoplasia in rats, were able to find only two reports of splenic tumors in rats. No primary tumors were found in this study.

10. Nervous System A. Brain

Although untreated hyperglycemia has severe functional effects on the peripheral nervous system, the central nervous system is relatively unaffected. On the other hand, hypoglycemia severely affects the CNS but usually has very little effect on the peripheral nerves (Brierly, 1981). Hypoglycemic brain damage has been seen as a result of insulin shock therapy for psychoses, irreversible coma after treatment of diabetes with insulin, insulinomas, infants of 90 diabetic mothers and idiopathic hypoglycemia in infants. Hypoglycemic brain damage morphologically bears a similarity to hypoxic brain damage. Pyramidal neurons in Sommer's sector of the hippocampi become necrotic and diffuse demyelination of the centrum ovale occasionally is seen. When the cerebellum is involved, the damage is predominantly to the Purkinje cells (Brierly, 1981; McCall, 1992). Although pure hyperglycemia appears to have little adverse effect on the brain, diabetic ketoacidosis can lead to massive diffuse cerebral edema as well as uncal and tonsillar herniation. The mechanism for this typically fatal complication is unknown. Suggested contributing factors include rapid correction of blood glucose and/or hyponatremia, increased CSF pressure due to increased polyol pathway activity in the brain, and cerebral hypoxia (McCall, 1992).

Finally, diabetes predisposes patients to thrombotic

(but not hemorrhagic) cerebrovascular disease. The risk of stroke is two-to-six-fold increased. Diabetes is believed to be responsible for at least 7% of stroke deaths, and cerebrovascular disease is seen in approximately 25% of patients dying with diabetes. The mechanism of this increased risk is unknown but is in part related to the increased risk of atherosclerosis and hypertension in diabetes (McCall, 1992; Standi et al., 1989). 91 CNS lesions were very uncommon in our study, and when they were present were usually iatrogenic in origin. Hypoglycemic brain damage (Agardh et al., 1981) was observed in two BBWd rats as a result of excessive insulin therapy. In both rats, convulsions preceded their sacrifice so hypoxia may have been a factor in addition to hypoglycemia. Both rats showed eosinophilia of the pyramidal neurons in

Sommer's sector of the hippocampus. The Purkinje cells of the cerebellum showed eosinophilia in one of the rats. The most interesting CNS finding was a lesion very similar to central pontine myelinolysis (CPM), a lesion in man characterized by a single focus of demyelination in the pons with relative sparing of nerve cells and axons. CPM is most frequently seen in alcoholism and malnutrition and is now believed to be an iatrogenic disorder due to the rapid correction of hyponatremia (Wright et al., 1979). This is a relatively rare disorder in man and has not been previously induced in experimental animals. However, Kleinschmidt- DeMasters and Norenberg (1982) have shown that the rapid correction of hyponatremia can cause a relatively diffuse demyelination of the brain in the rat.

CPM was observed at necropsy in an emaciated 333 day old BBWd rat that had been treated for dehydration by subcutaneous injection of a total of 15 mis of normal saline over a period of one week (Wright et al., 1983c, appendix

G). We can only speculate as to why CPM might occur in a 92 BBW diabetic rat yet has not been produced experimentally or observed incidentally in other animals. First, CPM is more common in chronic nutritionally deficient individuals. Poorly controlled diabetes mellitus is certainly such a condition. The emaciation and dehydration that frequently precedes the death of BBW diabetic rats may set the stage for the development of CPM. Secondly, the diabetic state causes additional osmotic stress. Aloia and Nilakantan (1973) have shown that fluctuations in the cerebral spinal fluid (CSF) sodium level are responsible for osmotic equilibrium between the CSF and the plasma. This prevents the abrupt fall in plasma glucose and osmolality following insulin treatment resulting in a shift of water into the brain (i.e. elevating CSF pressure). We postulate that the combination of these conditions followed by rapid treatment with relatively large doses of saline resulted in CPM in this case. Regardless of the mechanism, CPM has been reported in man several months after the onset of clinical diabetes and concurrent with fluid therapy for dehydration

(Behar et al., 1964). A second probable case although less severe than the first, was seen in another BBWd rat, but treatment sheets were not available for that animal. In addition to these iatrogenic lesions, a single case of acute meningitis was seen in a 289 day old BBWd rat.

Even though the incidence of inflammatory lesions of the ears and nose is high in most rat colonies, reports of 93 meningitis or other CNS infections in rats are relatively rare (Bullock et al., 1968). In fact, reports of central nervous system lesions in most srains of rats are exceedingly rare (Anver and Cohen, 1979), particularly before 18 months of age (Coleman et al., 1977). In the present study, no CNS lesions were observed in BBWnd or W rats. Anver and Cohen (1979) claim that the only clinically significant CNS lesion in rats is degenerative myelopathy secondary to radiculoneuropathy. Although this is common in many strains of aged rats, the etiology is unknown. All other lesions appear to be incidental findings. The most commonly reported age-associated brain lesion in rats is idiopathic vacuolation of the white matter of the thalamic area, pons, midbrain, and cerebellum (Burek, 1978; Anver and Cohen, 1979; Coleman et al., 1977). In fact, Coleman et al. (1977) reported an 80% incidence rate in Fischer 344 rats over 30 months of age.

High frequencies of various pigments have been reported (Burek, 1978). He observed a nearly 100% frequency of neuronal accumulation of intracytoplasmic lipofuscin in rats over 24 months of age. He also frequently found melanosis of the meninges and olfactory lobe of the brain. Melanosis was apparently not age associated as it was seen in both young and old rats. Burek (1978) also reported PAS-positive laminated extracellular basophilic bodies in thalamus or grey matter of the cerebellum of 20% of his rats. In 94 addition to these, age related membranous bodies were reported in the dendrites of cerebral cortical neurons of Sprague-Dawley rats (Anver and Cohen, 1979). Infrequent or single observations of hydrocephalus, atherosclerosis, infarction, focal hemorrhage, focal hemorrhagic necrosis, ischemic necrosis, and multifocal spongiform encephalopathy have been reported (Burek, 1978 and Coleman et al., 1977). A wide variety of brain tumors has been reported in various aged inbred or outbred rat strains, but the incidence of most of these is under 1%. Hollander et al. (1976) reported 12 primary granular cell myoblastomas in various strains of aged inbred rats. Ueberberg and Lutzen reported 9 glioblastomas in rats over 21 months of age and that male rats were more frequently afflicted. Coleman et al. (1977) were unable to find any CNS tumors in 144 male Fischer 344 rats of variable age. Burek (1978) reported a less than 1% incidence of ependymomas and oligodendrogliomas in BBW/Bi rats and of astrocytomas, glioblastoma multiforme, and meningiomas in WAG/Rij rats. All tumors were present in rats between 22 and 41 months of age. He also reported a 2-

7% incidence of granular cell tumors in these rats. Dagle et al. (1979) examined a total of 2,242 aged Wistar, Sprague-Dawley, and Osborne-Mendel rats and reported 1.3%, 1.2% and 1.9% incidences respectively. They reported predominantly astrocytomas but also observed several malignant ependymomas, meningiomas, pinealomas, neoplastic 95 reticulosis, as well as an isolated oligodendroglioma and a gliomatosis. No CNS neoplasms were observed in our study.

B. Peripheral Nerve Peripheral neuropathy is a common sequella of long­ standing diabetes mellitus (Greene et al., 1989; Greene et al., 1992). This entity has been extensively studied in BB Wistar rats (Sima, 1980; Sima and Hay, 1981; Sima and Thibert, 1982; Sima et al., 1982; Sima, 1983; Sima et al., 1983; Sima and Hinton, 1983; Greene et al., 1984; Sima,

1985; Yagihashi and Sima, 1985a; Yagihashi and Sima, 1985b;

Sima et al., 1986; Mendell et al., 1980; Mendell et al., 1981).

11. Other Various other organ systems are also adversely affected by the diabetic state. In this study, gross observations on other organ systems in the BB Wistar rat will be briefly presented. These organ systems include the skin, eyes, and adipose. a) Cutis Approximately 30% of long term human diabetics develop cutaneous lesions as a result of microangiopathy, arteriosclerosis, infection, metabolic disturbances, and/or the sequelae of chronic therapy. Integumentary diseases 96 associated with diabetes mellitus include: benign acanthosis nigricans, Kyrle's disease (hyperkeratosis follicularis et parafollicularis in cutem penetrans), necrobiosis lipoidica diabeticorum, ruberosis, carotenemia, scleredema, granuloma annulare, skin spots, and bullous diabeticum. In addition to these, improper insulin treatment may result in cutaneous disorders. The most frequent of these is insulin lipodystrophy (Huntley, 1993; Perez and Kohn, 1994). Skin disorders were infrequently observed in the BBW rat, but were not routinely examined with histology. A large proportion of the offspring of several litters became bald over the head, shoulders, and thighs. The skin appeared normal but with less than the normal amount of hair. The lesions were too bilaterally symmetrical for a parasitic infection and the pattern was suggestive of a temporary endocrine disorder (R. Bell, personal communication). All rats outgrew the disorder by approximately 90 days of age. Cutaneous ulcerations, seen very infrequently, were the other main integumentary lesions observed in BBW rats.

b) Eyes Occular disorders are some of the most frequent and disasterous complications of human diabetes mellitus. In fact, diabetic retinopathy is considered the leading cause 97 of blindness in adults under age 65 in the United States (Kohner, 1989). In addition to diabetic retinopathy, diabetics also have a higher incidence of cataracts and glaucoma. Occular histology of the BBW rat was not studied, but gross occular abnormalities observed in the BBW rats

included cataracts, microphthalmia, and corneal ulceration. The total incidence was less than 5%. There is also some evidence that BBWd rats develop lesions resembling diabetic retinopathy (Sima et al., 1985), but this was not examined in our study.

c) Soft tissues

One commonly-mentioned advantage of the BBWd rat as a model for juvenile-onset diabetes is the absence of obesity (Nakhooda et al., 1978). In fact, the animal appears somewhat emaciated when compared side-by-side with a Wistar rat of the same general body proportions. The anatomical sites that store excess fat in the Wistar rat (e.g. -

epididymal fat pat, omentum, etc.) grossly contain less

adipose tissue in the BBW diabetic rat. This is undoubtedly due to the limited insulin regimen and severe diabetes (4+ glucose) in our BBWd rats and is consistent with the emaciated appearance often seen in untreated human IDDM. Other studies have suggested that adipose differs from most other tissues by not showing capillary basement membrane thickening in diabetics (Warren et al., 1966). 98 One soft tissue tumor, a subcutaneous spindle cell sarcoma, was identified in a BBW rat of unknown age (Figure

4.42).

B. Hematology Until relatively recently, it was believed that there were no specific hematological changes in diabetes mellitus, because there are no characteristic morphological abnormalities in peripheral smears or bone marrow aspirates. However, metabolic studies have found a vast array of metabolic and biochemical abnormalities, but the clinical significance of many of these is unknown. Erythrocyte studies have demonstrated a relationship between the concentration of glycosylated hemoglobion (Ale) and the duration of hyperglycemia (Cole, 1978; McDonald and Davis, 1979). Glycosylation also suggests a mechanism for the development of basement membrane thickening as a sequelae of diabetes mellitus (Jones and Peterson, 1981).

Other findings pertaining to the diabetic RBC include abnormal oxygen affinity, glycosylation of the 2,3- diphosphoglycerate binding site, decreased concentration of inorganic phosphate, increased viscosity, and decreased deformability (Jones and Peterson, 1981).

The numerous white blood cell (WBC) defects that have been reported to be associated with diabetes mellitus have been reviewed by Jones and Peterson (1981). Polymorphonuclear neutrophil (PMN) defects include abnormal adherence, migration, chemotaxis, phagocytosis, and killing. Several enzymatic abnormalities have also been reported but alteration in the number of PMNs is not a characteristic of diabetes. Lymphocytes, particularly T-cells, have been shown to have abnormal metabolic properties, mitogen cell responses, and cell surface properties in both human and animal diabetes. The presence of defects in the eosinophils or basophils of diabetics has not been adequately studied to justify any conclusions. It is likely that some of these defects are responible for the increased susceptibility of poorly-controlled diabetics to infection (Robertson and Polk, 1974). There is good reason to suspect platelet abnormalities in diabetes. Some evidence suggests that the hemostatic system may be involved in the initiation or propagation of atherosclerotic lesions in diabetics (Tschoepe et al., 1993). In fact, frequently the results of in vitro studies seem to have little _in vivo significance. Although it is well established that there is no significant difference in platelet counts or platelet survival between diabetic patients and control subjects, hyperglycemia does lead to a hypercoagulable state, but this cannot be observed with standard clotting assays (e.g. prothrombin time, activated partial thromboplastin time, and thrombin time) (Glassman, 1993) . 100 In summary, there do appear to be hematological abnormalities associated with diabetes, but it is presently uncertain whether many of these have any clinical significance.

In the present study, there were many more strain- related differences (those present in both BBWd and BBWnd but not in control) than diabetes-related (those in BBWd but neither BBWnd nor control) differences (Wright et al., 1983d, appendix H). More specifically, the BBW strain had significantly decreased numbers of white cells and platelets, as well as markedly changed differential white cell counts. Differential counts revealed a pattern of marked lymphopenia, slight neutrophilia, monocytosis, and eosinophilia. Jackson et al., (1981) have shown that this marked lymphopenia is due to a T-cell deficiency and Greiner et al. (1986) have shown that this is due to depletion of the RT-6+ T-cell subset. No differences in gamma globulin concentrations were observed. Some statistically significant changes in red cell indices were seen, but none were outside of the normal range of other rat strains

(Ringler and Dabich, 1979). Red cell distribution width (RDW) values were significantly higher in the BBW strain (particularly BBWd), indicating a tendency toward anisocytosis. These strain-related changes are consistent with present knowledge since inbreeding of rats is known to change hematological parameters (Hulse, 1965) and since 101 diabetes does not significantly alter major red and white cell indices in man (Jones and Paterson, 1981). The changes in red cell indices indicate either a sex- related or random pattern and none of these values are very far out of ranges considered normal for rat hematological studies (Ringler and Dabich, 1979). Most of the white blood cell indices are changed in a strain-related pattern. Both BBWd and BBWnd have significantly lower white cell counts than control rats as a result of lymphocytopenia. This could partially explain the higher incidence of infections (Wright et al., 1980; Sima, 1980) and possibly the apparent shortened life expectancy of the BBW rat (unpublished observation). Reich and Dunning (1941) have reported a positive correlation between the mean life-span of inbred rat strains and their mean WBC count.

Marked eosinophilia was present in many of the BBW rats regardless of the presence of diabetes. Since no parasitic infections could be demonstrated, hypersensitivity to some unknown allergen or autoantigen may be responsible. This is supported by the observation that massive numbers of eosinophilic infiltrates in the tissue sections of some rats in our study (Figure 4.43). Eosinophilic infiltrates in the islets of some rats have also been reported (Nakhooda et al., 1977). This is interesting in light of the frequent occurrence of peri-insular and insular eosinophilic infiltrates in the offspring of diabetic mothers (Barresi et 102 al., 1978). The other major strain-specific observation in this study was that BBW rats have decreased platelet concentrations relative to W rats (p <0.01) (Table H.l). It is questionable whether a change of this small magnitude would have any effect on the process of atherogenesis in BBW rats. The mean gamma globulin concentrations in our study are well within the normal ranges reported in other studies (Ringler and Dabich, 1979; Coleman et al., 1977) and no group differences in the concentrations of gamma globulins were observed (Wright et al., 1983d). Therefore, hypogammaglobulinemia cannot account for the increased susceptibility to infection in the BBW strain. However, it is not possible to rule out some abnormality in the distribution of immunoglobulin fractions as an etiological factor.

C. DNA Repair and Longevity

The BBW rat appears to have a significantly shortened lifespan relative to other inbred or outbred rats. The oldest BBW rats in our study were a 496 day old BBWnd female and a 465 day old BBWd male which was non-diabetic until 38 days before its death. The average lifespan of the outbred Wistar rat maintained in standard animal housing is about 26 months (Hoffman, 1979). The maximum achievable lifespan for 103 an outbred specific pathogen free Wistar rat is about 46 months (Hoffman, 1979). The mean age (+ standard deviations) of BBWd (n=95) and BBWnd (n=12) rats that died

spontaneously in this study were 274.5 +87.1 days and 288.6 + 142.6 days, respectively. No Wistar rats died spontaneously in this study. Since it seemed likely that the high incidence of

pneumonia in BBW rats might have lowered these age values rather significantly, I calculated these lifespans again excluding rats with pneumonia and found that the rats with pneumonia lived on average much longer than those without it. This is probably because pneumonia occurs selectively

in older rats. Therefore, it is necessary to postulate other causes for this apparently shortened life expectancy in BBW rats. Undoubtedly, some of these animals died because of inappropriate insulin treatment, but this does not explain the similar trend in BBWnd rats.

A key question was to determine whether or not there is

a genetic basis for the decreased longevity in the BBW

strain. Since it is well known that maximal achievable lifespan correlates directly with cellular DNA repair capacity (Hart and Setlow, 1974), we compared the cellular DNA repair capabilities of the BBW strain and that of the

outbred Wistar strain from which they were originally

derived. (Wright et al., 1986, appendix J). In this study we compared ultraviolet (UV) induced excision (long-patch) repair in the two strains because this type of repair requires endonuclease, exonuclease, polymerase and ligase activities. Therefore, it should provide a sensitive screen for problems with DNA repair.

As shown in Figure 1.2, unscheduled DNA synthesis, a measure of excision repair, did not significantly differ between the two strains. Therefore, this cannot explain the higher incidence of degenerative, autoimmune, and neoplastic diseases nor the decreased longevity in this strain. TABLE 4.1 INCIDENCE OF HEPATIC LESIONS BY AGE*

BBWd RATS BBWnd RATS

age in days age in days

LESION 0-120 121-240 241-360 361-480 unknown total 0-120 121-240 241-360 361-430 481-600 unknown total n=8 n=42 n=59 n=18 n=1 n=128 n=3 n=2 n=5 n=12 ns1 n=7 n=30

Normal 8 35 51 13 1 102 2 2 5 11 1 6 27 (889%) (833%) (86.4%) (72.2%) (100%) (797%) (66.7%) (100%) (100%) (91.7%) (100%) (85.7%) (90 0%)

Fatty 0 4 4 4 0 12 1 0 0 0 0 0 1 Change (5.9%) (6.8%) (222%) (9.4%) (33.3%) (3.3%)

Accessory 0 0 1 0 0 1 0 0 0 1 0 0 1 Lobe (1,7%) (0.8%) (8.3%) (3.3%)

Marked bile 0 0 1 0 0 1 0 0 0 0 0 0 0 duct (1.7%) (0.8%) hyperplasia

Infarction 0 2 1 1 0 4 0 0 0 0 0 1 1 (4.8%) (1.7%) (5.6%) (3.1%) (14.3%) (3.3%)

Congestive 0 0 1 0 0 1 0 0 0 0 0 0 0 necrosis (1.7%) (08%)

Abscess 0 1 0 0 0 1 0 0 0 0 0 0 0 (24%) (0.8%)

A lota' of <5 W rals were examined. Isolaled findings included fatty change (2.2%). infarction (2.2%), and congestive necrosis (2.2%). TABLE 4.2 INCIDENCE OF PULMONARY LESIONS BY AGE

BBWd RATS BBWnd RATS age in days age in days

LESION 0-120 121-240 241-360 361-480 unknown total 0-120 121-240 241-360 361-480 481-600 unknown total n = a n = 49 n = 60 n = 18 n = 1 n = 136 n = 3 n = 2 n r 5 n = 12 n s 1 n = 1 n = 31

Normal 8 34 30 3 1 76 1 2 1 7 0 4 15 (100%) (69.4%) (50%) (16.7%) (100%) (55.9%) (33.3%) (100%) (20%) (583%) (50%) (484%)

Broncho 0 11 22 13 0 46 1 0 4 5 1 2 13 pneumonia (224%) (36 7%) (722%) (33 8%) (33.3%) ' (80%) (41.7%) (100%) (25%) (41.9%)

j Hemorrhagic 0 0 3 0 0 3 0 0 0 0 0 1 1 pneumonia (5.0%) (22%) (125%) (3-2%)

Aspiration 0 0 2 . 0 0 2 0 0 0 0 0 0 pneumonia (3.3%) (1.5%) •

Lobar 0 0 0 1 0 1 0 0 0 0 0 0 0 pneumonia (5.6%) (0.7%)

Bronchitis 0 0 1 1 0 2 1 .. 0 0 0 0 0 1 (1.7%) (5.6%) (1.5%) (33.3%) (32%)

Granutoma 0 0 1 0 0 1 0 0 0 0 0 1 1 (1.7%) (0.7%) (125%) (32%)

“Except for two instances of bronchitis, no pulmonary lesions were observed in 30W rats. 107 TABLE 4.3

INFECTIOUS ORGANISMS ISOLATED FROM BBW RAT TISSUES

PULMONARY

Bacilli Escherichia coli

Proteus mirabilis

Proteus vulcraris

Pseudomonas aerucrinosa

Cocci Staohvlococcus (coacr. nea.)

Streptococcus, aloha (histochemically not group

D)

Streptococcus faecalis

Veillonella parvula

OTHER SITES

Bacilli Clostridium perfrinaens

Escherichia coli

Klebsiella pneumoniae

Proteus vulcraris

Fungus Candida sp. not albicans TABLE 4.4

INCIDENCE OF CARDIAC LESIONS BY AGE‘

BBWd RATS BBWnd RATS age in days age in days

LESION 0-120 121-240 241-360 361-480 unknown total 0-120 121-240 241-360 361-480 481-600 unknown total n = 22 n = 7 n = 39 n = 52 n = 18 n = 1 n = 117 n = 3 n = 2 n = 4 n = 12 n = 1 n = 7

Normal 7 38 51 13 1 109 3 2 4 12 1 6 21 (100%) (97.4%) (98.1%) (72.2%) (100%) (93 2%) (100%) (100%) (100%) (100%) (100%) (85.7%) (95.5%)

Myocardial 0 0 ' 1 3 0 4 0 0 0 0 0 1 1 degeneration (1.9%) (16.7%) (3.4%) (14.3%) (4.5%)

Acute 0 0 0 2 0 2 0 0 0 0 0 0 0 myocarditis (11.1%) . (1.7%)

Endocardial 0 1 0 0 0 1 0 0 0 0 0 0 0 proliferation (26%) (0.9%)

•Myocardial degeneration was present in 2 of 5 (40%) W rats in the 481-600 day old age group but was not present in any younger W rats. No other cardiac lesions were observed in W rats. TABLE 4.5 INCIDENCE OF LYMPHATIC LESIONS BY AGE*

BBWd RATS BBWnd RATS age in days age in days

0-120 121-240 241-360 361-480 unknown total 0-120 121-240 241-360 361-480 481-600 unknown total LESION n s 8 n = 51 n = 62 n = 18 n = 1 n = 140 n = 3 n = 2 n = 5 n = 12 n = 1 n = 9 n = 32

Normal 8 47 53 18 1 127 3 2 5 11 1 9 30 (100%) (92.2%) (85.5%) (100%) (100%) (90.7%) (100%) (100%) (100%) (91.7%) (100%) (90%) (93.8%)

Lymphoma 0 1 5 0 0 6 0 0 0 0 0 1 1 (2.0%) (8.1%) (4.3%) (10%) (3.1%)

Granuloma 0 3 4 0 0 7 0 0 0 1 0 0 1 (6.0%) (65%) (5.0%) (8.3%) (3.1%)

*No lymphatic lesions were observed in W rats. TABLE 4.6 SITES OF LYMPHOMATOUS INVOLVEMENT

LYMPH NODES

GROUP R A T # SEX AGE MES. MED.OTHER THYMUSCOLON S.I. UVER PANC. SPLEEN LUNGS OVARY

BBWd 015 M 248 ♦ * ♦ + ♦

235 M 332 + +

241 F 198 ♦ +

294 M 255 ♦ + ♦

454 F 334 ♦ + ♦

1213 M 315 ■ ♦ ♦

BBWnd ND117 F — + ♦ + ♦ ♦ + +

abbrevixons: mes., mesemery; med.. medtastnunv. S t.. small w e s t o n e . pane, pancreas CHAPTER V GENERAL DISCUSSION (SUMMARY) Throughout this entire study, incidences of various lesions and other findings have been tabulated by group (BBWd, BBWnd, and W) and by age. In order to understand the significance of these lesions, these results have been classified as follows (Wright et al., 1983e, appendix J):

I. Strain-related lesions are those abnormalities found in BBWd and BBWnd rats but not in W rats.

II. Diabetes-related lesions are those abnormalities found in BBWd rats but not in BBWnd or W rats. The lesions are subdivided into three groups:

a. lesions strongly associated with human diabetes.

b. lesions weakly associated with human diabetes. c. lesions not specifically associated with human diabetes.

III. Rat-related lesions are those that occurred in BBWd, BBWnd, and W rat with similar frequencies.

IV. Organ systems virtually free of significant pathology.

111 112 Table J.l shows the incidence of strain-related lesions found in BBW rats. These include lymphoproliferative, inflammatory and developmental disorders. The presence of lymphoid hyperplasia, spontaneous granulomas in the absence of demonstrable organisms, lymphomas, eosinophilia, and lymphocytopenia all suggest an abnormal immune system. The high incidence of infections in BBW rats is consistent with this.

Tables J.2A and J.2B show the incidence of diabetes- related lesions that are strongly associated with diabetes

(insulitis, testicular atrophy, and cataracts) and those that are weakly associated with diabetes (hepatic fatty change, pancreatitis, lymphocytic thyroiditis, hypoglycemic brain damage, central pontine myelinolysis). The presence of these human diabetic sequelae in BBWd rats supports the validity of the model. Table J.2C also includes several lesions that are diabetes-related in the rat, but not specifically associated with diabetes in man. This category is comprised of stomach erosions and idiopathic megacolon. Lesions in this category, as well as those in the strain-related category, are undesirable and uncontrollable variables that may affect experiments utilizing this model. Of these, most are minor inconveniences because of either low incidences or minimal deleterious effects. Only the pulmonary infections and 113 gastric erosions are likely to be of practical consequence for most experiments. Low incidences of several other types of lesions (i.e. - myocarditis and hepatic infarction) appeared in all groups of rats. Finally, the absence of several important sequelae of human diabetes (i.e. - diabetic nephropathy, atherosclerosis, and severe microangiopathy) suggests a degree of infidelity as a model for human diabetes mellitus. Comparison of the findings in this study with the incidence of lesions in other strains of "aged" rats tends to downplay the significance of some of the lesions in the BBW rats. Generally speaking, only rats in excess of two or two and a half years of age are considered "aged". By this standard, none of the BBW rats in this study are "aged" because they do not live that long. Most outbred rat strains do not develop any significant lesions (except chronic respiratory infections) until they are over two years of age. This explains the paucity of lesions in our W rat series. The occurrence of a high incidence of typical senile lesions in the relatively young BBW rats suggests some mechanism for premature aging could be involved.

An important point that may be easily overlooked is that all of the lesion incidences are characteristic of BBW rats housed under the conditions present in the semi-barrier facilities at Wiseman Hall at The Ohio State University. It 114 is possible that animals housed elsewhere under different conditions may have somewhat different characteristics. This is particularly true since many of the BBW colonies around the country were started from a few mating pairs of outbred rats shipped from Ottawa and then were subsequently inbred by sibling mating within the new institutions. This geographic isolation would tend to firmly fix and amplify unusual traits (i.e. - similar to the process of speciation). In addition to inbreeding, the environmental conditions at the new institutions can have a significant effect on the expression of traits, even in highly inbred strains of laboratory animals. For example, 100% of inbred C3H-Avy mice developed both mammary and liver tumors when housed in the United States, but when this same colony was transferred to Adelaide, Australia, only 17% of the mice developed liver tumors. When food and bedding were imported from the United States to Australia, the 100% incidence was re-established (Sabine et al., 1973).

How Good is the BBW Model? What is a Model?

A model is something that represents something else. Several features are common to most models: (1) No model is perfect. If it were, then it would be the original.

Therefore, studies performed on models are less meaningful than studying the original. (2) There is no way to know 115 how good a model really is. Even if all examinable features of a model and the original appear to be the same; there is no guarantee that similar mechanisms are involved in producing those similarities. (3) There is an advantage to using multiple models. If results from several different models agree, this suggests that a similar mechanism may be involved. (4) Since results based on the use of models are less meaningful, models are used predominantly when the "cost" of studying the original is too great (i.e. - financial, technical, ethical, or legal reasons prevent studying the original but are not so limiting to studying the model). (5) Although models are only useful in so much as they mimic the real thing, differences can sometimes be exploited when trying to understand mechanisms. For example, the question can be asked: "Why does the model behave differently than the original when they appear to be so similar?" Even with all of these inherent limitations, animal models still have made major contributions to the understanding and treatment of human diseases (Jones, 1980). Animal models for human diseases are studied mainly because their environment, their heredity and their age or time of death can be more easily controlled than when dealing with humans. Table 5.1 shows the characteristics of a good animal model as outlined by Leader and Padgett (1980). 116 Based on these criteria, how good is the BBW rat as a model for IDDM? Several of the criteria are very easily satisfied by the BBW rat. First, it appears to reproduce accurately the metabolic features of the disease, the morphology of the disease, and some of the sequelae of the disease. Second, modest numbers of outbred rats are readily available to qualified investigators at no cost from Health and Welfare of Canada or larger numbers of inbred rats can be purchased from the University of Massachusetts colony. Third, it gives birth to multiple offspring - usually 5-10 pups per litter. Fourth, the BBW rat is docile and easily handled. Fifth, its size is greater than that of other rodent models. On the other hand, the BBW rat does not develop some of the major sequelae of diabetes, and it develops numerous other lesions or conditions that are not present in human diabetes that may serve as uncontrollable variables and may shorten the lives of the animals.

Lymphocytopenia is an important example of the latter. Second, it is not very easily maintained. Daily insulin dosages must be carefully determined. A full-time skilled technician is required. Breeding problems are common, in part as a result of developing testicular atrophy at a young age. Third, the BBW rat appears to have a significantly shortened lifespan. The latter of these seems to be one of the most severe limitations of the model. In conclusion, it is obvious that there are both advantages and disadvantages to the BBW model. As a result of our study, it is one of the best pathologically characterized models for diabetes research. It is hoped that this information will assist individual investigators to make an informed decision as to the suitability of the BBW rat for their experiments. This has become exceedingly important because of mounting pressure from both the scientific community and society to justify research using animal models. Societal and scientific norms agree that poorly designed research utilizing animals is unethical. However, animal models, if used appropriately, can provide important insights into many aspects of IDDM (Sieber and Traystman, 1993). 118

TABLE 5.1

CRITERIA FOR A GOOD ANIMAL MODEL (LEADER AND PADGETT. 1980)

1. Accurately Reproduce the Disease and Sequelae

2. Ready Availability

3. Multiple Offspring

4. Easily Handled

5. Relatively Large Size

6. Easily Maintained

7. Long Survival LIST OF REFERENCES

ABEND S, CORMIER J, APPEL MC, LUEPRASITSAKUL W, ALEX S, MORDES JP, USADEL KH, HANDLER ES, ROSSINI AA and BRAVERMAN LE: The effect of ciamexone on lymphocytic thyroiditis and insulin-dependent diabetes mellitus in the BB/Wor rat. Immunopharmacol. .19:163-168, 1990.

AGARDH CD, KALIMO H, OLSSON Y and SIESJA BK: Hypoglycemic brain injury: Metabolic and structural findings in rat cerebellar cortex during profound findings insulin-induced hypoglycemia and in the recovery period following glucose administration. J. Cerebral Blood Flow Metab. .1:71-84, 1981.

ALLEN EM: The effect of iodine on lipid peroxidation and ultrastructure in the thyroids of BB/Wor rats. J. Endocrinol. Invest. .15:519-23, 1992.

ALLEN EM, APPEL MC and BRAVERMAN LE: The effect of iodide ingestion on the development of spontaneous lympocytic thyroiditis in the diabetes-prone BB/W rat. Endocrinology 118:1977-81, 1986a.

ALLEN EM, RAJATANAVIN R, NOGIMORI T, CUSHING G, INGBAR SH and BRAVERMAN LE: The effect of methimazole on the development of spontaneously lymphocytic thyroiditis in the diabetes-prone BB/W rat. Am. J. Med. Sci. 292:267-71, 1986b.

ALLEN EM, APPEL MC and BRAVERMAN LE: Iodine-induced thyroiditis and hypothyroidism in the hemithyroidectomized BB/W rat. Endocrinology 121:481-5. 1987.

ALOIA JE and NILAKANTAN S: The cerebrospinal fluid in uncontrolled diabetes mellitus. Am. J. Med. Sci. 266:203-9. 1973.

ALTMAN NH and GOODMAN DG: Neoplastic Disease. In: The Laboratory Rat. Vol. 1. Biology and Diseases, edited by HJ Baker, JR Lindsey, SH Weisbroth, NY:Academic Press, 1979, pp.336-376.

119 ANDERSON DE: Some characteristics of familial breast cancer. Cancer 28:1500-4, 1971. ANDREW W: Age changes in the vascular architecture and cell content in the spleens of 100 Wistar Institute rats including comparisons with human material. Am. J. Anat. 79:1-74, 1946. ANVER MR and COHEN BJ: Lesions associated with aging. In: The Laboratory Rat. Vol. 1. Biology and Diseases, edited by HJ Baker, JR Lindsey, SH Weisbroth, NY:Academic Press, 1979, pp.377-99.

ASMAL AC, LEARY WP and THANDROYEN F: Diabetic heart disease. S. Afr. Med. Tydskif .1:788-90, 1980. AXELROD AA and LeBLOND CP: Induction of thyroid tumors in rats by a low iodine diet. Cancer 8.:339-67, 1955. BABA N and von HAAM E: Experimental carcinoma of the endometrium. Prog. Exp. Tumor Res. £: 192-260, 1967. BABA N and von HAAM E: Tumors of the vagina, uterus, placenta and oviduct. IARC Sci. Publ. 6.: 161-88, 1976.

BARRESSI 6, INFERRA C and DeLUCA, F: Eosinophilic pancreatitis in the newborn infant of a diabetic mother. Virchows Arch A. Path. Anat. Histol. 380:341-8. 1978.

BATTLE WM, SNAFE WJ, ALAVI A, COHEN S and BRAUNSTEIN S: Colonic dysfunction in diabetes mellitus. Gastroenterol. 79:1217-21, 1980.

BEHAR A, BENTLA E and AVIRAM A: Central pontine myelinolysis (A case report). Acta. Neuropathologica 3.:343- 50, 1964.

BENIRSCHKE K and KAUFMANN P: Pathology of the Human Placenta. 2nd ed., NY:Springer-Verlag, 1990.

BERG BN: Longevity studies in rats: II Pathology of ageing rats. In: Pathology of Laboratory Rats and Mice, edited by E Cotchin, FJC Roe, Oxford:Blackwell Scientific, 1967, pp.749-786.

BERNUAU D, GUILLOT R, DURAND-SCHNEIDER AM, POUSSIER P, MOREAU A and FELDMANN G: Liver perisinusoidal fibrosis in BB rats with or without overt diabetes. Am. J. Pathol. 120:38-45, 1985.

120 121

BIERMAN EL: Atherogenesis in diabetes. Atheroscler. Thromb. 12:647-56, 1992.

BIVIN WS, CRAWFORD MP and BREWER NR: Morphophysiology. In: The Laboratory Rat. Vol. 1. Biology and Diseases, edited by HJ Baker, JR Lindsey, SH Weisbroth, NY:Academic Press, 1979, pp.73-103.

BLISS M: The Discovery of Insulin, Chicago:University of Chicago Press, 1982.

BLOODWORTH JMB: A re-evaluation of diabetic glomerulosclerosis 50 years after the discovery of insulin. Hum. Pathol. .9:439-53, 1978.

BOITARD C, MICHIE S, SERRURIER P, BUTCHER GW, LARKINS AP and McDEVITT HO: In vivo prevention of thyroid and pancreatic autoimmunity in the BB rat by antibody to class II major histocompatibility complex gene products. Proc. Natl. Acad. Sci. USA 12:6627-31, 1985.

BONAR JR: Diabetes: A Clinical Guide. Flushing, NY:Medical Examination Publishing, 1977, pp.315-6.

BOORMAN GA, ZURCHER C, HOLLANDER CF and FERON VJ: Naturally occurring endocardial disease in the rat. Acta. Pathol. 96:39-45, 1973.

BOORMAN GA and HOLLANDER CF: Spontaneous lesions in the female WAG/Rij (Wistar) rat. J. Gerontol. .28:152-9, 1973.

BOORMAN GA and HOLLANDER CF: Medullary carcinoma of the thyroid. Am. J. Pathol. 13:237-40, 1976.

BOTTAZZO GF, FLORIN-CHRISTENSEN A and DONIACH D: Islet cell antibodies in diabetes mellitus with autoimmune polyendocrine deficiencies. Lancet 2.: 1279-83, 1974.

BRAS G and ROSS MH: Kidney disease and nutrition in the rat. Toxicol. Appl. Pharmacol. 1:2247-62, 1964.

BRIERLY JB: Brain damage due to hypoglycaemia. In: Hvpoglvcaemia. edited by V Marks, FC Rose, Oxford:Blackwell, 1981, pp.488-94.

BROWNLEE M: Glycation products and the pathogenesis of diabetic complications. Diabetes Care .15:1835-43, 1992. 122

BROWNSCHEIDLE CM and DAVIS DL: Diabetes in pregnancy: A preliminary study of the pancreas, placenta and malformations in the BB Wistar rat. Placenta (Suppl.) 3:203-16, 1981.

BULLOCK BC, BANKS KL and MANNING PJ: Common lesions in the aged rat. In: The Laboratory Animal in Gerontological Research, Publication 1591, Washington DC:National Academy of Sciences, 1968, pp.62-82.

BUREK JD: Pathology of Aging Rats. Boca Raton:CRC Press, 1978.

BUREK JD and HOLLANDER CF: Incidence patterns of spontaneous tumors in BN/Bi rats. J. Natl. Cancer Inst. 58:99-105, 1977. BUREK JD and MEIHUIZEN SP: Age-associated changes in the thymus of aging BN/Bi rats. In: Vth European Symposium on Basic Research in Gerontology, edited by UJ Schmidt, G Briischke, E Lang, A Viidik, D Platt, W Frolkis and FH Schulz, Erlangen:Verlaag Dr.med.D. Straube, 1977, pp.167-70.

CABARROU A, LAGUENS R, CAINO H, DORIA I, AUCIELLO N, CEDOLA N and PONCE de LEON H: The diabetic liver: Cytochemical, ultrastructural and enzymatic changes. Acta. Diabetologica Latina 20*1236-68, 1973.

CHAPPELL CJ and CHAPPEL WR: The discovery and development of the BB rat colony: An animal model of spontaneous diabetes melliltus. Metabolism 32(Suppl.1):8-10, 1983. CHESLER A and TISLOWITZ R: Effect of alloxan diabetes on the growing rat. Science 101:648. 1945.

CHIPS HD and DUFF GL: Glycogen infiltration of the liver cell nuclei. Am. J. Pathol. JL8:645-9, 1942.

CHRISTY M, DECKERT T and NERUP J: Immunity and autoimmunity in diabetes mellitus. Clinics on Endocrinology and Metabolism 6:305-22. 1977.

COHEN BJ, ANVER MR, RINGLER DH and ADELMAN RC: Age- associated pathological changes in male rats. Fed. Proc. 37:2848-50, 1978.

COLE RA: The significance of hemoglobion ALC in diabetes mellitus. Upjohn MS-6928E, 1978. 123

COLEMAN GL, BARTHOLD SW, OSBALDISTON GW, FOSTER SJ and JONAS AM: Pathological changes during aging in barrier-reared Fischer 344 male rats. J. Gerontol. 32.:258-78, 1977.

COLLE E, GUTTMAN RD and SEEMAYER TA: Association of spontaneous thyroiditis with the major histocompatibility complex of the rat. Endocrinology 116:1243-7, 1985.

COTRAN RS, KUMAR V and ROBBINS SL: Robbins Pathologic Basis of Disease, 5th edition, Philadelphia: WB Saunders, 1994, pp.640-1.

COUSER WG and STILMANT MM: The immunopathology of the aging rat kidney. J. Gerontol. 3^113-22, 1976.

CRAIN RC: Spontaneous tumors in the Rochester strain of the Wistar rat. Am. J. Pathol. 34:311-35, 1958.

CREUTZFELDT W, FRERICHS H and SICKINGER K: Liver diseases and disbetes mellitus. In: Progress in Liver Diseases. Vol.3, edited by H Popper, F Schaffner, NY:Grune & Stratton, 1970, pp.371-407.

CRISA L, MORDES JP and ROSSINI AA: Autoimmune diabetes mellitus in the BB rat. Diabetes/Metab. Rev. 8.:9-37, 1992.

DAGLE GE, ZWICKER GM and RENNE RA: Morphology of spontaneous brain tumors in the rat. Vet. Pathol. 16:318- 24, 1979.

DAVIS RK, STEVENSON GT and BUSCH KA: Tumor incidence in normal Sprague-Dawley female rats. Cancer Res. 16:194-7, 1956.

DONIACH D and BOTTAZZO GF: Polyendocrine autoimmunity. In: Clinical Immunology Update, edited by Franklin, Amsterdam:Elsevier, 1981, pp.96-121.

EISENBARTH GS: Type I diabetes mellitus. A chronic autoimmune disease. N. Engl. J. Med. 314:1360-8. 1986.

FEDERLIN K, SCHOFFLING K, NEUBRONNER P and PFEIFFER EF: Histometrische Untersuchungen am Hodengewebe des Diabetikers mit Keimdriisenunterfunktion. Diabetologia 1.: 85-90, 1965.

FERRIER TM: Radiologically demonstrable arterial calcification in diabetes mellitus. Aust. Ann. Med. 13:222- 8, 1964. 124 FOULIS AK and STEWART JA: The pancreas in recent-onset (insulin-dependent) diabetes mellitus: insulin content of islets, insulitis and associated changes in the exocrine acinar tissues. Diabetologia 26:456-61, 1984. FRALEY DS and TOTTEN RS: An autopsy study of endocrine organ changes in diabetes mellitus. Metabolism 17:896-900. 1968. FRIEDBERG EC: DNA Repair. NY: WH Freeman, 1985, pp.270-4.

GEPTS W: Sequential changes in the cytological composition of the pancreatic islets in juvenile diabetes. In: Diabetes. Proceedings of the IXth Congress of the International Diabetes Federation, edited by JS Bajaj, Amsterdam:Excerpta Medica, 1977, pp.299-306. GEPTS W and LECOMPTE PM: The pancreatic islets in diabetes. Am. J. Med. 70:105-14, 1981. GILBERT C and GILLMAN J: Spontaneous neoplasms in the albino rat. S. Afr. J. Med. Sci. 23.:2257-72, 1958.

GLASSMAN AB: Platelet abnormalities in diabetes mellitus. Ann. Clin. Lab. Sci. 23.:47-50, 1993. GLAVIN GB: Restraint ulcer: History, current research and future implications. Brain Res. Bull. 5 (Suppl.1):51-8. 1980. GOTO Y, SATO SI and MASUDA M: Causes of death in 3151 diabetic autopsy cases. Tohoku J. Exp. Med. 112:339-53, 1974. GOTTLIEB PA, HANDLER ES, APPEL MC, GREINER DL, MORDES JP, ROSSINI AA: Insulin treatment prevents diabetes mellitus but not thyroiditis in RT6-depleted diabetes resistant BB/Wor rats. Diabetologia 34:296-300, 1991.

GREENE DA, SIMA AAF and PFEIFER MA: Neuropathy. In: Prevention and Treatment of Diabetic Late Complications. Berlin:Walter de Gruyer, 1989, pp.93-149. GREENE DA, SIMA AAF, STEVENS MJ, FELDMAN EL and LATTIMER SA: Complications: neuropathy, pathogenetic considerations. Diabetes Care 15.: 1902-25, 1992. 125 GREENE DA, YAGIHASHI S, LATTIMER SA and SIMA AAF: Nerve Na+-K+-ATPase, conduction, and myo-inositol in the insulin- deficient BB rat. Am. J. Physiol. 247:E534-9, 1984.

GREINER DL, HANDLER ES, NAKANO K, MORDES JP and ROSSINI AA: Absence of the RT-6 T cell subset in diabetes-prone BB/W rats. J. Immunol. 136:148-51. 1986.

GUBERSKI DL: Diabetes-prone and diabetes-resistant BB rats: Animal models of spontaneously and virally induced diabetes mellitus, lymphocytic thyroiditis, and collagen-induced arthritis. ILAR News 35:29-37, 1994. HANSEN AK and HANSEN FD: An Introduction to the Use of Rodent Models in Diabetes Research. Skensved: Mollegaard Ltd. & Bomholtgaard Ltd., 1990.

HANSSON CG: Studies on the parathyroids in alloxan diabetic rats. Acta Endocrinol. 45.: 139-50, 1964.

HART RW and SETLOW RB: Correlation between deoxyribonucleic acid excision-repair and lifespan in a number of mammalian species. Proc. Natl. Acad. Sci. 71.:2169-73, 1974. HEBEL R and STROMBERG MW: Anatomy and Embryology of the Laboratory Rat. Worthsee:Biomed Verlag, 1986.

HEIFETZ SA: The placenta. In: Pediatric Pathology, edited by JT Stocker and LP Dehner, Philadelphia:JB Lippincott, 1992, pp.387-423.

HELLMAN R, JACOBSSON L and TALJEDAL IB: Endocrine activity of the testes in obese-hyperglycemic mice. Acta Endocrinol. 44:20-6, 1963.

HESLOP BF: Cystic adenocarcinoma of the ascending colon in rats occurring as a self-limiting outbreak. Lab. Anim. 2:185-95, 1969.

HIROKAWA K: Characterization of age-associated kidney disease in Wistar rats. Mech. Ageing Develop. 4.:301-16, 1975.

HOEFFEL JC, SENOT P, CHAMPIGNEULLE B and DROUIN P: Gastric retention and gastric ileus in diabetes mellitus. Radiologe 20:540-2, 1980. 126 HOFFMAN HJ: Survival distributions for selected laboratory rat strains and stocks. In: Development of the Rodent as a Model System of Aging. Book 2, DHEW Publication No (N1H) 79- 161, pp.19-34, 1979. HOLLANDER CF, BUREK JD, BOORMAN 6A, SNELL KC and LAQUEUR 6L: Granular cell tumors of the central nervous system of the rat. Arch. Pathol. Lab. Med. 100:445-7. 1976. HULSE EV: Laboratory animals in experimental haematology. Food Cosmet. Toxicol. 3>:735-48, 1965.

HUNT EL and BAILEY DW: The effects of alloxan diabetes on the reproductive system of young male rats. Acta Endocrinol. 38,:432-40, 1961.

HUNTLEY AC: The cutaneous manifestations of diabetes mellitus. Diabetes Metab. Rev. 9.: 161-76, 1993. IRISAWA S, SHIRAI M and MATSUSHITA S: [Sexual disturbances in diabetes mellitus: II. Histological examinations of the human thyroid, adrenal and sex glands in diabetes mellitus.] Jap. J. Urol. 51:225-35, 1966. IRVINE WJ, CLARKE BF, SCARTH L, CULLEN DR and DUNCAN LJP: Thyroid and gastric autoimmunity in patients with diabetes mellitus. Lancet 2:163-8, 1970.

JACKSON R, RASSI N, CRUMP T, HAYNES B and EISENBARTH GS: The BB diabetic rat. Profound T-cell lymphocytopenia. Diabetes 10:887-9, 1981. JOHANSEN K and 0RNSHOTT J: Frequency of diabetes after acute pancreatitis. Metabolism ,21:291-6, 1972. JONES RL and PETERSON CM: Hematologic alterations in diabetes mellitus. Am. J. Med. 70:339-52, 1981.

JONES TC: The value of animal models. Am. J. Pathol. 101 (Suppl.3):S3-S7. 1980.

KALANT N and SEEMAYER T: Malignant lymphoma in spontaneously diabetic rats. N. Engl. J. Med. 300:737, 1979.

KANNEL WB, HJORTLAND M and CASTELLI WP: Role of diabetes in congestive heart failure: The Framingham study. Am. J. Cardiol. 34:29-34, 1974. 127 KASTERN W, LANG F and KRYSPIN-S0RENSEN I: The genetics of insulin-dependent diabetes in the BB rat. Cur. Topics Microbiol. Immunol. 156:87-102. 1990.

KENDREY G and ROE FJC: Histopathological changes in the pancreas of laboratory rats. Lab. Anim. 3.: 207-20, 1969.

KIM U, CLIFTON KH and FURTH J: A highly inbred line of Wistar rats yielding spontaneous MA/MMO somatotrophic pituitary and other tumors. J. Natl. Cancer Inst. 24:1031- 55, 1960.

KLEINSCHMIDT-DeMASTERS BK and NORENBERG MD: Neuropathologic observations in electrolyte-induced myelinolysis in the rat. J. Neuropathol. Exp. Neurol. 4JL:67-80, 1982.

KOHNER EM: Diabetic retinopathy. In: Prevention and Treatment of Diabetic Late Complications. Berlin:Walter de Gruyter, 1989, pp.75-91.

KONEZ LK and BALODIMOS MC: Impotence in diabetes mellitus. Med. Times 98:159-70, 1970.

KOTHARE SN: Pathological lesions in diabetes mellitus - their changing concepts. Indian J. Med. Sci. 28:103-7. 1974.

KRAUS JE: The hypophysis in diabetes insipidus and mellitus. Urol Cutan. Rev. 48:417-21, 1944.

LAFFEL LMB and KROLEWSKI AS: Epidemiology and clinical impact of diabetic late complications in IDDM. In: Prevention and Treatment of Diabetic Late Complications. Berlin:Walter de Gruyter, 1989, pp.13-27.

LAMB D: Rat lung pathology and quality of laboratory animals: The user's view. Lab. Anim. j): 1-8, 1975.

LEADER RW and PADGETT GA: The genesis and validation of animal models. Am. J. Pathol. 101 (Su p p 1.3):S11-S16. 1980. LECOMPT PM and GEPTS W: The pathology of juvenile diabetes. In: The Diabetic Pancreas, edited by BW Volk and F Wellmark, NY:Plenum Press, 1977, pp.325-63.

LEDET T: Diabetic cardiomyopathy. Acta Path. Microbiol. Scand. Sect. A. 84.:421-8, 1976. 128 LEDET T, NEUBAUER B, CHRISTENSEN NJ and LUNDBAEK K: Diabetic cardiopathy. Diabetologia 16:207-9, 1979.

LEMA BE, F06LIA VG and COLLAZO EF: Lesiones histologicas testiculares en la rata diabetica. Rev. Soc. Argent. Biol. 41:197-203, 1965. LIKE AA, BUTLER L, WILLIAMS RM, APPEL MC, WERINGER EJ and ROSSINI AA: Spontaneous autoimmune diabetes mellitus in the BB rat. Diabetes 31 (Suppl. 1 P+21:7-13. 1982. LIKE AA, KISLAUSKIS E, WILLIAM RM and ROSSINI AA: Neonatal thymectomy prevents spontaneous diabetes mellitus in the BB/W rat. Science 216:644. 1982. LIKE AA and ROSSINI AA: Spontaneous autoimmune diabetes mellitus in the BioBreeding/Worcester rat. Surv. Synth. Path. Res. 3.: 131-8, 1984. LINDSEY JR, BAKER HJ, OVERCASH RG, CASSELL GH and HUNT CE: Murine chronic respiratory disease. Am. J. Pathol. 64:675- 716, 1971. LORENZ G and BARENWALD G: Histologic and electron- microscopic liver changes in diabetic children. Acta Hepato-Gastroenterol. 26.:435-8, 1979. LUKENS FWD: Alloxan diabetes. Physiol. Rev. 28:304-30. 1948.

LUTZEN L and UEBERBERG H: A study on morphological changes in the testes of old albino rats. Beitr. Pathol. Bd. 149:377-85, 1973.

MACHADO EA and LOZZIO BD: Accessory lobe of the liver in Gunn-derived rats. Teratology 5.: 361-6, 1972. MAGNUSSON G and RAMSAY CH: Urolithiasis in the rat. Lab. Anim. j>:153-62, 1971.

MAGNUSSON G, MAJEED SK and GOPINATH C: Infiltrating pituitary neoplasms in the rat. Lab. Animals 13:111-3. 1979.

MARLISS EB, NAKHOODA AF, POUSSIER P and SIMA AAF: The diabetic syndrome of the "BB" Wistar rat: Possible relevance to type I (insulin dependent) diabetes in man. Diabetologia 22:225-32, 1982. 129 MAUER SM, STEFFES MW and BROWN DM: The kidney in diabetes. Am. J. Med. 70:603-12, 1981. MCCALL AL: The impact of diabetes on the CNS. Diabetes 41:557-70, 1992. MCCOY GW: a prelilminary report on tumors found in wild rats. J. Med. Res. 2!l.:282-96, 1909.

MCDONALD JM and DAVIS JE: Glycosylated hemoglobins and diabetes mellitus. Human Pathol. 10:279-91, 1979.

MEEHAN CJ, FLEMING S, SMITH W and BAIRD JD: Idiopathic megacolon in the BB rat. Internatl. J. Exp. Pathol. 75: 37-42, 1994. MENDELL JR, SAHENK Z and WARMOTTS JR: Peripheral nerve abnormalities in the spontaneously diabetic BB-Wistar rat. Neurol. 30:434, 1980 (abstract).

MENDELL JR, SAHENK Z, WARMOTTS JR, MARSHALL JK and THIBERT P: The spontaneously diabetic BB Wistar rat. Morphologic and physiologic studies of peripheral nerve. J. Neurol. Sci. 52:103-115, 1981.

MERRIAM JC and SOMMERS SC: Mammary periductal hyalin in diabetic women: Report of twenty cases. Lab. Invest. 6.: 412- 20, 1957.

MOENS H and FARID NR: Hashimoto's thyroiditis is associated with HLA-DRw3. N. Engl. J. Med. 299:133-4. 1978.

MORDES JP, DESEMONE J and ROSSINI AA: The BB rat. Diabetes/Metab. Rev. 3.:725-50, 1987.

MUNTZING J, SUFRIN G and MURPHY GP: Prostatitis in the rat. Scand. J. Urol. Nephrol. .13:17-22, 1979.

MURAOKA Y, ITOH M, MOEDA Y and HAYASHI Y: Histological changes of various organs in aged SD-JCL rats. Exp. Anim. 26:1-12, 1977.

MURRAY FT, CAMERON DF and ORTH JM: Gonadal dysfunction in the spontaneously diabetic BB rat. Metabolism 32 (Suppl.1):141-7. 1983. 130 MURRAY FT, CAMERON DF, ORTH JM and KATOVICH MJ: Gonadal dysfunction in spontaneously diabetic BB rat: Alterations of testes morphology, serum testosterone and LH. Horm. Metabol. Res. 17.:495-501, 1985.

NAGORE N and SCHEUER PJ: The pathology of diabetic hepatitis. J. Pathol. 156:155-60, 1988.

NAKHOODA AF, LIKE AA, CHAPPEL Cl, MURRAY FT and MARLISS EB: The spontaneously diabetic Wistar rat. Metabolic and morphologic studies. Diabetes .26:100-12, 1977.

NAKHOODA AF, LIKE AA, CHAPPEL Cl, WEI CN and MARLISS EB: The spontaneously diabetic Wistar rat (the "BB" rat) studies prior to and during development of the overt syndrome. Diabetologia 14:199-207, 1978.

NERUP J and LERNMARK A: Autoimmunity in insulin-dependent diabetes mellitus. Am. J. Med. 70:135-41, 1981.

NISHIMURA M, YOKOYAMA M, TAGUCHI T and KITAMURA Y: Spontaneous gastric erosions in NOD and KK-Ay mice. Lab. Animal Sci. 33.:577-9, 1983. OKSANEN A: Testicular lesions of streptozotocin diabetic rats. Horm. Res. 6.: 138-44, 1975.

OLCOT CT: A transplantable nephroblastoma (Wilm's tumor) and other spontaneous tumors in a colony of rats. Cancer Res. 10:625-8, 1950.

ORCI L and PERRELET A: The islet of Langerhans revisited with immunofluorescence and freeze-fracture. Pfizer Spectrum 22:17-23. 1979.

OSTERBY R: Kidney structural abnormalities in early diabetes. In: Vascular and Neurological Changes in Early Diabetes. edited by RA Camerini-Davalos, HS Cole, NY:Academic Press, 1973, pp.323-40.

PARSON W, MACDONALD FW and SHAPER AG: African diabetics necropsied at Malago Hospital, Kampala, Uganda 1957-1966. East African Med. J. 4J5:89-99, 1968.

PEREZ MI and KOHN SR: Cutaneous manifestations of diabetes mellitus. J. Am. Acad. Dermatol. .30:519-31, 1994.

POLLARD M and KAJIMA M: Lesions in aged germfree Wistar rats. Am. J. Pathol. 61.:25-32, 1970. 131 PYdRALA K: Diabetes and heart disease. In: Prevention and Treatment of Diabetic Late Complications. Berlin:Walter de Gruyter, 1989, pp.151-68.

RAHIER J, GOEBBELS RM and HENQUIN JC: Cellular composition of the human diabetic pancreas. Diabetologia 24:366-71, 1983.

RAJATANAVIN R, APPEL MC, REINHARDT W, ALEX S, YANG Y-N, BRAVERMAN LE: Variable prevalence of lymphocytic thyroiditis among diabetes-prone sublines of BB/Wor rats. Endocrinology 128:153-7, 1991. RATCLIFFE HL: Spontaneous tumors in two colonies of rats of the Wistar Institute of Anatomy and Biology. Am. J. Pathol. 16:237-54, 1940.

REGAN TJ, ETTINGER PO, KHAN MI, JESRANI MU, LYONS MM, OLDEWURTEL HA and WEBER M: Altered myocardial function and metabolism in chronic diabetes mellitus without ischemia in dogs. Circ. Res. .35:222-37, 1974. REICH C and DUNNING WF: Leucocyte level and longevity in rats. Science 9^3:429-30, 1941.

RINGLER DH and DABICH L: Hematology and clinical biochemistry. In: The Laboratory Rat. Vol.1. Biology and Diseases. edited by HJ Baker, JR Lindsey, SH Weisbroth, NY:Academic Press, 1979, pp.105-21. ROBERTSON HD and POLK HC: The mechanism of infection in patients with diabetes mellitus: A review of leukocyte malfunction. Surgery 75.: 123-8, 1974. ROOT HF: The association of diabetes and tuberculosis. N. Engl. J. Med. 210:127, 1934.

ROSENMANN E, PALTI Z, TEITELBAUM A, and COHEN AM: Testicular degeneration in genetically selected sucrose-fed diabetic rats. Metabolism 23:343-81. 1974. ROSS R and AGIUS L: The process of atherogenesis - cellular and molecular interaction: from experimental animal models to humans. Diabetologia 35 (Suppl.2):S34-S40. 1992. ROSSINI AA, MORDES JP and LIKE AA: Immunology of insulin- dependent diabetes mellitus. Ann. Rev. Immunol. 3.:289-30, 1985. 132

ROSSINI AA, GREINER DL, FRIEDMAN HP and MORDES JP: Immunopathogenesis of diabetes mellitus. Diabetes Rev. 1:43-75, 1993.

ROWLATT UF: Pancreatic neoplasms of rats and mice. In: Pathology of Laboratory Rats and Mice, edited by E Cotchin, FJC Roe, Oxford:Blackwell Scientific, 1967, pp.85-103. RUSSFIELD AB: Pathology of the endocrine glands, ovary and testis of rats and mice. In: Pathology of Laboratory Rats and Mice, edited by E Cotchin, FJC Roe, Oxford:Blackwell Scientific, 1967, pp.391-467.

RUSSI S, BLUMENTHAL HT and GRAY SH: Small adenomas of adrenal cortex in hypertension and diabetes. Arch. Int. Med. 76:284-91, 1945.

SABINE JR, HORTON BJ and WICKS MB: Spontaneous tumors in C3H-A and C3H-AFB mice: High incidence in the United States and low incidence in Australia. J. Natl. Cancer Inst. 50:1237-42, 1973.

SAKSENA SK, LAU IF and CHANG MC: Age dependent changes in the sperm population and fertility in the male rat. Exp. Aging Res. 5.:373-81, 1979.

SASS B, RABSTEIN LS, MADISON R, NIMS RM, PETERS RL and KELLOFF GJ: Incidence of spontaneous neoplasms in F344 rats throughout the natural life-span. J. Natl. Cancer Inst. 54:1449-56, 1975.

SCAFFIDI L and ROTOLO G: II dannotesticolare nel diabete sperimentale da allossana nel ratto. Minerva Med. 65:4477- 84, 1974.

SCHOENBAUM SC: How to minimize infectious complications of diabetes. Geriatrics 21:51-61, 1979.

SCH&FFLING K, FEDERLIN K, DITSCHUNEIT H and PFEIFFER EF: Disorders of sexual function in male diabetics. Diabetes 12:519-27, 1963.

SCHdFFLING K, FEDERLIN K, SCHMITT N and PFEIFFER EF: Histometric investigations on the testicular tissue of rats with alloxan diabetes and Chinese hamsters with spontaneous diabetes. Acta Endocrinol. 54.:335-46, 1967. 133

SCHWARTZ CJ, VALENTE AJ, SPRAGUE EA, KELLEY JL, CAYATTE AJ and ROZEK MM: Pathogenesis of the atherosclerotic lesion. Implications for diabetes mellitus. Diabetes Care 15:1156- 67, 1992.

SCOTT J: The spontaneously diabetic BB rat: Sites of the defects leading to autoimmunity and diabetes mellitus. A review. Curr. Topics Microbiol. Immunol. 156:1-14, 1990.

SEEMAYER TA, KALANT N, SCHURCH W and COLLE E: Lympho- proliferative lesions in BB Wistar rats. Metabolism 32 (Suppl.1):97-100. 1983.

SEEMAYER TA, SCHURCH W and KALANT N: B cell lymphoproliferation in spontaneously diabetic BB Wistar rats. Diabetologia 23.:261-5, 1982.

SEEMAYER TA, TANNEBAUM GS, GOLDMAN H and 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. 106:237- 49, 1982a.

SHAIN SA,, MCCULLOUGH B and SEGALOFF A: Spontaneous adenocarcinoma of the ventral prostate of aged AXC rats. J. Natl. Cancer Inst. 55.: 177-80, 1975.

SHER SP: Mammary tumors in control rats: Literature tabulation. Toxicol. Appl. Pharmacol. 22.:562-88, 1972.

SIEBER FE and TRAYSTMAN RJ: Ethical issues involved in the development of animal models for type I diabetes. ILAR News 35:19-22, 1993.

SIMA AAF: Peripheral neuropathy in the spontaneously diabetic "BB"-Wistar rat. An ultrastructural study. Acta Neuropathol. .51:223-7, 1980.

SIMA AAF: The development and structural characterization of the neuropathies in the spontaneously diabetic BB Wistar rat. Metabolism 32 (Suppl.1):106-11. 1983. SIMA AAF: can the BB rat help to unravel diabetic neuropathy? Neuropathol. Appl. Neurobiol. jLl:253-64, 1985.

SIMA AAF, BOUCHIER M and CHRISTENSEN H: Axonal atrophy in sensory nerves of the diabetic BB-Wistar rat: a possible early correlate of human diabetic neuropathy. Ann Neurol. 13:264-72, 1983. 134 SIMA AAF, CHAKRABARTI S, GARCIA-SALINAS R and BASU PK: The BB rat - an authentic model of human diabetic retinopathy. Curr. Eye Res. 1087-92, 1985.

SIMA AAF and HAY K: Functional aspects and pathogenetic considerations of the neuropathy in the spontaneously diabetic "BB" wistar rat. Neuropath. Appl. Neurobiol. 7:341-50, 1981.

SIMA AAF and HINTON D: Hirano-bodies in the distal symmetric polyneuropathy of the spontaneously diabetic BB- Wistar rat. Acta Neurol. Scand. _6£: 107-12, 1983.

SIMA AAF, LATTIMER SA, YAGIHASHI S and GREENE DA: Axo-glial dysjunction. A noval structural lesion that accounts for poorly reversible slowing of nerve conduction in the spontaneously diabetic bio-breeding rat. J. Clin. Invest. 71:474-84, 1986.

SIMA AAF, LORUSSQ AC and THIBERT P: Distal symmetric polyneuropathy in the spontaneously diabetic BB-Wistar rat. An ultrastructural and teased-fiber study. Acta Neuropathol. 58:39-47, 1982.

SIMA AAF and THIBERT P: Proximal motor neuropathy in the BB-Wistar rat. Diabetes 31:784-8, 1982.

SINCLAIR-SMITH BC: The heart in diabetes mellitus. Texas Reports Biol. Med. 39:429-38, 1979.

SINGHAL KC, RASTOGI GK, AIKAT BC and CHUTTANI PN: Testicular histology in diabetics with impotence. Ind. J. Pathol. Bacteriol. 11:145-9, 1969.

SNELL KC: Spontaneous lesions of the rat. In: The Pathology of Laboratory Animals, edited by WE Ribelin, JR McCoy, Springfield, IL:Charles C Thomas, 1963, pp.241-302.

SOUADJIAN JV, MOLNAR GD, SILVERSTEIN MN and TITUS JL: Morphologic studies of the thymus in acromegaly, diabetes mellitus, and Cushing's syndrome. Metabolism 19:401-5. 1970.

SOULAIRAC A, DESCLAUX P and KATZ RF: Action du diabete alloxanique sur le tractus genital du rat male et femelle. Cr Soc. Biol. (Paris) 142:311-2. 1948. 135 SQUIRE RA and LEVITT MH: Report of a workshop on classification of specific hepatocellular lesions in rats. Cancer Res. 35^:3214-23, 1975.

STANDL E, STIEGLER H, JANKA HU and MEHNERT H: Cerebral and peripheral vascular disease. In: Prevention and Treatment of Diabetic Late Complications. Berlin:Walter de Gruyter, 1989, pp.169-98. STERNTHAL E, LIKE AA, SARANTIS K and BRAVERMAN LE: Lymphocytic thyroiditis and diabetes in the BB/W rat. A new model of autoimmune endocrinopathy. Diabetes 30:1058-61, 1981.

STONE BG and van THIEL DH: Diabetes mellitus and the liver. Seminars Liver Dis. 5.: 8-28, 1985. STOUT LC, KUMAR S and WHORTON EB: Insudative lesions - their pathogenesis and association with glomerular obsolescence in diabetes: A dynamic hypothesis based on single views of advancing human diabetic nephropathy. Hum. Pathol. 25:1213-27, 1994. SUFRIN G and PRUTKIN L: Experimental diabetes and the response of the sex accessory organs of the castrate male rat to testosterone propionate. Biochemical and electron microscopic observations. Invest. Urol. .11:361-9, 1974. THOMPSON SW and HUNT RD: Spontaneous tumors in the Sprague- Dawley rat and incidence rates of some types of neoplasms as determined by serial section versus single section techniques. Ann. N.Y. Acad. Sci. 108:832-45. 1963. THOMPSON SW, HUSEBY RA, FOX MA, DAVIS CL and HUNT RD: Spontaneous tumors in the Sprague-Dawley rat. J. Natl. Cancer Inst. 27:1037-57, 1961.

TSCHOEPE D, ROESEN P, SCHWIPPERT B and GRIES FA: Platelets in diabetes: the role in the hemostatic regulation in atherosclerosis. Semin. Thromb. Hemost. 19:122-8, 1993. UEBERBERG H and LUTZEN L: The spontaneous rate of tumours in the laboratory rat: Strain Chbb:TH0M (SPF). Arzneim.- Forsch./Drug Res. 29:1876-9, 1979. 136 VOORBY HAM, KABEL PJ, DeHAAN M, JEUCKEN PHM, Van der GAAG RD, DeBAETS MH, and DREXHAGE HA: Dendritic cells and class II MHC expression on thyrocytes during the autoimmune thyroid disease of the BB rat. Clin. Immunol. Immunopathol. 55:9-22, 1990. WARREN S and LeCOMPTE PM: The Pathology of Diabetes Mellitus. Philadelphia:Lea & Febiger, 1952, pp.208.

WARREN S, LECOMPTE PM and LEGG MA: The Pathology of Diabetes Mellitus, 4th edition, Philadelphia:Lea & Febiger, 1966.

WELLMANN KF and VOLK BW: Islets of Langerhans: Structure and function in diabetes. Pathobiol. Annual 10:105-33, 1980. WESTBERG NG: Diabetic nephropathy. Pathogenesis and prevention. Acta Endocrinologica 94 (Suppl.238):85-101. 1980.

WEXLER BC: Spontaneous coronary atherosclerosis in repeatedly bred male and female rats. Circ. Res. 14:32-43. 1964. WEXLER BC and FISCHER CT: Hyperplasia of the islets of Langerhans in breeder rats with arteriosclerosis. Nature 200:33-7, 1963a. WEXLER BC and FISCHER CT: Abnormal glucose tolerance in repeatedly bred rats with arteriosclerosis. Nature 200:133- 6, 1963b.

WEXLER BC and TRUE CW: Carotid and cerebral arteriosclerosis in the rat. Circ. Res. 12.:659-65, 1963.

WHITE NH, WATTMAN SR, KRUPIN T and SANTIAGO J: Reversal of neuropathic and gastrointestinal complications related to diabetes mellitus in adolescents with improved metabolic control. J. Pediatr. 99:41-5, 1981. WHITEHOUSE FW: Infections that hospitalize the diabetic. Geriatrics 28:97, 1973. WILENS SL and SPROUL EE: Spontaneous cardiovascular disease in the rat. I. Lesions of the heart. Anat. Pathol. 14:177- 216, 1938. 137

WILES PG: Erectile impotence in diabetic men: aetiology, investigation, and management. Diabet. Med. j>:888-92, 1992.

WILLIS RA: Carcinoma of the intestine in rats. J. Pathol. Bacteriol. 40:187-8, 1935.

WRIGHT DG, LAURENO R and VICTOR M: Pontine and extra pontine myelinolysis. Brain 102:361-85, 1979.

WRIGHT J, YATES A, SHARMA H AND THIBERT P: Histopathological lesions of the pancreas of the BB Wistar rat as a function of age and duration of diabetes. J. Comp. Pathol. 95:7-14, 1985. WRIGHT JR Jr: Atrophy, testis, rat. In: Genital System, edited by TC Jones, U Mohr, RD Hunt, Berlin:Springer-Verlag, 1987, pp.218-25.

WRIGHT JR Jr, SENHAUSER DA, YATES AJ, SHARMA HM and THIBERT P: Spontaneous thyroiditis in BB Wistar diabetic rats. Vet. Pathol. 20:522-30, 1983a.

WRIGHT JR Jr, SHARMA H, THIBERT P and YATES AJ: Pathologic findings in the spontaneously diabetic "BB" Wistar rat. Lab. Invest. 42.: 162, 1980 (abstract).

WRIGHT JR Jr, SHARMA HM, YATES AJ, THOMPSON BM and THIBERT P: Supradiaphragmatic accessory lobe of the liver in BB Wistar rats. Teratology 27.: 117-20, 1983b.

WRIGHT, JR Jr, STEPHENS RE, FORD KC, THIBERT P and YATES A: DNA excision repair in spontaneously diabetic BB Wistar rats. Mech. Ageing Develop. 36:109-116, 1986.

WRIGHT JR Jr, YATES AJ, SHAH NT, NEFF JC, COVEY DW and THIBERT P: Hematological characteristics of the BB Wistar rat. Vet. Clin. Pathol. 12:9-13, 1983d.

WRIGHT JR Jr, YATES AJ, SHARMA HM and THIBERT P: Spontaneous gastric erosions and ulcerations in BB Wistar rats. Lab Anim. Sci. .3i.:63-66, 1981.

WRIGHT JR Jr, YATES AJ, SHARMA HM, SHIM C, TIGNER RL and THIBERT P: Testicular atrophy in spontaneously diabetic BB Wistar rats. Am. J. Pathol. 108:72-9. 1982. 138 WRIGHT JR Jr, YATES AJ, SHARMA HM and THIBERT P: Central pontine myelinolysis following saline treatment of a diabetic rat for dehydration. J. Comp. Pathol. 93:509-14, 1983c. WRIGHT JR Jr, YATES AJ, SHARMA HM and THIBERT P: Pathological lesions in the spontaneously diabetic BB Wistar rat: A comprehensive autopsy study. Metabolism 32(suppl.1):101-5. 1983e. YAGAHASHI S and SIMA AAF: Diabetic autonomic neuropathy. The distribution of structural changes in sympathetic nerves of the BB rat. Am. J. Pathol. 121:138-47, 1985a. YAGAHASHI S and SIMA AAF: Diabetic autonomic neuropathy in the BB rat. Ultrastructural and morphometric changes in sympathetic nerves. Diabetes 34:558-64, 1985b. YANG YH, YANG CY and GRICE HC: Multifocal histiocytosis in the lungs of rats. J. Path. Bact. 92:559-63, 1966. FIGURE 4.1

Pituitary chromophobe adenoma in a 327 day old BBWd rat. The lesion was roughly spherical with a large central hematoma. Each large unit on rule is 1 cm. 140

FIGURE 4.2

Photomicrograph of the same adenoma. The tumor appeared somewhat pleomorphic with few mitotic figures. A portion of the neurohypophysis is seen on the left slightly above the lesion. (Reticulin, original magnification x25). 141

FIGURE 4.3 Inflamed seminiferous tubules in the periphery of an otherwise atrophic testis from a 397 day old BBWd rat. Tunica albuginea is markedly thickened. Rat had bilateral orchitis (HE, original magnification x25). FIGURE 4.4 BBWd rat with multifocal granulomatous lesions. Several necrotic foci are present. All seminiferous tubules were totally devoid of germ cells and Sertoli cells. Thickening of tunica albuginea is also present (HE, original magnification x25). FIGURE 4.5 Granuloma with several multinucleated giant cells (HE, original magnification x250). 144

FIGURE 4.6

Acute prostatitis in a BBWd rat (HE, original magnification x25) . FIGURE 4.7

Normal prostate from 551 day old W rat (HE, original magnification x25). FIGURE 4.8

Normal prostate from 551 day old W rat (HE, original magnification xlOO). 147

FIGURE 4.9

Normal prostate from 551 day old W rat. Epithelium is columnar and negative image of the golgi is visible at the apical end of the cell (HE, original magnification x630). Lumens contain pale eosinophilic secretions. 148

FIGURE 4.10

Inactive prostate from 177 day old BBWd rat. Epithelium is low cuboidal to simple squamous. No secretion in the lumens. (HE, original magnifications x25). 149

FIGURE 4.11

Inactive prostate from 177 day old BBWd rat. Epithelium is low cuboidal to simple squamous. No secretion in the lumens. (HE, original magnifications xlOO). 150

7 S 9 1011121314

* - >s •4:,

FIGURE 4.12 Large uterine stromal polyps in 372 day old BBWd rat. Histology showed squamous metaplasia and acute inflammation in the lumen. No inflammatory infiltrates were seen in the stroma. Each unit on rule is 1 mm. 151

FIGURE 4.13 Mammary adenocarcinoma found in milk line near base of tail of a 248 day old BBWnd rat. Tumor measures approximately 3 cm. in diameter. Tumor was firm, white, and lobulated. No metastases were seen. Photograph was taken after the tumor had been shaved to remove the hair. 152

FIGURE 4.14

Photomicrograph of the mammary adenocarcinoma. Epithelium has squamous elements. Ductules contain a deeply eosinophilic secretion (HE, original magnification xlOO). 153

FIGURE 4.15 Idiopathic megacolon in a 327 day old BBWd rat. The rat presented with abdominal distention and anorexia and was treated with enemas. Treatments were temporarily effective. At necropsy, the colon and caecum were observed to be distended (up to 5 cm in diameter) with hard fecal material. Nothing was found constricting the rectum when the pubic symphysis was removed. All tissue in the area of the rectum was examined histologically and was normal. The duration of diabetes in this rat was 205 days. 154

FIGURE 4.16 412 day old BBWd rat with erosions and ulcerations in both the thick-walled glandular portion and the thin-walled squamous portion of the stomach. The glandular portion is severely hemorrhagic. The lesions in the squamous forestomach are surrounded by white, raised borders. 155

FIGURE 4.17

Small erosion in the thin-walled squamous portion of the

stomach (forestomach). Area of coagulation necrosis extends to the muscularis mucosa. Epithelial surface has sloughed. (HE, original magnification xlOO). 156

FIGURE 4.18

Severe hepatic fatty change in the region of several portal triads in a 150 day old BBWd rat. Duration of diabetes was only 57 days. (HE, original magnification x250). 157

FIGURE 4.19

Several cysts consisting of dilated tubules filled with eosinophilic proteinacious fluid were seen in the kidney of a 412 day old BBWnd rat. The cyst just to the right of center is associated with focal chronic inf lamination of the tubule. This pattern was seen in several rats and is typical of some of the changes seen in chronic renal disease in the rat. Most cysts in rat kidneys were not inflamed. (HE, original magnification xlOO). 158

FIGURE 4.20

A small focal subcapsular granuloma in the renal cortex of a

200 day old BBWd rat (HE, original magnification xlOO). FIGURE 4.21 Chronic interstitial nephritis in a 210 day old BBWd rat

Mononuclear infiltrates, interstitial fibrosis, tubular atrophy, and relative sparing of glomeruli are prominent (HE, original magnification xlOO). FIGURE 4.22

Gross specimens from different lobes of a 325 day old BBWd rat with severe bilateral bronchopneumonia. The cut surfaces of the specimens on the left and right reveal large abscesses. The two center specimens show pleural adhesions. 161

FIGURE 4.23

Severely congested, edematous lung from a 192 day old BBWd rat. Presence of neutrophils in the alveoli indicate bronchopneumonia (HE, original magnification x250). 162

FIGURE 4.24

Murine chronic respiratory disease in a young W rat. Main features seen in this photomicrograph are peribronchial lymphoid hyperplasia, perivascular lymphoid aggregates, and chronic pneumonitis (HE, original magnification x25). 163

FIGURE 4.25

551 day old W rat with limited multifocal chronic myocarditis-predominately in the left ventricle. Antischkow cells are prominent in the chronic infiltrate. This picture is typical of myocardial degeneration in aged rats. (HE, original magnification x250). 164

FIGURE 4.26

Pregnant female BBWnd rat (age unknown) with vasculitis in all organ systems. A necrotizing vasculitis was seen in the pancreas (HE, original magnification x250). 165

FIGURE 4.27

Several less involved vessels were seen in the submucosa of the stomach (HE, original magnification xlOO). Other lesions in this rat included a small organizing mural thrombus in the heart, disseminated intravascular coagulation in the kidney (see Figure 4.29), severe extramedullary hematopoiesis in the spleen, and an infarct in a mesenteric node. Russell bodies were very prominent in the lymph nodes. The rat was also pregnant. 166

FIGURE 4.28 Focal vasculitis in the colon of a 320 day old BBWd rat. Histiocytes are seen predominantly in the intima and media and eosinophils are in the adventitia. This rat also had systemic eosinophilia. (HE, original magnification xlOO). 167

FIGURE 4.29 Two glomeruli with large microthrombi in the kidney of a near-term pregnant BBWnd rat (unknown age) which died with disseminated intravascular coagulation (DIC) secondary to the widespread vasculitis. (HE, original magnification x400) 168

FIGURE 4.30

Hyperplastic mesenteric lymph nodes from 320 day old BBWd rat. Nodes are enlarged and fused. Normal nodes are small and discrete. 169

*

FIGURE 4.31

Typical lymph node from a 300 day old W rat with prominent follicular germinal centers and abundant lymphocytes (HE, original magnification xlO). 170

FIGURE 4.32

Hyperplastic lymph node from a 412 day old BBWd rat. Sinusoidal dilatation and lymphocytic depletion are prominent histologic features. (HE, original magnification xlO). 171

FIGURE 4.33

Necrotizing granuloma in a pancreatic lymph node from a 125 day old BBWd rat. Additional sections were stained with acid fast blue, GMS, and gram stain. No organisms were seen. (HE, original magnification x25). 172

FIGURE 4.34

Lymphoma mass in a 334 day old female BBWd rat found at the junction of the cecum and colon. Tumor is firm, lobulated, and hemorrhagic. Tumor was also present in colon and pancreas. 173

FIGURE 4.35 Histologically, the lymphoma is diffuse with a total loss of normal nodal architecture. (HE, original magnification x25). 174

FIGURE 4.36

The tumor is histiocytic with plasmacytoid features. Several binucleated plasma cells are shown as well as a mitotic figure in the lower left corner. (HE, original magnification x400). 175

FIGURE 4.37 This lymphoma was found in a 248 day old BBWd rat. The tumor mass was found at the junction of the ileum and cecum. The three specimens in the photograph are organized proximal to

distal from left to right. The lumen is totally occluded in the specimen on the left. Each large unit on the rule is 1 cm. 176

FIGURE 4.38

Photomicrograph of the area on the left. The neoplasm in the submucosa of the ileum is a histiocytic lymphoma with plasmacytoid features. (HE, original magnification x25). 177

FIGURE 4.39 Single, small, well-circumscribed area of necrosis in the proximal ileum (50 cm from cecum) of a 315 day old BBWd rat. Normal mucosa seen to right of necrotic area. Histological examination revealed much necrosis and a histiocytic lymphoma with plasmacytoid features. The adventitia of the colon was also involved, but this was not grossly visible. Lymph nodes were grossly large and histologically hyperplastic, but no tumor was seen. Spleen showed severe extramedullary hematopoiesis, but was not markedly enlarged. 178

FIGURE 4.40

Lymphoid depletion in the spleen of a 320 day old BBWd rat. Distinction between red and white pulp is obscured. (HE, original magnification x25) 179

2 13 11 15 | (> 17 IS

FIGURE 4.41 Spleen from a BBWnd rat with a lymphoma (age unknown).

Spleen measures 7 cm long and weighs 7.5 gms. Normal values are 2-3 cm and 0.5-0.75 gm. Splenomegaly is a result of severe extramedullary hematopoiesis and tumor infiltration. 180

FIGURE 4.42

Massive subcutaneous spindle cell sarcoma in a BBW rat of unknown age. 181

FIGURE 4.43 Massive eosinophilic infiltrate in the submucosa of the colon of a 350 day old BBWd rat. The muscularis externa is also involved. Eosinophilic infiltrates were also seen in the other organs of this rat. (HE, original magnification xlOO). Appendix A

182 183

J . C o m i*. Pa i i i . IOH'l V o l . 'Jfi.

H1STOPATHO LOGICAL LESIONS IN THE PANCREAS OF THE BB YVISTAR RAT AS A FUNCTION OF AGE AND DURATION OF DIABETES

By

J . Y V r i g h t , A. Y a t e s , * H . S i i a r m a a n d P. T i m b e r t The Ohio State University College o f Medicine, Department o j I'athology, Columbus, Ohio 43210, U.S.A. and Animat Resources Division, Health I ’rotection ilranch. Health and Welfare Canada, Ottawa, Ontario, Canada A /A 0 1 2

INTRODUCTION The BB Wistar (BBW) rai, a useful animal model for juvenile-onsei diabetes (JOD), was developed in the late 1970s from an oulbred line of Wistar rats at the Bio Breeding Laboratories in Ottawa, Canada by Nakhooda, Like, Chappel, Murray and Marliss (1977). The BBW diabetic syndrome occurs spontaneously in the absence of obesity and is characterized by hypcrglycaemia, glycosuria, ketoacidosis, insulinopenia, glucagonaemia, and hyperlipaemia (Nakhooda el a l., 1977; Nakhooda, Like, Chappel, Wei and Marliss, 1978). Although the morphology of the pancreatic islets has been extensively studied in 60- to 120-day-old BBW rats at or near the time of onset of diabetes (Nakhooda el a l., 1977, 1978; Scemayer, Tannebaum, Goldman and Colic, 1982), (here has been no published report describing the incidence of pancreatic lesions in older BBW rats. In this study we have examined the incidence of both endocrine and exocrine pancreatic lesions in BBW Wistar diabetic (BBWd) rats, their non-diabetic siblings (BBWnd), and outbred Wistar rats.

MATERIALS AND METHODS A breeding colony of BBW rats was established in the Pathology Department at The Ohio Slate University from the parent colony at the Health Protection Branch of Health and Welfare of Canada. Detailed descriptions of our animal housing facilities and animal husbandry techniques have been reported previously (Wright, Yates, Sharma and Thibert, 1981). A necropsy study was initiated to determine the incidence and types o f lesions in our colony. Some o f the results o f these necropsies have been described elsewhere (Wright, Sharma, Thibert and Yates, 1980; Wright, Yates, Sharma and Thibert, 1983a). Complete necropsies were performed on 121 BBW d, 43 BBW nd, and 33 outbred Wistar rats. A ll major organs including pancreas were grossly examined and (ixed in either neutral phosphate buffered formalin or Bouin’s fluid. The pancreas was removed intact and spread over a piece of cardboard to maximize surface area during fixation. After fixation, specimens were removed from the cardboard, and divided into 2 portions. The entire pancreas and representative portions of other organs were

* Correspondence lo: D r A. J. Yales, Division of Neuropathology, The Ohio Siaie University, 111 Upliain Hall, 473 West I2lh Avenue, Columbus, Ohio 43210, U.S.A.

0021-9975/85/010007 + 08 803.00/0 IQ 1985 Academic 1'rcss Inc. (London) Limited j. w rigiit et al.

processed for liglil microscopy. Tissue seciions were routinely stained with hacmatoxy- lin and eosin (11 and Ef. Paraflin seciions o f specimens suspected o f antemortein infection were stained with acid fast and Gram’s slain, and Gomori’s methenamine silver (C M S ). M ultiple sections of all pancreas specimens were examined by light microscopy. Statistical analyses were by analysis of variance, £-icst, or chi-square test.

RKSUL'I'S Table 1 shows the incidence of inflammatory pancreatic lesions in BBWd, BBWnd, and W rats. Iusulitis was the most common inflammatory pancreatic lesion in both BBWd (14*0 per cent) and BBWnd (16*3 per cent) rats and is characterized as a mononuclear cell infiltrate involving the islets. In the majority, the infiltrate was predominantly peri-insular (Fig. 1) but the most severe lesions also included marked invasion of the islets (Fig. 2). A few eosinophils were often present at the periphery oflhc inflamed islets. The ages of BBWd rats with insulitis ranged from 107 to 317 days of age (mean ±S.D .= lG4±6l-9 days) but the change occurred most frequently in the younger age groups (Table 1). In the BBWd it was seen only in those rats that died relatively shortly after the onset of hypcrglycacmia and so its incidence was inversely related to the duration of diabetes (^ = 33-4, d.f. = 5, TcO-OOl) (Table 1). The longest period of lime following the onset ofdiabetes for which a BBWd rat still had severe insulitis was G9 days; the mean duration of diabetes for BBWd rats with insulitis when they died was 49-9 ± 16-9 days. In the non-diabetic rats, there was no specific age predilection for insulitis. It was observed in BBWnd rats ranging from 9G to 422 days of age (m can± S.D. = 306-5 ± 132-4). Only one BBWnd rat with insulitis was less than 120 days old. Several other types of inflammatory lesions also occurred (Table 1). Chronic interstitial inflammation (inflammation with little acinar or insular involvement) was the most frequent of these, occurring in all 3 groups of rats al incidences between 11 and 14 per cent. No other types of pancreatic lesions were seen in outbred Wistar rats. In BBW rats chronic pancreatitis (chronic inflammation with diffuse, extensive acinar involvement), acute pancreatitis, acute and chronic pancreatitis, acute interstitial inflammation, and acute and chronic interstitial inflammation all occurred but were not common. In the acute lesions eosinophils were often present either as the predominant cell type - (Fig. 3) or mixed with neutrophils. No age pattern was discernible for any of the inflammatory lesions. Granulomas were occasionally observed in BBW pancreatic tissue (Table 1). In no instance could any organisms be demonstrated with Gram, GMS, or acid fast stains. In 2 rats, granulomas and insulitis were present in close approximation (Fig. 4). Infiltrates of eosinophils were frequently associated with granulomas. In one rat several granulomas consisting almost entirely of reticuloendothelial cells and eosinophils were present in the pancreas. Islet morphology in BBWd rats without insulitis was markedly altered from that of young pre-diabetic BBW rats. Shortly after the onset of glycosuria, islets from BBVVd rats typically contained numerous endocrine cells with acidophilic cytoplasm and pyknotic nuclei (Fig. 5). Occasionally, hyperplastic TABLE A .l

AGE INCIDENCE OF PANCREATIC LESIONS IN BB WISTAR RATS*

BBW d rats BBW nd rets

Oto 120 121 to 240 211 to 360 36 / to 480 Oto 120 121 to 240 241 to 360 361 to 480 481 to 600 Ltsion days days dojs doys S ot tnoimt Totol days days dots dots dots Sot known! Totol

Normal 3 22 34 13 1 73 5 3 6 1 0 6 22 Insulitis* 2 13§ 2 0 0 17 1 0 2 3§ 0 1 7 Chronic pancreatitis 0 2 2 0 0 4 0 0 0 I 0 0 1 Acute pancreatitis 1 1 1 0 0 3 0 0 0 0 0 0 0 Acute and chronic pancreatitis 0 1 2 0 0 3 0 0 0 0 0 0 0 Chronic inflammation 1 2 7 4 0 14 1 3 1 0 1 0 6 Acute inflammation 0 2 0 1 0 3 0 0 0 0 0 0 0 Acute and chronic inflammation 0 0 3 0 0 3 0 1 0 0 0 0 1 Granuloma 0 15 1 0 0 2 0 0 0 i§ 0 0 1 n 7 44 52 18 1 *122 7 7 9 12 1 7 36

* Except for 4 rats with chronic non-specific interstitial inflammation, no pancreatic lesions were seen in 33 W rats. * Precise age not known but all were adults, at least 120 days o f age. * The incidence of insulitis in BBWd rats was inversely related to the duration of diabetes ( js* 3 3 -4 l,d .£ *5 , P < 0 ’0001). The incidence in the following duration groups was: 0 to 60 days {14 in 32; 61 to 120 days (3 in 26); 121 to 160 days (0 in 17); 181 to 240 days (0 in 25); 241 to 300 days (0 in 12); and 301 to 360 (0 in 8). §Both findings present in same rat. 185 J. W R IG H T ft al.

Fig. A. 1 Insulitis in a 115-day-old BBWnd rat. The mononuclear infiltrate is predominantly peri-insular with minimal islet invasion. Surrounding acinar tissue is not involved. HKxSO.

Fig. A .2 Islet from a 135-day-old BBWd rat showing marked invasion by mononuclear cells. Onset ofdiabetes was 55 days earlier. H K x 125.

Fig. A .3 Periductal eosinophilic infiltrate in the pancreas of a 2U0-day-old BBWd rat. HK x 125.

F ig . A . t> Pancreatic granuloma in a 3U5-day-old BBWnd rat. An islet in the upper left corner is surrounded by a predominantly lymphocytic infiltrate. Granuloma is also surrounded by lymphocytes. HK x 125.

islets were seen in young diabetic rats (Fig. 6). Islets from chronically diabetic BBW rats were generally small, decreased in number, and composed almost entirely of non-beta cells (Scemayer el al., 1982) (Fig. 7).

DISCUSSION

A wide variety of exocrine pancreatic lesions has been reported in human diabetics, but the significance of most of these remains obscure. Fibrosis, often with some degree of chronic-interstitial inflammation, is the most common finding, occurring in 50 to 60 per cent of necropsy cases (Warren, LeGomple and Legg, 1966; Kotharc, 1974). Pancreatitis, although an uncommon cause of diabetes, is seen in approximately 5 per cent of diabetic patients at necropsy (Goto, Sato and Masuda, 1974; Parson, MacDonald and Shaper, 1968). Other frequent findings include fatly infiltration and arteriosclerosis of the pancreas (W arren el a l., 1966). The types of pancreatic exocrine lesions in the BBW rat ('l'able 1) are PANCREATIC LESIONS IN DIAUETIC RATS

Fig. A.5 Islet frum a 137-duy-uld BllW d ral showing islet cells with pyknotic nuclei and acidophilic cytoplasm and intact islet cells. No leukocytic infiltrates were seen in any islets in this rat. Onset ofdiabcles was 47 days earlier. HE x 200.

F ig . A. 6 One of many hyperplastic islets in a 178-day-old UUWd ral showing a peri-insular mononuclear infiltrate. Onset ol'hyperglyeaeinia was 00 days earlier. HK x SO.

Fig. A.7 Atypical small cnd-stage islet consisting of only a few cells, and without evidence of necrosis. H E x 200.

generally similar to those in human diabetics. Chronic interstitial inflammation is common in BBWd rats but usually lacks the prominent fibrosis seen in man. However, it is equally common in non-diabetic BBW and Wistar rats, and is also common in many other strains of laboratory rats (Kendrey and Roe, 1969). Acute or chronic pancreatitis was most common in BBWd rats ( z = 1-34, /)=0-10) occurring in 10 BBWd (8-3 per cent), 1 BBWnd (2-3 per cent), and in none of the Wistar rats. Pancreatic granulomatous lesions were seen in approximately 2 per cent of the BBWd and BBWnd rats but were not present in Wistar rats. In several instances, the granulomas were in close approximation with insulitis. Granulomas were also common in other organs (Wright et a l., 1980; Wright el al., 1983a). The occurrence of spontaneous granulomatous lesions in the absence of infectious organism is consistent with the findings of others which suggest that the BBW rat has an abnormal cellular immune system (Jackson, Rassi, Crump, Haynes and Eisenbarth, 1981; Like, Rossini, Bugcrski, Appel and Williams, 1979; Like, Kislanskis, Williams and Rossini, 1982). Eosinophils were often present in the periphery oflhe granulomatous lesions. Indeed, the j. WHiciiT el al.

presence of an eosinophilic component in both the acute and the chronic pancreatic lesions in BBW rats was often striking. This is interesting in the light of several previous studies which show that (1) infiltrates of eosinophils are common in other organs of BBWd rats (Wright el al., 1980; Wright el al., 1983a), (2) marked cosinophilia (in the absence of parasitism) is frequently seen in the peripheral blood of BBWd and BBWnd rats but does not occur in the outbred Wistar line from which BBW rats were derived (Wright, Yates, Shah, Neff, Covey and Thibert, 1983b) and (3) small numbers of eosinophils are occasionally present in the infiltrates in insulitis in BBW rats (Nakhooda el a l., 1977). No eosinophilic infiltrates were observed in any of the Wistar rats. The significance ofeosinophilia in BBW rats is not known but is consistent with hypersensitivity to some unknown allergen or auto-antigen. No endocrine or exocrine pancreatic tumours were observed in this study, and are very rare in all strains of rats (Altman and Goodman, 1979). Lymphomas are relatively common in BBW rats (Kaiant and Seemayer, 1979) and lymphomatous infiltration of the pancreas was seen in two rats in our scries (W right el a l., 1983a). Insulitis is a lymphocytic infiltration selectively involving the pancreatic islets. It is frequently observed in some but not all islets of JO D patients who die within one year of clinical onset (Gepts, 1977). In most cases, only small lymphocytes arc seen, but the presence of macrophages, plasma cells, eosinophils, and neutrophils has also been reported (Wellmann and Volk, 1980). BBW rat islets typically show 3 general patterns: (1) “end-stage” islets in chronically diabetic rats are small, decreased in number, composed almost entirely of non-beta cells and a few granulated beta cells; (2) mononuclear cell insulitis is usually seen at the onset of diabetes and in the subsequent few weeks; (3) normal islet histology is typical of pre-diabetic and non-diabetic BBW rats (N akhooda el a l., 1978). This picture is confirmed by our study and that of others (Seemayer el a l., 1982). Most of our BBWd rats, regardless of the severity of diabetes, had at least a few typical small “end-stage” islets. These had numerous cells with acidophilic cytoplasm and pyknotic nuclei. Occasionally, hyperplastic isleis suggesting functional hyperactivity were seen in the pancreas of young diabetics. Most of these had died shortly after the onset of glycosuria, but some had been diabetic for several months. One of the latter also had insulitis. Seem ayer el al. (1982) found an orderly step-wise progression in the development of insular lesions which closely correlated with the onset of diabetes. They observed insulitis in none of four 50-day-old and 5 of seven 65- day-old BBWnd (“young normoglycemic”) rats. This suggests that the destructive process in the islets starts in advance of the clinical syndrome, but it has not been definitely established that glycosuria is always preceded by insulitis. They observed severe insulitis in “early” diabetic rats (1 to 3 days post-detection) but only end-stage islets and a few residual mononuclear cells in “unstable” and “stable” (7 to 22 and 41 to 63 days posl-delcction, respectively) BBWd rats. In contrast, our study reveals that insulitis may be quite prominent several months after the onset of glycosuria. A jiossible l’ANCREATIC LESIONS IN DIABETIC RATS

explanation for the dilfcrcnce between these 2 studies is that the rat populations were different. Seemayer et al. (1982) utilized small numbers of a relatively young population of rats over narrow age ranges for each group. All were the offspring of only 2 breeding pairs. Our rats were all from the same colony but most were not liltermales and there was a much wider range of ages at the onset of diabetes. To determine whether age of onset had any correlation with the persistence of insulitis, we examined all 45 rats that died or were killed within 70 days of onset of glycosuria (the maximal duration of diabetes in this study at which a rat was found to have severe insulitis). The mean age of onset of glycosuria in rats with persistent insulitis was 153* 1 ± 71 ’0 days, and in those without persistent insulitis was 129,8:fc58,7 days. There is no statistically significant difference in the ages of these groups [/•'( 1,18) = 0-62, n.s.]. Insulitis was observed in 16*3 per cent of our BBWnd rats. Nakhooda el al. (1978) also reported an incidence ranging from 17 to 25 per cent in BBWnd rats. This indicates that insulitis does not always cause the immediate onset of clinically apparent diabetes mcllilus. However, some rats may have progressive insular involvement which in time could lead to progressively more severe hypoinsulinacmia. Indeed, many older BBWnd rats do eventually convert into BBWd rats (Wright et a l., 1983a).

SUMMARY Pancreatic hislopalhology was studied in 121 BBWd, 43 BBWnd, and 33 Wistar rats. Insulitis was the most common inflammatory lesion in both BBW and BBWnd rats. The incidence was inversely associated with age and with duration of diabetes in BBWd rats, but there was no age-related pattern in BBWnd rats. Small end-stage islets were typical of BBWd rats but were not seen in BBWnd rats. Several BBWd rats showed hyperplastic islets months after the onset of diabetes, a pattern that is also seen in a small percentage of human JO D patients. Several non-specific exocrine inflammatory lesions occurred in both BBWd and BBWnd rats: acute and/or chronic pancreatitis, eosinophilic infiltrates, granulomatous lesions and acute and/or chronic interstitial inflammation. Only chronic interstitial inflammation was seen in outbred Wistar rats.

acknowledgments This work was supported by the Department of Pathology, the College of Medicine and the Graduate School of The Ohio Slate University; the Upjohn Company and N IH grant NS-18026. J. R. Wright, Jr was supported by the Samuel J. Roessler Research Scholarship. The authors would like to thank R. L. Tigner for statistical assistance, D . Covey and B. Thompson for their excellent maintenance o f the BBW colonies in Columbus and O ttaw a, and Pamela H orn for typing the manuscript. The authors wish to thank Drs T. Yasui, Y, Igarashi and S. Mikawa of The Regional Livestock Hygiene Service Centers of Hokkaido for providing help in the bacteriological examination.

REFERENCES

Altman, N. H., and Goodman, D. G. (1979). Neoplastic diseases. In The iMboratory Ral. Vol. 1: Biology and Diseases. H .J. Baker, J. R. Lindsey, and S. H. Weisbroth, Eds, Academic Press, New York, pp. 336-376. j . W r i g h t el al.

Gepts, W . (1977). Sequential changes in the eytological composition o f ihe pancreatic islets in juvenile diabetes. In Diabetes. Proceedings o f the IX lh Congress o f the International Diabetes Federation. J. S. Bajuj, Ed., Excerpta Medica, Amsterdam, pp. 299-30C. Goto, Y., Sato, S. 1., and Masuda, M . (1974). Causes ofdeath in 3151 diabetic autopsy cases. Tohoku Journal of Experimental Medicine, 112, 339-353. Jackson, R., Rassi, N., Crump, T., Haynes, B., and Eiscnbarlli, G. S. (1981). The BB diabetic rat: profound T-cell lymphocytopenia. Diabetes, 30, 887-889. Kalant, N., and Seemayer, T . (1979). Malignant lymphoma in spontaneously diabetic rats. New England Journal of Medicine, 300, 737. Kendrey, G., and Roe, F. J. C. (1969). Histopathological changes in the pancreas of laboratory rats. Ijiboralory Animal, 3, 207-220. Kothare, S. N. (1974). Pathological lesions in diabetes mellilus - their changing concepts. Indian Journal of the Medical Sciences, 28, 103-107. Like, A. A., Rossini, A. A., Bugcrski, D. L., Appel, M . C., and Williams, R. M . (1979). Spontaneous diabetes mcllitus: reversal and prevention in the BB/W rat with antiserum to rat lymphocytes. Science, 206, 1421-1423. Like, A. A., Kislanskis, E., Williams, R. M., and Rossini, A. A. (1982). Neonatal thymectomy prevents spontaneous diabetes mellitus in the BB/W rat. Science, 216, 644—646. Nakhooda, A. E., Like, A. A., Chappel, C. L, Murray, E. T., and Marliss, E. B. (1977). The spontaneous diabetic Wistar rat. Metabolic and morphologic studies. D ia­ betes, 26, 100-112. Nakhooda, A. E., Like, A. A., Chappel, C. I., Wei, C. N., and Marliss, E. B. (1978). The spontaneously diabetic Wistar ral (the “BB” rat). Studies prior to and during development of the overt syndrome. Diabelologia, 14, 199-207. Parson, W., MacDonald, F. W., and Shaper, A. G. (1968). African diabetics necropsied at Mulago Hospital, Kampala, Uganda 1957-1966. East African Medical Journal, 45, 89-99. Seemayer, T. A., Tannebaum, G. S., Goldman, H., and Colle, E. (1982). Dynamic lim e course studies of the spontaneously diabetic BB W istar rat. 111. Light- microscopic and ultrastruclural observations of pancreatic islets of Langerhans. American Journal of Pathology, 106, 237-249. Warren, S., LcCompte, P. M., and Legg, M. A. (1966). The Pathology of Diabetes Mellitus, 4th Edit., Lea and Febiger, Philadelphia, pp. 90-115. Wellmann, K. F., and Volk, B. W. (1980). Islets of Langerhans: structure and function in diabetes. Pathobiology Annual, 10, 105-133. Wright, J. R., Sharma, H., Thibert, P., and Yates, A. J. (1980). Pathologic findings in the spontaneously diabetic BB Wistar rat. Laboratory Investigation, 42, 162. Wright, J. R., Yates, A. J., Sharma, H. M., and Thibert, P. (1981). Spontaneous gastric erosions and ulcerations in the BB W istar rat. Laboratory Animal Science, 31, 63-66. Wright, J. R., Yates, A. J., Sharma, H. M ., and Thibert, P. (1983a). Pathological lesions in the spontaneously diabetic BB W istar ral. A comprehensive autopsy study. Metabolism, 32 (Suppl. 1), 101-105. Wright, J. R., Yates, A. J., Shah, N. T., NefT, J. C., Covey, D. W., and Thibert, P. (1983b). Hematological characteristics of the BB Wistar rat. Veterinary Clinical Pathology, 12, 9-13.

[Received fo r publication, August 26th, 1983] Appendix B 192

Vet. Pathol. 2 0 : 522-530 (1983)

Spontaneous Thyroiditis in BB Wistar Diabetic Rats

J. R. W right, Jr., D. A. S lniiausi:r, A. J. Y a t l s , H. M . S h a r m a , and P. T h i h l r t

Department o f Pathology, Ohio State University College of Medicine, Columbus, Ohio, and Animal Resources Division, Sir Frederick G. Banting Research Centre, Health Protection Branch, Health and Welfare Canada, Ottawa, Ontario, Canada

Abstract. Spontaneous lymphocytic thyroiditis was observed al necropsy in 36 BB Wistar diabetic rats (63.2%) and in eight o f their nondiabetic siblings (42.1%). The incidence of thyroiditis decreased both with age and with longer duration of diabetes. All rats with pancreatic insulitis (a manifestation o f the onset of diabetes) also had thyroiditis. BB Wistar rats with insulitis had more severe lymphocytic thyroiditis, characterized by lymphocytic, plasmacytic, and macrophage infiltration o f thyroid interstitium and follicles. A milder, mostly perivascular and interstitial lymphocytic thyroiditis was characteristic o f lesions in rats which did not have insulitis. The histological appearance o f the thyroiditis suggests that these rats may be subject to autoimmune disease at the onset o f diabetes which involves sites other than just the pancreas.

Although the exact mcchanism(s) responsible for the onset of human juvenile- onset diabetes is unknown, cell-mediated immunity probably is involved [2, 21]. This is supported by the occurrence of a mononuclear cell insulitis that selectively destroys the insulin-producing beta cells in the pancreatic islets concurrent with the onset of the diabetic syndrome in both the BB Wistar rat, a promising new animal model for human juvenile onset diabetes [19, 20], and man [8, 34]. The BB Wistar diabetic syndrome occurs in the absence of obesity and is characterized by hyperglycemia, glycosuria, ketoacidosis, insulinopcnia, glucagonemia, and hy­ perlipemia [19, 20, 22]. Spontaneous diabetes occurs in 30 to 55% of the offspring of diabetic matings. Onset is abrupt with rapid weight loss due to dehydration as a result of marked polyuria. In the absence of insulin therapy, death usually follows the onset of symptoms by one to two weeks. Although the pancreas of the BB Wistar rat has been studied extensively, relatively little is known about spontaneous diseases in other organs of this animal, therefore, we conducted an extensive necropsy study. Our results and the results of others have shown that the BB Wistar ral also develops a wide variety of inflammatory, degenerative, neoplastic, and developmental disorders [14, 29, 36- 41]. A significant proportion of these rats develop spontaneous lymphocytic thyroiditis—a finding that recently has been reported by another group [30], To SiJomaneous Thyroiditis in BBW Rats

Table B .l Character of thyroiditis in BB Wistar rats' Grade of Infiltrate interstitial Follicular Bilateral cell types3 infiltrate involvement lesions 0 1 2 3 L PC M All Diabetic 21 22 9 5 10 2 12 36 I I 12 7 Nondiabetic 11 7 1 0 1 4 8 4 1 1 1 Each value represents the number of rats in that category. Thyroids were examined in 57 diabetic and 19 nondiabetic rats. Histological evidence of thyroiditis was found in 36 diabetic and nine nondiabetic rats. 2 Only one lobe o f the thyroid was examined in three rats. 3 L = lymphocytes; PC = plasma cells; M = macrophages. All = all three cell types present. date there has been no histological description of this entity, nor has the incidence of thyroiditis been examined in relation to the age of the rats.

Materials and Methods A breeding colony of BB Wistar rats was established in the Pathology Department at The Ohio State University from the parent colony at the Health Protection Branch of Health and Welfare o f Canada. Detailed description o f our animal housing facilities and animal husbandry techniques has been reported previously [40]. Complete necropsies were done on 57 diabetic rats and 19 of their nondiabclic siblings which cither died spontaneously or were sacrificed for experimentation. All major organs were examined grossly, fixed in neutral phosphate buffered formalin, and processed for light microscopy. Tissue sections were stained with hematoxylin and eosin (HE). At least four sections from different pans o f the thyroid were examined in each rat. Thyroids were graded as cither normal or one o f three grades of severity of thyroiditis based on the extent of interstitial infiltrate. The presence of follicular involvement was noted when present. Lesions were determined to be either bilateral or unilateral— lesions involving only the isthmus were considered unilateral. The presence of ultimobranchial duct cysts also was tabulated. All statistical analyses were by chi-square test, binomial test, or one-tailed z-tcst.

Results Lymphocytic thyroiditis was observed in 36 diabetic rats (63.2%) and in eight nondiabetic rats (42.1%). Inflammatory lesions were bilateral in two-thirds of the diabetic rats and one-half of the nondiabetic rats (table I). Most of the lesions were characterized as predominantly perivascular, mild focal, or multifocal interstitial lymphocytic thyroiditis (grade 1). The most severe lesions (grade 3) were multifocal and involved over 25% of the intcrstitium. Those lesions intermediate in severity between grades 1 and 3 were classified as grade 2. The higher grades were less frequent and were characterized by extensive multifocal inflammatory infiltrates (fig. 1) frequently consisting of lymphocytes, plasma cells, and macrophages (fig. 2, 3). Follicular invasion was more common in these lesions (fig. 4). The latter type of lesion was characteristic of BB Wistar rats with insulitis. It is interesting that nine rats (seven diabetic and two nondiabetic) had insulitis, and all nine also hud Wright <7 al.

F ig . B. 1 Extensive multifocal mononuclear cell infiltrate in BB Wistar rat thyroid tissue. . Infiltrate is predominantly interstitial and perivascular. HE. F ig . B.2 Large numbers o f plasma cells and plasmacytoid lymphocytes composing some of the infiltrates. Infiltrate is in close association with thyroid follicles with some disruption of follicular basement membrane. HE. F ig . B.3 Mixed chronic inflammatory infiltrate adjacent to several thyroid follicles. F ig . B .4 Extensive follicular involvement by inflammatory infiltrate with apparent disrup­ tion of follicular architecture and dissolution of thyroid epithelial cells in the most severely affected thyroids. HE. pnaeu Thyodts n B Rats BBW in yroiditis h T Spontaneous

Table B.2 Incidence o f lymphocytic thyroiditis in BB Wistar rats by age and by duration of diabetes Age or duration o f diabetes in days

0-120 121-240 241-360 360-480 481-800 Diabetic Age 2/2 (1009c) 14/20 (70%) 17/27 (63.0%) 3/8 (37.5%) _ Duration 19/27 (70.4%) 13/21 (61.9%) 4/9 (44.4%) — --- Nondiabetic Age 1/2(50.0%) 0/2 1/2 (50.0%) 3/6 (50%) 0/1 Wright <•/ al.

thyroiditis (binomial test, p = 0.002). Furthermore, the severity of the thyroiditis, as manifested by the extent of interstitial involvement (X2 = 43.19, df = 3, p < .001) and follicular involvement (X2 = 16.73, df = 1, p < .001), was much greater in rats with insulitis. Rats with insulitis also had multiple inflammatory cell types present in the thyroid lesion more frequently (X2 = 20.57, df = 2, p < .001). Table II shows that the incidence of thyroiditis in diabetic rats decreases with age and with duration of diabetes, but this relationship was not statistically significant since most of the rats were in the two middle age groups. No age pattern was apparent in nondiabctic rats. Ultimobranchial duct cysts were observed in 13 diabetic (22.8%) and three nondiabctic (15.8%) rats. The margins of these cysts had a prominent chronic inflammatory infiltrate in six of the diabetic rats.

Discussion The histologic appearance of thyroiditis in BB Wistar rats is consistent with the features of autoimmune thyroiditis described in man [26, 32]. Although occasional parenchymal cells were enlarged and had eosinophilic cytoplasm, no unequivocal Askanazy cells were seen. Although Askanazy cells are a prominent feature in the adult form of Hashimoto's thyroiditis in man, they usually are not seen in all forms of human autoimmune thyroiditis [32] nor are they a prominent feature of experimentally induced or spontaneous autoimmune thyroiditis in other rat studies [10, 23]. Although the BB Wistar rat is very susceptible to infection [36, 38, 41], an infectious etiology for the thyroiditis seems unlikely because of the absence of acute thyroiditis, suppuration, or subaculc thyroiditis, features associated with infectious thyroiditis [32], and the absence of involvement of adjacent tissues. Furthermore, the rats were housed in a semi-barrier housing facility to minimize exposure to infectious agents. No thyroid lesions (except cysts) were seen in outbred Wistar rats kept in the same facility. Although spontaneous autoimmune thyroiditis is not rare in man, there are few naturally occurring animal models [4, 24]. Inbred strains of laboratory animals which develop lymphocytic thyroiditis include Obese Strain chickens [35], labora­ tory pure-bred beagles [9], marmosets of the genus C a lla lh rix [16], and the Buffalo strain rat [10]. Although the latter is the only strain of the rat which develops spontaneous thyroiditis, it has been induced in rats by several means of immuno­ suppression including chemical agents [15, 28], thymectomy [23, 28], and irradia­ tion [23]. This is particularly relevant because of findings showing that the BB Wistar rat is naturally immunocompromised [13, 38]. In an earlier study, a 59% incidence of lymphocytic thyroiditis in BB Wistar diabetic rats and an 11% incidence in their nondiabetic siblings was reported [30], but the functional significance of this lesion is uncertain since serum T3 and T^ levels were normal in rats with thyroiditis. In a second study, the effect of iodide supplementation on the incidence of thyroiditis in BB Wistar rats was examined Spontaneous Thyroiditis in BBW Rats

[31]. There was a slightly higher incidence (not statistically significant) of lympho­ cytic thyroiditis in diabetic and nondiabetic rats receiving tap water with iodide than in those receiving tap water without iodide supplementation. There are several pertinent facts concerning the pattern of occurrence of thyroid­ itis in BB Wistar rats that arc compatible with an autoimmune pathogenetic mechanism. First, the incidence is slightly higher (not statistically significant) in female than male diabetic (71.4% to 62.0%) and nondiabetic (50.0% to 28.6%) rats. It is well established that women have a higher prevalence of Hashimoto’s thyroiditis as well as most other autoimmune diseases [24]. Second, the incidence is higher in diabetic than nondiabctic rats (Z = 1.61, p = .05), a finding consistent with the earlier study [30]. This is interesting because several studies have reported an association between juvenile-onset diabetes and autoimmune thyroid disease [12, 21]. Relevant to this is the observation that HLA-B8 and DR3 histocompati­ bility antigens occur with a greater frequency in patients with insulin-dependent diabetes, Graves’ disease, and Hashimoto’s thyroiditis [2, 3, 18]. This pattern is characteristic of a small subgroup of the human juvenile-onset diabetic population and has been termed “the syndrome of polyendocrinc autoimmunity” [1,2, 5]. No lymphocytic thyroiditis has been reported in other strains of diabetic rodents in which the thyroid has been examined [27, 33]. Third, there is a higher incidence of thyroiditis in the younger rats (table II)—this also is true for insulitis. (Insulitis in BB Wistar rats occurs at the onset of diabetes, typically between 60 and 160 days of age). Finally, nine BB Wistar rats (seven diabetic and two nondiabetic) in this necropsy series had insulitis and all nine also had thyroiditis. Therefore, it seems likely that the onset of diabetes in BB Wistar rats is not only associated with a mononuclear cell insulitis, but also a mononuclear cell thyroiditis. Since the insulitis in BB Wistar rats is transient [19, 20], it was only seen in rats dying soon after onset of diabetes. On the other hand, data in table II show that the thyroiditis may be present for up to a year after the onset of diabetes. This is in contrast to the incidence of human autoimmune thyroiditis which increases with age [24]. It is well established that there is a tendency for more than one autoimmune disorder to occur in the same patient. When this happens, all of the autoimmune disorders tend either to be very organ specific as in Hashimoto’s thyroiditis or systemic with no organ specificity as in systemic lupus erythematosus [25], Autoim­ mune disorders associated with juvenile-onset diabetes are of the organ specific type [11]. Results of several studies suggest that both the insulitis and thyroiditis in BB Wistar rats may be autoimmune in nature. Serum immunofluorescence studies have not demonstrated circulating autoantibodies to pancreatic islet cells in BB Wistar rats [6, 7]. However, antibodies against the islet cell plasma membrane have been detected by radioligand assay in 12 of 14 (85%) BB Wistar diabetic rats studied three to eleven days after onset of hyperglycemia [6], Antithyroglobulin autoantibodies have been found by scrum immunofluorescence in BB Wistar rats. A second study reported no antithyroid microsomal antibodies but did demonstrate Wright cl ul.

antibodies to thyroid colloid antigens in 5% to 17.9% of the BB Wistar rats studied [7]. Another study reported thyroid colloid autoantibodies in 34 of 48 diabetic (71 %) and eight of 39 nondiabctic BB Wistar rats. In addition, they found evidence of autoanlibodics to thyroid follicular cells in some rats [17]. In spite of these results, there is still insufficient evidence to determine the specific pathogenetic mechanism responsible for thyroiditis and insulitis in BB Wistar rats. The concur­ rence of these lesions in the same rats suggests that these mechanisms are related closely.

Acknowledgements This work was supported by the Department of Pathology, the College of Medicine, and the Graduate School o f The Ohio State University; the Upjohn Company; and N .I.H . grant NS-18026. Mr. James R. Wright, Jr. was supported by the Samuel J. Roesslcr Research Scholarship. The authors also would like to thank Racchel Tigncr for her statistical assistance and Dave Covey and Bruce Thompson for their excellent maintenance of the BB Wistar colonies in Columbus and Ottawa. This work was presented in part at the Landacrc Society’s 25th Annivcrsity Medical Student Scientific Session, Columbus, Ohio, March 19, 1981.

References

1 Bottazzo, G.F.; Florin-Ciiristensen, A.; D o n ia c h , D.: Islet cell antibodies in diabetes mcllitus with autoimmune polycndocrinc deficiencies. Lancet 2:1279-1283, 1974 2 Cahill, G.F.; McDevitt, H.O.: Insulin-dependent diabetes mcllitus: the initial lesion. N Engl J Med 304:1454-1465, 1981 3 C h r is ty , M.; D e c k e rt, T.; N i:k u i\ J.: Immunity and autoimmunity in diabetes mcllitus. In: Clinics on Endocrinology and Metabolism, vol. 6, ed. Tattersall, pp. 305-332. WB Saunders Co., Philadelphia, 1977 4 Colacchio, T.A.; Li V o ls i, V.A.; Lo G e rfo , P.: Experimental thyroiditis: A review. In: Thyroiditis, ed. LiVolsi and Logerfo, pp. 179-190. CRC Press, Boca Raton, Florida, 1981 5 Doniach, D.; Bottazzo, G.E.: Polyendocrine autoimmunity. In: Clinical Immunology Update, ed. Franklin, pp. 96-121. Elsevier/North Holland, Amsterdam, 1981 6 D y k u e rg , T.; Nakiiooda, A.F.; Baekkestov,S.;Lernmark,A.; P ou ssie r, P.; M a rlis s , E.B.: Islet cell surface antibodies and lymphocyte antibodies in the spontaneously diabetic BB Wistar rat. Diabetes 31:278-281, 1982 7 E ld e r , M .; M a c la r e n , N.; R ile y , W.; M c C o n n e l, T.: Gastric parietal cell and other autoantibodics in the BB rat. Diabetes 31:313-318, 1982 8 G e p ts, W .; LeC om pte, P.M.: The pancreatic islets in diabetes. Am J Med 70:105-115, 1981 9 Gosselin, S.; Capen, C.C.; M artin, S.L.: Hashimoto's thyroiditis. Lymphocytic thy­ roiditis in the dog. Am J Pathol 90:285-288, 1978 10 H a jd u , A.; R o n a, G.: Spontaneous thyroiditis in laboratory rats. Experientia 25:1325- 1327, 1969 11 Ir v in e , W.J.: Organ-specific autoimmunity in diabetes mellitus and correlations with other endocrine disorders. In: Diabetes. Proceedings of the IX Congress o f the Interna­ tional Diabetes Federation, ed. Bajaj, pp. 261-272. Excerpta Medica, Amsterdam, 1977 12 Ir v in e , W.J.; C la r k e , B.F.; S c a r tii, L.; C u lle n , D.R.; D u n c a n , L.J.P.: Thyroid and gastric autoimmunity in patients with diabetes mellitus. Lancet 2:163-168, 1970 Spontaneous Thyroiditis in BBW Rats

13 Jackson, R.; Rassi, N.; C rum p, T.; Haynes, B.; Eisendartii, G.S.: The BB diabetic rat: profound T-ccll lymphocytopenia. Diabetes 30:887-889, 1981 14 K a la n t , N.; Seemayer, T.: Malignant lymphoma in spontaneously diabetic rats. N Engl J Med 300:737, 1979 15 K itc h e n , D.N.; T o d d , G.C.; M e y e rs , D.B.; P a g e t, C.: Rat lymphocytic thyroiditis associated with the ingestion o f an immunosuppresive compound. Vet Pathol 16:722- 729, 1979 16 Levy, B.M.; H am pton, S.; D re iz e n , S.; H am pton, J.K.: Thyroiditis in the marmoset. J Comp Pathol 82:99-103, 1972 17 Like, A.A.; B u tle r , L.; W illia m s , R.M .; Appel, M.C.; W e rin g e r, E.J.; Rossini, A.A.: Spontaneous autoimmune diabetes mellitus in the BB rat. Diabetes (suppl. 1) 31:7-13, 1982 18 M oens, H.; F a rid , N.R.: Hashimoto's thyroiditis is associated with HLA-DRw 3. N Engl J Med 299:133-134, 1978 19 N a k h o o d a , A.F.; Like, A.A.; C h ap p el, C.I.; M u r r a y , F.T.; M a rlis s , E.B.: The spontaneously diabetic Wistar rat. Metabolic and morphologic studies. Diabetes 26:100— 112, 1977 20 N a k h o o d a , A.F.; Like, A.A.; C h ap p el, C.I.; W ei, C.N.; M a rlis s , E.B.: The sponta­ neously diabetic Wistar rat (the “BB” rat). Studies prior to and during development of the overt syndrome. Diabetologia 14:199-207, 1978 21 N e ru p , J.; L e rn m a rk , A.: Autoimmunity in insulin-dependent diabetes mellitus. Am J Med 70:135-141, 1981 22 P a te l, S.T.; N ew m an, H.A.I.; Y a te s , A.J.; F a lk o , J.M.: Serum lipids and lipoprotein composition in spontaneously diabetic BB Wistar rats. Clin Chcm 27:1038, 1981 23 P e n h a le , W.J.; Farmer, A.; McKenna, R.P.; Irvine, W.J.; Spontaneous thyroiditis in thymcctomized and irradiated Wistar rats. Clin Exp Immunol 15:225-236, 1973 24 Q uim isy, F.W.: Lymphocytic thyroiditis Hashimoto’s disease. In: Spontaneous Animal Models of Human Disease, vol. I, ed. Andrews, Ward, and Altman, pp. 297-299. Academic Press, New York, 1979 25 R o it t, L: Essential Immunology, pp. 265-274. Blackwell Scientific Publications, Oxford, 1977 26 Senhauser, D.A.: Immunopaihologic correlations in thyroid diseases, hv. International Academy of Pathology Monograph, The Thyroid, ed. Hazard and Smith, pp. 167-182. Williams & Wilkins Co., Baltimore, 1964 27 Siluerherg, R.; Silueruerg, M.: Lesions in “yellow” mice fed slock, high-fat or high- carbohydrate diets. Yale J Biol Med 29:525-539, 1957 28 Silverman, D.A.; Rose, N.R.: Neonatal thymectomy increases the incidence of spon­ taneous and methylcholanthrenc-enhanced thyroiditis in rats. Science 184:162-163, 1974 29 Sima, A.A.F.: Peripheral neuropathy in the spontaneously diabetic BB Wistar rat: an ullrastructural study. Acta Neuropathol (Berl) 51:223-227, 1980 30 Sternthal, E.; Like, A.A.; S a ra n tis , K.; Braverman, L.E.: Lymphocytic thyroiditis and diabetes in the BB/W rat. A new model of autoimmune endocrinopathy. Diabetes 30:1058-1061, 1981 31 Sternthal, E.; Like, A.A.; S a ra n tis , K.; Braverman, L.: Spontaneous and iodide- induced lymphocytic thyroiditis in the diabetic Bio Breeding/Worcester (BB/W ) rat: a model of autoimmune endocrinopathy, p. 271. Program and Abstracts 63rd Annual Meeting of the Endocrine Society, 1981 32 V o lp e , R.: The pathology of thyroiditis. Hum Pathol 9:429-438, 1978 33 Von Dorsciie, H.H.: Histological investigations on the thyroid gland o f Sand rats W right el a/.

(Psammomys obesus) during the development o f diabetes mellitus. Anat Anz 150:343- 345, 1981 34 Wellmann, K.F.; Vot.K, B.W.: Islets of Langcrhans: Structure and function in diabetes. Pathobiol Annu 10:105-133, 1980 35 W ic k , G.: Mashimoto thyroiditis. Spontaneous hereditary autoimmune thyroiditis of Obese Strain (OS) chickens. Comp Pathol Bull 8:2, 1976 36 Wright, J.R.; Sharma, H.; Tiiiuert, P.; Y a te s , A.J.: Pathologic findings in the spontaneously diabetic BB Wistar rat. Lab Invest 4 2 :162, 1980 37 W r ig h t, J.R.; Sharm a, H .M .; Y ate s, A.J.; Thompson, B.M.; T iiiu e r t, P.: Supradia­ phragmatic accessory lobe of the liver in BB Wistar rats. Teratology 27: 117-120, 1983 38 W r ig h t, J.R.; Y ates, A.J.; Shah, N.T.; N e ff, J.C.; Covey, D.W .; T iiiu e rt, P.: Hema­ tological characteristics of the BB Wistar rat. Vet Clin Pathol 12: 9-13, 1983 39 W r ig h t, J.R.; Y ates, A.J.; Sharm a, H .M .; Shim, C.; T ig n e r, R.L.; T iiiu e rt, P.: Testicular atrophy in the spontaneously diabetic BB Wistar rat. Am J Pathol 108:72- 79, 1982 40 W r ig h t, J.R.; Y ates, A.J.; S harm a, H .M .; T iiiu e r t, P.: Spontaneous gastric erosions and ulcerations in the BB Wistar rat. Lab Anim Sci 31:63-66, 1981 41 W r ig h t, J.R.; Y ates, A.J.; S harm a, H .M .; T iiiu e r t, P.: Pathological lesions in the spontaneously diabetic BB Wistar rat. A comprehensive autopsy study. Metabolism 32 (suppt. I): 101-105, 1983 Appendix C 202

Testicular Atrophy in the Spontaneously Diabetic BB W istar R at

JAMES R. WRIGHT, Jr., BS, From the De/tarlment o f Pathology atui Department o f Psychiatry, ALLAN J. YATES, MD. PhD, College o f Medicine, The Ohio State University, Columbus, Ohio and HARI M. SHARMA, MD, FRCP(C), Sir Frederick C. Hunting Research Centre, Health Protection Hrattch, CHULL SHIM, MD. RAECHEL L. TICNER, AB, Health and Welfare, Tunney's Pasture, Ottawa, Ontano, Canada ■nd PIERRE THIBERT, DVM

Complete grots and microscopic postmortem examina­ ically similur to the senile testicular atrophy in the non­ tions were performed on 100 BU Wistar diabetic rats, diabetic rats. Histologic findings that were associated 27 BB W istar nondiabetic siblings, and 41 W istar rats, with increasing severity of atrophy were multinucle­ and the incidence of testicolar lesions was tabulated. ated giant cells in the lumens o f seminiferous tubules, Testicular atrophy was the predominant finding in all increased interstitial connective tissue, Leydig cell hy­ three groups of rats, but atrophy occurred at a much perplasia, and thickening of the tunica albuginea. Tes­ younger age in the diabetic rats. There was a strong re­ ticular atrophy has also been reported in human dia­ lationship between the duration of diabetes and the betics. Therefore, the BB W istar rat may be a useful presence o f atrophy, which was stronger than the rela­ model for investigating this aspect o f diabetes mellitus. tionship between age and atrophy. The testicular (Am J Pathol 1982, 108:72-79) atrophy observed in the diabetic rats was morpholog­

THE BB WISTAR (BBW) rai is an outbrcd sirain of tively, were examined. Because o f the extreme sus­ spontaneously diabclic rats that originated in the Bio ceptibility of BB Wistar rats to infection,1 all rats Breeding Laboratories Ltd. of Ottawa, Canada. The were maintained in a semibarrier housing facility. BBW rat is a promising model lor insulin-dependent The details of animal maintenance and cure have diabetes because it spontaneously develops beta cell been reported previously.* Every duy ull animals were destruction with subsequent insulinopcnia, hypergly­ weighed, and the urine o f diabetic rats was examined cemia, hyperglucagonemia, glycosuria, ketoacidosis, for ketones and glucose with the use o f Ketostix and hyperlipemia.1-1 However, except for the pan­ (Ames Co., Elkhart, Ind) and Testape (Eli Lilly, In­ creatic lesions, very little is known about other spon­ dianapolis, Ind), respectively. We administered pro­ taneous diseases that occur. Therefore, we conducted tamine zinc (U-40) insulin (Eli Lilly) subcutuneously a necropsy study to determine the types and frequen­ on a daily basis to maintain 4 + glycosuria without cies o f pathologic abnormalities in our breeding col­ ketonuria. ony of BB rats.1 The most frequent abnormality in the male rats was testicular atrophy. Although testi­ cular atrophy has been reported in both human Supported in part by the Department of Pathology, Col­ diabetics4** and in a wide variety o f diabclic animals,’ lege of Medicine, and the Graduate School of The Ohio there are few reports o f systematic detailed studies o f Slate University, the Upjohn Company, and N IH Grant NS-18026. Mr. Wright was supported by a Samuel J. this abnormality. Kocssler Research Scholarship. Presented in part at the American Association for Laboratory Animal Science Meeting, September 22, 1981, Materials and Methods Salt Luke City, Utah. A total o f 100 BB Wistar diabclic rats (BBW d), 27 Accepted for publication, February 24, 1982. Address reprint requests to Allan J. Yales, M D, PhD, BB W istar nondiabetic siblings (BBW nd), and 41 Neuropathology Unit, Department of Pathology, The Ohio standard Wistar (w) rats with mean ages o f 254 ± 100 Stale University, 105 Uphant Hall, 473 West 12lh Avenue, days, 241 ± 122 days, and 281 ± 162 days, respec­ Columbus, OH 43210.

0002-V44U/82/070V-0072SOO.VU © Amcncin Anodaiiou of Pnliologim 203

Vol. 108 • Nil. 1 TESTICULAR ATROPHY IN DIABETES

Severe Tevnculor Atrophy I I Minimal Testicular Atrophy LVVSI Hypocellulority

F i g . C . l Bar graph showing percentage ol rata with testicular lesions by age In Pays. Rats appear In the following order lor all age groups: BBWd, BBWnd, and Wistar. n»eo The number ol rats in each group is noted above each bpr.

nd W n>6 n>6 s*0 n*0

O-IZO 121 -2 4 0 241 - 3 6 0 361 -480 481-600 AGE IN DAYS

Necropsy was performed on all rals that died spon­ albuginea. All of these were rated subjectively us nor­ taneously or were sacrificed for experimentation mal or as having one of three grades o f severity und cither immediately or following storage at 4 C, usual­ related to the ages of the animals by the chi-square ly for no longer than 12 hours alter death. Thoracic test. Since pneumonia is known to all'cct spermato­ and abdominal organs were removed en block; and, genesis severely,’ " ' the presence or absence o f pneu­ following dissection, individual organs were weighed, monia was also noted. The duration o f diabetes in fixed in neutral phosphate-buirercd formalin, and days for each UBWd rat was taken from its treatment processed for light microscopy. Tissue sections were records. routinely stained with hematoxylin and eosin. Tissue and fluid specimens suspected o f antemortcm bac­ terial infection were placed in disposable collection Results tubes and cultured for aerobic and anaerobic micro­ organisms. Paraflin sections o f such tissues were Pigure I shows the distribution and severity o f stained with Cram’s stain, acid fast blue, and testicular lesions according to the ages o f the rats. All Gomori's methcnamine silver for bacteria and fungi. three types o f rats (BBW d, BBW nd, and W ) de­ In addition to determining testicular mass and veloped minimal and severe testicular atrophy. One- body mass, we determined the presence and distribu­ third o f BBWd rats in the youngest age group had tion of testicular atrophy (unilateral and bilateral) hypocellular testes. Both minimal testicular atrophy histologically. Testes from all three groups o f rats and hypoccllularity occurred in the 121-240-day-old were rated as normal, hypoccllular, or atrophic. age group of BBWd, and the frequency and severity Hypocellular testes had seminiferous tubules with de­ o f the lesions progressively increased with age in creased cellularity, but all retained some germ cells. these rats. Similar changes occurred in the BBWnd Atrophic testes were those that had seminiferous rats, but at an older age and with less severity. Testes tubules without any germ cells. We classified these as from Wistar rats showed no degenerative testicular Grades 1 or 2 on the basis of the extent of atrophy. changes until the second oldest age group, but all rats Testes were considered to have Grade 1 (minimal) had severely atrophic testes in the 4 8 1-600-day-old atrophy if there were fewer than 5 atrophic tubules age group. The relationship between the age o f the per cross-section; testes with Grade 2 (severe) atrophy rats and the degree o f atrophy was examined by a were those with more than 5 atrophic tubules per one-way analysis o f variance (unweighted means cross-section. We used chi-square analysis to com­ solution). Table I shows that the severity o f atrophy pare these categories with a variety o f specific histo­ increased significantly ( P < 0.001) with age in both logic abnormalities, including the presence o f multi­ BBWd and W rats. The same trend is seen in the nucleated giant cells in the lumens o f tubules, in­ BBWnd rats but is not statistically significant, be­ creased amounts o f interstitial tissue, increased den­ cause only 3 o f these rats developed atrophy. The sity o f Leydig cells, and increased thickness o f tunica presence o f atrophy related more strongly to dura­ 204

WRIGHT ET AL Ajr* • July IV82

Table C. 1 Distribution of Age and Duration ol Diabetes by Degree ol Testicular Atrophy*

Days Statistical ...... - relationship to Rat type 0-120 121-240 241-360 361-480 481-600 M ean ± SEMt degree ol atrophy!

BBWd Age 6 40 37 17 0 254 x 10.0 P < 001 n = 100 F(2, 97) = 21.25 Duration 49 30 21 0 0 147 x 10.0 P < 001 ft = 100 F{2, 97) = 38.29 BBWnd Age 6 7 7 6 0 240 x 23.8 NS ft e 26 F(2, 23) a 2.28 Wistar Age 6 6 6 6 5 280 x 30.1 P < 001 n = 29 F(2. 26) = 17.56

* Values are numbers ot rats for which precise ages and duration of diabetes are known. t SEM & standard error of the moan. 1 Statistical analysis by one way analysis of variance using the unweighted means solution. NS s not statistically significant.

tion of diabetes than to age alone (Table I), as in­ not seen in the lumens of normal seminiferous dicated by the higher F-valuc for the former (38.29, tubules, and the number present was strongly compared with 21.23) with the same degrees of free­ associated with the degree of atrophy, but less so dom. This is an important distinction, because many with age. An increase in testicular interstitial connec­ of the old BBWnd rats became diabclic shortly be­ tive tissue (Figure 4) was associated with the degree fore dying (the range for age at onset of diabetes was of atrophy at a low level of significance ( P< 0.0S) but 42 to 426 days; the mean was 107 ± 5 days). For in­ was not associated with age. There was a strong stance, the oldest BBW rat in the study (464 days) direct relationship between Lcydig cell hyperplasia was nondiabetic until 38 days before it died. This (Figure 4) and an increasing degree of atrophy in all may explain why some of the older diabetic rats did three types of rats (/*.< 0.01), but there was no not develop testicular atrophy and why the percen­ statistically significant relationship between age and tage of BBWd with atrophy does not reach 100°/o. Lcydig cell hyperplasia. There was also a strong posi­ Using an analysis of covariance with body weight tive relationship between the thickness of the tunica as the covariate (Table 2), we found that testicular albuginea (Figure 3) and the degree of atrophy (P < mass decreased significantly with increasing severity 0.001), but only in diabetic rats. No relationship be­ of histologic atrophy for all three groups of rats. tween the presence of pneumonia and testicular de­ Grossly, atrophic testes were small, soft, wrinkled, generative changes was present in BBWd or BBWnd and frequently light yellow (Figure 2). rats, and none of the Wistar rats had pneumonia. Or­ Table 3 shows the frequencies of a variety of mor­ chitis was present in 2 BBWd rats, and bilateral phologic features rated scmiquantitaiivcly on a scale testicular granulomas were observed in I BBWd rat, of severity ranging from 0 (normal) through 3 but no organisms were found. Adrenal and pituitary (severe). Multinucleaicd giant cells (Figure 3) were were hisiologicully normal in all three groups of rats.

Tuble C.2 Body and Testicular Weights

Testicular weights x SEM (grams)*

Body mass (grams)ti N orm al Minimal atrophy Severe atrop

BBWd 329 8 x 8.5 2.88 x 0.08 2.38 x 0.31 1.27 x 0.10 ft s* 87 n b 56 n = 8 n = 23 F (1, 83) s 42.81, P < 0.001 F (2, 83) > 68.79, P < 0.001 BBWnd 381.9 x 26.3 3.13 x 0.15 1.76 x 0 0 76 x 0 n b 19 n = 17 ft s 1 ft X 1 F Cl. 15) = 43 05, P < 0.001 F (2,15) b 4.96. P < 0 05 Wistar 449.7 x 27.8 3.56 x 0.10 2 60 x 1.24 1.61 x 0.36 ft s 33 n a 26 n b 2 n & 5 F (1, 29) s 40 64. P < 0.001 F (2. 29) = 36.36, P < 0.001 * Each measure is the combined mass of both testes from each rat. SEM = standard error of the mean. t Statisticat analysis examining testicular weight with respect to histologic severity of atrophy was by analysis ot covariance with body weight as covariate. Includes only rats tor which both body mass and testicular mass were known. I Correlations between testicular weight and body weight were 0.376, 0.796, and 0.534 for BGWd, BBWnd, and Wistar rats, respectively. 205

Vol. lua • No. I TESTICULAR ATROPHY IN DIABETES

Discussion

Structural changes in (he tcsics of human diabetics were first observed in 1878 by Paschutin'1 and later substantiated by others.s t- "■|1 Several of these in­ vestigators"" attempted to relate these histologic abnormalities to sexual impotence, a disorder present in approximately 50% of diabetic men of reproduc­ tive age." A study by Singhal et al" suggests that testicular abnormalities arc present in both impotent and sexually potent diabetics with similar frequency. Impotence is presently believed to be due to diabetic peripheral neuropathy and unrelated to these testic­ ular changes." Although the histologic changes in the testes have been studied extensively in chemically induced

diabetes,tlicy have been examined in F i g . C . 2 Transected rat testes. Normal Wistar rat on the left and only two other strains of spontaneously diabetic BB Wistar diabetic rat with bilateral atrophy on the rig h t. Line l& animals, the obese-hyperglycemia AO mouse"* and 1 cm. the Chinese hamster.11 Both studies described hypo- ccllularity or maturation arrest as the predominant finding, rather than a total absence of germ cells simultaneously compared them with the senile testic­ within seminiferous tubules. In addition to these ular atrophy that occurs in nondiabetic rats. milder changes, histologic changes observed in the Since testicular atrophy is prevalent in most strains BB Wistar rat include both mild and severe atrophy. of aged rats,1**" age was considered a likely con­ Furthermore, we have examined these changes with tributing factor to the high incidence of testicular le­ respect to both age and duration of diabetes and have sions in the BBW Wistar rats. Burck" reported that

Table C. 3 Numbers ot Rats With Different Degrees ot Severity ot Testicular Histologic Abnormalities and Their Relationship to Degree ol Atrophy and Age Degree ol severity Relationship to'

Rat type Normal 1 2 3 Degree of atrophy Age group

Giani cells BBWd 65 22 9 4 P < 0.001 P<0.01 X* = 26.24, t i t s 6 X* = 21.83, t i t b 9 BBWnd 23 4 0 0 NS NS X1 = 2.23, t i t b 2 X* b 6.61, t i t b 3 Wistar 36 2 0 0 P < 0.01 NS X * = 11.26, t i t = 2 X* = 3.24, t i t b 4

Increased interstitial BBWd 50 39 7 4 P < 0.05 NS connective tissue X 1 = 13.90, t i t b 6 X1 =14.00, t i t b 9 BBWnd 25 1 1 0 P < 0 ,05 NS X* = 13.05, t i t * 4 X1 = 5.86, t i t j= 6 Wistar 24 12 1 1 P < 0.05 P < 0.05 X 1 = 14.67, t i t b 6 X* = 27.01, t i t b 12

Loydig cells hyperplasia BBWd 92 6 1 1 P < 0.001 NS X 1 = 24.19, t i t b 6 X* = 6.12. t i t b 9 BBWnd 25 2 0 0 P < 0.001 NS X* = 19.71. t i t b 2 X* = 6.17. t i t b 3 Wistar 35 1 0 2 P < 0 .01 NS X* a 17.37. t i l b 6 x* = 21.50. t i t B 12

Tunica albuginea BBWd 45 27 14 10 P < 0 .001 P < 0 .001 thickening X * = 55.39. t i t = 6 X* = 39.77. t i t B 9 BBWnd 16 8 3 0 NS NS X * = 5.09, t it b A X* = 39.77, t i t b 9 Wistar 20 10 1 0 NS NS x* = 0.90, t it B 4 X* = 6.13, t it b 8

‘ All statistical analyses by chi-square test. Chi-square values and degrees ot Ireedom printed below P values. 206

WRIGHT ET AL AJP • July 1V82

■ T f - K ig . C .3 BB W islar diabetic rat with atrophy. Atrophic seminiferous tubule with several multinucleated giant cells in the lumen. Interstitial tissue shows Leydig cell hyperpla­ V 7 r ^ W _. - * ...ot ^ sia. Tubules located peripherally to the right, left, and abo ve the central tubule contain only Sertoli cell9. ( x 400)

at least limited atrophy was present in all rats over 18 some only Sertoli cells present.,,‘u This discrepancy months of age and was usually severe in rats older may possibly be explained by the severity of the than 24 months. He found that atrophic testes of old diabetic state in these different studies. Several in­ rats frequently had Lcydig cell hyperplasia, in­ vestigators have suggested that testicular atrophy is terstitial edema, and tubules with multinucleaied more frequent in poorly controlled diabetics."•” giant cells or only Sertoli cells. The changes described Several studies on diabetic men9-1-'1 and rats19 have by Burek were consistent with the histologic changes reported increased interstitial tissue in atrophic observed in all three types of rats in our study. testes. Only SchofHing et aP1 reported no increase in Several investigators have frequently reported benign interstitial tissue. We observed a slightly significant or malignant tumors associated with testicular (P < 0.05) increase in interstitial tissue in rats with atrophy in senile rats,19'1’ but these were not seen in testicular atrophy, regardless of the presence or any rats in our study. This is probably due to the absence of diabetes. We therefore feel that increased greater age of their animals. In our study, testicular interstitial tissue is probably a general feature of atrophy was first observed in BBWnd and W rats at testicular atrophy and is not specifically related to 355 and 361 days of age, respectively; the earliest age diabetes. of onset in the BBW diabetic rats was 148 days. In all three groups of rats, multinucleated giant Figure 1 shows clearly that the incidence of atrophy cells (Figure 3) were frequently seen in the lumens of increased with age in all rats but occurred at a much degenerating seminiferous tubules. Other studies younger age in BBWd rats. Saksena et aP* reported have reported similar findings in both senile19 and that sperm production in Sprague-Dawley rats re­ diabetic1' rats. These giant cells are believed to be mained maximal from 72 days of age to beyond 450 fused spermatids.19 To our knowledge, multinu- days of age. This appears consistent with our his­ clcaicd giant cells have not been reported in the semi­ tologic findings in standard Wislar rats but not in the niferous tubules of man. BBWd rats. Age is undoubtedly a very important fac­ Several studies utilizing either rats with chemically tor in determining the presence of atrophy in all induced diabetic or genetically obese (insulin-resis­ groups of rats, but the duration of diabetes is a much tant) mice have reported a decrease in the number of better predictor of atrophy in BBWd than is age Leydig cells.1”1119 It is possible that the diabetogenic (T able 1). chemicals used by some have a toxic effect on the The severity of testicular lesions in diabetic men Leydig cells, because alloxan is known to have severe and animals varies widely from study to study. toxic elleels on many organ systems, including the Several have reported only a hypoccllularity of ger­ testes.1*11 Lcydig cell hyperplasia was observed in this minal cells,1”1" others more severe atrophy,91 and study and in a study by Roscnmann el al11 Both his Vul tub • N.i 1 TESTICULAR ATROPHY IN DIABETES

study and ours utili/cd rais that were not obese and did not have chemically induced diabetes. Interest­ ingly, Ayad” has reported Leydig cell hyperplasia in a scries of 30 diabetic men. We also observed Leydig cell hyperplasia in nondiabclic rats with senile testicular atrophy. This is consistent with senile atrophy in nondiabclic rat strains.” -” Maturation arrest within seminiferous tubules has been reported in a variety of studies of human’ ”'” and animal’"-” diabetics. Most of these investigators have attributed this to hyposceretion of gonadotropic hormones. This explanation scents unlikely, because several recent studies have shown that testosterone levels and gonadotropin levels are normal in diabetic men,’, J“ but an cndocrinologic mechanism is sug­ gested by the observation that pancreatectomy or alloxan diabetes not only affects the germinal epithelium but also suppresses the development of secondary sexual characteristics in animals.” -” Some investigators have attributed testicular le­ sions in diabetics to diffusion or perfusion problems caused by basement membrane thickening of tunica propria, microangiopathy, or atherosclerosis. Since seminiferous tubules are avascular, diffusion from the interstitial vessels through the tunica propria is necessary to nourish the germ cells and Sertoli cells in each tubule. Only a few studies have reported base­ ment membrane thickening,” and others have re­ ported its absence.” In the present study, micro­ angiopathy and atherosclerosis were not observed by light microscopy and so could not be involved in (he pathogenesis of the lesions. Only a few rats had tubules with marked basement membrane thicken­ ing, and this thickening may have been secondary to the atrophy rather than causative. Spermatogenesis in man and the rat is generally similar, except that it proceeds in waves in the rat.” In both, proliferation of the germinal epithelium is controlled by a complex milieu of hormones and other factors. The number of spermatozoa produced F ig . C. 4 Testes trom 3 rats showing dlllerent degrees ol atrophy depends on temperature, available lighting, nutri­ and other variables. A -N o rm a l seminiferous tubules trom a 361-dayOld Wislar rat. Full complement ol germ cells present. 8 - tional factors, age, pituitary gonadotropin levels, Testls ol a Wistar rat 551 days old with severe atrophy, thickening ol testosterone levels, and other variables.” It is unlike­ Interstitial connective tissue (3 1, and Leydig cell hyperplasia (3 +■). C - Severely atrophic testis In a 387-day-old BBW diabetic ly that temperature or lighting played a significant rat. Most tubules have only Sertoli cells. Tubules retaining some role in the incidence of testicular atrophy, because germ culls show sloughed epithelium. There Is one relatively normal tubule in the Held. (H&E, x 100) (With a photographic reduction the housing facility was consistently maintained at 72 o l 30Y.) F with alternating 12-hour periods of artificial light and darkness. ruts, ihcir nondiabclic siblings, and standard Wislar In summary, a high incidence of testicular abnor­ rats. These lesions occurred at a much younger age in malities characterized by a partial or total loss of ger­ (he diabclic rais than in either llie BUWnd or W rats. minal epithelium, frequently with a relative sparing Similar findings have been reported in human dia­ of Sertoli cells, was observed in UU Wistar diabetic betics as well as other animal models for diabetes. Al- 208 WRIGHT ET AL Alt' • Jul) m2

F ig* C.5 A-Normal Wistar rat tes­ tis demonstrating normal cellularily of seminiferous tubules and normal thickness of tunica albuginea. (H&E, xlO ) B -B B Wistar diabolic rat testis with diffuse, severe atrophy ot seminiferous tubules, and a thickened tunica albuginea. (H&E, x 10)

though atrophic testes of diabclic animals are his­ 2. Nakhooda AF, Like AA, Chappel Cl, Wei C-N, tologically similar to those with senile atrophy, the Marliss UU: The spontaneously diubetic Wistar rat (the -BB" rat): Studies prior to and during development of exact etiology and pathogenesis of these lesions re­ the overt syndrome. Diabetologia 1978, 14:199-207 main to be elucidated. The 013 Wistar rat appears to 3. Wright J, Sharma H, Thibert P, Yates AJ:Pathologi­ be an excellent model for study of the pathogenetic cal findings in the spontaneously diabetic BB Wistar rat. Lab Invest 1980, 42:66 mechanisms involved. Furthermore, the significance 4. Paschutin VW: Vorlesungen der allgemeinen Patho­ of these findings to investigators attempting to breed logic, St. Petersburg, 1878 bis 1881 BB Wistar rats for other types of studies is reflected 5. Koch K: Zwischcnzellen und Hodenatrophie. Virchows Arch Pulhol Anat 1910, 202:376-406 in the lower rate of breeding success in BBWd rats. 6. Kraus EJ: Zur Pathogenese dcs Diabetes mellitus auf- grund morphologischcr Umersuchungcn der en- docrincn Organc. Virchows Arch Pulhol Anat 1923, R e feren ces 247:1 7. Schollling K: Hypogonadism in male diabetic subjects, I. Nakhooda AF. Like AA, Chappel Cl, Murray FT, On the Nature and Treatment of Diabetes. Edited by Marliss EB: The spontaneous diabetic Wistar rat: BS Leibel, GA Wrenshall. Amsterdam, Excerpta Metabolic and morphologic studies. Diabetes 1977, Medica, 1965, pp S05-S2I 26:100-112 8. Wright JR, Yates AJ, Sharma HM, Thibert P: Spon- 209

Vol. IDS • No I TESTICULAR ATROPHY IN DIABETES

laneous gastric erosions and ulcerations in BB Wistar 24. Warren S, LcComptc PM: The Pathology of Diabetes ruts. Lab Anim Sci 1981, 31:63-66 Mcllitus. Philadelphia, Lea and Febiger, 1952, p 208 9. Muclcod J: A testicular response during and following 25. Oksancn A: Testicular lesion's of streptozotocin a severe allergic reaction, Ferlil Slcril 1962, 13:531-543 diabclic rats. Horm Res 1975, 6:138-144 10. Mills KG: The pathological changes in the tcslcs in 26. Lukcns, FDW: Alloxan diabetes. Physiol Rev 1948, 28: epidemic pneumoniu. J Exp Med 1919, 30:505-535 304-330 11. Schollling K, Fcderlin K, Dilschuncil H, Pfeiller El-': 27. Chester A, Tislowitz R: Ellecl of alloxan diabetes on Disorders of sexual function in male diabetics. the growing rat. Science 1945, 101:468 Diabetes 1963, 12:519-527 28. Ayad H: Discussion of hypogonadism in male diabetic 12. Schooling K: Diabetes mcllitus and male gonadal func­ subjects,’ p 520 tion, Diabetes, Proceedings of the Seventh Congress of 29. Kaush-Strooman JG, Petry R, Mauss J, Hientz HA, the International Diabetes Federation. Edited by KK Jukubowski HD, Scnge T, Muller KM, Eckhardl B, Rodrigue/,, J Vallancc-Owcn. Amsterdam, Excerpta Bcrthold K, Sauer H: Studies of sexual function in Medica, 1971, pp 36-57 diabetes. Seventh Congress of (he International Di­ 13. Singhal KC, Kastogi GK, Aikul IIC, Chhultani PN: abetes Federation. Excerpta Medica Foundation Inter­ Testicular histology in diabetics with impotence. Ind J national Congress Scries 1970, 209:112 Pathol Bactoriol 1969, 12:145-149 30. Faerman J, Vilar O, Kivarola MA, Rosncr JM, Jadzin- 14. Roncz LK, Ualodimos MC: lni|x>tcucc in diabetes mcl- sky MN, Fox D, Llorcl AP, Bernstein-Hahn L, Sara- lilus. Medical Times 1970, 98:159-170 ccni D: Impotence and diabetes: Studies of androgenic 15. Burck JD: Pathology of Aging Kals. West Pahn Beach, function in diabetic impotent males. Diabetes 21: 1972, Florida, CKC Press, 1978, pp 129-138 23-30 16. Coleman GL, Barthold SW, Oshaldiston, GW, l-oster 31. LepkovskyS, Nalbrandcr AV, Dimick MK, McFarland SJ, Jonas AM: Pathological changes during aging in LC, Pcncharz K: Giowth and reproduction of dcpan- barrier-reared Fischer 344 male rats. J Gerontol 1977, creatized chickens. Endocrinology 1964, 74:207-211 32:258-278 32. Hunt EL, Bailey DW: The cllccls of alloxan diabetes 17. Lul/.en L, Ueberberg H: A study on morphological on the reproduction system of young mule rats. Acta changes in the testes of old albino rats. Beitr Pathol lid Endocrinol 1961, 38:432-440 1973, 149:377-385 33. Lenta BE, Foglia VG, Collazo EF: Lesiones hislo- 18. Sakscnu SK, Chang MC: Age dependent changes in the logicas lesliculares en la rata diabetica. Rev Soc Argent sperm population and fertility in the male rat. Exp Ag­ Biol 1965, 41:197-203 ing Res 1979, 5:373-381 34. Leblond CP, Clermont Y: Spermatogenesis of rat, 19. Heilman K, Jacobsson L, Taljedul IB: Endocrine ac­ mouse, hamster und guinea pig us revealed by the tivity of the testes in obese-hyperglycemic mice. Acta “periodic ucid-fuchsin sulfurous acid” technique. Am J Endocrinol (Copcnh) 1963, 44:20-26 Anat 1952, 90:167-215 20. Scullidi L, Kotolo G: II dunno testicolure ncl diubclc 35. Turner CD, Bugnnru JT: General Endocrinology. 6th spcrimenlaleda allossana ncl ratio. Minerva Med 1974, edition. Philadelphia, W. B. Saunders Co., 1976, pp 65:4477-4484 406-449 21. Schollling K, Federlin K, Schmitt N, Pfcitfcr EF: Histomclric investigations on the testicular tissue of rats with alloxan diabetes and Chinese hamsters with Acknowledgments spontaneous diabetes. A cta Endoerinol 1967 , 54:335- 346 The authors wish to thank Dave Covey and Bruce 22. Rosenntann E, Palli Z, Teitelbaum A, Cohen AM: Testicular degeneration in genetically selected sucrose- Thompson for the excellent maintenance of the BB Wislar fed diabetic rats. Metabolism 1974, 23:343-348 breeding colonies al Columbus und Ottawa, respectively; 23. Soulairac A, Desclaux P, Katz KF: Action du diabclc Dr. Leona Ayers for microbiological studies; and Drs. alloxanique sur le tractus genital du rut mule cl fcmcllc. Charles C. Capen and Henry A. Wise II for critically CK Soc Biol (Puris) 1948, 142:311-312 reviewing the manuscript. Appendix D

210 211

J.iilit-Vs R. W right, Jr.

Robbins SL, Cotran RS, Kumar V (1984) Pathological ba­ Yuan YD (1979) Correlation of testicular pathology and sis of disease, 3rd cdn. Saunders, Philadelphia reproductive performance in bulls used for artificial in­ Setchell UP, Wailes GMH (1973) The blood-teslis barrier. semination. PhD thesis, Cornell University In: Hamilton UW, Grecp RO (eds) Handbook of physi­ ology. VII. Endocrionology, vol 5. American Physiologi­ cal Society, Washington, chap6, pp 143-172

Atrophy, Testis, Rat

Jam es R. Wright, Jr.

Synonyms. Testicular degeneration, gonadal de­ rophy of chronic diabetes mcllitus, chronic vita­ generation. min E deficiency, chronic zinc deficiency, and chronic cryptorchidism. Cross Appearance. Severely atrophic testes are Hypocellularity refers to hypoplasia of all germ small, soft, and covered by a smooth but wrinkled, cell types. Spermatogonia, spermatocytes, sper­ rubbery, yellow-tan tunica albuginia. The testicu­ matids, and spermatozoa are present, but the lar mass may be decreased by more than 50%. numbers of each type are decreased, resulting in (Normal testes in a 500-g rat weigh approximately an overall thinning of the germinal epithelium 2-2.25 g each.) Upon cutting through the tunica and decreased tubular diameters (Figs.267, 268). albuginia, a scant amount of edema fluid may be Because there are fewer germ cells, there will ap­ released. The seminiferous tubules are loosely pear to be proportionally more Sertoli's cells. The held together. In very old rats of some strains interstitial tissue and Leydig's cells are generally (particularly the Fisher 344 strain), small gray and unremarkable. This pattern appears to precede white nodules may be present within the interstir complete atrophy when atrophic changes are tial tissue. Testicular atrophy is usually bilateral, progressing slowly. Hypocellularity is most com­ but marked unilateral atrophy can occur. Atroph­ mon in old rats and rats with diabetes of interme­ ic changes occur over a continuum of severity. diate duration. Maturation arrest refers to an abrupt cessation of Microscopic Appearance. Testicular atrophy is an spermatogenesis at a particular developmental acquired decrease in testicular size from a prior normal state. Histological appearance varies de­ pending on both etiology and duration, but sever­ al general patterns can be identified. Degenera­ tive changes involving seminiferous tubules may Kin. o.ib'ifupper left). Severe atrophy, testis, rat. Seminifer- ► be broadly categorized us complete atrophy (only ous tubules are lilted by Sertoli's cells; germ cells are ab­ sent; tubules are decreased in diameter. The space between Sertoli's cells remaining), hypocellularity, or mat­ the tubules and interstitial tissue is an artifact. Formalin uration arrest. fixation causes tubular shrinkage. H & E, x 400 - Complete atrophy, the most severe form of semi­ niferous tubular degeneration, refers to totul ab­ Fig. 0.264 (lower left). Atrophy, seminiferous tubules, testis, rut. Note the small distorted tubules without germinal epi­ sence of germinal epithelium and is irreversible. thelium with marked thickening of the tunica propria. Ley­ Only Sertoli's cells are present within these tu­ dig's cells contain PAS-posilive, acid-fast pigment. PAS, bules (Fig. 263). The diameter of involved tubules x 400 is markedly decreased. Thickening of the tunica Fig. v.ibifup/ier right). Focal atrophy, testis, 331-day-old propria and basement membrane is o'flen present Wistar rjt. At the bottom are some normal tubules and in­ (Fig. 264). “Sertoli-only" change may be either fo­ terstitial tissue. Severely atrophic tubules and markedly in­ cal, involving only a few tubules, or diffuse. Inter­ creased interstitial tissue (upper). Focal dystrophic calcifi­ stitial tissue is usually increased (Fig. 265). Ley- cation (arrow). H & E, x 63 dig’s cell hyperplasia may be present (Fig. 266). Fig. 0.266 (lower right). Leydig's cell hyperplasia, testis, This pattern is characteristic of senile atrophy, at­ 2-year-old Wistar rat. Nodular (A)und (0). H St E, x 400 Atrophy, Tertis, Rat 213

James R. Wright, Jr.

: * A

m s m m m M & m m m $ 1 sa S itia ste .

n f e f t

$ & , © - ± m i< iIU tm i

® 3 » m

& Atrophy, TcMi-s, Hat

stage. A full complement of immature cells is pres­ pigments (Fig. 264). Atrophic seminiferous tu­ ent, but no mature forms arc produced (Fig. 269). bules may contain extracellular deposits, especial­ Because cell division is diminished or absent, the ly near the basement membrane. These yellow- tubular diameter is decreased. Interstitial tissue is brown pigments exhibit autofluorescence when often unremarkable, but cdeina may be present. they are examined in unstained histological sec­ Leydig's cells are usually present in normal num­ tions under ultraviolet light. The pigments are typ­ bers. This pattern is characteristic of decreased ically acid-fast, negative for iron stains, and posi­ gonadotrophin levels, inanition, early chemically tive for FAS, oil red “O", and Sudan stains. induced diabetes mellitus, early zinc deficiency, Although lipofuscinosis is most characteristic of and manganese deficiency. vitamin E deficiency (Swensen and Telford 1973; The initial event common to these three histologi­ Nelson 1980), acid-fast pigments may also be cal patterns is sloughing of the more mature por­ found throughout the testes of aged rats. tions of the germinal epithelium into the tubular lumens. Often, the sloughed cells are cytologically Ultrastructure. The diagnosis of testicular atrophy unremarkable; however, marked necrosis is seen does not require electron microscopy. in vitamin E deficiency. Multinucleated giant cells arising from sloughed Differential Diagnosis. Hie diagnosis of testicular germinal epithelium are common to most types of atrophy is relatively straightforward und is depen­ degenerative changes in the testes (Mason 1933; dent upon an acquired decrease in testicular size Cohen and de Vries 1969). They consist of aggre­ and mass. Testicular hypoplasia is probably ex­ gations of secondary spermatocytes or spermatids ceedingly rare. whose cytoplasm has fused into an eosinophilic Chronic degenerative changes following testicu­ syncytium. The nuclei usually appear identical to lar infarction may mimic end-stage testicular atro­ those of normal secondary spermatocytes or sper­ phy. Spontaneous infarction occurs rarely but matids. However, giant cells have characteristic may be induced experimentally with surgery “crescnt-like” nuclei in E-avitaminosis (Fig. 270). (Steinbergcr 1970) or toxic doses of cadmium Multinudear giant cells are believed to be the (Gunn and Gould 1970). Cadmium appears spe­ product of cytoplasmic fusion o f adjacent cells cifically to damage the endothelium of the testicu­ rather than nuclear division without cytoplasmic lar capillaries, an effect that is prevented by simul­ division because of the rapidity of their formation taneous administration of zinc. (as early as 6 h after injury) and because evidence of mitotic activity is not seen in the multinucleat- Biological Features. Numerous factors including ed giant cells (Cohen and de Vries 1969). Bizarre temperature, available lighting, pituitary gonado­ hyperchromatic mononuclear giant cells have trophin levels, testosterone levels, age, and nutri­ been described following thermal injury to the tional factors affect the function and integrity of testis (Cohen and de Vries 1969). the germinal epithelium or rat testes. In some Sertoli's cells, interstitial macrophages, and Ley­ strains of seasonally breeding wild rats, tempo­ dig's cells sometimes contain abundant lipofuscin rary testicular degeneration characterized by mat­ uration arrest at the primary spermatocyte stage occurs during winter. Degenerative changes are accompanied by low levels of serum FSH, LH, and androgens. Reactivation of spermatogenesis occurs during the spring (Kerr et al. 1980; Irby et Fig. D. 267 (upper left). Seminiferous tubule, rat. Normal ger­ minal epithelium and tubular diameter al. 1984). At 30-40 days of age, rat testes descend into the Fig. D. 2(jdf/oKW left). Seminiferous tubule, diabetic BB scrotum; the inguinal canal remains open allow­ Wislar rat testis. Hypocellular germinal epithelium und de­ creased tubular diameter. H & E, x 400 ing the testes to retract and descend freely throughout life (Bivin et al. 1979). Spontaneous Fig. b.2ti9(upper right). Seminiferous tubules, rat. Changes typical of chronic vitamin A deficiency. Note reduced tu­ cryptorchidism occurs rarely. However, a strain bular diameter und preservation of a thin rim ofspcrmato- with a genetic predisposition has been described gonia and Senoli's cells. (M ason 1933, p 232) H &. E, x 400 (Fechheimer 1970). A high incidence has also been repotted in biotin-deficient rats (Russfteld Fig. D.llQflowerright). Seminiferous tubule, rat deficient in vitamin E. Note mullinucleated giant cells consisting of 1967). On the other hand, surgically induced, ex­ fused spermatid nuclei with typical cresentic degeneration. perimental cryptorchidism has been extensively (Slide courtesy o f Dr. James S. Nelson) H & E, x 400 studied in the rat. Typical atrophic changes occur 215

Jam es K. W right, Jr.

whether induced in prepubertal or adult rats. propria to nourish the germinal epithelium. Al­ “Sertoli-only” change requires approximately though severely atrophic tubules (Sertoli-only) 1 month (Davis and Firlil 1966; Lecson and Lcc- usually have thickened tunica propria and base­ son 1970). Temperature elevation of the cryptor- ment membranes, these changes may be second­ chid testis is presumed to be the insult initiating ary to atrophy rather than causative. Microan­ the degenerative changes. giopathy and atherosclerosis have not been dear­ Most rat strains reach sexual maturity at about ly demonstrated in diabetic rats and, therefore, 35 days of age (Russfield 1967). Maximal sperm arc unlikely etiological factors. Several studies production occurs from about 70 to 450 or more with alloxan diabetic rats appear to implicate the days of age (Saksena el al. 1979). Testicular atro­ pituitary. Testicular atrophy can be prevented or phy (hypocellularity type) occurs sporadically in ameliorated by the administration of cortisone. rats over 1 year of age and is nearly universal by HCG, insulin, or ascorbic acid, but not by testos- 18 months of age. Sertoli-only change and leronc-proprionate (Scaffidi et al. 1975; Chatter- markedly increased amounts of interstitial tissue jee 1966; Chatterjee and Mukherji 1966). (Fig. 265) are usually present in older rats (Wright According to Mason (1933), testicular atrophy et al. 1982; Uurck 1978; Coleman et al. 1977; An­ secondary to E-avitaminosis in the rat may be di­ ver and Cohen 1979). These are usually associated vided into five morphological stages: (1) chromo­ with Leydig's cell hyperplasia, which may be lysis and fusion of spermatozoa; (2) liquifaction string-like or nodular (Fig. 266). The frequency of and segregation of the chromatin of spermatids interstitial cell tumors with senile atrophy varies und secondary spermatocytes to one side of the from 0% to 100% depending on the rat strain. In nucleus (i.e., “cresent-like” nuclear changes) the F344 strain these interstitial cell tumors are es­ (Fig. 269); (3) fusion of cresent-like spermatids pecially frequent (Burek 1978; Coleman et al. and secondary spermatocytes into muliinudeated 1977; Lutzen and Ueberberg 1973). Senile atro­ giant cells (Fig. 270); (4) chromolysis of the pri­ phy is commonly associated with polyarteritis no­ mary spermatocyte and spermatogonia giant cell dosa of the testicular arteries und focal dystrophic nuclei; and (5) “Sertoli-only" change. Because the calcification of seminiferous tubules (Fig. 265). tubules are not affected simultaneously, different Inanition in the rat results in generalized weight stages may be seen in adjacent tubules. The initial loss and a slightly larger than proportional testic­ stages of damage occur after 50-100days of E- ular weight loss. Acute starvation causes slough­ avitaminosis; the entire degenerative process re­ ing of the distal portions of the germinal epitheli-’ quires an additional 35-50days. According to um, giant cell formation, and cessation of sper­ Mason (1933), the “cresent-like" nuclear changes matogenesis. This is followed by maturation and more extensive giant cell formation are pa­ arrest. Chronic starvation is characterized by thognomonic of E-avitaminosis. The degenerative small tubules lined by Sertoli's cells and inactive changes are irreversible even if vitamin E is re­ germ cells resembling those found in sexually im­ stored a week or more prior to any histological mature rats. These changes are completely und evidence of injury (Mason 1940). rapidly reversible when adequate food supplies Testicular atrophy secondary to A-avitaminosis is are restored (Mason 1933; Siperstein 1921). not as severe as that caused by E-avilaminosis Testicular atrophy of varying severity (maturation (Mason 1933). The more mature sperm-forming arrest, hypocellularity, and “Sertoli-only”) has cells are sloughed and most tubules appear in a been reported in rats with spontaneous diabetes state of maturation arrest with only the spermato­ mellitus (Wright et al. 1982; Murray et al. 1983), gonia and primary spermatocytes spared alloxan diabetes (SchofTling et al. 1967; Scaflidi (Fig. 269). However, spermatogenesis does not and Rotolo 1974), streptozotocin diabetes (Ok- cease entirely; a few tubules continue to attempt sanen 1975), diet-induced diabetes (Rosenmann an abortive type of sperm maturation (Mason et al. 1974), and diabetes secondary to 95% pan­ 1933). These degenerative changes are reversible createctomy (Lema et al. 1965). The mechanism of in 1 or 2 months with vitamin A supplementation. the degeneration change is not clear. In diabetic Complete atrophy does not occur because rats die men, testicular atrophy has been attributed to dif­ from the other sequelae of A-avitaminosis before fusion or perfusion problems secondary to base­ the residual germ cells are affected. If rats are sup­ ment membrane thickening, diabetic microan­ plemented with retinoids other than vitamin A, giopathy, or atherosclerosis. Because the seminif­ they do not die of these other sequelae and devel­ erous tubules are avascular, nutrients must diffuse op irreversible testicular atrophy (Thompson through the basement membrane of the tunica 1970). Alro|iliy, Testis, Rat

Several elemental deficiencies have profound ad* Testicular atrophy in the diabetic man is charac­ verse effects on spermatogenesis. Histochcmical terized by either maturation arrest of hypocellu­ studies indicate that zinc is incorporated into larity. Although Sertoli’s cells become relatively spermatids and spermatozoa. Acute zinc deficien­ more numerous, Sertoli-only change is uncom­ cies secondary to inadequate dietary zinc, pres­ mon. Some degree of tubular basement mem­ ence of zinc chelators, or increased levels of other brane or tunica propria thickening is usually pres­ related elements cause malurational arrest of the ent. Leydig's cells arc generally present in normal germinal epithelium. Chronic zinc deficiency numbers. Interstitial tissue may be normal or in­ causes irreversible complete atrophy in postpu- creased. These changes are often associated with hertal rats; however, the deficiency is reversible testicular arteriosclerosis (Fcderlin et al. 1965; Iri- by zinc repletion in prepubertal rats (Mason et al. sawa et al. 1966; Facrman et al. 1972; Singhal et 1982). Manganese and selenium are also incorpo­ al. 1969; Chokyu 1965; Schoffling 1971; Thiel et rated into spermatozoa. Chronic dietary manga­ al. 1981). Diabetic testicular atrophy has also been nese deficiency causes maturation arrest at the described in other species including obese hyper­ spermalozoal level (Orent and McCollum 1931), glycemic AO mice and diabetic Chinese hamsters its point of incorporation during spcrmiogenesis (Schem ing1971; Scheminget al. 1967). (Gunn and Gould 1970). Although testicular Vitamin E, selenium, zinc, und manganese dcfi- size may decrease to 50% of normal, atrophy is cencies are laboratory stales that are of theoretical reversible by manganese repletion (Gunn and significance; these deficiencies have not yet been Gould 1970). Dietary selenium deficiency in the conclusively demonstrated in man. Human vita­ rat causes spermatozoa! midpiece breaks and min A deficiency has not been associated with tes­ poor sperm motility, but seminiferous tubule his­ ticular pathology. The effects of inanition on the tology is essentially normal (Gunn and Gould human testis include mild atrophy with near-com­ 1970). plete absence of spermatozoa, peritubular fibro­ Ionizing radiation and numerous drugs prevent sis, and increased interstitial cells (Keys et al. spermatogenesis and cause testicular degenera­ 1950; Jackson 1925). Multinucleated giant cells tion in the rat. These have been reviewed else­ arising from germinal epithelium do not occur in where (Ellis 1970; Timmermans 1974; Gomes human testicular atrophy regardless of its etiolo­ 1970; Jackson 1959) and are outside the scope of gy- this review. Most nutritional deficiencies affect common lab- ■ oratory animals in a manner similar to that de­ Comparison with Other Species. Degenerative scribed for the rat. Vitamin E deficiency is an ex­ changes within rat testes have been more exten­ ception. In the rat, testicular damage secondary to sively studied than those in other species, includ­ E-avitaminosis is irreversible. On the other hand, ing man. Cryptorchidism is probably the most in­ degenerative changes occur slowly and are revers­ tensively studied mechanism of testicular atrophy ible in the hamster. Changes include progressive in man. Cryptorchid human testes undergo de­ reduction in the size of seminiferous tubules and generative changes similar to those in the rat but in the height of the germinal epithelium; multi­ end-stage damage requires many years. In man, nucleated giant cells, sloughing of the germinal lack of tubular growth and decreased numbers of epithelium, and in situ germ cell necrosis arc spermatogonia can be recognized by the 3rd year markedly decreased. Accumulation of acid-fast of life. “Sertoli-only" change does not occur until pigment in Sertoli's cells is markedly increased in after puberty. Eventually, thickening o f the base­ the hamster (Mason and Maucr 1975). Similar ment membrane and tunica propria become findings have been reported in the testes of vita­ prominent (Nistal et al. 1980; Wong et al. 1973; min E deficient guinea pigs and rabbits. The Levin 1979). Cryptorchidism in man is associated mouse testis is resistant to damage from vitamin E with an increased incidence of germ cell neo­ deficiency (Muson and Horwitt 1972). plasms in both the ipsilateral and cont(alaterul testes (Ruber 1982). Senile testicular atrophy occurs in man. It is char­ References acterized by decreased testicular mass, decreased or absent spermatogenesis, peritubular fibrosis, Anver MR, Cohen BJ (1979) Lesions associated with and mild Leydig's cell hyperplasia; Sertoli-only aging. In: Baker HJ, Lindsey JR. Weisbrolh SH (eds) change may also occur. Frequently, normal sper­ The laboratory rat, vol I. Academic, New York, chap matogenesis is present focally. 14 Jam es R. Wright, Jr.

Hivin WS, Crawford MP, Brewer NK (1979) Morphophy- Keys AB, Brozek J, Hcnschel A, Mickelsen O, Taylor HL siology. In: Baker HJ, Lindsey JR, Weisbrolh Sll (eds) (1950) The biology of human starvation, vol 1. Universi­ The laborato/y rat. vol I. Academic, New York,chap4 ty of Minnesota Press, Minneapolis Burek JD(1978) Pathology of aging rats. CRC, West Palm Kubcr W (1982) Testicular tumor and cryptorchidism. Eur Beach, pp 129-138 Urol 8: 280-283 Chatterjee A (1966) Prevention of testicular degeneration Lema BE, Foglia VG, Fernandez Collazo E (1965) Le- in ailoxanized rats by use of cortisone. Acta Anat (Basel) sioncs histologicas testiculares en la rata diabetica. Rev 63:591-595 Soc Argent Biol 41: 197-203 Chatterjee A, Mukherji M (1966) Prevention of gonadal Lccson TS, Leeson CR (1970) Experimental cryptorchi­ degeneration in male diabetic rats by the use of insulin. dism in the rat. A light and electron microscope study. Endokrinologie 50: 11-14 Invest Urol 8: 127-144 Chokyu K (1965) Studies on diabetes mellitus and func­ Levin IIS (1979) Testiculur biopsy in the study of mule in­ tions of the male sexual glands. Acta Urol Jpn 11: fertility: its current usefulness, histologic techniques, 850-876 and prospects forlhc future. Hum Pathol 10:569-584 Cohen BJ, de Vries MJ (1969) Giant ceils in the rat testes Lutzcn L, Ucbcrbcrg H (1973) A study on morphological after accidental and experimental thermal injury. Pathol changes in the testes of old albino rats. Beitr Pathol 149: Eur4:336-344 377-385 Coleman GL, Barthold SW, Osbaldiston GW, Foster SJ, Mason KE (1933) Differences in testis injury and repair af­ Jonas AM (1977) Pathological changes during aging in ter vitamin A-deficiency, vitamin E-deficiency and in­ barrier-reared Fischer 344 male rats. J Gerontol 32: anition. Am J Anat 52:153-239 258-278 Mason KE (1940) Minimal requirements of male and fe­ Davis JR, Firlit CF (1966) The germinal epithelium of male rats for vitamin E. AmJ Physiol 131:268-280 cryptorchid testes experimentally induced in prepuber­ Mason KE, Horwitt MK(I972) Effects of deficiency inan- tal and adult rats. Fertil Slcril 17:187-200 irnals. In: Scbrell WH Jr. Harris RS (eds) The vitamins: Ellis LC (1970) Radiation effects. In: Johnson AD, Gomes chemistry, physiology, pathology and methods, vol 5. WR, Vandemark NL (eds) Tire testis, vol 3. Academic, Academic, New York, chap 16, pp 272-292 New York, pp333-376 Mason KE, MauerSI (1975) Reversible testis injury in the Faerman I, Vilar O, Rivarola MA, Rosner JM, Jadzinsky vitamin E-deficient hamster. J Nutr 10S: 484-490 MN, Fox D, Uorct AP, Eemstcin-Hahn L, Saraceni D Mason KE, Bums WA,Smith JC Jr (1982) Testicular dam­ (1972) Impotence and diabetes: studies of androgenic age associated with zinc deficiency in pre- and postpu- function in diabetic impotent males. Diabetes 21: bcrtal rats: response to zinc repletion. J Nutr 112: 23-30 1019-1028 Fcchheimer NS (1970) Genetic aspects of testicular devel­ Murray FT, Cameron DF, Orth JM (1983) Gonadal dys­ opment and function. In: Johnson AD, Gomes WR, function in the spontaneously diabetic BB rat. Metabo­ Vandemark NL (eds) The testis, vol 3. Academic, New lism 32 (Suppl 1): 141-147 York, c h a p l, pp2-40 Nelson JS (1980) Pathology of vitamin E deficiency. In: Federiin 1C, Schbffling K, Neubronncr P, Pfeiffer EF Machlin LJ (ed) Vitamin E: a comprehensive treatise. (1965) Histometrische Unlersucliungen am Hodenge- Dekker, New York, chap 7, pp 397-428 webe des Diabetikers mil Keimdriisenunterfunktion. Nisial M, Paniagua K, Diez-Pardo JA (1980) Histologic Diabetologia 1:85-90 classification of undescended testes. Hum Pathol 11: Gomes WR (1970) Chemical agents affecting testicular 666-674 function and mate fertility. In: Johnson AD, Gomes Oksanen A (1975) Testicular lesions of streptozotocin dia­ WR, Vandemark NL (eds) The testis, vol 3. Academic, betic rats. Horm Res 6:138-144 New York,chap11,pp483-5S4 Orenl ER, McCollum EV (1931) Effects of deprivation of Gunn SA, Gould TC (1970) Cadmium and other mineral manganese in the rat. J Biol Chcm 92:651 -678 elements. In: Johnson AD, Gomes WR, Vandemark NL Rosenmann E, Palti Z, Teitelbaum A, Cohen AM (1974) (eds) The testis, vol 3. Academic, New York, chap 10, Testicular degeneration in genetically selected sucrose- pp 378-481 fed diabetic rats. Metabolism 23: 343-348 Irby DC, Kerr JB, Risbridger GP, de Krelser DM (1984) Russfteld AB (1967) Pathology of the endocrine glands, Seasonally and experimentally induced changes in the ovary and testis of rats and mice. In: Cotchin E, Roe testicular function of the Australian bush rat (RaUusfus- FJC (eds) Pathology of laboratory rats and mice. Black- cipes). J Reprod Fertil 70:657-666 well, Oxford chap 14 Irisawa S, Shirai M, Matsushita S, Kagayama M, Ichijo S Sakscna SK, Lau IF, Chang MC (1979) Age dependent (1966) Sexual disturbances in diabetes. Tohoku J Exp changes in the sperm population and fertility in the male Med 88:311-326 rat. Exp Aging Res 5: 373-381 Jackson CM (1925) Effects on the male reproductive tract. Scaffidi L, Rotolo G (1974) II danno testicolare nel diabete In: The effects of inanition and malnutrition upon spcrimcntale da allossanu nel ratio. Minerva Med 65: growth und structure. Blakislon, Philadelphia, chap 25 4477-4484 Jackson H (1959) Antifertility substances. Pharmacol Rev Scaffidi L, Rotolo G, Scaffidi A (1975) Effetlo protettivodi 11:135-172 alcuni ormoni (insulina, testosterone, HCG) sul danno Kerr JB, Keogh EJ, Hudson B, Wliipp GT, de Krctser DM testicolare nel diabete da allossanu nel ratio. Minerva (1980) Alterations in spermatogcnic activity and hor­ Med 66:4586-4591 monal status in a seasonally breeding rat, Rullus Jus- Schdffling K (1971) Diabetes mellitus and male gonadal cipes. Gen Com p Endocrinol 40: 78-88 function. In: Rodriguez RK, Vallance-Owen J (eds) Dia- 218

Atrophy, Testis, Rat

boles: proceedings of the seventh congress ol' the Inter­ Swensen SR, Telford IK (1973) Lipofuscin distribution and national Diabetes federation, Excerpta Medica, Am­ histological lesions in the vitamin U deficient cotton rat sterdam (Signiodwt luspiiius liis/iiJui). Arch Hislol Jpn 35: Scheming K, Federlin K, Schmitt W, Pfeiffer EF (1967) 327-341 Histometric investigations on the testicular tissue of rats Thiel W, L6wO, SchreiberG (1981) Hodcnbioptische Un- with alloxan diabetes and Chinese hamsters with spon­ tersuchungen bei milnnlichen Diabelikem. Z Gesamte taneous diabetes. Acta Endocrinol (Copcnh) 54: Inn Med 36:917-920 335-346 Thompson JN (1970) The role of vitamin A in reproduc­ Singhal KC, Kastogi GK, Aikal UK, Chhultani PN (1969) tion. In: DcLuca HI-, Sutlie JW (eds) The fat-soluble vi­ Testicular histology in diabetics with impotence. Indian tamins. University of Wisconsin Press, Madison, J Pathol llacicriol 12: 145-148 chap 18, pp 267-281 Siperstein DM (1921) Tlic effects of acute and chronic in­ Timmermans L (1974) Influence of antibiotics on sper­ anition upon the development und structure of the testis matogenesis. J Urol 112:348-349 in the albino rat. Anat Kec 20: 355-391 Wong TW, Straus FH II, Warner NE (1973) Testicular bi­ Sleinberger E (1970) Effect of altered blood blow on the opsy in the study of male infertility. I. Testicular causes lestis. In: Johnson AD, Gomes WR, Vandemark NL of infertility. Arch Pathol 95: 151-159 (eds) The lestis, vol 3. Academic, New York, chap 8, Wright JR, Yates AJ, Sharma HM, Shim C, Ttgner RL, pp313-332 Thibert P(1982) Testicular atrophy in the spontaneously diabetic UU Wistar rat. Am J Pathol 108: 72-79 Appendix E 220

laboratory Anifii.il S ch trc Copyright ' 19M by th i’ A n u rica n A.v«K.T

Spontaneous Gastric Erosions and Ulcerations in BB Wistar Rats1,2,3

James R Wright Jr, Allan J Yates, Hari M Sharma, and Pierre Thibert

Summitry | One hundred thirty-four BB Wislar diabetic ruts, 31 nondiubctic siblings, and 30 Wistar ruts were nccropsied. Guslric erosions and ulcers were observed in 43 (32.1%)ofthcdiubetic ruts und three (9.7%) of their nondiubetic siblings. None of the Wislar rats showed evidence of gastric mucosal injury. Lesions were most frequently found in the thick-walled, glundular portion of the stomach and were morphologically consistent with stress ulcers. BB W istar rats may be valuable for studies on gustric stress ulcers.

Key W ords | Diubeles mellitus — Gustric mucosa — Ulcer — Hattun species

The BB Wistar (Bbk(WI)] rut, a promising ani­ the lleulth Protection Branch of the Cunudiun Govern­ mal model for human diubetes mellitus, was developed ment. Animuls were shipped by commercial uirlines from sevcrul years ago from un outbred line of Wistur rats ut the Ottuwu to Columbus, Ohio where they were originully Bio Breeding Laboratories in Ottuwu, Cunudu (1). Diabetes housed in a conventional animul housing facility, and later in the BBW rat occurs spontaneously between GO and 120 in a semi-barrier housing facility. The animal room was days of age in 30 to 55% of the offspring of diabetic parents. m aintained under slight positive pressure a t 22°C and was The mechanism of transmission is unknown. supplied with fresh air pussed through high efficiency The diabetic syndrome is churucterized by purticlo filters at a ruto of 16 uir changes per hour. Humid­ hyperglycemia, glycosuria, ketoacidosis, insulinopcniu, ity vuried between 45% und 55%, and a constant photo­ glucagoncmiu, and hyperlipemiu. Without insulin ther­ period was maintained with a 12-hour light-durk cycle. apy, diabetic rats develop rapid weight loss with polyuriu Rats were housed as mating pairs in 27.9 x 48.3 x 17.8-cm resulting in dchydrution. Deuth frequently follows within polycurbonutc cugcs with hardwood chip bedding.* All rats 1 to 2 weeks. In addition to these feulurcs, the eurly onset of hud uccess to commercial rut diet1 und wuler ud libitum. diabetes in the absence of obesity more closely mimics the All rats were weighed and monitored daily for clinicul syndrome of juvenile-onset diabetes in mun than do ketones* und glucose’ in thu urine with laboratory test other animul models. Unfortunately, the BBW rut is sus­ strips. Glycosuria wus estimated semiquuntitutively on u ceptible to u wide vuricly of degenerative, inllummulory, sculc from 0 to 4 plus; ketosuriu wus recorded us zero, smull, und neoplastic disorders thut uppeur unrelulcd to diubetes or lurge. Proluminc zinc (U-40) insulin* wus administered mcllitus (2). We report here in dctuil one of these complica­ subcutunoously each morning to maintain 4-1- glucose tions, gastric ulceration. (without kotosis) in the urine. This treatment regimen was used because our colony wus being employed in other stud­ Materials and Methods ies involving the complications of long term, poorly control­ A breeding colony of BB Wistur ruts was estab­ led diabetes. The initial dosage was determined on the lished at Ohio Stale University from the parent colony ut basis of body weight and then adjusted according to indi­ vidual responses to therapy determined by urinalysis. Non­ diabetic siblings, BB Wistar ruts thut did not develop 'Front the Ohio Statu University, Colicuc of Medicine, Doimrlniciil of FuOiolutty.ColumbuH.OH 43210. (Wriphl.Yttlurt.Shurmut.unu the Animul diubetes, and commercially purchused outbred Wistur ruts’ ItvMiurcva Uivuiuii , Hcullh Frolcclion lirumli, 1 leultlt und Wellure Cunudu, were housed under identicul conditions but did not require Ottawu, Ontario, Ctmudu K1A 0L2 tTliibort). * 'Supported in purt by Ute Department ol Fulhulupy, the Collceo of Medicine, insulin therupy. und the Graduate School or the Ohio Suite University und by the Upjuhn Compuny. All rats which cither died spontaneously or were 'I'hc author* wmh to thank Uuve Covey. Cindy Oiaon, und llruce Thunip*on killed for experimentation were necropsicd. Most major for llivtr ullentivu uuiiniouunco of (lie ill! WinUir column* in t ’uluiubun und Ottuwu. organs were examined for gross ubnormulitics, weighed, ‘Suni'Chipu. FJ Murphy Co, Moonuchie, NJ. fixed in neutral phosphute buffered formalin, und proces­ *Kudcnl Luborutury Chow #5001 *. Itui&ton I’urinu Co, Hichlund, IN. "Ketoatix*. Ames Co. Klkhurt, IN. sed for light microscopy. Thu stomach wus opened anterior­ ^leaiupo*. Eli Lilly, lndiunupoln>, IN. "Eli Lilly, lndiunupolih. IN. ly by means of a longitudinal incision from the curdiuc "llurlun Indublnvtt, Indiunupolis, IN. sphincter to the proximul duodenum. Lumcnul contents 221

Vol Jl, No I l.ul>0Tdt4iry A t i i i h .i I S c i e n c e Kebruary 1^1

were examined, and the mucosal surface was washed and diahctcs was 92±24 days for rats with gastric lesions and carefully inspected with the naked eye. Because a dissect­ 108252 duys fur rats without lesions. ing lens was not employed for this examination, it is possi­ Lesions were either pinpoint in size and shape, or ble that very small or completely healed lesions were over­ long (up to 10 mm) and narrow (Figure 1). They were looked. The intestines also were opened and carefully ex­ multifocal in ull but three rats, usually present only in the amined. Rcd-tintcd lumenal contents were frequently thick-wulled glandular portion of the stomach, and most tested for blood with lest strips,1'1 but this wus not routinely frequently occurred at the bases of the rugae. However, two done. Tissue from any abnormal areas us well us other ruts ulso hud erosions in the thin-walled squamous portion representative portions of the stomach, lurge and small of the stomach. The ulcer base frequently was stained dark bowels were fixed and processed for light microscopy. When brown by the ucid digestion products of the exuded erythro­ multiple gastric lesions were present, three to five ureas cytes. The surrounding mucosa was occasionally representative of these were sampled. I’urufiin embedded hypcremic. The gastrointestinal luminu of severely tissues were sectioned und routinely stained with hema­ affected rats usually were virtually devoid of solid food toxylin und cosin. Five serial sections of each tissue sample particles but contuincd vuriublc umounls of blood stained with a lesion, and one section of samples with no gross fluid. lesion were examined histologically. No attempt was made to quuntitute the number of lesions in cuch animal, nor were distinctions mude between erosions (those lesions involving the mucosa only) and ulcerutions (those lesions perforating the muscularis mucosae). Several representa­ tive sections of stomach lesions with u light brown pigment were also stained with Gomori stum for iron, oil,red 0 fur lipofuscin, and periodic ucid Schilf (PAS) stain. Sovcrul comparisons were made to determine whether the quulity of diubelic control uffectcd the inci­ dence of gustric lesiuns. The frequencies of gustrointeslinul lesions were compared in animals dying spontaneously with those killed for experimentation. Treulment records for individual diubetic rats ulso were examined. Since stress ulcers are considered un ucule process, we utilized only the records for the duy prior to dculh for statistical analysis. The following purumctcrs were compared: (1) in- Bulin doeugu recoived, (2) level of glycosuriu, and (3) pres­ Fig. E.l ence or absence of kctosuria. All frequency comparisons Multifocal gastric erosions and ulccrntionB with poorly defined were made using u two-tuilcd standard normal (Z) test for margins in thick-walled glandular portion of stomach. Many le­ the difference between two proportions (31. Meuns were sions are ut buse of rugae und are darkly sluined. Eudi unit on rule compured by two-tuiled Student’s (-test. - 1 mm.

Results The smull lesions wero restricted to the mucosu Necropsies were performed on 134 BBW diabetic and histologicully consisted only of coagulation necrosis rats(107 mule und 37 female), 31 non-diabetic siblings(17 (Figure 2). Murked hyperemiu usually wus found peripher­ mule und 14 fcmulc), und 30 Wislar ruts (ull mule). Gastric al to thuse foci which frequently contuincd u light brown erosions and/or ulcers were observed in 43 132. 1%) of the pigment thut wus negutivo with Gomori stain for iron, oil diabetic ruts und three (9.7%) of the non-diabetic siblings, red 0 for lipofuscin, und PAS. This suggested thut the but none were seen in the Wistur ruts. Dillerences in the pigment was aposiderin, un iron-free breukdown product of frequencies of lesions in the two groups of BUW rats were hemoglobin (4). Leukocytic marginution was found in the significant (Z=2.51, p<0.05), but there wus no difference in submucosal blood vessels in these m ild lesions, but in more the frequencies in male (31.8%) and female 133.3%) diabe­ severely affected foci, a neutrophil-rich exudate was pres­ tic BBW rats. Two of the three non-diabetic BBW rats with ent throughout the necrotic mucosa. In still more severe lesions were female. lesions, the necrotic tissue had sloughed, resulting in No significant age differences were ubserved be­ either un erosion or an ulcer. Although no distinction be­ tween groups of ruts. The meun ages of BBW rats with tween erosions und ulccrutionB wus mude for quantitative lesions, BBW ruts without lesions, und Wistar rats were studies, lesions of both types were observed, und it is our 2652:91 duys (range was 103 to 454 days), 243293 days impression that erosions predominated. (range was 91 to 464 days), and 2552200 days (range was A significantly higher incidence (39.3%.) of le­ 60 to 551 days), respectively. The meun age of unset of sions was observed in the 102 rats which died spontaneous­ ly when compured to the incidence (9.3%) in the 32 diabetic ruts killed for experimentation (Z-3.51, p<0.001). The Amen Co, Elkhurl, IN. meun dosage of insulin on the day prior to death (Table 1) Gastric lesions in Rats

entirely responsible for the gustric lesions in the BB W istar rat. Numerous factors affect the resistance of gastric mu­ cosa to injury (71, and although it is theoretically possible that some of these mechanisms may be compromised by the diabetic slate, this has not been established. In fact, a decreased incidence of gastrointestinal ulcers in human diubetus has been reported (8) which may be due to u subnormal capucity to secrete gustric acid (9-101. A recent study may offer some insight into the mcchunism of the mucosal injury in the D13 Wistar rat. It reported u significant increase in anlrul und serum gastrin levels in ubese geneticully diubclic mice when cumpurcd to litter mate controls (11). The ussuciution of gastrin hypersecretion und the occurrence of multifocal gustric ulcers, for example, the Zollinger-Fllison Syndrome in man Fig. E.2 (S), is well established. A lthough gastrin levels huvc not Small gastric erosion comprised of well circumscribed urea of been studied in the BB Wistur rut, in view of our findings necrosis involving gustric rnucosu. Murgins of lesion are shown by this would be very interesting. ligher arrows. Durkerurrows poinllo region of necrosis containing brown pigment with staining properties of uposiderin. Hematoxy­ Gastric erosions are infrequent findings in lin and eosin stain. Line - 9G pm. routine necropsies of laboratory ra ts (12—14). Therefore, it is nut surprising tliul we found no lesions in the thirty Table E.l Wistur rats that were housed with the BBW rats. Since similar gustric lesions were observed in both diabetic und Compunawta of tnNtUtictil uni hvillli Mstlu* vt IIUW diaUlK ntU with non-diubctic BBW rats, it is likely that the BBW strain, and without tfa«tnc Kmkmui regardless of the diabetic condition, has an increased sus­

l^ivt insulin ceptibility to gustric mucosal injury. However, the differ­ Number iklM-" Kn^ueiH'y uf ence in incidence of the lesions in the BBW siblings indi­ uf rut* m il (ilvtinunub kt-uwiunrf* cutes that the diabetic stule favors the expression of these lUta with lesions. Furthermore, gustric lesions were more frequently lOOUIIS lib 1.47 s OH*- * 2 34 s 1 tO* JI4 U observed in diubclic ruts that died spontaneously than in Rats witlwut those that were killed for experimental reusons. This sug­ Inoons H3 a v J i o a y 2 U 2 : 1 71 22 *K gests thut poorly controlled diabetes may serve as a stress­ ‘Insulin duse is in itiU-nwltuiuil units *Urgrt* of gljuaurU *a on a acalv uf U tu 4 |4ua ful condition thut initiates the necrotic process in the gus- 'Kvfawunu w«s drionnitiod aa «nlhrr prtM*ni or uU*nt tric mucosa in these rats. 'M ean x atandanj deviation •SttfiuftconUy difUtvtil fnmi rata wiUiuut Uwiuna Since all of the BBW diabetic animals in this study had glycosuria, they were by most stundurds poorly controlled. Indeed, we adjusted their insulin on a daily was significantly higher in unimais with ll.47i0.82 1U) basis to maintain them with glycosuria but withuut tliun without (0.93±0.89 1U) gustric losiuns (p-:0.01). ketosuriu. However, some deviations from these levels Moan lovols of glycosuria wore essentially the sume for were unuvoidublc. In order to determine if gastric lesions both groups. Kclosuriu wus more frequently observed in were related to the degree of glycosuria, kclusuriu, or in­ rats with ulcers, but this wus not statistically significant. sulin dosage, the relevant dulu (rum the treatment records for each rut's finul duy of life were grouped uccording to Discussion whether lesions were present or absent. Both groups uf rats Gastric erosions und ulcers cun be divided into hud similar levels uf glycosuria, and ketosuriu was only two categories, ucute and chronic. Acute lesions (ulcers and slightly more frequent in diabetic ruts with ulcers iTubte erosions) usuully have poorly defined margins and fre­ 1). However, meun insulin dosage for ruts with ulcers wus quently ure stained dark brown due to the ucidic digestion significantly higher on their final day uf life thun for those of erythrocytes. They ure commonly multilbcul undhuvc no without ulcers. It is u well-known fact thut a large vuriety preferred site in the gastric mucosa. Chronic lesions, on the of stressful stimuli leud to increased insulin requirements other hand, usually ure solitary und are almost exclusively in diulietics (51. Therefore, the larger insulin dosage in located in the gustric antrum or lesser curvature. They those unimais which died with gastric lesions possibly re­ typically have u sharply demarcated border und a clcnn- flects that they were in u greater degree of stress prelermi- appeuring base (5). The uppeurunce of the lesions in the lift nully thun those without lesions. The nature of this stress is Wistur rats was typical of ucute mucusal injury. currently unknown. This interpretation seems mure likely The pathogenesis of such gustric lesions is com­ than the alternative possibility that the larger amuunt of plex und still has not been completely elucidated. Diabetes insulin administered caused the erosions und ulcers, be­ is known loulfect gastrointestinal tract structure and func­ cause there is no evidence indicating a rnlcofcithcr insulin tion (til, but it is unlikely that the diabetic state /n r sc is or hypoglycemia in the pulhugcncsis uf stress ulcers. 223

Vul 31, No I Liilwntlory Animal Science February 19H1

Because of the clinical importance of stress 4. L illie KD, Fullm erH M . Histopathologic Technic and ulcers, there has been considerable effort to develop a reli­ Practical Histochemistry. New York: McGraw-Hill Hook Co. 1976. 0. Bobbins SL, Colran \{S. Pathologic llusts o f Disease. able experimental model for gastric ulceration. Erosions Philadelphia: WB Saunders, 1979. have been induced in rodents by a variety of experimental 6. Katz LA, Spiro HM. Gastrointestinal manifesta­ techniques including electric shock with and without tions of diabetes. N Engl J Med 1966; 275:1350-61. 7. Kivilaak.so E, Silen W. Pathogenesis of ex­ starvation, pyloric ligation, variable means of physical re­ perimental gustric tnucosal injury. H Engl J Med 1979; 301:364- straint with and without starvation, and restraint with 69. 8. Dolcvull G. Incidence uf peptic ulcer in diabetes cold ambient temperature (15). However, all of these have meililus. Ai7a Med Scund 19f>9; 164:463-77. shortcomings as models of the human disease. Kor exam­ 9. Marks IN , Shuman CK, Shav H. Gastric acid secre­ ple, many of them produce lesions predominately in the tion in diabetes mellitus. Ann Intern Med 1959; 51:277-37. 10. Angcrvul! L, DotevullG, Lehmann KK.The gastric non-glandutar portion of the stomach and most of the le­ mucosa in diabetes inellitus. A functional and hislopathological sions are restricted to the mucosa. It is thus apparent that a study. Acta Med Scund 1961; 169:339-19. spontaneous model more closely mimicking human acute 11. Lichlenberger LM, Hainaswamy K. Alterations in serum und unlral gastrin levels in genelicully diulietic mice. G\i.s- gastric stress ulcers would be quite valuable. troenterology 1979; 77:1270-82. 12. Coleman GL, Barthold SW, Osbuldi.slon GW, <7 al. Pathological changes during aging in barrier-reared Fischer 344 male ruts. J Cierontol 1977; 32:258-78. References 13. Pollard M, Kajima M. Lesions in aged germfree 1. Nukhoodu AF, Like AA.Chuppel C l, c7 a/. The spon­ Wistur ruts. Am J Pathol 1970; 61:25-36. taneous diabetic Wistur rat: Metabolic and morphologic studies. 14. Dullock BC, Hunks KL, Manning PJ. Common le- Diabetes 1977; 26:100-12. biuns in the uged rat. In: The lAiborulory Animal in Gerontological 2. W right JK, Sharmu H ,T h ib crl P, etal. Pathological Research. Washington: National Academy ofSciencc, Publication findings in the spontaneously diabetic HU Wistar rat. Lab invent 1591, 1968:62-82. 1980; 42:66. 15. Gluvin GH. Kcslrninl ulcer: History, current re­ 3. Ferguson GA. Statistical Analysis in Psychology and search und future implications. Ilrain Res Hull 1980; 5 (Suppl Education. New York: McGraw-Hill Book Co, 1971. 11:51-58. Appendix F TKKATOLOGY 27:117-120(1983)

Supradiaphragmatic Accessory Lobe of the Liver in BB Wistar Rats

JAMKS n. WHIUIIT, JH , HAW M. SHARMA. ALLAN J. YATKS, IIHUCK M. THOM PSON, a n i i I’IKUKK THIIIKKT Department o f Pathology, The Ohio State University, College o f Medicine, Columbus, Ohio 43210tJ.H. W., Jr., It.M.S., A.J.Y.), and Animal Resources Division, Health Protection llruneh. Health and Welfare Canudu, Ottauiu, Canada K tA OL2 (U. M. T, P. 77 ABSTRACT Supradiaphragmatic accessory livers were observed in two closely related rats in u series of 172 necropsies on BB Wistar ruts. The gross und histologicul appearance of both accessory lobes ure described. This abnor­ mality has been reported in only one other inbred struin of ruts where it ulso arose with a very low incidence. As in the previous report, the pattern of occurrence of these accessory lobes suggests a mode of inheritance thut is either polygenic or uulosoinal recessive with low pcnelrunce.

A wide variety of gross developmental ab­ KKSULTS AND DISCUSSION normalities of the liver has been reported infrequently in man (Lcevy et al., '79; Cul­ Two supradiaphragmatic accessory livers len, ’25), but these appear to be exceptionally were observed in the 172 BBW ruts necrop- rare in the rat (Wilson, ’78). The following is sied in our study (i.e., 1.16%). Grossly, the a report of two observations of supradia­ accessory livei's in both ruts were attached to phragmatic accessory livers in BU Wistar the rostral right hcmidiuphrugmulic surface (BBW) rats. (Fig. 1). In both ruts the accessory lobe was Thu BB Wistur (BBW) rut, u struin thut upproxiinulely 1.5 cm in diameter. The dia­ wus developed from un outbred line of Wistur phragm nppeurcd grossly normal. Histologi­ ruts at the Bio Breeding Laboratories in Ot­ cally, there wus u transition from skcletul tawa, Cunadu, develops a syndrome that muscle fibers to dense rcgulur connective tis­ closely mimics human juvenilo-onset diabe­ sue to dense irregular connective tissue neur tes. The BBW syndrome is characterized by the site of attachment of the uccessory lobe. hyperglycemia, glycosuria, ketoacidosis, in- The histological appeurunce of the accessory sulinopcniu, and hyperlipemia (Nakhoodu ct lobes differed murkedly in the two ruts. al., '77; Nukhoodu el al., ’78). Onset of diubc- The first observation was in a 341-day-old tes occurs spontaneously ut 60-160 duys of mule BBW diabetic rat that died sponta­ uge in 30% to 55% of the ollspring uf diabetic neously. The accessory liver was severely patents. Onset is abrupt with rupid weight congested, multifocully hcmorrhugic und had loss due to dehydration as a result of poly­ numerous atypical large blood vessels. Ex­ uria. Death occurs within scverul weeks if tensive areas of fibrosis partitioned the he­ insulin therapy is not initiated. patic parenchym a (Fig. 2). Some fibrous ureas showed focal bile duct hyperplasiu, nonspe­ MATKItlALS AND MKTUODS cific chronic periportal inflammation, and hemosiderin-filled macrophuges (Fig. 3). The A total of 172 BBW rats (139 diabetic and capsule of the accessory liver was thickened 33 nondiabetic) that died spontaneously or and merged with the dense connective tissue were sacrificed for experimentation were necropsicd. Organs were removed en block, dissected, weighed, fixed in neutral phos- phate-builTcred formalin, und processed for llifvm il DtvuinW r 10,198); utvcjilud June 23,1982 light microscopy. Tissue sections were Addn, Neuiopulliol- stained with hematoxylin and cosin. otfy Unit. 473 W. !2lh Avenue, Oolumtnn., UH 43210.

0040-3709/S3/2701-OU7$01.50 <0 1983 ALAN R. L1SS. INC. 226

JR . WRIGHT, Jll. KT AL

The second observation was in a 415-day- old BBW nondiabetic rat that was sacrificed. Although grossly very similar to the first case, the histological picture wus quite differ­ ent. The accessory liver wus surrounded by a thin capsule; fibrosis, bile duct hyperplasia, and congestion were not present; and only a few foci uf insignificant pcriportul chronic inflammation were seen. The parenchyma n wus essentially normal. Several areas showed limited extrumcdullary hematopoicsis. This second accessory liver was histolopicully sim­ ilar to normal liver tissue. Complete pedigrees were not available for either of these rats, but we were able to de­ termine that both rats were closely related. Both of the parents of the nondiubetic rut with the accessory lobe were litter mates of the diabetic rut. A wide variety of developmental abnormal­ ities of the liver have been reported in man (Lecvy et ul., ’79, Cullen, ’25). Two types of accessory lobes were described by Grosfetd and Clutworthy C79u): the Reidel lobe, u tongue-like downward extension of the right lobe of the liver, and accessory lobes project­ ing into abdominal wall defects (e.g., omphul- ocele). In addition, ectopic liver tissue in the gall bladder, suspensory ligament, adrenal glund, splenic capsule, umbilicus, or lung F ig . F. 1 Thoracic cavity of the acconJ tnondiobclic) rut have been reported as incidental findings ih opened and tlie thoracic content* ure reflected cuu- dally to luvcul a auprudiuphraginulic ucceiMury liver lobe (Lecvy et ul., '79, Cullen, ’25). Hcputic ha­ (urrowu outline the marginal in the right hentilhorax. martomas are occasionally seen in man, but leaded blood ia uluo preaenl in the thurucic cavity. A, lull these ure typically loeuted in the right lobe lung; [I. right lung; C, liver; I), heart. of the liver (Crosfeld und Clutworthy, ’79b). All of these differ from the supradiaphrag­ matic accessory livers present in the BBW ruts. ut the site of ultuchmcnt to the diuphrugm. Congonitul diaphragmatic herniation can Occasional hcpalocytes and bile ducts were result in an intruthoracic liver surrounded incorporated into the connective tissue por­ by incompletely developed diuphragm mus­ tion of the diaphragm near the site of attach­ cle. This pattern was not seen in the BBW m ent (Fig. 4). The hepatocytos w ere not rats but has been reported in about 20% of arranged into the normal hepatic plates thut the offspring of irradiated short-eared mice radiate from the central veins of hepatic lob­ (Snell, '35). ules. This histologic picture is consistent with Developmental hepatic abnormalities are the diagnosis ofheputic humurtoma. The his­ extremely rare in the rat (Wilson, ’78). Only tologic uppcarance of the abdominal liver wus one previous study (Muchudo und Luzzio, ’72) basically siinilur to that of the ucccssory liver has reported uccussory supradiaphragmatic except thut lobular architecture wus normal lobes in that species. These occurred in the and fibrosis was minimal. Additional patho­ right hemithorax in 9 of 2881 (0.31%) Gunn- logical findings in this rut were biluterul tes­ derived ruts. The accessory lobe was indepen­ ticular atrophy, a pancrculic granuloma, dent of both the jaundice and hydronephrosis chronic thyroiditis, and plusmu cell lymphad­ traits frequently seen in these animal. Since enitis. All of the additional findings ure com­ one of our two unimais wus a nondiubetic mon in the BBW strain (Wright et al’., '80; sibling, this indicates that this is u strain- W right e t al., '82). related abnormality rather than being linked SlIPItADIAPIMAtiMATIC ACCESSORY I.IVEH IN HUW KM'S

wLJ u 1 >■+...<>}' r’VafiflWmS^a^

F ig . F • 2 Photomicrograph of an uccujatury liver lube in und ubnormul lubulur architecture H tU K ululn, original the llr»l(diulK.’licJ rat showing extensive ureuauf fibrosis magnification, x 25.

Fig* F* 3 Fibrous septum shewing bite duel hypcrplusiu in the center of the septum. Severe congestion is seen in tn the ncivtisnry U»U* of the first rat. Ilemosntcrin-lu* the diluted hepatic sinusoids. H & B atuin, originul mug- den niucrophugcs ure present near the lurgc blued vessel nilicutiun, x 25. J II WK1GHT, JII ET AL

ACKNOWLEDGMENTS This work was supported in part by N1H v , ■ . j> grant NS-18026, the Department of Pathol- V - ^ Of^y, the College of Medicine and the Grudu- r* *-'■ , »*• ’’ ‘ : ate School of the Ohio Stute University. Additional support was received from the Upjohn Company. Mr. James R. Wright, Jr. was supported by the Samuel J. Roessler Re­ search Scholarship. The authors would like to thank Dr. Martha E. Sucheston for review­ ing the manuscript.

I.ITKItATUUK CITED F ig , F.4 Fibrous portion of the diuphrugm of tliu first rut containing hepatic clement* ueur the site uf attach­ Cullen, T.S. UU2M Accessory lubes of the liver. Arch, m ent of th e accessory loin:. N um erous bile duel* are been Surg., //.7IH-7G4. ill ihe center und lo ihe right. Severul hcpatucytes ure Grosfctd, J.L., und II.W. Clutworthy (1979a) Accessory present in the loll portion of the diuphrugni. 11 4t 1C slain, hcpulic lobes. In: Ruth Defects Compendium. B. originul magnification, x 400. Bergsmu, ed., Alan II. Liss, NY, p. 520. Grosfeld, J.L., und II.W. Clutworthy (1979b) Liver he- murtomu. In: Uirlh Defects Com)>cndiuin, D. Uergsmu, Kd. Alun 11. l.iss, NY, pp. 059-660. U-evy, C.M., II. Popper, und S. Sherlock (1979) Congcni- lul or perinulul malformations of Ihe liver. In: Diseases with diabetes, the trait for which the BBW of the Liver und Biliary ’lYacl. Standardization of No- strain was inbred. inendulure, Diagnostic Criterut, und Diagnostic Meth­ Machado and Lozzio's (’72) description of odology. Custle House Publications, London, pp. 73- the microscopic appearance of the accessory 76. Machado, E.A., and D.D. Lozzio (1972) Accessory lobe of lobes in Gunn-derived ruts is very similar to the liver in Gunn-derived ruts. Tcrutology, 5:361-366. the picture in our lirst rut except that they Nukhoodu, A.K., A.A. Like, C.I. Chuppel, P.T. M urray, did not sec marked fibrosis or bile duct hy­ und E.B. Murliss (1977) The spontaneously diabetic perplasia. They reported poorly defined he­ W utur rut: Mclubolic und morphologic studies. Uiubc- patic plutes, congestion, hemorrhugc, and tes, 26:100-112. Nukhoodu A.K., A.A. Like, C.I. Chuppel, C.N. Wei, und limited fibrosis around utypicul blood vessels K.tl. M urliss (1978)1*110 spontaneously diubclic W istur as their predominant findings. rat (the "HR" rail: Studies prior lo und during devel­ Machado and Lomu ('721 concluded that opment of the overt syndrome. Diuhclulogia, jj.199- this may be an inheritable trail in Gunn ruts 207. Snell, 1).C. (1936) The induction of X-rays by lieredilury and if so muy involve u mode of inheritance change in mice. Genetics, 26.545-567. thut is either polygenic or autosoinul reces­ Wilson, J.A. (1978) Developmental abnormalities: Huts. sive with low penetrunce. Our study oilers In: Pathology of Laboratory Animuls. I. Uenirschke, further evidence for a genetic etiology. It is P.M. Gurner, und T.C. Jones, eds., Springvr-Vcrlug, NY, vol. 2, pp. 1840-18-17. well established thut inbreeding raises the Wright, J.K., H. Shurmu, P. Thibert, und A.J. Yules frequency of expression of rare genes. Since (1980) Pathologic findings in the spontaneously dia­ this trait has only been observed in two stud­ betic BB Wistur rut. Lab. lnvcst.,42;162 (ubslrucl). ies and both were with highly inbred struins Wright, J.K., A.J. Yutes, II.M. Sliurmu, C. Shim, K.L. Tigncr, and P. Thil>erl, (1982)Testicular utrophy in the of ruts, this is strong circumstantial evidence s|Kinlune

229 230

J . C o m i'. P a t h . 1983. V o l . 93.

CENTRAL PONTINE MYELINOLYSIS FOLLOWING SALINE TREATMENT OF A DIABETIC RAT FOR DEHYDRATION

liy

J . R . W r i g h t , J r., A .J . Y a te s , H . M . S iia rm a and P. T i i i b e r t The Ohio .State U ni le n ity College o f Medicine, Department o f Pathology, Divisions o f A'euiopdtliology and Anatomical Pathology, Columbus, Ohio 43211), U .S.A. and Animat Resources Division, Health Protection Jlranch, Health and Welfare, Canada, Ottawa, Ontario, Canada K IA O I.2

INTRODUCTION The BB Wistar (BBW) rat, a promising new model for human juvcnilc-onsct diabetes mellitus, was derived from an outbrcd line of Wistar rats at the Bio- Breeding laboratories in Ottawa, Canada (Nakhooda, Like, Chappel, Murray and Marliss, 1977). The diabetic syndrome in BBW rats is characterized by hypcrglycacmia, glycosuria, ketoacidosis, insulinopcnia, glucagonaemia, and hypcrlipaemia (Nakhooda et a l., 1977; Nakhooda, Like, Chappel, Wei and Marliss, 1978). Spontaneous diabetes occurs in 30 to 55 per cent of the ollspring of diabetic matings. The onset is abrupt with rapid weight loss due to dehydra­ tion as a result of polyuria. If insulin therapy is not initiated immediately, death occurs within 1 to 2 weeks. Usually, the rats arc easily maintained on insulin .for G months to 1 year after onset of diabetes. However, at this point maintenance often becomes difficult; the rats lose weight rapidly (up to 25 to 40 per cent over a few weeks), become dehydrated and emaciated, and glyco­ suria disappears (unpublished observation). Ten of these rats were treated with subcutaneous injections of normal saline and their insulin treatment was dis­ continued. Despite therapy the animals died. One such casualty is the basis of this brief report.

MATERIALS AND METHODS

A male BBW insulin-dependent diabetic rat, 333 days old at the time of death, had been progressively losing weight over several weeks, and became severely dehydrated. The animal received a total of 15 ml of normal saline (0‘9 per cent NaCl) as 2 to 3 ml subcutaneous injections per day during the week before its death in an attempt to rehydrate it. The rat died despite therapy (238 days after the onset of diabetes). A l necropsy, the rat was thin, emaciated, and weighed only 255 g (normal is approximately 450 g) at death. A drop of blood placed on a Dextroslix (Ames Co., Elkhart, Indiana) reagent strip was negative for sugar. A complete gross posl-inoricm examination revealed only testicular atrophy, a common finding in older rats of this strain (Wright, Yales, Shanna, Shim, Tigner and Thibert, 1982). The brain appeared grossly normal and was of normal weight (1 -67 g). Tissue specimens from all organ systems were fixed in neutral phosphate bullered form alin, and processed for light microscopy. Sections of each organ were stained with hacmatoxylin and eosin (H E ); 0021-9975/83/010309 -|- 0li S03.00/0 © 1903 Academic l'ress Inc. (London) Limited 231

J. R. VVUICHT f/ «/.

sections o f brain were also stained w ith luxol fast blue and Bodian’s slain for axons. A ll organs were histologically normal except the testes (which were atrophic) and the brain. A lesion in the pons involved most o f the basal portion and extended into the tegmentum and the base of the middle cerebellar peduncles (Fig. 1). The myelin in

Fie . c .i Transverse section of pons and cerebellum showing marked pallor of inyctin in basal region and tegmentum. Note the margin of normal myelin (arrow). Luxol fast blue. X 10.

this area was almost completely destroyed and macrophages were common (Fig. 2). Necrotic cells with nuclei demonstrating pyknosis and karyorrhexis were scattered throughout the lesion but no oligodendrocytes could be identified with certainty. There were a few cells with large nuclei and a moderate amount of granular eosino­ philic cytoplasm which may have represented degenerating neurons; however, neurons were numerous und a few axons traversed even the most damaged urea (Fig. 3). There were no reactive astrocytes or evidence of either acute or chronic inflammation and the vessels were normal. One small focus with similar histological features was seen in the corpus callosum. Nine other rats were treated for dehydration with subcutaneous normal saline, but none o f these received more than 9 m l. A fter a single dose o f saline (2 lo 8 m l) 4 died w ithin 1 day and 1 w ithin 2 days; 3 lived for 3 to G weeks after such treatment. O ne other received 9 ml normal saline over the last 4 days of life. Brains were examined histologically in all of these rats, but none had evidence of central pontine myelinoly- sis.

DISCUSSION Central pontine myelinolysis (CPM), a rare disease of unknown aetiology, was first reported by Adams, Victor and Mancall (1 9 5 9 ). Since then, over 150 F ig. C,2 Central pons with macrophages and several unidentifiable necrotic cells showing karyorrheclic and pyknolic nuclei. Luxol last blue. xC30. F ig. C.3 Central pons with axons traversing the area. Bodian's silver slain. xG30.

human cases have been reported (Wright, Laureno and Victor, 1979). The typical lesion is a single symmetrical focus of grey softening in the centre of the upper and middle pons. Histologically, there is a sharply defined area of myelin breakdown with relative sparing of nerve cells and axons, although some axonal degeneration and nerve cell loss may be present in the very centre j. u. w uioirr (/ a/. of the lesion. Macrophages containing sudauophilic lipids and reactive astro­ cytes may be numerous within the lesion, but oligodendrocytes arc absent. Neither vascular disease nor inflammation is seen in CPM and therefore cannot be involved in the pathogenesis (Oppenheimcr, 1976). The histological findings in this case are similar to those seen in human cases of CPM. In man, CPM is frequently associated with alcoholism and malnutrition. Recently, electrolyte abnormalities have been implicated in the pathogenesis of CPM (Wright el a l., 1979). In fact, many investigators now believe that it is an iatrogenic disorder due to the rapid correction of hyponatraemia (Finlayson, Snider, Olivia and Gault, 1973; Tomlinson, Picridcs and llradlcy, 1976; Leslie, Robertson and Norcnbcrg, 1980; Pogacar, 1980). Some evidence sug­ gests that slower correction of hyponatraemia does not lead to CPM (Leslie et al., 1980). This pattern is consistent with the clinical course (i.e. rapid correc­ tion of dehydration) seen in this BBW rat prior to its death. A recent study by Kleinschmidt-DcMasters and Norenbcrg (1982) has shown that CNS demyelination can be induced by fluctuations in the serum electrolyte concentrations in rats. In that study, rats were made hyponatraemic by the administration of vasopressin and water for 3 days and then treated with hypertonic saline. Rats were killed 5 days later. They reported demyelination with sparing of neurons and axons in regions of the brain with a rich admixture of grey and white matter. Sites of lesions included corpus striatum, claustrum, external capsule, cerebral neocorlex, anterior commissure, hippocampus and its fimbria, thalamus, periaqueductal grey, mid-brain reticular formation, red nucleus, and brainstem. In severely affected rats, neuronal loss and necrosis were present in the centre of the lesions. Although the lesions produced in that study are histologically similar to CPM, they are diffuse rather than localized to the pons as in our case. CPM has been reported in man several months after the onset of clinical diabetes and concurrent with fluid therapy for dehydration (Behar, Bental and Aviram, 1964). We can only speculate as lo why CPM might occur in this BBW diabetic rat and yet has not been produced experimentally or observed inciden­ tally in other animals. First, CPM is more common in chronic nutritionally deficient individuals (Oppenheimcr, 1976), and poorly controlled diabetes mellitus is such a condition. The emaciation and dehydration that frequently precede the death of BBW diabetic rats may set the stage for the development of CPM. Secondly, the diabetic state causes additional osmotic stress (Aloia and Nilakantan, 1973). We postulate that the combination of these conditions followed by rapid treatment with relatively large doses of saline caused CPM in this case. Nine other diabetic rats which did not develop CPM were treated with normal saline injections but none received as much saline as this animal, and their durations of therapy were all shorter. This suggests that both the dosage schedule of saline and the survival lime following initiation of therapy may be critical. It is probable that many factors were involved in this incidental finding in a diabetic rat, but the significance of this report is that CPM has not, so far, been experimentally induced in animals (Oppenheimcr, 1976). This study suggests that this can be done. 234

l'O N TI N E M YKLINOl.YSIS

SUMMARY

Central pontine myelinolysis (CPM ) was observed at necropsy in an ema­ ciated 333-day-old 1111 Wistar insulin-dependent diabetic rat that had been treated for dehydration by subcutaneous injection of normal saline. There was a large area of nearly complete demyelination involving most of the basal portion of the pons, part of the tegmentum, and the base of the middle cere­ bellar peduncles. There was a relative sparing of neurons and axons, but some unidentifiable large cells with degenerative changes in the cytoplasm were present. No oligodendrocytes or reactive astrocytes were seen and there was no evidence of inflammation or infarction. Although diffuse demyelination can be induced in rats by rapid fluctuation of serum electrolytes, no studies have experimentally induced central pontine myelinolysis nor has it been observed incidentally in animals previously.

AOKNO W I.IiUCM E N T S

This study was supported in pari by the Department of Pathology, the College of Medicine, and the Graduate School of the Ohio Stale University, by the Upjohn Company, and by N IH Grant NS-18026. J. R. Wright was supported by the Samuel J. Rocssler Research Scholarship.

It E E It E N C E S Adams, R., Victor, M ., and Mancoll, 1s. L. (1959). Central pontine myelinolysis. A hitherto undescribed disease occurring in alcoholic and malnourished patients. Archives o f Neurology (Chicago), 81, 154-172. Aloia, J. E., and Nilakantan, S. (1973). The cerebrospinal fluid in uncontrolled diabetes mellitus. American Journal oj' the Medical Sciences, 266, 203-209. Behar, A., Bcntla, E., and Aviram, A. (1964). Central pontine myelinolysis (A case report). Acta Neurvpathologica, 3, 343-350. l'inlayson, M ., Snider, S., Oliva, L. A. and Cault, M . (1973). Cerebral and pontine myelinolysis. Two cases with fluid and electrolyte imbalance and hypotension. Journal o f the Neurological Sciences, 18, 399-409. Klcinsclunidl-DcMasicrs, B. K., and Norenbcrg, M . 1). (1982). Neuropalhologic observations in electrolyte-induced myelinolysis in the rat. Journal o f Neuro­ pathology and Experimental Neurology, 41, 67-80. Leslie, K. O., Robertson, A. S., and Norenbcrg, M . D. (1980). Central pontine myelinolysis: an osmotic gradient pathogenesis. Journal of Neuropathology and Experimental Neurology, 39, 370. Nakhooda, A. P., Like, A. A., Chappel, C. I., Murray, E. T., and Marliss, E. B. (1977). The spontaneously diabetic Wistar rat. Metabolic and Morphologic studies. Diabetes, 26, 100-112. Nakhooda, A. 1\, Like, A. A., Chappel, C. I., Wei, C. N., and Marliss, E. B. (1978). The spontaneously diabetic Wistar rat (the “BB” rat). Studies prior to and during development of the overt syndrome. Diubelologia, 14, 199-200. Oppenheimcr, D. R. (1976). Demyclinating diseases, in Greenfield's Neuropathology. W . Blackwood and J. Corsellis, Eds, Edward Arnold, London, pp. 495-496. Pogacar, S. (1980). Iatrogenic precipitation of central pontine myelinolysis with formation of Rosenthal libers. Journal o f Neuropathology und Experimental Neurology, 39, 383. Tomlinson, B., Picrides, A., and Bradley, W . (1976). Central pontine myelinolysis. 235

j. k. \vright et al.

Two cases witli associated electrolyte disturbance. Quarterly Journal of Medicine (New Series), 40, 373-3116. Wright, D. (J., Laurcno, R., and Victor, M. (1979). Pontine and extrapontine myelinolysis. Jlraia, 102, 3G1-3115. Wright, J. R., Yates, A. J., Sharina, 11., Shim,C.., Tigner, R. L., anil Thibert, 1*. (19112). Testicular atrophy in spontuneotisly diabetic 1111 W istar rats. American Journal of Pathology, 108, 72-7‘J.

[Received for publication, July l(j///, 191)2 J Appendix H 237

Hematological Characteristics of the BB Wistar Rat

Jam es R. Wright, Jr.; Allan J. Yates; Pierre Thibert Nlranjan T. Shah; John C. Neff; David W. Covey Chief, Animal Resources Division Ohio State University Health Protection Branch College of Medicine Health and Welfare — Canada Department of Pathology Ottawa, Ontario Columbus, Ohio 43210 Canada K1A0L2 Supported by Samuol J. Hoessler Research Scholarship.

Summary

Complete blood counts, differential white changed differential white cell counts. Differ­ blood cell and platelet counts were per­ ential counts showed a pattern of lympho­ formed on male and female BB Wistar diabe­penia, neutrophilia, monocytosis and eoslno- tic rats (BBWd), their nondiabetic siblings philia. It is possible that these white blood (BBWnd) and outbred Wistar rafs of the line cell changes contribute fo the increased sus­ from which the BB Wistar rats were derived. ceptibility to infection reported tor the BBW Most of the observed changes were strain- strain. No significant difference in serum related (those present In both BBWd and immunoglobulin concentrations was found BBWnd but not in control rats) rather than in any of these three groups of rats. There­ diabetes-related (those in BBWd but neither fore, hypogammaglobulinemia cannot ac­ BBWnd nor control rats) and therefore prob­ count for the increased susceptibility to In­ ably due to the inbreeding process. The BBW fections, but it is not possible to rule out an strain had significantly lower numbers of abnormality in the distribution of immuno­ white cells and platelets, as well as markedlyglobulin fractions as an etiological factor.

KEY WORDS: BB Wistar rat, spontaneous diabetes, complete blood counts, leukopenia, eosinophilia, platelets, immunoglobulin

Introduction

Supported in pea by the Depart meat of Pathology, the College of Medicine and th e G rad u ate S chool of th e Ohio S ta te University an d by Ih e Upjohn C om pany. The BB Wistar (BBW) rat, a strain that was de­ Presented In part al the meeting of the Canadian Association ot Medical Bio­ ch e m ists in O ttaw a, O ntario, C an ad a (6-SttO). veloped from an outbred line of Wistar rats at the

VETEIUNAVY CLINICAL PATHOLOGY • Vol. XII, No. I • 238

Hematological Characteristics ol the BB Wistar Bat

Bio Breeding Laboratories in Ottawa, Canada, de­ were rarely present. All rats were housed together velops a syndrome that closely mimics human and were therefore equally exposed to any infec­ insulin-dependent (juvenile-onset) diabetes. The tive organisms that may have been present. Our BBW syndrome is characterized by hyperglyce­ barrier facility and animal husbandry techniques mia, glycosuria, ketoacidosis, insulinopenia, glu- have been previously described in detail.6 cagonemia and hyperlipemia.12 Onset of diabetes Rats used in hematological studies were anes­ occurs spontaneously between 60 and 160 days of thetized with ether, and blood (300 //I) was col­ age in 30 to 55% of the offspring of diabetic par­ lected in capillary whole blood collectors (Micro- ents. Onset is abrupt with rapid weight loss due to tainers R:Becton-Dickinson, Rutherford, New Jer­ dehydration as a result of polyuria. Death occurs sey) by inserting a heparinized capillary tube within several weeks if insulin therapy is not ini­ (American Hospital Supply Corp., Miami, Florida) tiated. into the periorbital plexus of vessels at the inner The present study was designed to determine canthus of the eye. This technique was chosen several hematological parameters of the BBW rat because it offers more consistent results than and was prompted by several observations made tail vein bleeding.6 Complete Blood Counts (CBC), during a previous study.3 First, the BBW rat, both differential white cell counts and platelet counts diabetic (d) and nondiabetic (nd), is very suscepti­ were performed on BBWd (16 male and 11 female), ble to a variety of infections. Thus, they may have BBWnd (12 male and 10 female) and control (9 an abnormality of the immune system which male and 10 female) rats. (Control rats were a Wis­ could be reflected in the numbers of circulating tar line obtained from Bio Breeding Laboratories leukocytes or concentrations of serum immuno­ in Ottawa, Canada.) CBC's and platelet counts globulins. Second, the frequent presence of were accomplished with a Coulter S-plus. Platelet eosinophilic infiltrates and/or mast cells in tissue counts in excess of 700,000 were diluted 1:2 until sections of both BBWd and BBWnd rats may be there was agreement in two of three simultane­ associated with abnormal numbers of these cells ous measurements. in blood. Third, the high incidence of lymphomas Blood for serum studies was also collected by in this model3'4 suggested that leukemia might retro-orbital bleeding. Samples were allowod to also occur. Finally, histological changes sugges­ clot and then centrifuged in a tabletop centrifuge tive of autoimmunity in some rats prompted us to (Model HN:lnternational Equipment Co.) with a examine the protein electrophoretic pattern of the fixed-angle rotor at 1500 rpm. The supernatant serum for a change in the concentrations of was pipetted into polypropylene micro sample gammaglobulins. tubes (Kew Scientific, Columbus, Ohio) and fro­ In addition to these specific questions, another zen at — 40°C until use. Protein electrophoresis goal of this study was to determine whether any was performed on agarose plates at pH 8.6. Total hematological parameters of the BBW strain protein was measured by the biuret method. Sam­ differ from those of the original Wistar stock from ples from five rats in each of the six categories which they were derived, and whether these were compared by two-way analysis of variance. changes (if present) are a result of the inbreeding process or of the diabetic condition. Results Table 1 shows the average body weight, CBC Materials and Methods and platelet values for both male and female rats A breeding colony of BBW rats was established in the three groups studied (BBWd, BBWnd, con­ by the Pathology Department of the Ohio State trol). Average rat weights for the three groups did University College of Medicine in June of 1979. not differ significantly within the same sex but Rats were transported to Columbus by commer­ were significantly different across sexes. Al­ cial airlines from the parent colony at Health-and though exact birth dates were not known for some Welfare of Canada in Ottawa. Rats were originally of the rats, all were adults of similar ages. Inter­ housed in a conventional animal facility but were group variation for each variable was examined soon moved to a semi-barrier housing facility fol­ by two-way analysis of variance (ANOVA) using lowing an outbreak of bronchopneumonia. At the the unweighed means solution for unequal cell time of this study, the pulmonary infections were sizes, and paired comparisons were performed under control, but lowgrade pinworm infections by the Waller-Duncan Bayes exact test.' The re­

• Vol. XII, No. I • VETERINARY CLINICAL PATHOLOGY 239

Hematological Characteristics of th e BS W istar R at

T a b ic H . l Complete Blood Count and Platelet Values*

Body Weight WBC RBC Hgb HCT MCV MCH MCHC PLT Strain Sex n (gm) (x103) (x10‘) (in g/dl) <%) (II) (/lA fl) (Qfdl) RDW (x10>)

BBWd M 16 380.08 5.86 ‘ 8.22* 16.30 41.96 52.29* 19.91 38.11 10.72*' 767.69* 59.06 2.31 0.61 0.80 4.25 1.81 1.15 1.90 1.65 149.69

F 11 276.00 8.15 7.64* 16.24* 42.46 55.54 21.28 38.30 9 .3 3 " 832.23* 27.51 3.72 0.32 0.65 1.57 1.96 0.74 1.43 1.09 209.92

BBWnd M 12 443.54 7.80‘ 8.09 16.00 42.15* 52.07* 19.86 38.17 9.25** 902.21 * 63.79 2.40 0.42 0.65 3.09 1.50 0.45 1.50 0.70 174.29

F 10 292.00 5.66* 7.75 16.10 42.69* 55.11 20.82 37.78 7.30* 748.30* 41.51 3.99 0.34 0.37 1.47 1.71 0.77 0.87 0.99 112.63

Control M 9 415.33 12.57 7.76 16.68 43.54 56.27 21.59 38.39 7.13 1224.44 17.49 2.33 0.69 0.82 2.27 3.17 1.16 0.89 1.31 80.22

F 10 277.00 9.18 7.32 15.38 39.47 53.96 21.04 38.98 5.70 1189.60 17.09 3.07 0.25 0.60 1.47 2.39 1.06 1.21 0.72 216.97

Abbreviations: n^num beioluts; WBC^ white blood cell couni/^i; RBC* red blood cell count/ rod ceil distribution width; PIT ** platelei&/;rl. a The seco n d value in ea c h pair re p rese n ts th e stan d ard deviation. • Value tor BBW rat is significantly different (p 0 0 1 ) from th e co n tro l rat value. * Value tor BBWd rat Is significantly different (p •* 0.01) Irom Ihe BBWnd ral value. suits of these tests have been summarized in Ta­ counts. Differential counts revealed a pattern of ble 1. All of the differential white cell count vari­ marked lymphopenia, slight neutrophilia, mono­ ables were compared as above except for the cytosis and eosinophilia. Some statistically sig­ monocyte and eosinophil counts, which were nificant changes in red cell indices were also examined using the Z test (Table 2). This test was seen, but none were outside of the normal range chosen because the intra-group number of mono­ of other rat strains.' Red Cell Distribution Width cytes or eosinophils per rat had a bimodal distri­ (RDW) values were significantly higher In the BBW bution rather than a normal distribution. There­ strain (particularly BBWd), indicating a tendency fore, rat blood smears were considered to show toward anisocytosis. monocytosis or eosinophilia when they exceeded Gamma globulin concentrations were also com­ the control rat sex-matched mean monocyte or pared. The mean concentrations (±S.E.M.) in eosinophil counts by four standard errors of the mg/dl for male rats in the three groups (BBWd, mean. The ratio of the number of rats with eosino­ BBWnd and control) were 0.624 ±0.055, 0.944± philia or monocytosis within each group (BBWd 0.069, and 0.687 ±0.037 respectively, while in fe­ and BBWnd) was then compared to the same ratio male rats these values were 0.938 ±0.088,0.948 ± for the sex-matched control group by the Z test 0.186, and 0.866 ±0.115 respectively. There were (Table 2). no statistically significant group or sex differ­ A general observation is that there are many ences. more strain-related differences (those present in both BBWd and BBWnd but not in control) than diabetes-related (those in BBWd but in neither Discussion BBWnd nor control) differences (Tables 1 & 2). More specifically, the BBW strain had significant­ It is difficult to establish general baseline he­ ly lower numbers of white cells and platelets, as matological data for experimental rats because well as markedly changed differential white cell of the large number of strains, stocks and sub­

VETEK1NAKY CLINICAL PATHOLOGY • Vol. XU. No. I • 240

Hemato/ogicaf Characteristics of (ho 66 Wistar flat

Tabic H .2 Differential White Cell Counts**6

Strain Sex n Total WBC PMN Lymph Monoc Eosd Baso

BBWd M 16 5,860* * 2,280* 2,760* * 108 526 7.1 2,320 1,320 1,020 129 934 20.9 5/16 6/16

F 11 8,150 3,130* 3,000* * 177 1,830* 4.3 3,720 1,320 1,030 147 2,140 14.2 6/11 8/11

BBWnd M 12 7,800** 3,000* 3,260“ 481* 1,050” 3.2 2,400 1,260 935 507 942 11.1 7/12 10/12

F 10 5,660* * 2,110* 2,610“ 151* 778 0 3,990 1,750 934 113 1,470 — 6/10 3/10

Control M 9 12,600 1,620 10,800 29.0 126 0 2,330 885 2,220 86.7 111 - 1/9 1/9

F 10 9,180 1,560 7,410 58.9 149 0 3,100 630 2,580 58.6 143 — 1/10 1/10

Abbreviations:;)* number olrt'ts; PMN* neutrophils/pl; Lymph «lymphocyle&s/rl; Mono*monocytes/jil; £oa = eosinophlls/pl; Baso* basophils/pt. * T h e se c o n d value In ea c h pair re p rese n ts th e stan d ard deviation.

^Statistical analyses were performed to test lor differences between BBW end sex-matched control rets. Single estetiek (•) Indicates Ihe p *'0.06. Double asterisk (**) lndicatesp<0.0t.

c Ret blood smears were considered to show monocytosis when they exceeded Ihe sex-mutched control mean rnonocyle count by 4 standard eriors of Ihe mean (i.e., c o u n ts -*144.6 for m a le s an d co u n ts -» >33.0 tor fem ales). R atio fa Ihe num ber of ra ls with m o n o cy to sis by th is criterion. S ta tistica l an aly sis by 2-leal.

d Rat blood smears were considered to show eosinophilia when they exceeded the sex-matched control rat mean eosinophil count by 4 standard errors of the mean (i.e., counts > 273.7 tor m ales and counts >t330.2 lor Iumatex). Ratio Is the number ot rats with eosinophilia by this criterion. Statistical analysis was by 2-lest.

stocks recognized within the species* For this values are very far out of ranges considered nor­ reason, we chose non-diabetic Wistar rats from mal for rat hematological studies.* The degree of Bio Breeding Laboratory (from which the BBWd anisocytosis is greatest in the BBWd, lower in was derived) as our control animal for strain com­ BBWnd, and least in control rats. This is what is to parison. The comparison of BBWd with BBWnd al­ be expected since the degree of anisocytosis is lows us to determine diabetes-related hematolo­ increased in several disease states.* gical changes while the comparison of BBW(d Most of the white blood cell indices are changed and nd) with control rats allows us to determine in a strain-related pattern, but we saw no evidence strain-related changes due to the inbreeding pro­ of leukemia. White coll indices are of great Inter­ cess. Tables 1 and 2 show that strain-related est for two reasons: (1) the BBW strain (both dia­ changes are more frequent than diabetes-related betic and non-diabetic) has an increased suscep­ changes. This Is consistent with present know­ tibility to infection;1'10 and (2) autoimmunity has ledge since inbreeding of rats is known to change been implicated in the onset of the diabetic syn- hematological parameters* and since diabetes drome.11'12 Both BBWd and BBWnd have signifi­ does not significantly alter major red and white cantly lower white cell counts than control rats cell indices in man.* (Table 1). This could partially explain the higher in­ The changes in red cell indices indicate either cidence of infections and possibly the apparent a sex-related or random pattern and none of these shortened life expectancy of the BBW rat (unpub­

• Vol. XU. No. 1 • VETEK1NAKY CLINICAL PATHOLOGY 241

Hematological Characteristics of the BB Wistar Rat

lished observation). Reich and Dunning have re­ (Table 1). This was significant at the p < 0.01 level ported a positive correlation between the mean in both BBWd and BBWnd rats. Although this is a life-span of inbred rat strains and their mean WBC significant change, none of the values are mar­ count.'3 kedly different from values reported in hematolo­ The differential white blood cell pattern of BBW gical studies of other rat strains.' shows marked alterations from the control rat The mean gamma globulin concentrations in pattern (Table 2). The typical pattern in the BBW our study are well within the normal ranges re­ rat shows increased numbers of PMNs, mono­ ported in other studies.'-'5 Since we observed no cytes and eosinophils as well as a large decrease group differences in the concentrations of gam­ in the number of lymphocytes. Basophils are rare ma globulins, hypogammaglobulinemia cannot in rats' as they were in all rats in this study. account for the increased susceptibility to infec­ Most of these differential patterns were rea­ tion in the BBW strain. However, it is not possible sonably consistent (i.e., approximated a normal to rule out some abnormality in the distribution of distribution), but eosinophilia was either present immunoglobulin fractions as an etiological fac­ or absent (i.e., bimodal distribution). This, at first, tor. ■ suggested that parasitic infections were respon­ sible. Several lines of reasoning indicate that this

is not the case. Since all three groups of rats were ACKNOWLEDGEMENTS housed together, one would expect a parasitic The author* would like to thank Raechel Tlgner tor her statistical expertise end infection to affect all three groups equally rather Ors. O.A. Senhauset and S. Patel tor their advice and assistance. than the strain-related pattern seen in this study. Secondly, mild pinworm infection (Syphacia obve- iate) was the only parasitism observed in our col­ REFERENCES

ony, even following complete necropsy. Pinworm 1. Nakhooda AF, Like AA, Chappel Cl, Murray FT, Marliss EB: The Spontaneously infections are ubiquitous in most rat colonies'4 Diabolic Wistat Rat. Metabolic and Morphologic Studies. Diabetes 26:100-112,1077. 2. Nakhooda AF, Like AA, Chappel Cl, Wei CN, Marliss EB: The Spontaneously but eosinophilia has not been reported In associa­ Oiabetlc w ntar Rat (The "BB" Rat). Studies Prior to and During Development ol the Ovett Syndiome. OmbetotOQta 14 199-200,1978. tion with such infections. Furthermore, our colony 3. Wright JR, Shaima H, Thlberl P, Vates AJ: Pathologic Findings in the Spon­ (BBWd, BBWnd and control rats) was occasional­ taneously Diabetic BU Wistar Rat. Lab Invest 42:162,1980. 4. Kalanl N, Seemayer T: Malignant Lymphoma in Spontaneously Diabetic Rats. ly treated for pinworms with Piperazine (W.A. But­ N Eng J Med 300:739,1079. 6. Wiight JR, Yates AJ, Sharma HM, Thibert P: Spontaneous Gastric Erosions and ler Co.) in the drinking water. Thirdly, eosinophilic Ulcerations m the BOW Wistar Rat. Lab Anlm Scl 31:63-66,1061. 6. Hulse EV: Laboratory Animals In Experimental Hematology. Fed Cosmel Toxi­ infiltrates are occasionally present in tissue sec­ co l 3 736 748.1966. tions from many organ systems in BBW rats,3 7. Waller RA, Duncan OB: A Bayes Rule tor the Symmetric Multiple Comparisons Problem. AmStat Assoc J 64:1463-1603,1969. rather than just the site of pinworm infection (i.e., 6. Rlngier OH, Dabich L: Hematology and Clinical Biochemistry. The Laboratory Rat, Vol 1, Biology and Diseases, ed by Baker HJ, Undsey JR. Weisbroth SH. New the rectum and colon). Others have also reported York, A cadem ic P ress, 1970, p p 106-121. occasional eosinophilic infiltrates In the pancrea­ 9. Jones RL, Peterson CM: Hematologic Alterations In Diabetes MetUtue. Am J M od 70:339-362,1981. tic islets of BBW rats.'-" Finally, the eosinophilic 10. 8ima AAF: Peripheral Neuropathy In the Spontaneously Dlabetio BB Wistar Rat. An Ullresiructural Study. Acta Neuropelhol (Berl) 61:223-227,1060. infiltrates are frequently seen in rats with large I t . Like AA, R ossini AA, G uberski DL, A ppel MC, W illiam s RM: S p o n ta n eo u s Diabetes Mellilus: Reversal and Prevention in the BB/W Rat with Antiserum to Rat numbers of mast cells in tissue sections. This pat­ Lymphocytes. Science 208:1421-1423,1079. tern is more suggestive of strain-related hyper­ 12. Like AA, W illiam s HM, K islauskls E, R ossini AA: N eo n alal T hym ectom y P re­ vents Spontaneous Diabetes In the UiOBrueding/Worcesler (BB/W) Rat. Clin Hes 29: sensitivity to some unknown allergen or auto- 642A, 1961. 13. Reich C. Dunning WF; Leukocytu Level and Longevity In Rats. Science 91:430- antlgen. 431,1941. The other major strain-specific observation in 14. Tails LF: Pinworm Infections in Laboratory Rodents: A Review. Lab Animals 10 1-13.1076. this study was that BBW rats have decreased 16. Coleman GL, Barthold SW, Osbaldision GW, Foster SJ, Jonas AM: Pathologi­ cal Changes During Aging in Barrier-reared Fischer 344 Male Rats. J Gerontol 32: platelet concentrations relative to Wistar rats 266 276.1077.

VjrrEUlNAKY CLINICAL PATHOLOGY • Vol. XU, No. 1 • Appendix I Mechanisms of Ageing and Development, 36 (1986) 109-116 Elsevier Scientific Publishers Ireland Ltd.

DNA EXCISION REPAIR IN SPONTANEOUSLY DIABETIC BB WISTAR RATS

JAMES R. WRIGHT, Jr.*-*, RALPH E. STEPHENS*, KENNETH C. FORD*, PIERRE T H IB E R T b and A LLAN J. YATES*

^Department o f Pathology, T h e Ohio State University College o f Medicine, Columbus, Oil 4 3 2 1 0 (U.S.A.) an d A n i m a l Resources Division, Sir Frederick G, Banting Research Center, Health Protection Branch, Health an d Welfare Canada, Ottawa, Ontario KJ A O L 2 (Canada) (Received January 14th, 1986) (Revision received May 19th, 1986)

SUMMARY

BB Wistar rats develop a syndrome characterized by spontaneous diabetes mellitus as well as a wide variety o f autoimmune, neoplastic, and degenerative disorders which do not occur in the outbred Wistar strain from wltich they were derived. This syndrome also includes elements o f premature ageing (i.e., a markedly shortened lifespan and premature occurence o f diseases associated with ageing). Excision DNA repair capacity which has been reported lo be directly proportional to maximal achievable life span was estimated in neonatal BB Wistar and outbred Wistar rats. Excision repair was assayed autoradiographically by determining unscheduled DNA synthesis following U V radi­ ation o f passage 3 cultured skin fibroblasts. No difference in excision repair capacity could be demonstrated between the two strains.

Key words: Ageing; BB rat; Diabetes mellitus; D NA repair; Fibroblast

INTRODUCTION

The BB Wistar (BB) rat develops a syndrome that closely mimics human juvenile- onset diabetes [1 ]. In addition to the spontaneous onset of hyperglycemia and keto- nuria, young BB rats also develop a wide variety o f developmental, autoimmune, neo­ plastic, and degenerative disorders which do not occur in the outbred Wistar (W ) rat

Address all reprint requests to: Allan J. Yates, M.D., Ph.D., Department of Pathology (Neuropath­ ology), The Ohio State University, 473 West 12th Avenue, 111 Uphum Hall, Columbus, Oil 43210, U.S.A. *Present address: Department of Pathology, Washington University School of Medicine, 660 S. Euclid Avenue St. Louis, MO 63110, U.S.A.

0047-6374/86/J03.50 © 1986 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland 244

strain from which the BB strain was originally derived. Many of these occur in a strain- related pattern (i.e., both diabetic and non-diabetic BB Wistar rats) but are exacerbated by the expression o f the diabetic trait. This suggests that the BB strain is not simply a Wistar strain that develops diabetes mellitus. Rather, diabetes appears to be only a part of a larger “ BB Wistar syndrome” [2 ], a syndrome which in many ways also demon­ strates elements of premature ageing (i.e., a markedly shortened lifespan and premature occurrence of diseases o f ageing). Although the onset o f hyperglycemia is heralded by an autoimmune insulitis [3,4], the underlying cause o f the “ BB Wistar syndrome" is not known. Various human syndromes characterized by premature ageing and symptoms of diabetes mellitus have been attributed to faulty D NA repair [5 -1 0 ]. In this study, we tested whether an excision DNA repair defect could account for the “ BB Wistar syndrome” , but the results also offer insight into a mechanism of ageing. We chose to investigate U V induced excision (long-patch) repair which requires endonuclease, exonuclease, polymerase, and ligase activities, because the results reflect on multiple DNA repair processes rather than just one type o f DNA repair. Differing maximal achiev­ able lifespans in various species have been related to differing excision D NA repair capac­ ities [11]. Because the Wistar rat strain and the BB Wistar rat strain are presumably identical except for the expression o f the “ BB Wistar syndrome” in the latter, it is in many ways an excellent model to examine this relationship.

MATERIALS AND METHODS

BB Wistar (BB) rats and outbred Wistar (W ) rats from which the BB Wistar strain was originally derived were obtained from Health and Welfare o f Canada. Breeding colonies o f both strains were established at The Ohio State University. Our animal facilites and maintenance procedures have been described extensively elsewhere [12]. Primary fibroblast cell lines were established from each o f two neonate offspring of single BB or W rat mating pairs. These cells lines were called BBi, BBj, W it and W3, respectively. Because BB rats do not develop diabetes until 3 0 -6 0 days o f age, the BB pups were not yet diabetic. Neonatal rats were utilized in this study to maximize repair levels since evidence exists that DNA repair efficiency in rats decreases with ageing [13]. The neonates were sacrificed and their abdomens cleaned with 70% ethanol prior to excision of 1 cm3 skin samples. Samples were washed three times in phosphate buffered saline (PBS), placed in 0.5 ml o f 0.01% of trypsin, and then minced between two razor blades on a ground glass cutting block for 10 min. Minced dermis was placed in 25-ml digestion flasks containing 5 ml of 0.01% trypsin and then incubated for 30 min in a 37°C water bath/shaker. The contents o f the flasks were poured into 50-ml plastic centri­ fuge tubes and the trypsin was inactivated by the addition of 40 ml o f B medium (Eagles MEM supplemented with 1.5X essential amino acids, 2X non-essential amino acids and 1.5X vitamins), containing 10% fetal bovine serum and 0.02% gentamycin (B 10X 2). 245

Tubes were centrifuged at 1100 rev./min for 15 min and the medium decanted. Pellets were resuspended in 8 ml o f B(0 2X medium, divided into four T jS flasks (Corning Glass Works, Corning, N .Y .), and incubated at 37°C with 5% carbon dioxide. After 12 days, the cells had reached near confluency and were harvested with 0.01% trypsin, split 1:2, and plated (i.e., passage 1). The cells were split 1:2 for each successive passage. Passage 3 cells were used for the experiment because they appeared relatively free o f epithelial cell contamination and because passage 3 rodent cell lines have not yet undergone a senescent decline in DNA repair capacity 114]. Passage 3 cells for each cell line were suspended at a concentration of 9 X 104 cell/ml into six 100-mm diameter petri dishes containing ten 22 mm X 11 mm, coverslips with 0.5 m l cell suspension/cover slip. Cells were permitted to attach and grow for 2 days. The medium was then aspirated and replaced with U medium containing 2 X 10~3 M hydroxyurea. After 15 h, the medium was aspirated and the cells washed twice with PUS. For each of the cell lines, the petri dishes were divided into 3 groups: (a) 2 control plates receiving no radiation, (b) 2 plates exposed to 20 J/m 2 ultraviolet (U V ) radiation, and (c) 2 plates exposed to 40 J/m 2 U V radiation. Irradiated and unirradiated plates were then exposed to medium containing 2 X 1(T3 M hydroxyurea and 2 jtCi/ml tritiated thymidine QHdThd, spec. act. = 20 Ci/mmol, New England Nuclear) for 0, 2, 4 ,6 , and 8 h. After their respective incubations, coverslips were removed from the dishes, washed three times in PBS, fixed for 5 min in freshly prepared Carnoy’s solution three times, dehydrated through an ethanol series, mounted on labeled glass slides cell-surface up, and placed overnight in a drying oven. The glass slides were then dipped in a photo­ graphic emulsion (0.67% glycerol + 32.67% distilled water + 66.67% Kodak NTB-2) and stored in opaque slide boxes containing calcium carbonate at 4°C. After 6 days, the emulsion coated coverslips were developed with Kodak D 19, washed, fixed, and lightly stained with toludine blue. Cells were magnified 630X and nuclear grains were counted electronically with a Docuval camera and an Artex 880 counter. The frequency o f nuclear grains was determined for approximately 50 non-S phase cells from each coverslip. Corrections for background on each coverslip were made by subtracting the mean number o f grains present in adjacent equivalent areas without cells. All analyses were performed on these adjusted grain counts. Three coverslips per cell line were exam­ ined for most group and dose interactions. However, only two coverslips were available for several group-dose interactions. The magnitude o f scheduled DNA synthesis for each group and dose was equivalent to the average number o f grains per nucleus for irradiated cells minus the values for unirradiated cells.

RESULTS

Grains were counted in 11 452 nuclei. In 1767 of these, the grains were too numerous to count reliably, so those nuclei were not included in the analysis. The sample size was thus 9685. The 1767 nuclei dropped from the analysis were predominantly from the higher dosage/longer incubation period conditions. Therefore, the mean grain counts 246

T a b le I . l

ANALYSIS OP VARIANCE PARAMETERS

Source of variation SS df MSF

Animal 12441.3 3 4147.1 22.2 Time interval 2200198.2 4 550049.6 2950.0 Dosage 2270096.4 2 1135048.2 6087.4 Animal X time 21399.9 12 1783.3 9.6 Animal X dosage 8288.7 6 1381.5 7.4 Time x dosage 1003768.1 8 125471.0 672.9 Animal X time X dosage 53607.8 24 2233.7 12.0 Error 1794676.9 9625 186.5

Three-way Analysis ol Variance with grain count as the dependent variable and animal, dosage of UV radiation, and time interval as the independent variables; SS = sum of squares; df = degrees of free­ dom; MS — mean square; I' - E-score. Effect o f heterogeneity of variances was judged to be minimal because the cell sizes (i.e. n for each treatment) were large and uncorrelated with variances (r = 0.09).

in those conditions are underestimated. A three-way A N O VA was performed on grain counts (Table 1). Figure 1 shows that the magnitude o f unscheduled D N A synthesis for both BB Wistar and Wistar rats increased from 0 to 20 J/m 3, but did not increase at the higher dose. Post-hoc comparisons using the Tukey method confirmed that tri- tiated thymidine uptake at 40 J/m 2 was not significantly greater than at 20 J/m 2 (P >

6 5 -

5 5 -

_ j N £ 4 5 -

* 2 M. 5 tr h-U l U < Ul

° BB Wistar • Wistar

O 2 4 6 B TIME (hrs) Fig. I .l Unscheduled DNA synthesis (average grains/nucleus) in BB and W rats as a function of fluence at 0, 2,4 ,6 , and 8 h of incubation. Repair at 40 J/m 1 does not exceed that at 20 J/m 3. o BB Wistor i l r5 • • Wislar65

8 hr 55 • s 6 hr 2 U)

0 hr 0 20 40 FLUENCE ( J / m 2 ) Fig. 1,2 Unscheduled DNA synthesis (average grains/nucleus) in BB and W rats as a function of time at a dosage of 20 J/m1. Fooled standard errors of the means at 0, 2, 4, 6, and U h are 0.28, 0.67, 0.65, 0.63 and 0.72 grains/'nucleus, respectively.

0.05, one-tailed). Therefore, the responses of the BB Wistar and Wistar rats for the different incubation periods were graphed using the data for 20 J/m 2. Figure 2 shows the average amount o f unscheduled DNA synthesis as a function of time in BB Wistar and Wistar rats. The average number o f grains increased with incubation time in an approximately linear fashion with saturation beginning to occur at longer time periods. A t every time interval, unscheduled D NA synthesis was slightly greater in the BB Wistar rats than in the Wistar rats. To determine whether the two control animals had significantly less uptake o f tritiated thymidine (i.e., unscheduled D N A synthesis) than the two BBW diabetic animal, post-hoc comparisons were performed. The Tukcy method was used to test the hypothesis that the grain counts for the two control rats were significantly less than the grain counts for the two diabetic rats at each o f the 15 dosage X incubation time levels. None o f these tests showed significant differences at the 0.05 level (one-tailed).

DISCUSSION

Our experience suggests that a shortened lifespan is part of the “ BB Wistar syndrome” . The average lifespan o f the outbred Wistar rat is about 26 months, and the maximum achievable lifespan approximately 46 months [15], Although similar vital statistics are not available for the BB strain, these animals rarely survive longer than 14 months 248

regardless of the expression o f the diabetic trait. Interestingly, many of the hemato­ poietic and degenerative disorders characteristic o f the BB Wistar strain also occur in out- bred rats, but not unless the rats are “aged” ( 2 - 3 years old). The presence of a wide variety o f spontaneous senile lesions occuring in “young" BB Wistar rats is consistent with premature ageing. For instance, BB Wistar rats commonly develop lymphomas [2,16]. In our autopsy series, the mean age of BB rats with lymphomas was 280 days with the youngest being only 6.S months old. Primary lymph node neoplasms occur sporadically in 2-3-year-old rats [1 7 - 1 9 ], but have never been reported in a rat younger than 17 months o f age [17]. The youngest rat with a lymphoma was a 20-month-old Sprague-Dawley rat [20]. The presence o f a markedly shortened lifespan and the pre­ mature occurrence o f diseases o f ageing are strong evidence o f premature ageing in BB Wistar rats. It has been speculated that human diabetes may represent a form o f premature ageing [2 1 -2 3 ] and many human syndromes o f premature ageing manifest symptoms o f dia­ betes mellitus [6 ], It is well known that human diabetics also have decreased lifespans and experience some o f the diseases o f ageing prematurely. Although some recent clinical studies suggest that precise control o f blood glucose will delay or prevent diabetic sequel- lae, there is a large body o f evidence suggesting that there is a major genetic component. Numerous studies have shown that fibroblasts from human diabetics or prediabetics (individuals with strong familial tendency toward diabetes but have not themselves expressed hyperglycemia) are markedly different from fibroblasts from normal subjects. Non-diabetic fibroblasts grown in cultures exhibit greater plating efficiencies [2 4 -2 6 ], numbers o f passages or cell population doublings before senescence [2 4 ,2 6 -2 9 ], DNA synthetic abilities [25,26], and log-phase doubling rates [26,30]. Because rat fibroblasts transform rather than senesce [31] these types o f studies would be a less meaningful way to evaluate ageing in diabetic rats than in diabetic humans. However, excision DNA repair studies with neonatal rat fibroblasts permit evaluation o f maximum achievable lifespan and ageing in BB Wistar rats. Several studies have demonstrated that a mammalian species capacity for excision DNA repair bears a direct linear relationship to its maximal achievable lifespan [11,32]. This has also been demonstrated in both inbred and outbred rodents o f differning long­ evities. Hart et al. [3 3 ], compared excision repair of UV damage in early-passage cultured fibroblasts from two species o f outbred mice with differing maximum achievable life­ spans, Mus musculus (1250 days) and Peromyscus leucopus (3000 days). Both maximum achievable lifespan and capacity for excision D NA repair were greater in Peromyscus by a factor o f 2.5 [3 3 ]. Similar results were obtained using fibroblasts from fetal inbred NZB, C3Hf, and CBA/H mice [34] and PHA stimulated splenic lymphocytes from adult inbred NZB and CBA mice [35]. Our findings demonstrate that no abnormality o f excision D N A repair exists in BB Wistar rats. This suggests that the differences in longevities are not due to abnormalities o f this multiple enzyme requiring system. Obviously, this does not rule out the possibility o f this disorder being due to abnormalities in DNA through other mechanisms, or that 249

there could be isolated repair defects involving certain essential genes. A t this point the specific molecular genetic basis for the BB Wistar syndrome remains obscure.

ACKNOWLEDGEMENTS

Supported in part by the Department o f Pathology, The College of Medicine, and the Graduate School of the Ohio State University, the Upjohn Company, and N il! Grant NS-18026. James R. Wright, Jr. was supported by the Samuel J. Roessler Research Scholarship fund. The authors thank Raechel Wright for her statistical expertise and Eleanor Borem for typing this manuscript.

REEERENCES

1 The Juvenile Diabetes Foundation Workshop on the spontaneously diabetic Dll rat: Its potential for insight into human juvenile diabetes. Metabolism, 32, Supplement I (19113) 1-1 6 6 . 2 J.R. Wright, A.J. Yates, It.M. Sharma and P. Thibert, Pathology of the BB Wistar Rat. Metabolism, 32, Supplement I (1983) 101-105. 3 T.A. Seemayer, G.S. Tannebaurn, 11. Goldman and E. Colie, Dynamic time course studies of the spontaneously diabetic BB Wistar rat. III. Light-microscopic and uluastructural observations of pancreatic islets of Langerhans. A m . 3. Pathol., 106 (1982) 237 - 249. 4 J. Wright, A. Yates, >1. Sharma and P. Thibert, Histopathological lesions in the pancreas of the BB Wistar rat as a function of age and duration of diabetes./ C o m p . Pathol., 95 (1985) 7-14. 5 D.S. Schalach, D.E. McFarlin and M.H. Barlow, An unusual form of diabetes mellitus in ataxia telangiectasia. N, Engl. J. Med., 28 2 (1970) 1396-1402. 6 S. Goldstein, Human genetic disorders Uiat feature premature onset and accelerated progression of biological ageing. In E.L. Schneider (ed.), The Genetics of ageing. Plenum Press, New York, 1978, 171-224. 7 R.B. Setlow, Repair deficient human disorders and cancel. Nature, 271 (1978) 713-717. 8 B. Lambert and U. Ringborg, DNA repair and human disease. Acta Med. Scand., 2 0 0 (1976) 4 3 3 -4 3 9 . 9 K.H. Kraemer, Progressive degenerative diseases associated with defective DNA repair: xeroderma pigmentosum and ataxia telangiectasia. In W.W. Nicholds and D.G. Murphy (eds.), D N A Repair Processes, Stratton, New York, 1977, 37-71. 10 P.K. Poon, J.W. Parker and K.L. O'Brien, Faulty DNA repair following ultraviolet irradiation in Fanconci's anemia. In P.C. Ilanawatt and R.B. Setlow (eds.), Molecular Mechanisms for Repair o f DN A (Part A), Plenum Press, New York, 1975,821-824. 11 R.W. Hart and R.B. Setlow, Correlation between deoxyribonucleic acid excision-repair and life­ span in a number of mammalian species. Proc. Natl. Acad. Sci. US A , 71 (1974) 2169-2173. 12 J.R. Wright, A.J. Yates, H.M. Sharma and P. Thibert, Spontaneous gastric erosions and ulcer­ ations in the BB Wistar rat. Lab. Anim. Sci., 31 (1981) 63-66. 13 F.L. Fort and P.A. Cerutti, Altered DNA repair in Fibroblasts from aged rats. Gerontology, 2 7 (1981) 306-312. 14 A.C. Chan and l.G. Walker, Loss of DNA repair capacity during successive subcultures of primary rat Fibroblasts./. Cell Biol., 74 (1977) 365 - 370. 15 II.J. Hoffman, Survival distributions for selected laboratory rut strains and slocks. In D.C. Gibson, ll.l', Adtdiiiun and I'. Ilioli (eds.), IterelopmetU o/ the Ituilenl a Almlel ,Vr tle m i»/ uni'tiia (Book 2). DIIEW Publication No. 79-161 (NIH), 1979, 19-34. 16 T.A. Seemayer, W. Schurch and N. Kalant, B cell lymphoproliferation in spontaneously diabetic BB Wistar rats. Diabetologia, 23 (1982) 261-265. 17 J.D. Burek and C.F. Hollander, Incidence patterns of spontaneous tumors in BN/Bi ruts. / Natl. Cancer Inst., 58 (1977) 99-105. 250

18 G.A. Boorman and C.F. Hollander, Spontaneous lesions in the female WAG/Ry (Wistar) rat. J. Gerontol., 20(1973) 152-159. 19 11. Uebergerg and L. Lutzen, The spontaneous rate of tumors in the laboratory rat: strain Chbb: THOM (SPF). Arzneim.-Forsch./Drug Res., 29{1979) 1876-1879. 20 S.W.Thompson, R.A. Iluseby, M j\. l-'ox, C.L. Davis and K.D. Hunt, Spontaneous tumors in the Sprague-Dawley rat. J. Natl. Cancer Inst., 27(1961) 1037-1057. 21 S. Goldstein, Analytical review: the pathogenesis of diabetes mellitus and its relationship to biological ageing. Humangenetik, 12 (1971) 83-100. 22 B.P. Shagan, Is diabetes a model for ageing? AW. Clin. N. Am., 60 (1976) 1209-1211. 23 S. Warren, P.M. LeCompte and M.A. Legg, The Pathology of Diabetes Mellitus (4th edn.), Lea and Fcbiger, Philadelphia, 1966. 24 S. Goldstein, B.J. Moerman, J.S. Soeldner, R.B. Gleason and D.M. Barnett, Diabetes mellitus and prcdiabetes: decreased replicative capacity of cultured fibroblasts. J. Clin. Invest., 53 (1974) 2 7 a-2 8 a. 25 i. Archer and R. Kaye, Cultured skin fibroblasts and juvenile diabetes: senescence and collagen synthesis. Diabetes, 26, Supplement (1977) I 361. 26 B.V. Howard, R.M. Fields, D.M. Mott, P.J. Savage, M. Nagulesparan and P.H. Bennett, Diabetes and cell growth - lack of differences in growth characteristics of fibroblasts from diabetic and non-diabetic Pima Indians. Diabetes (1980) 119-124. 27 R. Vracko and B.P. Benditt, Restricted replicative life-span of diabetic fibroblasts in vitro: its relation to microangiopathy. Fed. Proc., 34 (1975 ) 68 - 70. 28 S. Goldstein, B.J. Moerman, JJS. Soeldner, R.B. Gleason and D. M. Barnett, Chronologic and physiologic age affect replicative life-span of fibroblasts from diabetic, prediabetic, and normal donors. Science, 199 (1978) 781-782. 29 R.E. Gleason and S. Goldstein, Age affect and replicative life span of fibroblasts of diabetic, prediabetic, and normal donors: another look at the data. Science, 202 (1978) 1217-1218. 30 D.W. Rowe, B.J. Starman, W.Y. Fugimoto and R.H. Williams, Abnormalities in proliferation and protein synthesis in skin fibroblast cultures from patients with diabetes mellitus. Diabetes, 26 (1977) 284-290. 31 R.L. Meek, P.D. Bowman and C.W. Daniel, Establishment of rat embryonic cells in vitro. Relation­ ship of DNA synthesis, senescence, and acquisition of unlimited growth potential. Exp. Cell Res., 127 ( 1980) 127-132. 32 A.A. Francis, W.H. Lee and J.D. Regan, The relationship of DNA excision repair of ultraviolet- induced lesions to the maximum life span of mammals. Mech. Ageing Dev., 16 (1981) 181-189. 33 R.W. Hart, C.A. Sacher and T.L. Hoskins, DNA repair in a short- and a long-lived rodent species. J. Gerontol., 34 (1979) 808-817. 34 V. Paffenholz, Correlation between DNA repair of embryonic fibroblasts and different life span of 3 inbred mouse strains. Mech. Ageing Dev., 7 (1978) 1 3 1 -1 5 0 . 35 K.Y. Hall, K. Bergman and R.L. Wall'ord, DNA repair, H-2, and ageing in NZB and CBA mice. Tissue Antigens, 16(1981) 104-110. Appendix J

251 251

Pathological Lesions in the Spontaneously Diabetic BB Wistar Rat: A Comprehensive Autopsy Study

James R. Wright, Jr. Allan J. Yates, Hari M. Sharma, and Pierre Thibert

A total of 145 BB Wistar diabetic rats, 46 of their nondiabatic siblings, and 43 outbred Wistar rats were autopsied and the frequency of lesions in all organ systems were determined. Common strain-related lesions included pulmonary infections, granulomas, lymphoid hyperplasia, lymphomas, lymphocytopenia, eosinophilia, supradiaphragmatic accessory lobes of the liver, and prostatic atrophy. These suggest some basic strain-related abnormalities of the immune system that were selected by the process of inbreeding. Diabetes-related lesions were insulitis, testicular atrophy, cataracts, hepatic fatty change, pancreatitis, lymphocytic thyroiditis, hypoglycemic brain damage, central pontine myelinolysis, stomach erosions, and idiopathic megacolon. Many of these are sequelae of human juvenile-onsat diabetes and support the validity of the BB Wistar rat as an animal model for human diabetes mellitus. The absence of several important sequelae of the human disease (i.e., diabetic nephropathy, atherosclerosis, and severe microangiopathy) suggests a degree of infidelity as a model for human diabetes mellitus. Nonspecific lesions occurring in all three groups of rats includad myocardial degeneration and fibrosis, splenic extramedullary hematopoiesis. and chronic progressive glomerulonephropathy.

UR APPROACH to examining fhe validity of the susceptible than arc nondiabclics to pulmonary infec­ O BB Wistar rat as an animal model for human tions," but in the BB rat these arc strain-, not diabe­ juvcnile-onsct diabetes has been to perform detailed tes-, related. The absence of pneumonia in Wistar rats autopsies on 145 BB Wistar diabetic, 46 of their housed in the same facility indicates that the BB strain nondiabetic siblings, and 43 outbred (non-BB) Wistar has an increased susceptibility to pulmonary infec­ rats that died spontaneously or were sacriliced for tions. The incidence diminished considerably when the experimentation. The types and frequencies of all colony was housed in semibarricr facilities. Myco­ lesions were tabulated by age and group. Some of the plasma appears to be the usuul causative ugcnt, but results of this study have been published in detail other organisms were occasionally present. elsewhere.1 10 Here we present an overview of our Granulomatous lesions were moderately common in findings. The significance ofihcsc is best demonstrated BB rats, but no organisms were seen histologically. by utilizing the following classification scheme: Organs frequently involved were lymph nodes, kidney, 1. Strain-related lesions are abnormalities found in and pancreas; those infrequently involved were lungs diabetic and nondiubctic BB but not in Wistar and testes. Although granulomas were usually small, rats. some large lesions occurred in lymph nodes. 2. Diabetes-related lesions are abnormalities found Strain-specific lymphadcnopathics were common in in BB diabetic rats but not in nondiubctic or BB rats in contrusl to the uniformly normal lymph Wistar rats. These lesions were subdivided into nodes in the Wistar rats. Mesenteric lymph nodes in three groups: Wistar rats were small and distinct, while in BB rats a. Lesions strongly associated with human dia­ they were usually fused into one or more longitudinal betes. nodes, some of which were very large. Although there b. Lesions weakly associated with human diabe­ was no uniform histological appearance in these tes. enlarged nodes, most had sinusoidal hyperplusia. c. Lesions not specifically associated with Three cell populations predominated in hyperplastic human diabetes. nodes. In some, sinusoids were massively dilated with 3. Nonspecific lesions arc those that occurred in plasma cells and differed from plasmacytomas only by nondiabetic und diabetic BB and Wistar rats the presence of occasional cortical follicles. Others with similar frequencies. 4. Organ systems virtually free of significant pathology. From the Ohio State University College o f Medicine, Department o f Pathology, Columbus, Ohio; and Animal Resources Division, Health Protection Branch, Health and Welfare Canada, Ottawa, STRAIN-RELATED LESIONS Ontario, Canada. Supported in part by the Department o f Pathology, the College o f Table I shows the frequency of strain-related lesions Medicine, and the Graduate^ School o f the Ohio State University, in BB rats. These include lymphoproliferative, inflam­ the Upjohn Company, and NIH Grant NS-18026. James R. Wright matory, degenerative, and developmental disorders. Of was supported by the Samuel J, Roessler Research Scholarship. Address reprint requests to Allan J. Yates, M.D., Ph.D., Neuro­ these, bronchopneumonia is of practical importance pathology Unit, Department o f Pathology, The OhioState Universi­ because of its relatively high incidence and mortality. ty, 105 Upham Hall, 47JWI2th Ave, Columbus, OH 43210. Older studies suggest that human diabetics arc more 0026-04 ^5/83/3206-00205 / .00/0

Meiabotism, Vol. 32, No. 7, 1 (July). 1903 252

WRIGHT ET AL

i a b iu J . l Frequency of Strain-Related Lesions (Percent) absence of lumcnal secretions. Similar changes were Non-BU not seen in Wistar rats regardless of age. Interstitial 0«atMi*c B8 Non<*ab«l»c (10 Wistar inflammation was common in all three groups; acute Pneumonia 32 2 46.2 0 hemorrhagic prostatitis was present in one diabetic BB Granulomas 6 4 9 4 0 rat. Lymphomas 4.3 3.1 0 Several closely related BB rats were observed to Eosinophilia* 61.9(27) 59.1 (22) 10.5 (19) Hepatic accessory lobe 0.8 3.1 0 have supradiaphragmatic accessory lobes of the liver, a ‘Results from examinations of peripheral blood smears. Value in rare strain-related developmental abnormality that parentlieses represents number of rets studied. has been reported in only one other strain of inbred rat.14 This suggests a genetic mode of inheritance that showed sinus histocytosis. A third group hud mixed cell may be either polygenic or autosomal recessive with types in diluted sinusoids. Most were lympho-dcplctcd, low penetrance. The gross and histologic appearances and germinal centers were rarely seen in the cortical of these lesions has been described elsewhere* follicles. Lymph nodes from the Wistar rats in our study had distinct follicles, frequently with germinal DIABETES-RELATED LESIONS centers, and never showed the sinusoidal hyperplasia or lympho-depletion typical of BU rats. Table 2, part A, shows the frequency of lesions that Lymphomas were the most common neoplastic are related to diabetes in the BB rat and are also lesions. They were found in approximately 4% of older strongly associated with human diabetes mellitus. One BB diabetic and BB nondiubctic rats but were not seen of the most significant is insulitis, a manifestation of in Wistar rats. The mesenteric lymph nodes and the the onset of JOD in man and in the BB rat. Since colon were the most frequent sites of involvement. insulitis is discussed in detail elsewhere in this mono­ Other sites included the mediastinal lymph nodes, graph, we will mention only briefly a few observations thymus, ileum, liver, pancreas, spleen, lungs, and ova­ from our study.1" The frequency of insulitis in diabetic ry. Lymphomatous nodes were firm, lobulated, white, BB Rats was inversely associated with age and dura­ and usually hemorrhagic. Histologically, they had a tion of diabetes, but was seen in diabetic BB rats up to diffuse pattern and the cells were histiocytic with 317 days old and up to 70 days post detection. Since plasmacytoid differentiation. Some were so large that insulitis was present in 16.3% of our nondiabetic BB they caused complete bowel obstruction. Kulant and rats, it is apparent that insulitis does not always result Seemayer have previously reported u 12.6% and 1.2% in the immediate onset of clinically apparent diabetes incidence of mesenteric lymphomas in diabetic and mellitus. Indeed, many older nondiubctic BB rats nondiubctic BB rats, respectively. These tumors histo­ develop diabetes which may be a consequence of logically were cither plasmacytoid or had features of progressive insular involvement. immunoblastic sarcomas.13 Strain-specific hematological differences were also Table J.2 Fnqutncr ol Diibai«B*flalat*ilLciioni IPtrctml found.2'7 Compared with Wistar rats, BB had signifi­ N 00*08 cantly fewer white cells ar.d platelets, as well us Dub* lie BB NondMtwuc BB W«*t« different differential white-cell counts. There was A. Strongly associated with marked lymphocytopenia, slight neutrophilia, monocy­ human diabetes tosis, and eosinophilia. None studied had leukemia. Insuihtis 14.0* 16.3 0 Testicular atrophy 36.0* 11.1 24.4 Recently, it has been shown that the lymphocytopenia B. Weakly associated with is due to a T cell deficiency.11 These findings are human diabetes consistent with the lympho-dcplctcd appearance of the Hepatic tatty 9.4 3.3 2.2 lymph nodes and the frequent presence of an eosino­ change philic infiltrate in inflammatory lesions. No significant Pancreatitis 8.3 2.3 0 Hypoglycemic brain 0 0 changes in red-cell indices or gummu-globulin concen­ 2.1 damage trations were found. The presence of lymphoid hyper­ Central pontine my- 1.1 0 0 plasia, spontaneous granulomas in the absence of elinolysis demonstrable organisms, lymphomas, eosinophilia, Lymphocytic thy- 63.2* 42.1 0 and lymphocytopenia all suggest an abnormal immune rouklis C. Not associated with system. The high incidence of infections in BB rats is human diabetes also consistent with this. Gastric erosions 32.1 9.7 0 A common degenerative strain-specific lesion in BB Idiopathic megaco- 1.4 0 0 rats was prostatic atrophy (data not shown) character­ km ized by low cuboidul glandular epithelium and the near •Occurred a younger ago m diabetic BB than in other two strains. 253

PATHOLOGICAL LESiONS IN BB WISTAR RATS

Another lesion strongly associated with diabetes thyroiditis in BB diabetic rats suggests that BB rats both in man and BB rats was testicular atrophy.1'5 may closely mimic this syndrome of polycndocrinc Testicular atrophy occurred in all three groups; but autoimmunity. In most instances, the infiltrates were occurred at much younger ages in diabetic rats. Mild mild and predominantly interstitial. Severe thyroiditis testicular atrophy was observed in diabetic BB rats as was strongly associated with the concurrent presence early as 148 days of age, but was not apparent in of insulitis; and as with insulitis, the prevalence of nondiabetic BB or Wistar rats until one year of age. thyroiditis decreased with both age and duration of Testicular atrophy associated with diabetes and that diabetes.’ with old age were histologically identical. Table 2, part C, shows the frequency of several Cataracts were clinically uppurcnt in less than 5% of lesions that arc diabetes-related in the rat but not diabetic BB rats but were not seen in either nondia- associated with diabetes in man. In this category are bclic or BB Wistar rats. An increased incidence of stomach erosions and idiopathic mcgacolon. The most cataracts is strongly related to hyperglycemia in man. common lesions in (he gastrointestinal tracts of BB rats Several types of lesions present in BB rats arc were gastric erosions and ulcerations.4 Gastric erosions weakly associated with human diabetes (Table 2, part were observed in 32.1% of the diabetic BB and 9.7% of B). Hepatic fatly change occurred slightly more fre­ the nondiabetic BB rats. None were seen in Wistar quently in diabetic than in nondiabclic BB or Wistar ruts. The stom ach of the rat is divided into two distinct rats, but in man is usually more closely associated with portions by a transverse ridge—an orud portion lined maturity-unset diabetes and obesity than with JOD.15 by stratified squamous epithelium and a cuudad por­ Acute and/or chronic pancreatitis was likewise some­ tion lined by secretory glandular epithelium. Although what more common in diabetic BB rats,10 but is only erosions of the squamous forcstoinach arc not uncom­ weakly associated with diabetes in man.** mon in some strains of rats,22 mucosal lesions in the BB CNS lesions other than those due to vascular disease rat were almost exclusively confined to the glandular arc uncommon in human JOD. When present, these stomach. Stress ulcerations of this portion may be are often iatrogenic. Two rats with hypoglycemic brain induced in rats experimentally, but they rarely occur damage due to excessive insulin therapy had convul­ spontaneously. We are unaware of similar reports in sions prior to death, nosinophilic neurons in Sommer's any other rat strains. sector of the hippocampus (indicative of hypoglycemic There were a few diabetic ruts with markedly dis­ damage) were present in both rats, but eosinophilic tended colons. Several of these were due to bowel Purkinjc cells were seen in only one. A more interesting obstruction caused by bowel necrosis, lymphomalous ncuropathological linding was a lesion similar to cen­ infiltrates, or prostatitis; but two of these had no tral pontine myclinolysis (CPM). In man this occurs as physical obstruction and were termed idiopathic ntcgu- a single focus of demyelination in the pons with relative colon. Both of the latter were about I year old and had sparing of nerve cells and axons. CPM is frequently been diabetic more than 200 days. We examined the associated with alcoholism and malnutrition and may rectum and colon of these rats with multiple histologi­ be due to the rapid correction of hyponatremia with cal sections, but the ganglion cells of the myenteric intravenous fluids.11 This is a relatively rare disorder in plexus were present in both animals. This rules out man and has not been previously induced in experi­ aganglionic mcgacolon as a cause. Since the distended mental animals, although it has been shown that the abdomen did not appear soon after birth, these gan­ rapid correction of hyponatremia can cause scattered glion cells must have been functional for some lime. foci of demyelination in the rat."1 The animal in which One possible pathogenetic mechanism is an autonomic we made this observation was an emaciated diabetic neuropathy. This concept is supported by some studies BB rat that had been treated for dehydration by suggesting that BB Wistar rats develop ultraslruclural multiple subcutaneous injections of saline.* CPM in and functional evidence of peripheral neuropathy (re­ man has been reported in association with diabetes ceived by Sinta21). mellitus and fluid therapy for dehydration.1’ Lymphocytic thyroiditis was observed in a large NONSPECIFIC LESIONS proportion of diabetic BB but a smaller proportion of Table 3 lists lesions that occurred with similar nondiabetic BB rats. Spontaneous thyroiditis has been frequencies in all three groups of rats. Therefore, they reported in only one other rat strain.20 An association arc neither strain- nor diabetes-related. Hepatic between human JOD and lymphocytic thyroiditis of infarction showed no age-related pattern, but “myo­ other autoimmune thyroid disorders is well-established cardial degeneration and fibrosis”22'24 was seen only in in a small population of JOD patients and has been Wistar rats over 500 days old and occurred at a termed “the syndrome of polycndocrinc autoimmuni- younger age in BB rats. Splenic extramedullary henta- ty."2‘ The occurrence of a spontaneous lymphocytic lopoicsis was also common in all three groups. This was 254

WRIGHT ET AL

Tabic* J .J Fr«()u«ncY ol Noncpecific U tiont tPwtctnll gests a degree of infidelity as a model for human Non-BB diabetes mellitus. The lesions represented in Tables I, Oiatmfic BB Nunsfcabatic BB Wibtar 2 part C, and 3 are undesirable and uncontrollable Hepatic infarction 3.1 3.3 2 .2 variables that may aflccl experiments utilizing this Myocardial degeneration 3 .4 * 4.5 * 10 .5 model. Of these, most are minor inconveniences •Present al a younger age in BB than in W rats. because of low incidences or minimal deleterious cfTects. Only pulmonary infections and gastric erosions associated willi severe splenomegaly only in rats bear­ arc likely to be of practical consequences for most ing tumors or large granulomas. Acute necrotizing experiments. vasculitis was seen occasionally in older BB rats, but All the disorders and frequencies in this study arc this is common in older rats of many strains.2*'2* Some characteristic of BB rats housed under the conditions of the older BB rats had early “chronic progressive present in our housing facilities. It is possible that glomcruloncphropathy" (cortical cysts, glomerular animals housed elsewhere under different conditions basement membrane thickening, and chronic intersti­ may have somewhat different characteristics and inci­ tial nephritis),22 a lesion common to most aged rats and dences of diseases. This is particularly true, since often cited as the most common cause of death in some several BB colonies were started from a few mating rat strains.21 Changes characteristic of diabetic glo­ pairs of outbred rats shipped from Ottawa and were merulosclerosis were not seen. then inbred by sibling matings. This geographic isola­ tion would lend to fix and amplify unusual trails. In addition to inbreeding, the environmental conditions at ORGANS VIRTUALLY FREE OF SIGNIFICANT LESIONS dilferent institutions could have a significant effect on Organs virtually free of significant lesions (except the expression of traits, even in highly inbred strains of those mentioned above) included adrenals, aorta, blad­ laboratory animals.211 der, brain, intestines, kidney, ovaries, parathyroid, From these results we have drawn several general pituitary, spleen, thymus, and uterus. Rare or inciden­ conclusions. First, the BB rat is not simply a “normal” tal findings in BB rats were a pituitary chromophobe Wistar rat that develops bcla-ccll necrosis with subse­ adenoma, ectopic adrenal tissue, three adrenal glands, quent glycosuria and kctosis. Ilistopathologica) subacute orchitis, uterine stromal polyps, mammary changes arc consistent with there being some busic udenocarcinoma, hepatic bile duct hyperplasia, un abnormalities of the immune system which have hepatic abccss, small-bowel necrosis, disseminated apparently been selected, along with the diabetic trait, intravascular coagulation associated with pregnancy, by the process of inbreeding. It seems likely that cystitis, bronchitis, endocardial proliferation, a splenic diabetes is only, a sm all component of a larger BB developmental abnormality, and a subcutaneous spin­ syndrome. Second, the BB rat is prone to a variety of dle cell sarcom a. -inflammatory, degenerative, lymphoproliferativc, and The lesions represented in Table 2, parts A and B, developmental disorders not present in outbred Wistar support the validity of considering BB rats as a model rats. Some of these may adversely affect experiments of human JOD. However, the absence of several utilizing this model. Therefore, quality control should important sequelae of human diabetes (ie, nephropa­ be assured by postmortem examination of animals thy, atherosclerosis, and severe microangiopathy) sug­ involved in experiments.

REFERENCES 1. Wright JR. Sharma It, Thibert P, cl al: Pathologic findings in 6. Wright JK,Sharma HM , Yates AJ.ctal: Supradiaphragmatic the spontaneously diabetic UU Wistar rat. Lab Invest 42:162, i 980 accessory lobe of the liver in UU Wistar rats. Teratology 27:117-120, (abstract) 1983 2. Wright JR, Yales AJ, Sharma HM, cl al: Hematological and 7. Wright JK, Yates AJ, Shuh NT. ct al: Hematological charac­ pathological findings in the spontaneously diabetic BB Wistar rat. teristics of the BB Wistar rat. Vet Clin Pathol 12:9-13,1983 Ann R Coll PhysSurg (Canada) 13:144, 1980 (ubslract) 8. Wright JK, Yales AJ, Sharma HM, ct al: Central pontine 3. Wright JK, Yates AJ, Sharma HM , ct al: Testicular pathology myclinolysis following saline treatment for dehydration: A case in the spontaneously diabetic UU Wistar rat. Diabetes 29(suppl study in a diabetic rat. J Comp Pathol, in press 2):I3I A, 1980 (abstract) 9. Wright JK, Yales AJ, Sharma HM , cl al: Spontaneous 4. Wright JK, Yules AJ, Sharina IIM , ct al: Spontaneous gastric thyroiditis in UU Wistar diubclic rats (submitted) erosions and ulcerations in the BB Wistar rat. Lab Anim Sci. 10. Wright JK, Yales AJ, Sharma HM, el al: Histopalhological 31:63-66, 1981 lesions in the pancreas of the UU Wistar rat as a function of age and 5. Wright JK, Yales AJ, Sharma l(M , cl al: Testicular atrophy duration of diabetes (submitted) in the spontaneously diabetic UB Wistar rat. Am J Path 108:72-79, 11. Root l ! l :: The association of diabetes and tuberculosis. N 1982 l-ngl J Med 210:127-147, 1934 255

PATHOLOGICAL LESIONS IN BB WISTAR RATS

12. Kalanl N, Seemayer T: Malignant lymphoma in sponta­ Franklin ED (cd): Clinical Ir.nnunology Upday, Amsterdam, FJsc- neously diabetic rats. N Engl J Med 300:737, 1979 (letter) vicr/North Holland, 1981 13. Jackson K, Rassi N, Crump T, cl al: The 111) Diabetic rat: 22. llurck JD: Pathology of aging rats. West Palm Beach, Fla, Profound T-ccl( lymphocytopenia. Diabetes 30:887-889, 1981 CRC Press. 1978, pp 70-71 14. Machado EA, Lozzio BD: Accessory lobe of the liver in 23. Sima, AF: The development and structural characterization Gunn-derived rats. Teratology 5:361-366, 1972 of the neuropathies in the spontaneously diabetic BB Wistar rat. 15. Creutzfeldt W, Frerichs 11, Sickinger K: Liver diseases und This volume. diabetes mellitus, in Popper H, Schalfner F (eds): Progress in Liver 24. Anver MR, Cohen BJ: Lesions associated with aging, in Diseases, New York, Grunc & Stratton, 1970, pp 371-407 Baker HJ, Lindsey JK, Wcisbrolh SH (eds): The Laboratory Rat, 16. Warren S, LeCompte PM, Legg MA: The Pathology of vol. 1. New York. Academic P, 1979, p 381 Diabetes Mellitus, ed 4. Philadelphia, Lea 4k I'cbigcf, 1966, pp. 25. Anver MR, Cohen BJ: Lesions associated with aging, in 90-115 Baker HJ, Lindsey JR, Wcisbrolh SH (eds): The Laboratory rat, 17. Wright DG, Laurcno R, Victor M: Pontine and extrapontinc vol. I . New York, Academic P, 1979, pp 383-384 myelinolysis. Brain 102:361-385, 1979 26. Burek JD: Pathology of aging rats. West Palm Beach, Fla, CRC Press, 1978, pp 86-89, 94 18. Kleinschmidl-DcMasters BK, Norenberg MD: Neuropath©- 27. Burek JD: Pathology of aging rats. West Palm Beach, Fla, logic observation in electrolyte-induced myelinolysis in the rat. J CRC Press, 1978, pp 99-100 Ncuropathol Exp Neurol 41:67-80, 1982 28. Anver MR, Cohen BJ: Lesions associated with aging, in 19. Bchar A, Benul E, Aviram A: Central pontine myelinolysis Baker HJ, Lindsey JR, Wcisbrolh SH (eds): The Laboratory Rat, (a case report). Acta Neuropath 3:343-350,1964 vol. I. New York, Academic P, 1979, pp 378-380 20. Hajdu A, Kona G: Spontaneous thyroiditis in laboratory rats. 29. Sabin JR, Horton BJ, Wicks MB: Spontaneous tumors in Expercntia 25:1325-1327, 1969 C3H-A'7 and CJJf-A'TB mice. High incidence in the United Slates 21. Doniuch D, Bottazzo GF: Polycndocrinc autoimmunity, in und low incidence in Australia. JNCI 50:1237-1242, 1973