Citation
Prevention and transfer of autoimmune diabetes following allogeneic bone marrow transplantation

Material Information

Title:
Prevention and transfer of autoimmune diabetes following allogeneic bone marrow transplantation
Creator:
La Face, Drake Maurice, 1959-
Publisher:
University of Florida
Publication Date:
Language:
English
Physical Description:
v, 146 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Bone marrow ( jstor )
Bone marrow transplantation ( jstor )
Cells ( jstor )
Diabetes ( jstor )
Diabetes complications ( jstor )
Diseases ( jstor )
Islets of Langerhans ( jstor )
Neonates ( jstor )
Spleen cells ( jstor )
Type 1 diabetes mellitus ( jstor )
Autoimmune Diseases -- prevention & control ( mesh )
Bone Marrow -- transplantation ( mesh )
Diabetes Mellitus, Type I -- etiology ( mesh )
Disease Models, Animal ( mesh )
Dissertations, Academic -- Pathology -- UF ( mesh )
Pathology thesis, Ph.D. ( mesh )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1988.
Bibliography:
Bibliography: leaves 133-145.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Drake Maurice La Face.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
025477327 ( ALEPH )
20345308 ( OCLC )

Aggregation Information

UFIR:
Institutional Repository at the University of Florida (IR@UF)
UFETD:
University of Florida Theses & Dissertations
IUF:
University of Florida

Downloads

This item has the following downloads:

AA00054903_00001.pdf

00006.txt

00026.txt

00047.txt

00080.txt

00058.txt

00105.txt

00060.txt

00054.txt

00092.txt

00051.txt

00055.txt

00061.txt

00153.txt

00137.txt

00067.txt

00142.txt

00037.txt

00033.txt

00100.txt

00096.txt

00145.txt

00108.txt

00062.txt

00002.txt

00112.txt

00146.txt

00076.txt

00057.txt

00148.txt

00087.txt

00066.txt

00073.txt

00075.txt

00007.txt

00127.txt

00027.txt

00063.txt

00114.txt

00091.txt

00071.txt

00120.txt

00059.txt

00136.txt

00150.txt

00042.txt

00012.txt

00125.txt

00023.txt

00039.txt

00122.txt

00133.txt

00072.txt

00081.txt

00020.txt

00038.txt

ENSXEDJJX_AP8ESB_xml.txt

00151.txt

00101.txt

00011.txt

00034.txt

00010.txt

00083.txt

00143.txt

00024.txt

00110.txt

00093.txt

00117.txt

00152.txt

00022.txt

00119.txt

00111.txt

00019.txt

00126.txt

00135.txt

00070.txt

00032.txt

00138.txt

00068.txt

00107.txt

00128.txt

00140.txt

00064.txt

00008.txt

00035.txt

00095.txt

00090.txt

00016.txt

00116.txt

00118.txt

00005.txt

00103.txt

00017.txt

00139.txt

00097.txt

00050.txt

00121.txt

00085.txt

00018.txt

00098.txt

00113.txt

00052.txt

00144.txt

00084.txt

00069.txt

00134.txt

00004.txt

00088.txt

00029.txt

00074.txt

00132.txt

00077.txt

00041.txt

00053.txt

00104.txt

00115.txt

00078.txt

00149.txt

00141.txt

00131.txt

00021.txt

00028.txt

00031.txt

00009.txt

00046.txt

00147.txt

00044.txt

00013.txt

00001.txt

00109.txt

00099.txt

00102.txt

00040.txt

00129.txt

00094.txt

00014.txt

00086.txt

AA00054903_00001_pdf.txt

00130.txt

00049.txt

00079.txt

00048.txt

00123.txt

00065.txt

00106.txt

00015.txt

00056.txt

00045.txt

00030.txt

00089.txt

00082.txt

00036.txt

00124.txt

00043.txt

00025.txt

00003.txt


Full Text
PREVENTION AND TRANSFER OF AUTOIMMUNE DIABETES FOLLOWING ALLOGENEIC BONE MARROW TRANSPLANTATION
BY
DRAKE MAURICE LA FACE
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR
THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA 1988




ACKNOWLEDGMENTS
I thank my mentor, Dr. Ammon Peck for his guidance and support during my graduate studies and the members of my supervisory committee, Drs. A. Kimura, N. Maclaren, S. Normann, W. Winter and J. Zucali for their guidance in my dissertation Special thanks go to Ms. Marlene Wiley and Ms. Gladys Thompson for help in the preparation of the histological section. I also wish to thank Drs. W. Winter, R. Hacket, R. Braylon and B. Croker for their advice in the interpretation of the pathology.
ii




TABLE OF CONTENTS
Pae
ACKNOWLEDGMENTS ii
ABSTRACT iv
INTRODUCTION 1
Insulin Dependent Diabetes Mellitus in Man . . 1
Non-Obese Diabetic (NOD) Mouse: An Animal
Model for IDDM 13
MATERIALS AND METHODS . . . . . . . . 35
THE USE OF NEONATAL SPLEEN CELLS TO INHIBIT GVH DISEASE FOLLOWING SEMI-ALLOGENEIC AND ALLOGENEIC BONE MARROW TRANSPLANTATION . . . . 41
Introduction 41
Results 45
Discussion 71
PREVENTION AND TRANSFER OF AUTOIMMUNE DIABETES FOLLOWING ALLOGENEIC BONE MARROW TRANSPLANTATION . 78
Introduction 78
Results 86
Discussion 124
SUMMARY AND FUTURE PERSPECTIVES . . . . . 130
REFERENCES 133
BIOGRAPHICAL SKETCH . . . . . . . . 146
iii




Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
PREVENTION AND TRANSFER OF AUTOIMMUNE DIABETES USING ALLOGENEIC BONE MARROW TRANSPLANTATION By
Drake Maurice La Face
August 1988
Chairman: Dr. Ammon B. Peck Major Department: Pathology
Current evidence supports an autoimmune pathogenesis for type I insulin dependent diabetes mellitus (IDDM) in which the pancreatic beta cell is the target for humoral and/or cell mediated immune responses. Diabetes in the nonobese diabetic mouse (NOD) provides an excellent animal model as it shares many pathological features with human type I diabetes. I have carried out reciprocal allogeneic bone marrow reconstitutions between diabetes susceptible NOD and diabetes non-susceptible strains of mice to further delineate the role of the endogenous immune response and host environment in the development of diabetes. I report here that NOD mice reconstituted with a hematopoietic cell system from diabetes non-susceptible mice remained totally free of insulitis and diabetes, whereas NOD mice reconstituted with a syngeneic hematopoietic system showed a 100% prevalence of both insulitis and diabetes.
iv




surprisingly, 100% of diabetes non-susceptible C57BL/6 and B10.BR/cd mice reconstituted with a NOD hematopoietic system developed insulitis with approximately 1 of 10 progressing to overt diabetes. These data emphasize the role of the reconstituting hematopoietic progenitor cell population in the pathogenesis of this autoimmune disease.
v




INTRODUCTION
Insulin Dependent Diabetes Mellitus in Man
Diabetes has been recognized and recorded as a clinical disease in man since ancient times. Recorded descriptions of the symptoms date back as far as 1500 B.C. as inscribed on the Ebers papyrus of Egypt and was first named by Aretaeus of Cappadocia in the second century A.D. (reviewed in Notkins, 1979). Clinical and basic research have revealed that diabetes consists of a heterogeneous set of clinical disorders, all of which are characterized by hyperglycemia (CaHill and McDevitt, 1981; Bennett, 1983).
Diabetes mellitus has recently been categorized into
distinct classifications including Type I, insulin dependent diabetes mellitus (IDDM) and Type II, noninsulin-dependent diabetes mellitus (NIDDM) by an international workshop sponsored by the National Diabetes Data Group of the National Institutes of Health (National Diabetes Data Group, 1979). A third classification of diabetes mellitus includes diabetes associated with pancreatic disease, hormonal etiology, drug or chemical induction, insulin receptor abnormalities and certain genetic syndromes. The focus of this discussion will be specifically on Type I, insulin
1




2
dependent diabetes mellitus (IDDM) in this dissertation. IDDM in man is characterized by an abrupt onset of symptoms which includes elevated blood glucose levels (> 200 mgs/dl), insulinopenia, and frequently ketoacidosis (National Diabetes Data Group, 1979; Keen and Ng Tang Fui, 1979). Persons with IDDM are dependent on insulin treatment to prevent ketosis and death. The onset of IDDM generally occurs in youth but may occur in later life as well. There is a strong correlation between the development of IDDM with certain HLA alleles and there are experimental data which support an autoimmune etiology including evidence for the presence of specific autoantibodies and cellular immunity to pancreatic islet cells.
Epidemiology, morbidity and mortality. Epidemiologic
studies indicate the prevalence of IDDM in the United States to be approximately one in 300 under the age of 30 (Rosenbloom, 1983). The incidence of IDDM appears to vary among developed countries (Norris et al., 1987). The highest incidence was reported in Scandinavia. In Finland, about 29 per 100,000 people per year develop IDDM. In contrast, only 0.8 per 100,000 people per year develop IDDM in Japan. The incidence in the United States lies between these two extremes at approximately 14 per 100,000 per year. Geographic variation in incidence suggests an influence of environmental factors on the development of IDDM. An alarming statistic is that the incidence of IDDM has




3
increased in recent decades (Krolewski et al., 1987; Stewart-Brown et al., 1983). For example, in the first half of the twentieth century, the incidence rate of IDDM in the Northeastern United States remained relatively constant but during the last 30 years there has been a nearly 3-fold increase in the incidence rate (Krolewski et al., 1987). Likewise, in England there was a 2-fold increase in the incidence of IDDM from 1958 to 1980 (Stewart-Brown et al., 1983). However, the rising incidence of IDDM remains controversial since increasing numbers of patients surviving beyond the reproductive age would not be sufficient to account for the dramatic increases in incidence rate.
Prior to the availability of insulin, IDDM patients
generally died within a few years of onset (Dorman et al., 1984; Crabbe, 1987). Most deaths were associated with ketosis induced coma and infection (Rossini and Chick, 1980). There was a dramatic increase in life expectancy after the onset of IDDM associated with the advancement of insulin treatment. However, the development of long-term sequelae became evident and IDDM patients still experience a life expectancy reduced by one-third as compared to nondiabetic individuals.
A 40 year follow-up study of over 300 patients
diagnosed before 1933 at less than 21 years of age was carried out at The Steno Memorial Hospital in Denmark (Deckert et al., 1973). Only 40% of the patients were




4
still alive in 1973. The calculated mortality rate was 2-6 times that of an age matched non-diabetic population. The cause of death in 31% of the patients was kidney failure and 25% died of myocardial infarction. Further complications included blindness in 16% and severely impaired vision in another 14%. Twelve percent of the patients also suffered gangrene or previous amputation.
Clinical and experimental evidence clearly indicates that vascular disease is the major cause of morbidity and mortality in diabetic individuals (Rossini and Chick, 1980; Levin and O'Neal, 1983). Macroangiopathy and arteriolar vascular disease can lead to myocardial infarction, stroke, and gangrene of the lower extremities. Microangiopathy can lead to nephropathy and retinopathy. The etiology of vascular disease abnormalities is not clearly understood but may involve hyperglycemia, hyperlipidemia, rheological disorders, and hypertension secondary to nephropathy. Physical manifestations associated with diabetes include connective tissue changes in arterial walls capable of altering normal blood flow and basement membrane thickening of capillaries of the kidney, eye, peripheral nerves, and skeletal muscle (McMillan, 1983; Rossini and Chick, 1980).
The two long-term complications most strongly associated with mortality in IDDM patients are diabetic renal and heart disease (Rossini and Chick, 1980; Crabbe, 1987). It is estimated that there is a 20-fold increase in the risk




5
of renal failure in diabetics as compared to non-diabetics. Pathological changes associated with renal disease in diabetics include basement membrane thickening in glomerular capillary walls, nodular intercapillary glomerulosclerosis and diffuse glomerulosclerosis. Clinical features associated with IDDM include coronary atherosclerosis, myocardial hypertrophy and interstitial fibrosis (Fein and Scheuer, 1983).
Autoimmune pathogenesis and insulitis. A fundamental characteristic that separates IDDM from other forms of diabetes is that abnormal immune responses and autoimmunity are thought to play a role in the pathogenesis of IDDM (National Diabetes Data Group, 1979). Reports of mononuclear cell infiltration of the islets of Langerhans (insulitis) in patients who died suddenly after an abrupt onset of IDDM triggered a widespread search for evidence of autoimmunity against specific elements of the endocrine pancreas (Gepts, 1965; Nerup et al., 1984).
The pathognomonic lesion of insulitis can be more clearly understood after a brief review of the normal histologic morphology of the islets of Langerhans (pancreatic islets) (Ito, 1977; Volk and Wellmann, 1985). The normal pancreatic islet is comprised of a compact, rounded mass of endocrine cells surrounded by a connective tissue sheath (comprised of reticular fibers and a basement membrane). The islets are irregularly distributed within




6
the lobules of the exocrine portion of the pancreas but are more or less demarcated from the surrounding acinar tissue by the connective tissue sheath. The islets are highly vascular with numerous capillaries which form glomeruluslike networks that allow close proximity with every endocrine cell. The cell cord mass is made up of 4 cell types: 1) P-cells comprise about 70% of the islet cells and their endocrine function is to produce and secrete insulin, 2) a-cells account for about 15-20% of the islet cell population and are responsible for the production and secretion of glucagon, 3) 6-cells which produce somatostatin account for about 10% of the islet cell population, while 4) PP-cells comprise about 1% of the islets cells and produce pancreatic polypeptide. The a-cells and 6-cells are primarily located in a narrow peripheral zone of the islet.
Lymphocytic infiltration of the islets of Langerhans in diabetic patients was initially described at the beginning of the twentieth century and in 1940 Von Meyenburg coined the term "insulitis" to describe this inflammatory lesion (reviewed in Gepts and LeCompte, 1981). Insulitis was initially thought to be a rare lesion as the prevalence was calculated from studies of a heterogeneous set of diabetic pancreata. However, more selective studies of pancreata from IDDM patients who died early in the course of their disease indicated that insulitis was not a rare lesion. (LeCompte, 1958; Gepts, 1965; Gepts and LeCompte, 1981).




7
These findings encouraged more specific studies of the possible role of insulitis in the pathogenesis of IDDM. Insulitis was observed in 68% (15/22) and 78% (47/60) of cases in two relatively large studies of pancreata from young diabetic patients with an acute onset of disease (Gepts, 1965; Foulis et al., 1986).
Histologic examination disclosed several pathological features characteristic of pancreatic islets from patients with acute onset of IDDM (Foulis and Stewart, 1984; Gepts and LeCompte, 1985; Foulis et al., 1986). The majority of the islets were composed of narrow cords of small endocrine cells arranged in a fibrous stroma with irregular outlines. The cytoplasm of the cells was not abundant and the small nuclei had dense chromatin. The cells appeared atrophic but were actually composed of active endocrine cells. Immunocytochemical staining techniques showed that these pseudoatrophic islets were devoid of pancreatic p-cells but contained normal numbers of a-cells, 6-cells and PP-cells. The less numerous set of unique pancreatic islets was relatively large and contained large hypertrophic cells with large nuclei. Immunocytochemical staining techniques showed the hypertrophic islets to be composed of degranulated pancreatic P-cells and normal numbers of a-cells, 6-cells and PP-cells. Some of the hypertrophic islets showed marked reduction in the number of P-cells. Quantitative studies indicated a marked reduction in the total mass of endocrine




8
tissue in IDDM patients as compared with normal controls and the P-cell mass was reduced to less than 10% of normal (Gepts and LeCompte, 1981; Rahier et al., 1983). Further evidence for P-cell specific autoimmunity was derived from immunocytochemical staining methods which revealed that the lymphocytic infiltrations were primarily surrounding or within pancreatic islets that contained P-cells (Gepts and LeCompte, 1981; Foulis and Stewart, 1984; Gepts and LeCompte, 1985; Foulis et al.,1986). Quantitative analysis showed that about 20% of islets containing insulin (P-cells) were affected by insulitis but only 1% of islets deficient in insulin were affected.
Morphological analysis of mononuclear cell infiltrates of pancreatic islets from IDDM patients indicated that the infiltrating cells consisted predominantly of lymphocytes and occasional polymorphonuclear leukocytes (Gepts, 1965; Foulis and Stewart, 1984). Immunofluorescence staining techniques on frozen sections from a diabetic pancreas indicated that the majority of the infiltrating lymphocytes were T-cells (Bottazzo et al, 1985). Further characterization of the T-lymphocytes showed the predominant subpopulation to be cytotoxic/suppressor T-cells. T-helper cells and natural killer (NK) cells were also shown to be present in small numbers. The majority of the infiltrating T-lymphocytes were shown to be positive for HLA-DR expression indicating that they were activated.




9
There was an apparent progression of inflammatory cell infiltration and p-cell degeneration. Pancreatic islets affected by early stages of cellular infiltration still maintained normal islet morphology and contained relatively well preserved p-cells. With the progression of insulitis, the islets began to develop a pattern of small cell cords, a collapsed islet framework and reduced numbers of p-cells. The lymphocytes gradually disappeared from the islets as all the p-cells were destroyed leaving collapsed islets with cords of insulin-deficient endocrine cells separated by fibrous tissue. A few of the collapsed islets showed a residual inflammatory cell infiltrate but most were totally devoid of insulitis. The observation of the progressive and specific destruction of pancreatic p-cells (leaving a-cells, 6-cells and pp-cells intact) supports the hypothesis of autoimmune mediated destruction of pancreatic p-cells in the pathogenesis of IDDM.
Cellular autoimmunity. Methods to measure specific cellular immunity to pancreatic islet autoantigens have provided direct evidence for the autoimmune pathogenesis of IDDM. Initially, organ-specific cellular immunity was indicated by the ability to induce leukocyte migration inhibition of cells from IDDM patients in the presence of pancreatic cell extracts (Nerup et al., 1971; Nerup et al., 1974). In vivo evidence of specific cellular immunity was indicated by the ability to induce a delayed-type




10
intracutaneous reaction in IDDM patients by subcutaneous injection of pancreatic cell extracts (Nerup et al., 1974). No delayed-type reaction was seen in normal controls or in IDDM patients who did not show a positive leukocyte migration inhibition.
Experiments have also been designed to show lymphocyte and antibody dependent cellular immunity to pancreatic fl-cells (Huang and Maclaren, 1976). Lymphocytes from IDDM patients had a markedly higher cytoadherence to human insulinoma cells after 15, 40 and 60 hours co-culture as compared to normal individuals. Furthermore, enriched T-lymphocytes (erythrocyte rosette-forming cells) were shown to be significantly cytotoxic to human insulinoma cells in chromium-51 release assays. Significant antibody-dependent cellular cytotoxicity was also observed when T-cell depleted leukocyte populations (non-rosette-forming cells) and serum from IDDM patients were cultured with chromium-51 labelled insulinoma cells.
An effect of cellular immunity on the production of insulin by pancreatic fl-cells in IDDM patients has been investigated (Boitard et al., 1981). Pancreatic islets were isolated from DBA/2 mice and stimulated to increase insulin production in culture by the addition of glucose and theophylline. The addition of lymphocytes from IDDM patients significantly inhibited the stimulation of insulin production as compared to lymphocytes from normal controls.




The experiments discussed above provided definitive evidence for specific cellular immunity against pancreatic #-cells in the pathogenesis of IDDM.
Humoral autoimmunity. The detection and
characterization of organ-specific autoantibodies suggested an autoimmune etiology and a possible role for humoral immunity in the pathogenesis of IDDM. The first significant detection of autoantibodies to pancreatic islet cells were in a select group of diabetic patients with associated autoimmune endocrine diseases (Bottazzo et al., 1974; MacCuish et al., 1974). Circulating islet cell antibodies (ICA) were detected in patient sera by indirect immunofluorescence techniques using unfixed frozen sections from human pancreas as the binding substrate. ICA from positive patient sera gave uniform cytoplasmic immunofluorescence involving all the pancreatic islet cell types, were complement fixing, and were of the IgG class. Studies were then done to detect the frequency of ICA in young IDDM patients of recent onset (Lendrum et al., 1975). Of the 105 IDDM patient sera tested by immunofluorescence assays on frozen human pancreas sections, 51 (49%) had detectable ICA antibody. Further studies indicated that the frequency of ICA was 75% in caucasian IDDM patients within 3 months of diagnosis but then declined markedly (Neufeld et al., 1980).
The use of living cells as the binding substrate has made it possible to detect cell surface binding antibodies




12
in the sera of IDDM patients. Initially, cell surface binding of circulating antibody from IDDM patient sera was demonstrated using indirect immunofluorescence methods on human-insulinoma cells (Maclaren et al., 1975). Positive binding of antibody to the cell surface of the insulinoma cells was reported in 34 of 39 IDDM sera. These results were later confirmed using viable insulin-producing islet cells from rats as the binding substrate (Lernmark et al., 1978). The cell surface binding autoantibodies were termed islet cell surface antibodies (ICSA). Complement-mediated cytotoxicity of chromium-51 labeled rat islet cell cultures (Dobersen et al., 1980) and cloned rat islet cell monolayers (Eisenbarth et al., 1981) was detected in ICSA+ sera from IDDM patients suggesting a possible role for ICSA in the pathogenesis. Furthermore, ICSA+ sera produced significant antibody-dependent cellular cytotoxicity of cloned human pancreatic a-cells in the presence of normal human lymphocytes (Maruyama et al., 1984a).
Genetics of IDDM. The association of genetic susceptibility for IDDM to genes within the major histocompatibility complex (MHC or the HLA complex in man) supports the hypothesis for a role of the immune system in the pathogenesis. Serotyping of blood cells from IDDM patients in population and family studies have indicated a strong association of HLA-DR3 and HLA-DR4 alleles with diabetes susceptibility (Platz et al., 1981; Cudworth and




13
Wolf, 1982; Wolf et al., 1983; Henson et al., 1986). In a study done in the United States (Henson et al., 1986) with over 1000 IDDM probands, 95% of probands were positive for either HLA-DR3 or HLA-DR4 while only 50% of controls were positive for at least one of these antigens. Approximately 40% of the probands possessed both the HLA-DR3 and HLA-DR4 antigens compared to only 3% of the general population.
Population and family studies have also indicated a strong association of HLA-DR2 and HLA-DR5 antigens with resistance for the development of IDDM. Only 4% of IDDM patients were positive for HLA-DR2 compared to 28% of controls and only 1% of IDDM patients were positive for HLA-DR5 compared to 17% of normal controls (Cudworth and Wolf, 1982; Wolf et al., 1983). The association of diabetes susceptibility and resistance with certain HLA-DR haplotypes suggests a possible role for the immune system in the pathogenesis of the disease since class II gene products are important in the regulation of the immune response.
Non-Obese Diabetic (NOD) Mouse: An Animal Model for IDDM Development and characterization of the NOD mouse. As indicated above, IDDM in man is a significant health care problem associated with severe long-term complications. Experimental evidence indicates that the etiology is very complex and may entail a genetic predisposition and an autoimmune pathogenesis. However, experimental approaches




14
are relatively limited when working with human IDDM patients. Representative experimental animal models for IDDM have been important in developing an understanding of this very complex disease. Of particular interest was the recent development of the NOD mouse which spontaneously develops an autoimmune response against the p-cells within the islets of Langerhans resulting in their destruction and loss of the insulin production with progression to insulin dependent diabetes (IDD) (Makino et al., 1980). These phenotypic features strongly resemble those observed in human Type I insulin dependent diabetes mellitus. The NOD mouse shares many pathological features common to human type I diabetes including hyperglycemia, insulinopenia, hyperglucagonemia, glycosuria, polydipsia, polyuria, ketoacidosis, weight loss, hypercholesteremia, and the pathognomonic lesion of insulitis (Makino et al., 1980; Leiter et al., 1987). These similarities make the NOD mouse an excellent model for the examination of human IDDM.
The NOD mouse was initially discovered and developed at the Shionogi Research Laboratories in Osaka, Japan. It was developed from a subline of ICR mice which were being selectively bred for cataracts (Makino et al., 1980). Since one of the clinical manifestations of IDD is cataract formation, the ICR-derived mice were screened beginning at the 6th generation for hyperglycemia. After selective breeding for 13 additional generations, two sister sublines




15
were established: a normoglycemic (100 mg/dl blood glucose) subline and a slightly hyperglycemic (150 mg/dl blood glucose) subline (Tochino, 1986). One of the female offspring from the normoglycemic subline spontaneously developed diabetes and died. The surviving progeny from that diabetic female were then inbred to produce a diabetic subline that resulted in the development of the non-obese diabetic (NOD) mouse. NOD mice have been established as an inbred strain which continues to develop IDD in a consistent manner. A non-obese non-diabetic (NON) strain has been established from the slightly hyperglycemic subline as an inbred strain. Examination of histocompatibility and plasma enzyme markers revealed that the NOD mice and NON mice were significantly different strains of mice.
Histologic examination indicated that there was not a
significant difference in the overall frequency of insulitis between male and female NOD mice (Komeda and Goto, 1986), as insulitis was observed in nearly 100% of both males and females by 30 weeks of age. However, the development of insulitis appeared to be delayed in male NOD mice as compared to females. Tochino reported that female NOD mice developed overt diabetes more frequently than the males. There was about an 80% frequency of overt diabetes in female mice but only a 20% frequency in male mice (Tochino, 1976). The overall nature of these differences is not clearly understood, but investigators have suggested that hormonal




16
differences are involved since castrated males have a higher incidence of diabetes while oophorectomized females have a lower incidence of diabetes (Makino et al., 1981).
The blood glucose levels of NOD mice rose at the onset of diabetes from the normal 100-180 mgs/dl to levels of 600-800 mgs/dl correlating with decreased levels of plasma insulin. Several changes from the normal physiology of NOD mice were observed after the onset of diabetes (Makino et al., 1980). A decrease in body weight was observed despite an increase in food consumption and a four fold increase in water consumption. The volume of urine excretion increased to 15-20 times normal and the urine contained up to a 100-fold higher concentration of glucose (glycosuria). Plasma cholesterol levels were also increased significantly and ketonuria could be observed in diabetic mice. NOD mice did not undergo spontaneous remission from diabetes and the mice died due to ketoacidosis if they were not administered insulin treatment. Daily injection of insulin induced an increase in body weight and a marked prolongation of life in the diabetic mice.
Autoimmune pathogenesis and insulitis. The most prominent histopathologic lesion of insulin dependent diabetes in both man and the NOD mouse is the inflammatory response of insulitis, characterized by the infiltration of leukocytes into pancreatic islets. The development and progression of insulitis over time has been examined by




17
light microscopy in the NOD mouse (Fujita and Yui, 1976; Fujino-Kurihara et al., 1985; Leiter et al., 1987).
Histological examination of pancreata of NOD mice from parturition to 3 weeks of age showed no cellular infiltrates by light microscopy. However, by 4-5 weeks of age some of the animals showed mononuclear cells accumulating in or near the pancreatic islets. The cellular infiltration was also observed around the periductal and perivascular structures within the surrounding connective tissue which contains excretory ducts, arterioles, venules capillaries and small lymphatics. The mononuclear infiltrates usually invaded the islet from the side of the periductal connective tissue and congregated on that side of the connective tissue sheath. Eventually, the lymphocytes formed a multilayered aggregation around the perimeter of the islets. The infiltrating lymphocytes gradually increased in number until more and more islets were infiltrated. The boundary formed by the connective tissue sheath was eventually broken as some of the mononuclear cells invaded the territory of the islet. The P-cells adjacent to the infiltrating mononuclear cells began to degenerate and were replaced by the infiltrating cell mass. The original P-cells within the pancreatic islets decreased in number and eventually were entirely replaced by the mononuclear infiltrating cells. When all the P-cells were destroyed the infiltrating cell masses gradually dispersed and then disappeared leaving a




18
residual islet containing primarily a-cells, 6-cells and some PP-cells as determined by immunohistochemical findings. These findings indicate that the immune response was specific for the pancreatic P-cells since the other pancreatic islet cells remained intact. Overt diabetes developed when most of the P-cells had been destroyed.
A sequential analysis of pancreatic sections stained with fluoresceinated antisera against Thy-l.2 and immunoglobulin showed that T-lymphocytes, including helper/inducer and cytotoxic/suppressor T cells, were the predominant infiltrating cells in the early stages of insulitis (Miyazaki et al., 1985; Miyazaki et al., 1986; Hanafusa et al., 1986; Toyota et al., 1986). Furthermore, double fluorescence staining showed that more than one-half of the T-lymphocytes were positive for class II molecule (Ia+) expression. Class II molecules are not usually expressed on murine T-lymphocytes, thus, this may indicate that the T-lymphocytes present in the insulitis were activated. Analysis of the T-lymphocyte subsets showed that Lyt-l bearing cells (primarily T-helper/inducer cells) were more commonly present than those expressing the Lyt-2 antigen (primarily T-cytotoxic/suppressor cells). The T-lymphocytes were located close to the islets and also infiltrated deep within the islets. B-lymphocytes were also present at all stages of insulitis and gradually increased in numbers as the severity progressed. One-half or more of




19
the infiltrating lymphocytes were B-lymphocytes in severe insulitis. The B-lymphocytes were located primarily near the blood vessels and at the periphery of the T-lymphocytes. Some natural killer (NK) cells were shown to be present close to pancreatic islet cells. Close examination of the P-islet cells next to the infiltrating lymphocytes revealed evidence of cellular degeneration and necrosis as well as cell death (Fujino-Kurihara et al., 1985; Fujita and Yui, 1986; Fujita and Fujino-Kurihara, 1986). The histological and immunohistochemical studies described above suggest that the immune system plays an important role in the pathogenesis of diabetes in the NOD mouse, indicating an autoimmune etiology.
Cellular autoimmunity. Data from several experimental systems support the hypothesis that diabetes observed in NOD mice has an autoimmune etiology. For example, generalized immunosuppression induced by sublethal irradiation (870 Rads) reduced the frequency of diabetes to 25% (2 of 8 NOD mice) within 29 weeks of age (Harada and Makino et al., 1976). Furthermore, a reduced frequency of diabetes was observed in neonatal thymectomized NOD mice (Hanafusa et al., 1986) suggesting a role for the cellular immunity since the thymus is involved in development and education of T-lymphocytes.
Islet specific T-cell lines have been isolated from NOD pancreatic islets bearing mononuclear cell infiltrates




20
(Hattori et al., 1986), indicating that T-lymphocytes are involved in the development of insulitis. The resulting cell lines were shown to be Thy-1.2+ and Lyt-2+ by flow cytometric analysis and to proliferate in response to NOD islet homogenates. Furthermore, intravenous injection of rabbit anti-mouse thymocyte antisera to deplete T-lymphocytes markedly decreased the incidence of overt diabetes in NOD female mice (Hanafusa et al., 1986) and reduced the severity of insulitis (Yokono et al., 1986) indicating an important role for thymus-derived lymphocytes in the development of diabetes.
The development of diabetes in NOD mice was also shown to be dependent on the L3T4+ subpopulation of T-lymphocytes by intravenous injection of L3T4-specific antibody (GKl.5) for 12 weeks starting at 2 weeks of age (Koike et al., 1987). Monoclonal antibody GK1.5 binds to the cell surface antigen L3T4 which is primarily associated with T-helper cells reactive against antigen in the context of self class II molecules (Dialynas et al., 1983). Histological examination of pancreata from GK1.5 injected NOD mice showed no insulitis and the development of diabetes was prevented. In another study, diabetic NOD mice were engrafted with pancreatic islets from BALB/c mice and injected intravenously with GKl.5 for four consecutive days to prolong engraftment (Wang et al., 1987). The engraftment of pancreatic islets and the temporary depletion of L3T4+




21
cells resulted in a return to normoglycemia that persisted for 2-4 weeks. Histologic examination of the grafted pancreatic islet tissue, at the time when the animals again became hyperglycemic and the L3T4+ T-cells returned to normal numbers, showed that the islet tissue had been destroyed and replaced by mononuclear infiltrating cells. Considered together, these studies indicate an important role for T-helper cells in the pathogenesis of insulitis and diabetes in NOD mice.
Theoretically, the binding of antigen by T-lymphocyte receptors requires the antigen to be presented on the surface of a cell in the context of a self MHC class I or class II molecule (Kronenberg et al., 1986). Therefore, initiation of the response by T-helper cells against the pancreatic -cells may require that an autoantigen be presented on the surface of a 1-cell in the context of a self class II molecule. Initially, cells of the macrophage lineage were considered to be responsible for antigen presentation to T-helper cells in the context of self class II MHC molecules (Unanue 1981; Ziegler and Unanue, 1981). However, other cell types have been hypothesized to be capable of antigen presentation in the context of self class II molecules including: 1) Langerhans cells (Tanaki et al., 1979), 2) B-lymphocytes (Chesnut and Grey, 1981; Chesnut et al., 1982), 3) thyroid cells (Hanafusa et al., 1983) and pancreatic B-cells (Bottazzo et al., 1983; Bottazzo et al.,




22
1985; Londei et al., 1984). The first reports of aberrant class II molecule expression on endocrine cells associated with autoimmune disease were from patients with thyroid autoimmunity (Hanafusa et al., 1983) and type I IDDM (Bottazzo et al., 1983). It was suggested that the class II expressing a-cells may present auto- antigens in the context of self class II and thus initiate the autoimmune response (Bottazzo et al., 1983; Bottazzo et al., 1985; Londei et al., 1984). This theory would provide a possible explanation for the association of type I diabetes with the DR 3 and DR 4 histocompatibility phenotypes.
Immunofluorescence studies have been performed to
determine if I-A molecules might be expressed on the surface of pancreatic a-cells in NOD mice (Hanafusa et al., 1987) as described in human IDDM patients. Positive islet cells were seen in all 23 NOD mice examined. Double immunofluorescence staining showed that the I-A positive cells were all insulin producing cells but were negative for production of glucagon or somatostatin indicating that only the #-cells of the pancreatic islets expressed I-A. All of the islets with mononuclear cell infiltration and about half of the islets with no infiltration were positive for I-A molecule expression. Perhaps part of the genetic susceptibility in organ specific autoimmunity is related to a predisposition for aberrant expression of class II MHC molecules on the surface of the target cells resulting in sufficient




23
presentation of autoantigens to L3T4+ lymphocytes.
Further evidence for an autoimmune pathogenesis of diabetes in NOD mice comes from adoptive transfer experiments with which a pathological contribution by T-lymphocytes has been implied. In one experimental system, adoptive transfer procedures in NOD mice could induce rapid onset of diabetes at an age when spontaneous diabetes is rarely observed (Wicker et al., 1986a; Wicker et al., 1986b). NOD mice between 6 and 13 weeks of age were sublethally irradiated (775 Rads) and injected intravenously with spleen cells or enriched T-cells from overtly diabetic mice. In contrast to the very low percentage of female mice that would have become spontaneously diabetic by 13 weeks of age, 95% (79 of 82 mice) developed diabetes within 12-22 days after sublethal irradiation and injection of autoreactive cells from overtly diabetic NOD mice. To determine the subset(s) of T-cells responsible for the adoptive transfer of autoimmunity, Hanafusa and co-workers utilized T-cell depleted NOD mice which were injected with L3T4+ depleted spleen cells or Lyt-2+ depleted spleen cells from cyclophosphamide treated NOD mice (Hanafusa et al., 1988). The mice from the donor population were pre-treated with cyclophosphamide since this treatment has been shown to promote the onset of overt diabetes in NOD mice (Harada and Makino 1984). T-cell depletion in NOD mice was achieved by thymectomy at 4-5 weeks of age, followed by injection of




24
antithymocyte antiserum 2 days later and sublethal irradiation (800 Rads) 3 weeks after thymectomization. T-cell depleted NOD mice had a reduced incidence of insulitis (23%) as compared to untreated NOD mice (80%) or to T-cell depleted NOD mice which were adoptively transferred with either whole spleen cell populations (69%) or L3T4+ enriched cell populations (81%). However, adoptive transfer of L3T4+ depleted spleen cells resulted in a reduced incidence of insulitis (20%). These data have been interpreted to mean that the L3T4+ T-lymphocytes are crucial in the pathogenesis of insulitis in NOD mice.
Humoral autoimmunity. A possible role for humoral immunity in the development of diabetes has also been indicated from studies showing the presence of islet cell antibodies (ICA) and islet cell surface antibodies (ICSA) in NOD mice. The presence of ICA in NOD mice was detected by indirect immunofluorescence using serum samples from NOD mice reacted with pancreatic tissue sections from ICR mice followed by an incubation with a fluoresceinated rabbit anti-mouse antibody (Toyota et al., 1982; Takei et al., 1986). ICA was shown to be present in 75% of NOD mice at 4 weeks of age, a time prior to the presence of insulitis detectable by light microscopy. A possible role for ICA in autoimmunity is not clear since the antigen is present in the cytoplasm rather then on the cell surface. The detection of ICSA in NOD mice was reported by two groups in




25
Japan by indirect immunofluorescence using cultured islet cells (Kanazawa et al., 1984; Matsuba et al., 1986). Kanazawa and co-workers determined the presence of ICSA in NOD mice at various ages. ICSA was first present at 6 weeks of age in less than 10% (1/13) of the NOD mice and reached a peak prevalence of 50% (3/6) by 12 weeks of age. Subsequently, the detectable ICSA decreased as the mice became older. In contrast, Matsuba and co-workers reported a 69% (34/49) overall incidence of ICSA in NOD mice from all ages. Hybridoma cell lines secreting monoclonal antibody specific for ICSA have been produced by fusing lymphocytes from ICSA+ NOD mice with FO myeloma cells (Yokono et al., 1984). Prior to fusion the NOD lymphocytes had been transferred to sublethally irradiated (600 Rads) ICR mice by intravenous injection since this increased the frequency of ICSA reactive hybridoma production. The specificity of purified ICSA reactive monoclonal antibody (3A4) was determined in a radiolabelled protein-A binding assay using Syrian golden hamster insulinoma cells (In-lll cells). Immunoprecipitation of radiolabelled proteins from In-lll cells using monoclonal antibody 3A4 allowed the isolation of a 64 kilodalton cell surface protein and a 28 kilodalton cytoplasmic protein (Hari et al., 1986). The immunoprecipitated 64 kilodalton protein may represent the auto-antigen recognized by ICSA. Further evidence that ICSA may play a role in autoimmunity was indicated by positive killing in




26
antibody-dependent cellular cytotoxicity (ADCC) assays using 3A4 antibody.
Abnormal immune functions. Functional studies of the immune system from NOD mice have revealed an altered immune cell repertoire and a relatively diminished immune responsiveness. Experiments using 3 month old NOD mice showed possible abnormalities of cellular or humoral immunity when compared with results from ICR mice (Kataoka et al., 1983; Maruyama et al., 1984b; Toyota et al., 1986b). NOD mice appeared to have significantly lower numbers of mononuclear blood cells and mononuclear spleen cells as well as a significantly reduced number of T-lymphocytes as compared with ICR mice. A more specific examination of the T-lymphocyte subpopulations indicated a normal level of Lyt-1+ (T-helper cells) and a markedly reduced number of Lyt-2+ (T-cytotoxic/suppressor cells) (Toyota et al., 1986b). The number of B-lymphocytes in NOD mice was normal, which translates to a relative increase in the percentage of B-cells. However, there was a significantly reduced number of phagocytizing macrophages in NOD mouse spleens.
The ability of NOD immune cells to mount a response in vitro and in vivo was also measured using various immunological assays. Natural killer (NK) cell activity against YAC-1 target cells, antibody-dependent cell-mediated cytotoxicity (ADCC) activity against chicken red blood cells, and the induction of cytotoxic T-lymphocytes (CTL)




27
against EL-4 cells were all significantly reduced in NOD mice indicating a diminished ability to mount a cellular immune response. Furthermore, NOD spleen cells produced significantly less interleukin-2 (IL-2) in response to Concanavalin A stimulation than ICR spleen cells, and NOD mice showed a diminished resistance to a virulent herpes simplex virus following intraperitoneal inoculation (Toyota et al., 1986b). However, delayed type hypersensitivity (DTH) responsiveness to sheep erythrocytes was normal indicating normal cellular immunity for certain immune functions. Furthermore, NOD mice showed a significantly increased ability to produce antibodies against T-cell dependent antigens such as sheep erythrocytes, even though NOD mice possess normal numbers of B-cells, indicating a possible enhancement of T-helper cells that regulate B-cell responsiveness. These data indicate that there are alterations of immune response functions in NOD mice but that the specific defects are not well defined.
Genetics of IDD in the NOD mouse. Breeding studies
have been designed by several laboratories to determine the inheritance of genetic susceptibility for the development of autoimmune diabetes in NOD mice (Makino et al., 1985; Hattori et al., 1986; Prochazka et al., 1987; Wicker et al., 1987). Makino and co-workers examined the inheritance characteristics of insulitis in breeding studies between NOD and C57BL/6 mice. Insulitis was used as a marker since




28
there is a 100% incidence of insulitis in NOD mice and the lesion is believed to be important in the development of diabetes (Makino et al., 1985). Insulitis was not observed in (NOD x C57BL/6)Fl or (Fl x C57BL/6) backcross progeny indicating insulitis had a recessive functional inheritance. However, insulitis was observed in 3.9% of (Fl x F1)F2 and 23.7% of (Fl x NOD) backcross female mice as compared to 89.7% of (NOD x NOD)P1 mice. The 1/16 and 1/4 ratios for the incidence of insulitis in the F2 and backcross mice with respect to the parental (Pl) mice indicated to the authors that the development of insulitis was controlled by two independently segregated recessive genes.
Hattori and co-workers specifically examined the possibility of MHC-linked susceptibility genes for the development of overt diabetes in NOD mice (Hattori et al., 1986). The MHC region of the NOD mouse was shown to be of a recombinant haplotype. Serological analysis indicated that the class I histocompatibility molecules were of the H-2Kd and H-2Db alleles. However, the initial screening with a panel of antisera to class II antigens indicated that the NOD I-A molecule was unique and that NOD did not express an I-E molecule. Assays to test the ability of NOD splenocytes to stimulate class II specific T-cell hybrid clones as well as mixed lymphocyte cultures using NOD mice and various C57BL/10 congenic mouse lines confirmed that the NOD class II I-A molecules were unique to the NOD mouse line.




29
Analysis of mRNA from NOD mice showed no detectable E. mRNA, thus confirming the serologic data which indicated a lack of I-E molecule expression. Out breeding studies were then carried out with NOD and C3H (H-2k) mice to examine a possible linkage of diabetes to the inheritance of the unique NOD I-A molecule (Hattori et al., 1986). A difference in the restriction fragment length polymorphisms (RFLP) of the Ap genes from NOD and C3H mice using southern blot analysis enabled the investigators to distinguish which I-A genes were inherited by the progeny. None of (C3H x NOD)F1 female mice developed diabetes as compared with 80% of NOD females. Three of 117 (Fl x F1)F2 female progeny and 12 of 57 (Fl x NOD) backcross female mice developed diabetes. All of the diabetic mice that were analyzed by southern blot analysis for Ap RFLP were found to be homozygous for the NOD I-A molecule indicating a recessive nature for the NOD MHC-linked susceptibility gene. The authors suggested that there were probably two or more diabetes susceptibility genes in addition to the MHC-linked gene since only 15.6% of the (Fl x NOD) backcross females developed diabetes.
Sequence analysis of cDNA clones from NOD A. and Ap
genes has been carried out to examine the possible role of NOD I-A molecules as diabetes susceptibility genes (AchaOrbea and McDevitt, 1987). Sequences from the NOD A. cDNA clone were identical to that of the I-A d gene. However, sequences from the first external domain of the NOD Ap cDNA




30
clone were unique as compared to allelic sequences of I-Ap k,b,d,s,u,f,q. Only one region, corresponding to amino acids 56 and 57, was completely unique to the NOD I-Ap sequence. Acha-Orbea and McDevitt suggested that the unique sequence of the NOD I-Ap gene could be responsible for the association of diabetes susceptibility and the I-A region of the NOD mouse.
Prochazka and co-workers carried out breeding studies
between NOD and the related non-obese normal (NON) strain of mice to more closely examine the polygenic nature of diabetes susceptibility (Prochazka et al., 1987). None of the (NOD x NON)F1 progeny developed diabetes over a 12 month period. When 200 (F1 x NOD) backcross progeny were examined over a 12 month period, 19 (9.5%) developed overt diabetes. These data indicated to the authors that at least 3 unlinked autosomal recessive genes were involved in the development of autoimmune diabetes in the NOD mouse line. The investigators showed that all 19 of the diabetic backcross progeny were homozygous for the H-2Kd allele of the NOD haplotype indicating that one of the diabetogenic recessive genes was MHC-linked (chromosome 17). The investigators screened the backcross mice for a series of polymorphic genetic markers capable of distinguishing NOD from NON mice and showed that a second susceptibility gene mapped to chromosome 9 in the vicinity of the Thy-1 gene. However, the third susceptibility gene could not be mapped.




31
A more complex role of susceptibility genes with
respect to insulitis and the development of diabetes was inferred from breeding studies between NOD and B10 mice (Wicker et al, 1987). None of 200 (NOD x B10)F1 mice developed diabetes but 1 of 54 examined histologically developed insulitis and 17 had small lymphoid infiltrations of the exocrine pancreas associated with vascular elements. These data confirmed a recessive inheritance for diabetes but suggested that insulitis may be controlled in part by an incompletely dominant gene that appears to have low penetrance in (NOD x B10)F1 mice. None of the (NOD x B10)F2 mice examined developed diabetes but most showed pancreatic lesions including mild to severe insulitis on histological examination. These data suggested to the authors that pancreatic inflammation and insulitis may be controlled by a single dominant or incompletely dominant gene and that fewer genes may be required for the inflammatory responses than for the full development of overt diabetes. There was a 12.5% (4 of 61 females) incidence of overt diabetes in the (Fl x NOD)BC1 backcross progeny by 7 months of age and insulitis was observed in all the backcross mice examined. These data indicated that at least 3 functionally recessive genes or gene complexes control the development of diabetes in the NOD mice. The data discussed above on the genetic susceptibility for diabetes, suggests that inheritance of autoimmunity in NOD mice is of a complex polygenic nature.




32
Experimental protocols for treatment and prevention of diabetes in the NOD mouse. Several experimental systems for the treatment and prevention of diabetes in NOD mice have recently been examined that have not yet been discussed. Takayoshi and co-workers gave 17 female NOD mice weekly intraperitoneal injections of a streptococcal preparation (OK-432) from 4-24 weeks of age (Toyota et al., 1986b). The OK-432 streptococcal preparation is a potent immunomodulator capable of activating both macrophages and killer T-cells and increasing interleukin-2 (IL-2) production. None of the 17 female NOD mice treated with OK-432 developed diabetes, whereas 14 of 17 NOD mice given only saline developed diabetes by 24 weeks of age. Histological examination of pancreatic sections from OK-432 treated NOD mice showed 98% of the islets were intact or only mildly infiltrated, whereas pancreatic sections from saline treated mice exhibited 79% of total islets affected with severe insulitis. Immunological characterization revealed that the number of mononuclear cells in the spleen and the reactivity of NK cells was significantly increased in OK-432 treated NOD mice.
Oldstone examined the possibility that certain
lymphotropic viruses may be capable of inducing a selective suppression of an autoimmune response (Oldstone, 1988). A lymphotropic variant of lymphocytic choriomeningitis virus (LCMV) was used that results in the infection of the




33
T-helper subset. LCMV infection of NOD mice at birth or in adulthood prevented the development of overt diabetes. Adoptive transfer experiments were done to demonstrate the requirement for LCMV infection of lymphocytes to prevent diabetes. When non-diabetic NOD mice were sublethally irradiated (400 Rad) and injected intravenously with splenic lymphocytes from uninfected NOD mice, 80% developed diabetes within 30 days. In contrast, adoptive transfer of lymphocytes from LCMV infected NOD mice into sublethally irradiated NOD mice prevented the development of diabetes in all recipients. Lymphocytes from donors did not release infectious virus and no virus was detectable in the sera of recipients. These data indicated to Oldstone that the prevention of diabetes was caused by a virus-induced inactivation of potentially autoimmune reactive lymphocytes.
Experiments have also been designed to examine the
ability of preventing diabetes in NOD mice with large-dose treatment with nicotinamide (Yamada et al., 1982). Nicotinamide (0.5 mg/g body weight) was injected subcutaneously into 3 month old female NOD mice every day for 40 days. None of the nicotinamide treated NOD mice developed diabetes. In contrast, nearly 70% of untreated mice developed diabetes within the 40 day trial. Histological examination of pancreatic sections from treated mice showed normal numbers of islets and inflammatory cells were scanty and limited to the periphery of the islets.




34
Bone marrow transplantation procedures have also been carried out to prevent and/or treat insulitis and overt diabetes in NOD mice. Ikehara and co-workers lethally irradiated NOD mice (greater than 5 months of age) and reconstituted them with bone marrow cells from BALB/cnu/nu mice (Ikehara et al., 1985; Ikehara et al., 1987). The allogeneic bone marrow reconstituted NOD mice showed a normal glucose tolerance profile at 3 months posttransplantation. In contrast, 8 month old untreated NOD mice exhibited impaired glucose tolerance similar to that seen in IDDM. Furthermore, insulitis was not present in the pancreatic islets of the allogeneic bone marrow reconstituted NOD mice.
The allogeneic bone marrow transplantation was
sufficient to treat NOD mice which still had sufficient numbers of pancreatic -cells to produce enough insulin for the prevention glucose intolerance and hyperglycemia. However, NOD mice with overt diabetes required treatment with concomitant allogeneic bone marrow transplantation plus pancreas transplantation (Yasumizu et al, 1987). Four of seven treated NOD mice survived over 90 days posttransplantation, as compared to, control NOD mice which all died within 2 weeks without insulin treatment. Furthermore, the transplanted NOD mice showed an improvement in glycosuria and a normal glucose tolerance profile.




MATERIALS AND METHODS
Animals. Inbred strains of mice used in this study
were bred and maintained in the animal facility located in the Department of Pathology, University of Florida, Gainesville, Florida. They included AKR/J, B10.BR/cd, B10.BUA16, B10.Q, B0lO.RIII(71NS), BlO.S, B10.SAA48, C57BL/6, C57BL/10, CBA/J, DBA/2J, PL/J, non-obese diabetic (NOD) mice, (C57BL/10 x B10.BR/cd)F1, and (NOD X B10.BR/cd)Fl. In bone marrow reconstitution experiments, female mice 8-12 weeks of age were used for donor bone marrow cell populations. All host mice receiving donor bone marrow were female mice 7 weeks of age.
Antisera. Monoclonal antisera 15-3-1s (anti-H-2k), 31-3-4s (anti-H-2Dd), 28-13-3s (anti-H-2b), MK-D6 (anti-I-Ad), 31-3-4s (anti-H-2Kd), 10-2.16 (anti-I-Ak) and 20-10-5s (anti-Thy-1), obtained from American Type Culture Collection, Rockville, MD., were used to type for the histocompatibility antigens expressed on spleen cells of the chimeric animals. Alloantisera D.32 (anti-H-2Dk) and K.333 (anti-H-2Kb) provided by Dr. E.K. Wakeland (Department of Pathology, University of Florida), were also used.
35




36
Spleen cells preparations. Single-cell suspensions of newborn and adult splenocytes were prepared by gently pressing freshly explanted tissues through wire mesh screens and washed with phosphate-buffeted saline (PBS). For mixed leukocyte culture reactions, the red blood cells were lysed with 0.84% ammonium chloride. The leukocytes were then washed once and resuspended in PBS to the appropriate concentrations.
Bone marrow cells preparations. Adult bone marrow
cells were prepared by cutting the epiphyses from freshly explanted femurs and tibias and then flushing the contents from the lumens with PBS using a 27 gauge needle and syringe. The tissue was dispersed to a single cell suspension using gentle pipetting.
Bone marrow reconstitution of lethally irradiated
hosts. Protocols used in bone marrow reconstitutions of lethally gamma-irradiated adult mice included a novel system utilizing newborn spleen cells to mediate engraftment. Newborn spleen-associated null suppressor-inducer cells (Peeler et al., 1983) were used to mediate bone marrow engraftment and to prevent graft-versus-host disease. Newborn spleen cell cultures were established by culturing
6 x 106 spleen cells/35cm2 dish for 24 hours in 3 ml eagle's high amino acid (EHAA) medium supplemented to 0.5% normal mouse serum (Click et al., 1972). This culture period




37
permitted suppressor factors to be produced and secreted in the supernate. After 24 hrs, 60 x 106 freshly explanted adult donor bone marrow cells were added to each dish and co-cultured an additional 24 hrs. The cell mixtures were collected, washed and resuspended to a concentration of 100 x 106/ml.
In a few experiments, as an alternative to the bone marrow/newborn spleen cell mixtures, animals were reconstituted with hematopoietic stem cells from long-term bone marrow cell cultures. Primary bone marrow cultures were established and maintained according to Dexter and co-workers (Dexter et al., 1977). Briefly, 10 x 106 freshly explanted bone marrow cells were cultured in 10 mls Fisher's medium for leukemic cells of mice supplemented to 20% horse serum and 10-6 molar hydrocortisone within a 75 mm2 culture flask. Weekly, 50% of the medium was discarded and replaced with fresh medium and on the third week the cultures were reseeded with an additional 10 x 106 bone marrow cells. At time of reconstitution, non-adherent cells were collected, washed and resuspended to 100 x 106/ml. These dexter cultures did not contain mature lymphocytes (Dexter and Spooncer, 1978; Spooncer and Dexter, 1983) and cells derived from these cultures failed to induce GVH reaction. Furthermore, the cells from these cultures would not be capable of passively transferring mature lymphocytes capable of inducing autoimmunity in the transplanted hosts.




38
Each recipient mouse received 0.2 ml of reconstituting cells via the tail vein within 4 hrs after lethal irradiation (970-1050 Rads). Reconstituted animals were usually placed in laminar flow hoods and given water containing 10 mg/L polymyxin B plus 100 mg/L neomycin. In a few experiments, the mice were maintained under normal colony conditions and given acidified water to drink, but no differences in survival time/rates were observed. In any one experiment, all groups of animals were treated similarly. At various time points recipient mice were killed for histological and functional studies. Single cell suspensions of splenocytes were prepared as described above and tested for functional reactivities in mitogen stimulation and mixed leukocyte culture assays.
Lymphocyte cultures. Mixed leukocyte cultures (MLCs) were performed according to Peck and Bach (Peck and Bach, 1973) and consisted of 0.5 x 106 splenic leukocytes co-cultured in flat bottom plates with an equal number of gamma-irradiated (2500 Rad) stimulating splenic leukocytes in 0.2 ml EHAA. At various time points, as indicated in the figures, cells were pulsed with 1.0 gCi of tritiated thymidine, harvested 8 hours later and 3H-uptake measured using standard scintillation procedures. Mitogenic responses were measured in a similar manner following stimulation with either 8 Ag concanavalin A (Con A) or 25 Mg




39
lipopolysaccharide (LPS). Data were reported as the means of triplicate cultures minus background.
Histocompatibility typing for chimerism. Leukocytes
were serotyped using a two-step cytotoxicity assay. Spleen cells at 1.0 x 106 cells/ml were incubated with antisera against host specific or donor specific histocompatibility antigens for 45 min at 4C. The cells were then washed, resuspended in rabbit complement and incubated for 45 min at 37*C. All tests were carried out in RPMI 1640 medium. Cell viability was assessed by trypan blue dye exclusion.
Alternatively, when leukocytes differed only at class II loci, spleen cells were serotyped by immunofluorescence. The spleen cells were incubated with histocompatibility antigen specific antisera for 45 minutes at 4 C. The cells were washed and then incubated an additional 45 minutes with fluoreseinated rabbit-anti-mouse antisera at 37 C. The cells were then analyzed in a fluorescence activated cell sorter (FACS) to detect positive binding by specific antisera.
Histological examinations. Freshly removed organs and skin tissue were fixed in 10% formalin for 18-24 hr, followed by dehydration in 80% ethanol. The tissues were embedded in paraffin, sectioned and stained with hematoxylin-eosin (H&E) dye.




40
Measuring of blood glucose levels to determine the onset of overt diabetes. Blood glucose levels were determined using chemstrip bG test strips. A drop of blood drawn from the tail was placed on the test strip indicator, wiped off after one minute, and the glucose concentration was indicated by the color change in comparison with a calibrated scale.
Glucose tolerance testing. Mice which had fasted for 12 hours were injected intraperitoneally with a glucose solution (1.5 grams per kilogram weight). Blood glucose levels were measured at the time of injection and subsequently at 15, 30, 60, 120 and 180 minutes as described above.




THE USE OF NEONATAL SPLEEN CELLS TO INHIBIT
GVH DISEASE FOLLOWING SEMI-ALLOGENEIC AND
ALLOGENEIC BONE MARROW TRANSPLANTATION
Introduction
Bone marrow transplantation provides a practical means to manipulate the immune system. Successful engraftment with pluripotent stem cells from bone marrow provides a continuous source of differentiated leukocytes, platelets, erythrocytes, fixed macrophages of the reticuloendothelial system and osteoclasts (Desnick, 1987). Bone marrow can be readily ablated from the host animal by lethal irradiation (Johns, 1966) and then replaced by injection of donor bone marrow cells into the venous system. Replacement of the hematopoietic stem cell system provides a means to examine the effects of certain genes expressed by hematopoietic derived cells.
The normal procedure for bone marrow transplantation is to ablate the hematopoietic progenitor cells of the host with a lethal dose of irradiation (Gale, 1982; O'Reilly, 1983). Donor bone marrow is then injected into the irradiated host to reconstitute the hematopoietic progenitor cell population. However, freshly explanted donor bone marrow also contains mature T-lymphocytes capable of
41




42
inducing an immune response. Therefore, allogeneic bone marrow transplantation usually results in the induction of an immune response to allogeneic histocompatibility antigens resulting in graft-versus-host (GVH) disease and the eventual death of the animal if left untreated. A successful allogeneic bone marrow transplantation system must be capable of dealing with the normal occurrence of GVH disease.
Spleens of newborn mice less than 3-4 days of age contain a naturally-occurring population of monocytes capable of suppressing T-dependent and T-independent immune responses of third-party adult cells both in vitro and in vivo. The presence of naturally occurring newborn spleen associated suppressor cells capable of inhibiting T-dependent and T-independent immune responses of third-party adult cells in vitro is now well established (Skownon-Cendrzak and Ptak, 1976; Ptak and Skownon-Cendrzak, 1977; Argyris, 1978). Despite earlier claims that this suppression was mediated by T suppressor cells (Argyris, 1978; Murgita et al., 1978; Ross and Pilarski, 1981), recent reports (Rodriguez et al., 1979; Hooper and Murgita, 1980; Peeler et al., 1983; Piquet et al., 1981) have identified the important cell population as a monocyte. These suppressor, or suppressor-inducer, monocytes elicit their suppressor activity in part through the secretion of soluble materials capable of activating the T suppressor




43
limb of the immune response (Basset et al., 1977; Argyris, 1981; Peeler et al., 1983; Jadus and Peck, 1984). While no biological role for this phenomenon has yet been determined, it may be important in establishing maternal non-responsiveness to an allogeneic fetus (Peeler et al., 1983; Jadus and Peck, 1986) .
Recently, Jadus and Peck (Jadus and Peck, 1984) were
able to utilize these naturally occurring suppressor-inducer monocytes to inhibit graft-versus-host (GVH) disease in sublethally gamma-irradiated hosts reconstituted with semi-allogeneic or allogeneic splenic T lymphocytes. Host animals which failed to develop GVH disease were shown to contain a mixture of host and donor leukocytes for 3-4 weeks but only host phenotype cells by 8-10 weeks. These data suggested that in sublethally irradiated animals, the newborn monocytes suppressed donor cell reactivity long enough for the host system to recover from the effects of irradiation. The ability to suppress allogeneic reactivity in the hosts unfortunately proved to be restricted by two genetic elements (Jadus and Peck, 1984): first, newborn monocytes and adult T cells had to be compatible at a H-2-linked region, and second, the newborn cells had to express a strongly stimulating non-H-2 phenotype.
In the present section, I discuss the extension of this protocol to enhance the success of allogeneic bone marrow transplantation of lethally irradiated hosts.




44
This protocol is based on the use of naturally occurring suppressor cells from the spleens of 2-3 day old mice (see methods). I have utilized this population of newborn suppressor cells to prevent acute GVH disease in lethally irradiated adult hosts reconstituted with semi-allogeneic or even fully allogeneic bone marrow cells. The advantage of this bone marrow reconstitution protocol is that it does not require any intervention with immunosuppressive drugs and does not require the elimination of T-cell subsets from the donor bone marrow. Spleen cells, rich in mature T-lymphocytes, were added to the donor cell population to enhance the development of GVH disease.
The mice included B0lO.BR/cd, CBA/J, C57BL/6 and
(C57BL/10 x B0lO.BR/cd)Fl. B0lO.BR/cd mice were the source of donor bone marrow cells and were co-cultured with newborn spleen cells from 2 day old CBA/J mice since this combination was previously shown to induce optimal suppression of GVH disease in sublethally irradiated animals. The bone marrow recipients utilized were C57BL/6 and (C57BL/10 x B0lO.BR/cd)Fl mice since they were readily available and provided sufficient host environment for the induction of GVH disease. The breeding efficiency of CBA/J mice was very good making the newborn suppressor cell mediated bone marrow transplantation procedure an efficient means for carrying out large numbers of allogeneic bone marrow reconstitutions.




45
Pretreatment of the reconstituting cell populations
with this newborn suppressor cells reduced the incidence of GVH disease from 100% to 20% in semi-allogeneic and from 100 to 40% in allogeneic combinations. Long-term surviving reconstituted hosts proved immunologically unresponsive to both donor and host histocompatibility antigens, yet possessed a fully chimeric lymphoid system responsive to T and B cell mitogens as well as unrelated third-party alloantigens.
Results
Newborn spleen cells prevent GVH disease in hosts
engrafted with semi-allogeneic bone marrow. I first tested the ability to prevent GVH disease in semi-allogeneic bone marrow reconstituted hosts with the newborn suppressor cells since there would only be a 50% relative dosage of allogeneic histocompatibility antigens. In the first set of experiments, lethally gamma-irradiated (C57BL/10 x BIO.BR/cd)F1 hosts were reconstituted with BIO.BR/cd bone marrow cells. As shown in Table 1, all irradiation control mice injected with PBS, but no reconstituting hematopoietic cells, died within 15 days. Likewise, control animals injected with BlO.BR/cd bone marrow plus spleen cells cultured 24 hours alone died within 16 days. However, 80% of hosts reconstituted with B0O.BR/cd bone marrow plus spleen cells co-cultured 24 hrs with CBA/J newborn suppressor cells were long-term survivors, living >60 days




46
Table 1. Suppression of lethal GVH disease by newborn
spleen cells in lethally irradiated (C57BL/10 x
BlO0.BR/cd)Fl hosts reconstituted with semiallogeneic B10.BR/cd bone marrow cells.
Reconstituting cell Survival times % Long-term
population (days) survivors
None 9, 10, 10 0 %
BlO0.BR/cd adult 12, 14, 15 0 % bone marrow cells
and spleen cells 8, 8, 8, 8, 8, 8, 13 16 0%
Adult B0lO.BR/cd 8, >60, >120, bone marrow cells >170, >190 80 % and spleen cells
co-cultured 24 hrs 8, 8, >35, >55, with CBA/J newborn >103, >103, >369, spleen cells >375, >375, 535, 556 80 %
S> indicates the animal was sacrificed for
characterization.




47
(Figure 1). Long-term surviving hosts have been followed as long as 18 months after reconstitution before being sacrificed for functional studies. These data show that newborn suppressor splenocytes are capable of mediating successful development of semi-allogeneic bone marrow engrafted hosts.
Newborn spleen cells prevent GVH disease in hosts engrafted with allogeneic bone marrow. The ability to suppress GVH disease in fully allogeneic bone marrow reconstituted hosts was then examined. In a second set of experiments, lethally gamma-irradiated C57BL/6 hosts were reconstituted with BlO.BR/cd bone marrow cells. As shown in Table 2, all irradiation control animals, injected with PBS but no reconstituting hematopoietic cells, died within 16 days. Similarly, animals injected with untreated B0O.BR/cd bone marrow plus spleen cells died within 19 days, with one animal dying at 27 days. Under the experimental and environmental conditions used 90-100% of C57BL/6 mice reconstituted with syngeneic bone marrow plus spleen cells survived long-term, whereas 55-60% of hosts reconstituted with fully allogeneic BlO.BR/cd bone marrow plus spleen cells following co-culture with CBA/J newborn suppressor cells were long term survivors (Figure 1 and Table 1). These data indicate that the newborn spleen-associated suppressor cells are also capable of suppressing GVH disease in fully allogeneic engrafted hosts.




Figure 1. Cumulative survival curves of reconstituted and
non-reconstituted lethally irradiated hosts.
a) Lethally irradiated (C57BL/10 x B0lO.BR/cd)F1 and C57BL/6 mice were reconstituted with syngeneic bone marrow (filled triangles) or injected with PBS but not reconstituted (open circles).
b) Lethally irradiated (C57BL/10 x B0O.BR/cd)F1 mice were reconstituted with untreated (open triangles) or newborn spleen cell pretreated (filled circles) semiallogeneic BlO0.BR/cd bone marrow plus spleen cells.
c) Lethally irradiated C57BL/6 mice were reconstituted with untreated (open triangles) or newborn spleen cell pretreated (closed circles) fully allogeneic B0lO.BR/cd bone marrow plus spleen cells.




49
SRmVAL. RATES A. SYGJJ RECONSTflhUTION
TIME (DAYS)
B. SYNEMI-ALOGENEIC RECONSTIONITUTION
10 + newborn splic roenocyteo 100
4 70
0 I I I I I 70 n-omfg 80
10
0 10 0 o 40 so 80 70 0 0 100 110 12o TIME (DAYS) C. SE-ALLOGENEIC RECONSTITUTION loo
eao
0 -+ newborn splenic monocyte 40
41
=o newrn splnic ffo ota
0 10 20 0o 40 so50 so 70 0 0 1o00 110 10o TIME (DAYS)
C. ALLOGENEIC RECONSTITUTION
100
70
so
so + newborn spleic monocytee
20 newborn splenic monocytes 10
o 10 s0 80 Lo IL0 Lo 70 '0 90 00 110 120 TIME (DAYS)




50
Table 2. Suppression of lethal GVH disease by newborn
spleen cells in lethally irradiated C57BL/6 hosts reconstituted with semi-allogeneic B10.BR/cd bone
marrow cells.
Reconstituting cell Survival times % Long-term
population (days) survivors
None 3, 3, 4, 10, 10, 11, 11, 11, 12,
12, 13, 13, 13, 14 0%
5, 6, 11,
12, 13, 16 0%
BlO0.BR/cd adult 9, 11, 11 0 % bone marrow cells and spleen cells 7, 7, 9, 9 0 %
7, 11, 15, 19, 27 0 %
Adult B0lO.BR/cd 8, 9, 11, >21 bone marrow cells >77, >87 50 % and spleen cells
co-cultured 24 hrs 7, 9, 9, 166, with CBA/J newborn >323, >375 50 % spleen cells
17, 20, >15, >33, 71% >55, 195, 350
> indicates the animal was sacrificed for
characterization.




51
Histologcical examination of experimental mice.
Physical examinations and histological studies were carried out to confirm that death of experimental animals following reconstitution resulted from GVH disease and not irradiation, per se. As depicted in figure 2, experimental mice which failed to thrive following bone marrow reconstitution showed signs of severe GVH disease: emaciation, exaggerated hunched appearance, patches of hair loss, sloughing of the tails, development of diarrhea, and hypothermia.
Photomicrographs of histology sections from skin
biopsies as well as liver and spleen sections are presented in Figures 3-5. Irradiation control mice showed fairly normal skin biopsies (Figure 3a) with the exception of noticeable loss of fat cells between connective tissue of the dermis and muscle. Similarly, liver histology appeared normal (Figure 3b). However, spleens from irradiation control mice showed severe generalized cytopenia in both white and red pulp, a collapsed architecture disrupting the white and red pulp areas, hemosiderin, and markedly visible connective tissue septae (Figure 3c & 3d). Experimental mice which where successfully reconstituted with bone marrow pretreated with newborn suppressor cells displayed normal skin (Figure 4a), liver (Figure 4b), and spleens (Figure 4c & 4d). Occasionally, slight leukocytic cell infiltration could be seen around portal tracts of the livers.




Figure 2. Physical appearance of mice undergoing lethal
GVH disease.
Mice with GVH disease appear emaciated and maintain an exaggerated hunched position. Some mice also exhibit patches of hair loss and sloughing of tails.




53 GH C




Figure 3. Photomicrographs of H&E stained histology
sections from irradiation control mice which
were lethally irradiated and injected with PBS
but not reconstituted with hematopoietic
progenitor cells.
a) Skin biopsies revealed fairly normal tissue morphology with the exception of a paucity of normal skin appendages and a loss of fat cells between the dermis connective tissue and muscle. (250 X magnification).
b) Livers from irradiation control mice exhibited normal tissue morphology. (100 X magnification).
c) Spleens from irradiation control mice exhibited severe cytopenia throughout the white and red pulp regions. (250 X magnification).
d) The architecture of the spleens had a collapsed appearance and markedly visible connective tissue septae. (100 X magnification).




55
A
1
At *1




,-J.
'1 CD L~)
C.)
0 '-I
I-i.
CD
CM) Cal




Figure 4. Photomicrographs of H&E stained histology
sections from experimental mice which where successfully reconstituted with bone marrow
plus spleen cells pretreated with newborn
suppressor cells.
a) Histologic sections of skin biopsies exhibited normal tissue morphology and skin appendages. (250 X magnification).
b) Liver sections exhibited normal tissue morphology. (100 X magnification).
c) Spleens sections from successfully reconstituted mice had re-established normal tissue morphology. (250 X magnification).
d) The architecture of spleens from successfully reconstituted mice appeared normal with distinct red and white pulp areas.




58
A
B
4 4e
- 4




OO
44
0
I
I
*-.




Figure 5. Photomicrographs of H&E stained histology
sections from experimental mice reconstituted with untreated bone marrow plus spleen cells.
a) Skin biopsies from lethally irradiate mice reconstituted with untreated bone marrow plus spleen cells exhibited an atrophic epidermal layer with a hyperkeratotic surface and focal infiltrates within the epithelium. (250 X magnification).
b) Liver sections exhibited many inflammatory cell infiltrates within the portal tracts. (250 X magnification).
c) Fibrosis associated with inflammatory cell infiltrates and necrosis at the periphery of fibrotic regions was observed in a few liver sections. (250 X magnification)
d) Spleen sections exhibited expanded and delineated white pulp regions as well as lymphocyte depletion. (250 X magnification).
e) The architecture of the spleens appeared collapsed with no discernable red and white pulp regions. (100 X magnification).




61
A




62
B
C
Figure 5--continued.




:43
40
*It
6 gO




64
In contrast, skin biopsies from mice reconstituted with untreated bone marrow plus spleen cells revealed an atrophic epidermal layer and a hyperkeratotic surface as well as a loss of skin appendages (Figure 5a). Upon closer examination, there were focal lymphocyte infiltrates within the surface epithelium and there were increased mononuclear cell numbers in and around blood vessels and the remains of skin appendages. The livers (Figure 5b & 5c) contained marked inflammatory lesions: inflammatory cell infiltrates were evident within the portal tracts whereas focal cell infiltrates were present in the sinusoids of the lobules and associated with patchy hepatocyte necrosis. A few livers showed fibrosis with inflammation and necrosis at the periphery of fibrotic regions. Spleens from these mice (Figure 5d & 5e) had expanded and delineated white pulp regions which contained cells with morphologic characteristics of blast transformation. At the time of severe cachexia, the white pulp showed lymphocyte depletion. The red pulp had collapsed, and Russell bodies, septae and hemosiderin were visible. The results from histological analysis indicated that lethally irradiated mice transplanted with untreated allogeneic bone marrow plus spleen cells had developed GVH disease. Furthermore, pretreatment with newborn spleen cells appeared to suppress the development of GVH disease resulting in successful engraftment and reconstitution of lethally irradiated hosts.




65
Full chimerism in long-term survivors. Serological
analysis of spleen cells from long-term surviving hosts was carried out to confirm that they were bone marrow chimeras. Specific antisera against donor and host histocompatibility antigens were used in complement mediated cytotoxicity assays to determine the origin of the hematopoietic cells within the reconstituted hosts. Data presented in Table 3 show that the surviving semi-allogenic and allogeneic bone marrow reconstituted hosts were fully chimeric. The leukocyte populations residing in the spleens of (C57BL/10 x BlO0.BR/cd)F1 hosts reconstituted with B0lO.BR/cd, as well as the C57BL/6 hosts reconstituted with B0lO.BR/cd, serotyped positive for H-2k but negative for H-2b indicating that the hematopoietic cells were of donor origin.
Functional studies on splenocytes from long-term survivors. Semi-allogeneic and allogeneic bone marrow reconstituted hosts were killed at various times and their splenocytes assayed for immune reactivity to mitogenic stimulation and MLC reactivity (Figures 6 and 7). Cells from (C57BL/10 x B0lO.BR/cd)F1 recipients reconstituted with BlO0.BR/cd cells responded strongly to Con A and LPS stimulation, thus demonstrating immunocompetence of the T-cell and B-cell compartments at 60 days (Fig 6a). In addition, they responded to third-party alloantigens, e.g., DBA/2J (H-2d, Mlsa) and B0O.SAA48 (H-2w3), while failing to




66
Table 3. Serotyping of splenocytes from long-term surviving
reconstituted hosts.
% Cytotoxicity Experimental
group Normal
anti- H-2k anti- H-2b mouse serum
BlO0.BR/cd 85 % 21% 25 %
C57BL/6 16 % 99%
C57BL/6 reconstituted 15 % 80 % 15 % with syngeneic C57BL/6 bone marrow
(C57BL/10 x B0lO.BR/cd)Fl 96 % 13 % reconstituted with adult BlO0.BR/cd bone marrow and spleen cells co-cultured with CBA/J newborn spleen cells
C57BL/6 reconstituted 80 % 15 % 15 % with adult B0lO.BR/cd bone marrow and spleen cells co-cultured with CBA/J newborn spleen cells




Figure 6. In vitro immune responses of spleen cells from
lethally irradiated (C57BL/10 x B10.BR)F1 mice
reconstituted with semiallogeneic B0lO.BR/cd
bone marrow.
a) Spleen cell responses following mitogenic stimulation with 25 Ags/ml LPS (open circles) or 8 ggs/ml Con A (filled squares).
b) Mixed lymphocyte culture responses of spleen cells against gamma-irradiated B10.SAA48 (filled squares), DBA/2J (filled triangles), CBA/J (open triangles), C57BL/10 (filled circles), and B0lO.BR/cd (open circles) spleen cells.




68
#(C57BL/IO Sn xBIO.BR/cd)FI [BIO. BR/cd] A. Mitogenic responses B. MLC responses
150- 300 BIO.SAA48
a.
100- ConA 20C
.9 DBA/2J
o
C.
0
1 0CBA/J
.E 50- 10E
C57BL/O0
rLPS BI.BR/cd
0 0 ,, 024487296 24487296
Hours of culture




Figure 7. In vitro immune responses of spleen cells from
lethally irradiated C57BL/6 mice reconstituted
with fully allogeneic B0lO.BR/cd bone marrow.
a) Spleen cell responses following stimulation with 25 Ags/ml LPS (filled circles, unbroken line) or in unstimulated culture (closed circle, dashed line).
b) Mixed lymphocyte culture responses of spleen cells against gamma-irradiated DBA/2J (filled triangles), AKR/J (open triangles), B10.Q (filled squares), C57BL/6J (filled circles), and CBA/H (open circles) spleen cells.




70
C57BL/6J [BIO. BR/cd]
A. Mitogenic responses B. MLC responses
60- 60
DBA/2J
10
o
S40- LPS 40
C
. AKR/J
o
0
0.
0
Cr
20 20
.CBA/H
medium
Flo cl 5C57BL/6
0 J _ 0 J ___ d
II I I I I I I I I I I I 24 48 7296 24 48 72 96 120
Hours of culture




71
respond to B0O.BR/cd and C57BL/10 stimulator cells (Fig 6b). In addition, these cells responded against B0O.RIII(71NS) (H-2r), B0O.S (H-2s) and B0O.BUAl6 (H-2w22) stimulators (data not shown). These data indicate that the cells residing in the (C57BL/10 x B0O.BR/cd)F1 chimeric hosts were tolerant to both donor and recipient histocompatibility antigens but not to third party alloantigens. This substantiates the immunocompetence of the T-cell compartment of these hosts.
Similarly, C57BL/6 hosts reconstituted with BlO.BR/cd
bone marrow cells were assayed for mitogenic stimulation and MLC reactivity. Figure 7 shows data for mice sacrificed on day 90. The cells proved unresponsive to CBA/H and C57BL/6 cells indicating a tolerance to both donor and recipient haplotypes. In contrast, the cells were reactive to both LPS and histocompatibility alloantigens, e.g., expressed by BlO0.Q (H-2q), AKR/J (H-2k, Mlsd) and DBA/2J (H-2d, Mlsa). Thus, these reconstituted hosts also exhibited immunocompetence of the T and B cell compartments.
Discussion
The results showed that lethally-irradiated hosts could be successfully reconstituted with semi-allogeneic (80%-90% long-term survival) or allogeneic (55%-60% long-term survival) bone marrow if the donor cells were first co-cultured 24 hours in the presence of newborn monocytes.




72
Routine histological examination (Figure 8) of mice which had died or had been killed for functional testing revealed several points: 1) lethal irradiation caused complete destruction of the lymphoid compartment of the host, 2) hosts which had been reconstituted with histoincompatible bone marrow showed successful engraftment but subsequent development of severe organ and tissue lesions characteristic of classic graft-versus-host disease, and 3) hosts which had been successfully reconstituted with histoincompatible bone marrow pretreated with newborn suppressor cells showed normal lymphoid and tissue histology. Thus, pretreatment of histoincompatible donor bone marrow cells with newborn spleen-associated suppressor cells prevented the development of lethal GVH disease observed in host animals engrafted with untreated bone marrow cells.
Functional studies using spleen cells from
reconstituted chimeric hosts surviving >60 days showed that the cells were capable of proliferating in response to both T and B cell mitogens as well as histocompatibility alloantigens on third party cells. In contrast, these chimeric mice proved tolerant in their T cell responses to both donor and host histocompatibility antigens. Serotyping confirmed that the hosts were fully chimeric in that all the splenocytes expressed the donor histocompatibility antigens. These results are in marked contrast to those reported by




Figure 8. Composite of photomicrographs for comparison of
skin biopsies, liver and spleen sections from
irradiation control mice (a-c), long-term
surviving chimeras reconstituted with newborn
suppressor cell treated allogeneic bone marrow (d-f), and allogeneic bone marrow-reconstituted
animals undergoing GVH disease (g-i).




74
SKIN LIVER SPLEEN
hi
Mag h i
F
2:2 S
A




75
Jadus and Peck (Jadus and Peck, 1984). In their study, cells from sublethally-irradiated, T cell engrafted hosts examined at day 60 post-engraftment remained tolerant to host cells but responded to donor and third- party cells. Furthermore, at day 60 all of the splenocytes serotyped as host cells.
One interesting and unexpected result was the response of the chimeric (C57BL/10 x BlO.BR/cd)F1 hosts toward adult CBA/J cells. Although syngeneic with the newborn suppressor-inducer population, CBA/J stimulated cells of the reconstituted hosts. This lack of tolerance may result from an insufficient quantity of CBA/J alloantigens present during the development of the chimeric immune system, or alternatively, tolerance toward non-MHC alloantigens, e.g., Mls, may never develop in such a protocol. This point needs further examination.
Traditional protocols for bone marrow reconstitution in which bone marrow is "cleansed" the donor marrow of mature T-cells, primarily using anti-lymphocyte or anti-T cell antisera. Construction of semi-allogeneic parent F1 hybrid chimeras through the pretreatment of donor bone marrow and spleen cells with anti-theta antiserum plus complement (Tyan, 1973; Sprent et al., 1975; Vallera et al., 1981; Thiefelder et al., 1983) has resulted in about 80-100% of the hosts surviving greater than 100 days. Such




76
chimeric animals were totally populated with leukocytes of the donor strain and proved capable of either rejecting third-party skin allografts while remaining tolerant to skin grafts syngeneic with the donor mice or mounting an MLC response against third-party strains while showing no proliferative response against donor cells. Von Boehmer and co-workers (von Boehmer et al., 1975) obtained similar success rates in the production of tetra-parental chimeric mice.
As would be predicted, construction of fully allogeneic chimeric mice using anti-lymphocyte antiserum has not proven as successful as in semi-allogeneic combinations: Theifelder and co-workers (Thiefelder et al., 1983) found only a 55% success rate while Vallera and co-workers (Vallera et al., 1981) had up to an 80% success rate. Successfully reconstituted hosts proved tolerant toward skin grafts of the donor while rejecting third-party unrelated grafts.
In conclusion, bone marrow reconstitutions using either protocols in which the donor bone marrow is cleansed or pretreated with newborn suppressor-inducing cells appear similar. The major difference is that in the experimental protocol described herein, mature T cells have purposely been left in the reconstituting population, and have even been added without detrimental results. Thus, use of the




77
newborn suppressor-inducer population in bone marrow reconstitution may provide us with a unique and natural method to control GVH disease even in allogeneic transplantations.




PREVENTION AND TRANSFER OF AUTOIMMUNE DIABETES
FOLLOWING ALLOGENEIC BONE MARROW TRANSPLANTATION
Introduction
Current evidence supports an autoimmune etiology for
IDDM (Cahill and McDevitt, 1981; Eisenbarth, 1986) with the pancreatic beta (p) cell as the target for humoral and/or cell mediated responses. Since genes of the major histocompatibility complex (MHC) have been associated with susceptibility to IDDM (Platz et al., 1981; Cudworth and Wolf, 1982; Hattori et al., 1986) and have been shown to strongly influence immune responsiveness (Klein et al., 1981; Klein et al., 1983; Kaufman et al., 1984), it is possible that specific MHC molecules permit the development of aberrant immunity to autoantigens given the proper environmental conditions.
NOD mice were utilized to more specifically examine the role of the immune system in the pathogenesis of autoimmune diabetes. NOD mice spontaneously develop insulitis (Figure 9) and a progressive destruction of the insulin producing pcells within the islets of Langerhans leading to insulin dependent diabetes (Makino et al., 1980; Fujino-Kurihara et al., 1985). As depicted in Figure 10, there is a progressive lymphocytic infiltration starting with
78




79
Figure 9. Photomicrograph of an H&E stained pancreatic
histology section from an untreated NOD female
mouse. Note the presence of insulitis within the
pancreatic islet characteristic of autoimmune
diabetes (400 X magnification).




Figure 10. Photomicrographs of H&E stained pancreata from
untreated NOD mice depicting the progression
of lymphocytic infiltration (400 X magnification).
a) Peri-islet lymphocytic infiltration abutting several clustered islets.
b) Penetration of the pancreatic islet connective tissue sheath resulting in moderate insulitis.
c) Increasing numbers of infiltrating cells resulting in the progression to severe insulitis.
d) Advanced stage of insulitis is characterized by the destruction of the majority of the pancreatic islet cells and loss of normal islet architecture.




S8*




82
C
D
Figure 10--continued.




83
peri-islet infiltration, which advances to moderate insulitis as the infiltrating cells penetrate the pancreatic islet, and then to severe insulitis which results in the destruction of pancreatic islet.
Several experimental protocols have recently been
developed for the treatment and prevention of diabetes in NOD mice involving modulation of the immune system. Nonspecific immunosuppressive regimens such as sublethal irradiation (Harada and Makino, 1986), neonatal thymectomy (Hanafusa et al., 1986), and injection of anti-thymocyte antisera to deplete T-lymphocytes (Hanafusa et al., 1986) have all been shown to markedly reduce the incidence of diabetes in NOD mice.
Bone marrow transplantation procedures have also been utilized to manipulate the immune system of NOD mice. Ikehara and co-workers carried out allogeneic bone marrow transplantations to prevent and/or treat insulitis and overt diabetes in NOD mice (Ikehara et al., 1985; Ikehara et al., 1987). In these studies, 5-6 month old NOD mice were lethally irradiated and reconstituted with bone marrow cells from BALB/cnu/nu mice. The allogeneic bone marrow reconstituted NOD mice showed normal glucose tolerance profiles at 3 months post-transplantation. Furthermore, histological examination showed that insulitis was not present in pancreatic islets of the allogeneic bone marrow reconstituted NOD mice. In contrast, 8 month old untreated




84
NOD mice exhibited impaired glucose tolerance similar to that seen in IDDM. Thus, allogeneic bone marrow transplantation was sufficient to treat NOD mice which possessed sufficient numbers of pancreatic a-cells to produce enough insulin for the prevention of glucose intolerance and hyperglycemia. However, NOD mice with overt diabetes required treatment with allogeneic bone marrow transplantation concomitant with pancreas transplantation (Yasumizu et al, 1987).
In the present study, the newborn suppressor cell
mediated bone marrow transplantation protocol discussed in the previous section was utilized to do reciprocal allogeneic bone marrow transplantations between diabetes susceptible NOD mice and non-susceptible strains of mice. The reconstitution of NOD mice with allogeneic bone marrow from diabetes non-susceptible strains of mice provided experimental animals with non-diabetogenic immune systems, within a diabetes prone host environment. Elements within the diabetes prone host environment include: the expression of potential MHC and non-MHC-linked diabetes susceptibility genes, appropriate auto-antigens capable of inducing an autoimmune response to pancreatic P-cells, and a thymus that does not appear to delete or suppress the pertinent autoreactive clones during thymocyte development. Reconstitution of diabetes non-susceptible strains of mice




85
with bone marrow from NOD mice provided experimental animals with diabetogenic immune systems in a normally nonsusceptible host environments.
Bone marrow transplantation was carried out on female NOD mice at 7 weeks of age, prior to a large amount of pancreatic P-cell destruction, to determine if it would avert the progression of insulitis and prevent development of autoimmune diabetes. Likewise, the diabetes nonsusceptible strains of mice were reconstituted with NOD bone marrow at 7 weeks of age to determine if the diabetogenic immune system was sufficient for the development of autoimmunity between 2 months and 7 months post-transplantation. I have compared the newborn suppressor cell mediated protocol with one using a long-term bone marrow culture system initially developed by Dexter and co-workers (Dexter et al., 1977) as a source of hematopoietic stem cells for bone marrow reconstitutions. The long-term bone marrow culture conditions allow for the proliferation of hematopoietic progenitor cells to be maintained in vitro for up to 6 months. It takes approximately 6 weeks to establish a stable primary bone marrow culture during which time the cultured cells lose their immunoreactivity (Dexter and Spooncer 1978). The T-lymphocytes, which are responsible for the initiation of alloreactivity, do not survive in the long-term cultures. Therefore, it is possible to reconstitute a lethally




86
irradiated mouse with allogeneic bone marrow culture cells without inducing lethal GVH disease (Spooncer and Dexter, 1983). Long-term bone marrow cultures, devoid of mature Tlymphocytes allowed us to control for the possibility that the transfer of autoimmunity to non-susceptible hosts was dependent on direct adoptive transfer of autoimmunity. Furthermore the use of an alternate source of hematopoietic progenitor cells allowed for us to control for the possible effects of newborn suppressor cells on autoimmunity.
Results
Normal onset of IDD in the NOD mouse following
syngeneic bone marrow reconstitution. Female NOD mice, 7 weeks of age and prior to the onset of severe insulitis, were lethally irradiated and reconstituted with either freshly explanted NOD bone marrow cells that had been cocultured with newborn spleen cells or syngeneic NOD hematopoietic progenitor cells from long-term bone marrow cultures. Lethal irradiation was confirmed as all of the animals not reconstituted died within 15 days of irradiation. Of 13 NOD mice reconstituted with syngeneic bone marrow, all survived long-term ( > 2 months).
Starting 4 weeks post-transplantation, blood glucose levels were routinely screened using test-strips. Blood glucose levels greater than 200 mgs/dl, for 3 consecutive days, were considered to be diagnostic for overt diabetes. As




87
shown in Figure 11, all 13 of the syngeneic reconstituted NOD mice developed overt diabetes within 26 weeks posttransplantation (blood glucose levels of 400-800 mgs/dl).
Nine of the diabetic animals from this experimental group were sacrificed and their pancreata were examined histologically for the presence of insulitis. All nine of the syngeneic reconstituted NOD mice exhibited insulitis (Figure 12). As shown in Figure 13, there was a progression of lymphocytic infiltration from peri-islet infiltration, to moderate insulitis, and finally to severe insulitis resulting in the destruction of the pancreatic islets. The normal development of insulitis and progression to overt diabetes in the syngeneic reconstituted NOD mice indicated that the irradiation and bone marrow infusion procedures did not alter the pathogenesis of autoimmune diabetes. Furthermore, the same results were observed when NOD mice were reconstituted with newborn suppressor cell treated bone marrow or cultured bone marrow indicating that newborn spleen cells did not suppress the development of diabetes.
Prevention of IDD in the NOD mouse with allogeneic bone marrow reconstitution. B0lO.BR/cd mice were chosen as bone marrow donors since they are fully allogeneic to NOD mice. Seven week old female NOD mice were lethally irradiated and reconstituted with BO0.BR/cd bone marrow cells which had been co-cultured with newborn spleen cells to enhance engraftment and to suppress the development of GVH disease.




88
Reconstitution: NOD ~ NOD /
BM reconstitution
(7 wks of age)
TTIME (wks post-transplantation)
0 4 8 12 16 20 24 28
i I I I I I I I
Tested for elevated BGLs
Time mice were killed: 11 t t
* Detection of insulitis ++ 4+ 4+ + +
Onset of Diabetes: A A AA A A A A A AA
AA
Figure 11. Lethally irradiated NOD mice reconstituted with syngeneic hematopoietic stem cells at 7 weeks of age were routinely screened for elevated blood glucose levels (horizontal arrow) to detect overt diabetes. Individual mice were sacrificed at various time points during the experiment (vertical arrows) and examined histologically for the presence of insulitis (+ symbols represent individual animals in which insulitis was detected in pancreatic sections on H&E stained slides). The black triangles indicate the time point when mice developed overt diabetes.




89
Figure 12. A photomicrograph showing a pancreatic islet
from a lethally irradiated NOD mouse reconstituted with syngeneic NOD bone marrow. Note the
cellular infiltrate characteristic of insulitis
(H&E stained sections, 400 X magnification).




Figure 13. Photomicrographs of H&E stained pancreata from
NOD mice reconstituted with syngeneic NOD bone
marrow mice depicting the progression of
lymphocytic infiltration.
a) Peri-islet lymphocytic infiltration abutting a pancreatic islet (400 X magnification).
b) Penetration of the pancreatic islet connective tissue sheath resulting in moderate insulitis (250 X magnification.
c) Increasing numbers of infiltrating cells resulting in the progression to severe insulitis (250 X magnification).
d) Advanced stage of insulitis characterized by the infiltrating cell almost completely covering the pancreatic islet (400 X magnification).




91
A
B




92
**
C
D
Figure 13--continued.




93
Spleen cells from adult B10.BR/cd mice were not added to the reconstituting cell population since I wanted to increase the percentage long-term surviving reconstituted hosts. Of 14 mice reconstituted, 11 survived long-term (>2 months).
As shown in Figure 14, all long-term survivors
exhibited normal blood glucose levels (80-120 mg/dl) up to 6 months post-transplantation in a weekly screening with teststrips. At various times after 3 months posttransplantation, individual animals were killed for functional studies and histological examination. None of the pancreatic islets examined histologically showed the presence of insulitis (Figure 15). The use of young NOD mice as hosts for the reconstitution allowed us to avert the progression of insulitis before many pancreatic #-cells were destroyed. Thus, bone marrow reconstitution of the NOD mouse using cells from the IDD non-susceptible B10.BR/cd mouse was sufficient to prevent the onset of insulitis, hyperglycemia and diabetes.
Development of insulitis and diabetes in C57BL/6 and B10.BR/cd mice reconstituted with the NOD hematopoietic system. Diabetes non-susceptible C57BL/6 and B10.BR/cd mice were reconstituted with allogeneic bone marrow from NOD mice. Seven week old female C57BL/6 or B10.BR/cd mice were lethally irradiated and reconstituted with NOD bone marrow, which was previously co-cultured with newborn spleen cells. Animals which were irradiated but not reconstituted with




94
Reconstitution: B 1 O.BR/cd -- NOD/
BM reconstitution
(7 wks old)
VTIME (wks post-transplantation)
0 4 8 12 16 20 24 28
SI I I I I
Tested for elevated BGLs
Time mice were killed: T
Detection of Insulltis - - -
Onset of Diabetes: . . . . NONE
Figure 14. Lethally irradiated NOD mice reconstituted with allogeneic hematopoietic stem cells from B10.BR/cd mice at 7 weeks of age were routinely screened for elevated blood glucose levels (horizontal arrow) to detect overt diabetes. Mice were sacrificed at various time points during the experiment (vertical arrows) and examined histologically for the presence of insulitis (- symbols mean insulitis was not detected in pancreatic sections on H & E stained slides), (* symbols indicate no histology). Note that none of the animals developed diabetes.




95
~A
Figure 15. Photomicrographs of H&E stained pancreatic
islets from lethally irradiated NOD mice
reconstituted with allogeneic B10.BR/cd bone
marrow.
a) Examination of histological sections revealed healthy appearing pancreatic islets with no insulitis (400 X magnification).
b) A 1000 X magnification of a second healthy appearing pancreatic islet with no signs of insulitis.




Full Text

PAGE 1

PREVENTION AND TRANSFER OF AUTOIMMUNE DIABETES FOLLOWING ALLOGENEIC BONE MARROW TRANSPLANTATION BY DRAKE MAURICE LA FACE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1988

PAGE 2

ACKNOWLEDGMENTS I thank my mentor, Dr. Ammon Peck for his guidance and support during my graduate studies and the members of my supervisory committee, Ors. A. Kimura, N. Maclaren, S. Normann, w. Winter and J. zucali for their guidance in my dissertation Special thanks go to Ms. Marlene Wiley and Ms. Gladys Thompson for help in the preparation of the histological section. I also wish to thank Ors. w. Winter, R. Hacket, R. Braylon and B. Croker for their advice in the interpretation of the pathology. ii

PAGE 3

ACKNOWLEDGMENTS ABSTRACT INTRODUCTION TABLE OF CONTENTS Insulin Dependent Diabetes Mellitus in Man Non-Obese Diabetic (NOD) Mouse: An Animal Model for IDDM . MATERIALS AND METHODS THE USE OF NEONATAL SPLEEN CELLS TO INHIBIT GVH DISEASE FOLLOWING SEMI-ALLOGENEIC AND ALLOGENEIC BONE MARROW TRANSPLANTATION. Introduction Results .. Discussion PREVENTION AND TRANSFER OF AUTOIMMUNE DIABETES FOLLOWING ALLOGENEIC BONE MARROW TRANSPLANTATION. Introduction Results ... Discussion SUMMARY AND FUTURE PERSPECTIVES REFERENCES BIOGRAPHICAL SKETCH iii ii iv 1 1 13 35 41 41 45 71 78 78 86 124 130 133 146

PAGE 4

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PREVENTION AND TRANSFER OF AUTOIMMUNE DIABETES USING ALLOGENEIC BONE MARROW TRANSPLANTATION By Drake Maurice La Face August 1988 Chairman: Dr. Ammon B. Peck Major Department: Pathology Current evidence supports an autoimmune pathogenesis for type I insulin dependent diabetes mellitus (IDDM) in which the pancreatic beta cell is the target for humeral and/or cell mediated immune responses. Diabetes in the nonobese diabetic mouse (NOD) provides an excellent animal model as it shares many pathological features with human type I diabetes. I have carried out reciprocal allogeneic bone marrow reconstitutions between diabetes susceptible NOD and diabetes non-susceptible strains of mice to further delineate the role of the endogenous immune response and host environment in the development of diabetes. I report here that NOD mice reconstituted with a hematopoietic cell system from diabetes non-susceptible mice remained totally free of insulitis and diabetes, whereas NOD mice reconstituted with a syngeneic hematopoietic system showed a 100% prevalence of both insulitis and diabetes. iV

PAGE 5

surprisingly, 100% of diabetes non-susceptible C57BL/6 and B10.BR/cd mice reconstituted with a NOD hematopoietic system developed insulitis with approximately 1 of 10 progressing to overt diabetes. These data emphasize the role of the reconstituting hematopoietic progenitor cell population in the pathogenesis of this autoimmune disease. V

PAGE 6

INTRODUCTION Insulin Dependent Diabetes Mellitus in Man Diabetes has been recognized and recorded as a clinical disease in man since ancient times. Recorded descriptions of the symptoms date back as far as 1500 B.C. as inscribed on the Ebers papyrus of Egypt and was first named by Aretaeus of Cappadocia in the second century A.D. (reviewed in Notkins, 1979). Clinical and basic research have revealed that diabetes consists of a heterogeneous set of clinical disorders, all of which are characterized by hyperglycemia (CaHill and McDevitt, 1981; Bennett, 1983). Diabetes mellitus has recently been categorized into distinct classifications including Type I, insulin dependent diabetes mellitus (IDDM) and Type II, noninsulin-dependent diabetes mellitus (NIDDM) by an international workshop sponsored by the National Diabetes Data Group of the National Institutes of Health (National Diabetes Data Group, 1979). A third classification of diabetes mellitus includes diabetes associated with pancreatic disease, hormonal etiology, drug or chemical induction, insulin receptor abnormalities and certain genetic syndromes. The focus of this discussion will be specifically on Type I, insulin 1

PAGE 7

dependent diabetes mellitus (IDDM) in this dissertation. IDDM in man is characterized by an abrupt onset of symptoms which includes elevated blood glucose levels (> 200 mgs/dl), insulinopenia, and frequently ketoacidosis (National Diabetes Data Group, 1979; Keen and Ng Tang Fui, 1979). Persons with IDDM are dependent on insulin treatment to prevent ketosis and death. The onset of IDDM generally occurs in youth but may occur in later life as well. There is a strong correlation between the development of IDDM with certain HLA alleles and there are experimental data which support an autoimmune etiology including evidence for the presence of specific autoantibodies and cellular immunity to pancreatic islet cells. Epidemiology. morbidity and mortality. Epidemiologic studies indicate the prevalence of IDDM in the United States to be approximately one in 300 under the age of 30 (Rosenbloom, 1983). The incidence of IDDM appears to vary among developed countries (Norris et al., 1987). The highest incidence was reported in Scandinavia. In Finland, about 29 per 100,000 people per year develop IDDM. In contrast, only 0.8 per 100,000 people per year develop IDDM in Japan. The incidence in the United States lies between these two extremes at approximately 14 per 100,000 per year. Geographic variation in incidence suggests an influence of environmental factors on the development of IDDM. An alarming statistic is that the incidence of IDDM has 2

PAGE 8

increased in recent decades (Krolewski et al., 1987; Stewart-Brown et al., 1983). For example, in the first half of the twentieth century, the incidence rate of IDDM in the Northeastern United States remained relatively constant but during the last 30 years there has been a nearly 3-fold increase in the incidence rate (Krolewski et al., 1987). Likewise, in England there was a 2-fold increase in the incidence of IDDM from 1958 to 1980 (Stewart-Brown et al., 1983). However, the rising incidence of IDDM remains controversial since increasing numbers of patients surviving beyond the reproductive age would not be sufficient to account for the dramatic increases in incidence rate. Prior to the availability of insulin, IDDM patients generally died within a few years of onset (Dorman et al., 1984; Crabbe, 1987). Most deaths were associated with ketosis induced coma and infection (Rossini and Chick, 1980). There was a dramatic increase in life expectancy after the onset of IDDM associated with the advancement of insulin treatment. However, the development of long-term sequelae became evident and IDDM patients still experience a life expectancy reduced by one-third as compared to nondiabetic individuals. A 40 year follow-up study of over 300 patients diagnosed before 1933 at less than 21 years of age was carried out at The Steno Memorial Hospital in Denmark (Deckert et al., 1973). Only 40% of the patients were 3

PAGE 9

still alive in 1973. The calculated mortality rate was 2-6 times that of an age matched non-diabetic population. The cause of death in 31% of the patients was kidney failure and 25% died of myocardial infarction. Further complications included blindness in 16% and severely impaired vision in another 14%. Twelve percent of the patients also suffered gangrene or previous amputation. Clinical and experimental evidence clearly indicates that vascular disease is the major cause of morbidity and mortality in diabetic individuals (Rossini and Chick, 1980; Levin and O'Neal, 1983). Macroangiopathy and arteriolar vascular disease can lead to myocardial infarction, stroke, and gangrene of the lower extremities. Microangiopathy can lead to nephropathy and retinopathy. The etiology of vascular disease abnormalities is not clearly understood but may involve hyperglycemia, hyperlipidemia, rheological disorders, and hypertension secondary to nephropathy. Physical manifestations associated with diabetes include connective tissue changes in arterial walls capable of altering normal blood flow and basement membrane thickening of capillaries of the kidney, eye, peripheral nerves, and skeletal muscle (McMillan, 1983; Rossini and Chick, 1980). The two long-term complications most strongly associated with mortality in IDDM patients are diabetic renal and heart disease (Rossini and Chick, 1980; Crabbe, 1987). It is estimated that there is a 20-fold increase in the risk 4

PAGE 10

of renal failure in diabetics as compared to non-diabetics. Pathological changes associated with renal disease in diabetics include basement membrane thickening in glomerular capillary walls, nodular intercapillary glomerulosclerosis and diffuse glomerulosclerosis. Clinical features associated with IDDM include coronary atherosclerosis, myocardial hypertrophy and interstitial fibrosis (Fein and Scheuer, 1983). Autoimmune pathogenesis and insulitis. A fundamental characteristic that separates IDDM from other forms of diabetes is that abnormal immune responses and autoimmunity are thought to play a role in the pathogenesis of IDDM (National Diabetes Data Group, 1979). Reports of mononuclear cell infiltration of the islets of Langerhans (insulitis) in patients who died suddenly after an abrupt onset of IDDM triggered a widespread search for evidence of autoimmunity against specific elements of the endocrine pancreas (Gepts, 1965; Nerup et al., 1984). The pathognomonic lesion of insulitis can be more clearly understood after a brief review of the normal histologic morphology of the islets of Langerhans (pancreatic islets) (Ito, 1977; Volk and Wellmann, 1985). The normal pancreatic islet is comprised of a compact, rounded mass of endocrine cells surrounded by a connective tissue sheath (comprised of reticular fibers and a basement membrane). The islets are irregularly distributed within 5

PAGE 11

the lobules of the exocrine portion of the pancreas but are more or less demarcated from the surrounding acinar tissue by the connective tissue sheath. The islets are highly vascular with numerous capillaries which form glomeruluslike networks that allow close proximity with every endocrine cell. The cell cord mass is made up of 4 cell types: 1) fi-cells comprise about 70% of the islet cells and their endocrine function is to produce and secrete insulin, 2) a-cells account for about 15-20% of the islet cell population and are responsible for the production and secretion of glucagon, 3) 6-cells which produce somatostatin account for about 10% of the islet cell population, while 4) PP-cells comprise about 1% of the islets cells and produce pancreatic polypeptide. The a-cells and 6-cells are primarily located in a narrow peripheral zone of the islet. Lymphocytic infiltration of the islets of Langerhans in diabetic patients was initially described at the beginning of the twentieth century and in 1940 Von Meyenburg coined the term "insulitis" to describe this inflammatory lesion (reviewed in Gepts and Lecompte, 1981). Insulitis was initially thought to be a rare lesion as the prevalence was calculated from studies of a heterogeneous set of diabetic pancreata. However, more selective studies of pancreata from IDDM patients who died early in the course of their disease indicated that insulitis was not a rare lesion. (Lecompte, 1958; Gepts, 1965; Gepts and Lecompte, 1981). 6

PAGE 12

These findings encouraged more specific studies of the possible role of insulitis in the pathogenesis of IDDM. Insulitis was observed in 68% (15/22) and 78% (47/60) of cases in two relatively large studies of pancreata from young diabetic patients with an acute onset of disease (Gepts, 1965; Foulis et al., 1986). Histologic examination disclosed several pathological features characteristic of pancreatic islets from patients with acute onset of IDDM (Foulis and Stewart, 1984; Gepts and Lecompte, 1985; Foulis et al., 1986). The majority of the islets were composed of narrow cords of small endocrine cells arranged in a fibrous stroma with irregular outlines. The cytoplasm of the cells was not abundant and the small nuclei had dense chromatin. The cells appeared atrophic but were actually composed of active endocrine cells. Irnrnunocytochemical staining techniques showed that these pseudoatrophic islets were devoid of pancreatic ~-cells but contained normal numbers of a-cells, 6-cells and PP-cells. The less numerous set of unique pancreatic islets was relatively large and contained large hypertrophic cells with large nuclei. Immunocytochemical staining techniques showed the hypertrophic islets to be composed of degranulated pancreatic ~-cells and normal numbers of a-cells, 6-cells and PP-cells. Some of the hypertrophic islets showed marked reduction in the number of ~-cells. Quantitative studies indicated a marked reduction in the total mass of endocrine 7

PAGE 13

tissue in IDDM patients as compared with normal controls and the p-cell mass was reduced to less than 10% of normal (Gepts and Lecompte, 1981; Rahier et al., 1983). Further evidence for p-cell specific autoimmunity was derived from immunocytochemical staining methods which revealed that the lymphocytic infiltrations were primarily surrounding or within pancreatic islets that contained p-cells (Gepts and Lecompte, 1981; Foulis and Stewart, 1984; Gepts and Lecompte, 1985; Foulis et al.,1986). Quantitative analysis showed that about 20% of islets containing insulin (P-cells) were affected by insulitis but only 1% of islets deficient in insulin were affected. Morphological analysis of mononuclear cell infiltrates of pancreatic islets from IDDM patients indicated that the infiltrating cells consisted predominantly of lymphocytes and occasional polymorphonuclear leukocytes (Gepts, 1965; Foulis and Stewart, 1984). Immunofluorescence staining techniques on frozen sections from a diabetic pancreas indicated that the majority of the infiltrating lymphocytes were T-cells (Bottazzo et al, 1985). Further characterization of the T-lymphocytes showed the predominant subpopulation to be cytotoxic/suppressor T-cells. T-helper cells and natural killer (NK) cells were also shown to be present in small numbers. The majority of the infiltrating T-lymphocytes were shown to be positive for HLA-DR expression indicating that they were activated. 8

PAGE 14

There was an apparent progression of inflammatory cell infiltration and p-cell degeneration. Pancreatic islets affected by early stages of cellular infiltration still maintained normal islet morphology and contained relatively well preserved p-cells. With the progression of insulitis, the islets began to develop a pattern of small cell cords, a collapsed islet framework and reduced numbers of p-cells. The lymphocytes gradually disappeared from the islets as all the P-cells were destroyed leaving collapsed islets with cords of insulin-deficient endocrine cells separated by fibrous tissue. A few of the collapsed islets showed a residual inflammatory cell infiltrate but most were totally devoid of insulitis. The observation of the progressive and specific destruction of pancreatic p-cells (leaving a-cells, 6-cells and pp-cells intact} supports the hypothesis of autoimmune mediated destruction of pancreatic p-cells in the pathogenesis of IDDM. Cellular autoimmunity. Methods to measure specific cellular immunity to pancreatic islet autoantigens have provided direct evidence for the autoimmune pathogenesis of IDDM. Initially, organ-specific cellular immunity was indicated by the ability to induce leukocyte migration inhibition of cells from IDDM patients in the presence of pancreatic cell extracts (Nerup et al., 1971; Nerup et al., 1974). In vivo evidence of specific cellular immunity was indicated by the ability to induce a delayed-type 9

PAGE 15

intracutaneous reaction in IDDM patients by subcutaneous injection of pancreatic cell extracts (Nerup et al., 1974). No delayed-type reaction was seen in normal controls or in IDDM patients who did not show a positive leukocyte migration inhibition. Experiments have also been designed to show lymphocyte and antibody dependent cellular immunity to pancreatic ~-cells (Huang and Maclaren, 1976). Lymphocytes from IDDM patients had a markedly higher cytoadherence to human insulinoma cells after 15, 40 and 60 hours co-culture as compared to normal individuals. Furthermore, enriched T-lymphocytes (erythrocyte rosette-forming cells) were shown to be significantly cytotoxic to human insulinoma cells in chromium-51 release assays. Significant antibody-dependent cellular cytotoxicity was also observed when T-cell depleted leukocyte populations (non-rosette-forming cells) and serum from IDDM patients were cultured with chromium-51 labelled insulinoma cells. An effect of cellular immunity on the production of insulin by pancreatic ~-cells in IDDM patients has been investigated (Boitard et al., 1981). Pancreatic islets were isolated from DBA/2 mice and stimulated to increase insulin production in culture by the addition of glucose and theophylline. The addition of lymphocytes from IDDM patients significantly inhibited the stimulation of insulin production as compared to lymphocytes from normal controls. 10

PAGE 16

The experiments discussed above provided definitive evidence for specific cellular immunity against pancreatic fi-cells in the pathogenesis of IDDM. Humoral autoimmunity. The detection and characterization of organ-specific autoantibodies suggested an autoimmune etiology and a possible role for humoral immunity in the pathogenesis of IDDM. The first significant detection of autoantibodies to pancreatic islet cells were in a select group of diabetic patients with associated autoimmune endocrine diseases (Bottazzo et al., 1974; MacCuish et al., 1974). Circulating islet cell antibodies (ICA) were detected in patient sera by indirect immunofluorescence techniques using unfixed frozen sections from human pancreas as the binding substrate. ICA from positive patient sera gave uniform cytoplasmic immunofluorescence involving all the pancreatic islet cell types, were complement fixing, and were of the IgG class. Studies were then done to detect the frequency of ICA in young IDDM patients of recent onset (Lendrum et al., 1975). Of the 105 IDDM patient sera tested by immunofluorescence assays on frozen human pancreas sections, 51 (49%) had detectable ICA antibody. Further studies indicated that the frequency of ICA was 75% in caucasian IDDM patients within 3 months of diagnosis but then declined markedly (Neufeld et al., 1980). The use of living cells as the binding substrate has made it possible to detect cell surface binding antibodies 11

PAGE 17

in the sera of IDDM patients. Initially, cell surface binding of circulating antibody from IDDM patient sera was demonstrated using indirect immunofluorescence methods on human-insulinoma cells (Maclaren et al., 1975). Positive binding of antibody to the cell surface of the insulinoma cells was reported in 34 of 39 IDDM sera. These results were later confirmed using viable insulin-producing islet cells from rats as the binding substrate (Lernmark et al., 1978). The cell surface binding autoantibodies were termed islet cell surface antibodies (ICSA). Complement-mediated cytotoxicity of chromium-51 labeled rat islet cell cultures (Dobersen et al., 1980) and cloned rat islet cell monolayers (Eisenbarth et al., 1981) was detected in ICSA+ sera from IDDM patients suggesting a possible role for ICSA in the pathogenesis. Furthermore, ICSA+ sera produced significant antibody-dependent cellular cytotoxicity of cloned human pancreatic ~-cells in the presence of normal human lymphocytes (Maruyama et al., 1984a). Genetics of IDDM. The association of genetic susceptibility for IDDM to genes within the major histocompatibility complex (MHC or the HLA complex in man) supports the hypothesis for a role of the immune system in the pathogenesis. Serotyping of blood cells from IDDM patients in population and family studies have indicated a strong association of HLA-DR3 and HLA-DR4 alleles with diabetes susceptibility (Platz et al., 1981; Cudworth and 12

PAGE 18

Wolf, 1982; Wolf et al., 1983; Henson et al., 1986). In a study done in the United States (Henson et al., 1986) with over 1000 IDDM probands, 95% of probands were positive for either HLA-DR3 or HLA-DR4 while only 50% of controls were positive for at least one of these antigens. Approximately 40% of the probands possessed both the HLA-DR3 and HLA-DR4 antigens compared to only 3% of the general population. Population and family studies have also indicated a strong association of HLA-DR2 and HLA-DR5 antigens with resistance for the development of IDDM. Only 4% of IDDM patients were positive for HLA-DR2 compared to 28% of controls and only 1% of IDDM patients were positive for HLA-DR5 compared to 17% of normal controls (Cudworth and Wolf, 1982; Wolf et al., 1983). The association of diabetes susceptibility and resistance with certain HLA-DR haplotypes suggests a possible role for the immune system in the pathogenesis of the disease since class II gene products are important in the regulation of the immune response. Non-Obese Diabetic (NOD) Mouse: An Animal Model for IDDM Development and characterization of the NOD mouse. As indicated above, IDDM in man is a significant health care problem associated with severe long-term complications. Experimental evidence indicates that the etiology is very complex and may entail a genetic predisposition and an autoimmune pathogenesis. However, experimental approaches 13

PAGE 19

are relatively limited when working with human IDDM patients. Representative experimental animal models for IDDM have been important in developing an understanding of this very complex disease. Of particular interest was the recent development of the NOD mouse which spontaneously develops an autoimmune response against the fi-cells within the islets of Langerhans resulting in their destruction and loss of the insulin production with progression to insulin dependent diabetes (IDD) (Makino et al., 1980). These phenotypic features strongly resemble those observed in human Type I insulin dependent diabetes mellitus. The NOD mouse shares many pathological features common to human type I diabetes including hyperglycemia, insulinopenia, hyperglucagonemia, glycosuria, polydipsia, polyuria, ketoacidosis, weight loss, hypercholesteremia, and the pathognomonic lesion of insulitis (Makino et al., 1980; Leiter et al., 1987). These similarities make the NOD mouse an excellent model for the examination of human IDDM. The NOD mouse was initially discovered and developed at the Shionogi Research Laboratories in Osaka, Japan. It was developed from a subline of ICR mice which were being selectively bred for cataracts (Makino et al., 1980). Since one of the clinical manifestations of IDD is cataract formation, the ICR-derived mice were screened beginning at the 6th generation for hyperglycemia. After selective breeding for 13 additional generations, two sister sublines 14

PAGE 20

were established: a normoglycemic (100 mg/dl blood glucose) subline and a slightly hyperglycemic (150 mg/dl blood glucose) subline (Tochino, 1986). One of the female offspring from the normoglycemic subline spontaneously developed diabetes and died. The surviving progeny from that diabetic female were then inbred to produce a diabetic subline that resulted in the development of the non-obese diabetic (NOD) mouse. NOD mice have been established as an inbred strain which continues to develop IDD in a consistent manner. A non-obese non-diabetic (NON) strain has been established from the slightly hyperglycemic subline as an inbred strain. Examination of histocompatibility and plasma enzyme markers revealed that the NOD mice and NON mice were significantly different strains of mice. Histologic examination indicated that there was not a significant difference in the overall frequency of insulitis between male and female NOD mice (Komeda and Goto, 1986), as insulitis was observed in nearly 100% of both males and females by 30 weeks of age. However, the development of insulitis appeared to be delayed in male NOD mice as compared to females. Tochino reported that female NOD mice developed overt diabetes more frequently than the males. There was about an 80% frequency of overt diabetes in female mice but only a 20% frequency in male mice (Tochino, 1976). The overall nature of these differences is not clearly understood, but investigators have suggested that hormonal 15

PAGE 21

differences are involved since castrated males have a higher incidence of diabetes while oophorectomized females have a lower incidence of diabetes (Makino et al., 1981). The blood glucose levels of NOD mice rose at the onset of diabetes from the normal 100-180 mgs/dl to levels of 600-800 mgs/dl correlating with decreased levels of plasma insulin. Several changes from the normal physiology of NOD mice were observed after the onset of diabetes (Makino et al., 1980). A decrease in body weight was observed despite an increase in food consumption and a four fold increase in water consumption. The volume of urine excretion increased to 15-20 times normal and the urine contained up to a 100-fold higher concentration of glucose (glycosuria). Plasma cholesterol levels were also increased significantly and ketonuria could be observed in diabetic mice. NOD mice did not undergo spontaneous remission from diabetes and the mice died due to ketoacidosis if they were not administered insulin treatment. Daily injection of insulin induced an increase in body weight and a marked prolongation of life in the diabetic mice. Autoimmune pathogenesis and insulitis. The most prominent histopathologic lesion of insulin dependent diabetes in both man and the NOD mouse is the inflammatory response of insulitis, characterized by the infiltration of leukocytes into pancreatic islets. The development and progression of insulitis over time has been examined by 16

PAGE 22

light microscopy in the NOD mouse (Fujita and Yui, 1976; Fujino-Kurihara et al., 1985; Leiter et al., 1987). Histological examination of pancreata of NOD mice from parturition to 3 weeks of age showed no cellular infiltrates by light microscopy. However, by 4-5 weeks of age some of the animals showed mononuclear cells accumulating in or near the pancreatic islets. The cellular infiltration was also observed around the periductal and perivascular structures within the surrounding connective tissue which contains excretory ducts, arterioles, venules capillaries and small lymphatics. The mononuclear infiltrates usually invaded the islet from the side of the periductal connective tissue and congregated on that side of the connective tissue sheath. Eventually, the lymphocytes formed a multilayered aggregation around the perimeter of the islets. The infiltrating lymphocytes gradually increased in number until more and more islets were infiltrated. The boundary formed by the connective tissue sheath was eventually broken as some of the mononuclear cells invaded the territory of the islet. The p-cells adjacent to the infiltrating mononuclear cells began to degenerate and were replaced by the infiltrating cell mass. The original p-cells within the pancreatic islets decreased in number and eventually were entirely replaced by the mononuclear infiltrating cells. When all the p-cells were destroyed the infiltrating cell masses gradually dispersed and then disappeared leaving a 17

PAGE 23

18 residual islet containing primarily a-cells, 6-cells and some PP-cells as determined by immunohistochemical findings. These findings indicate that the immune response was specific for the pancreatic p-cells since the other pancreatic islet cells remained intact. Overt diabetes developed when most of the fi-cells had been destroyed. A sequential analysis of pancreatic sections stained with fluoresceinated antisera against Thy-1.2 and immunoglobulin showed that T-lymphocytes, including helper/inducer and cytotoxic/suppressor T cells, were the predominant infiltrating cells in the early stages of insulitis (Miyazaki et al., 1985; Miyazaki et al., 1986; Hanafusa et al., 1986; Toyota et al., 1986). Furthermore, double fluorescence staining showed that more than one-half of the T-lymphocytes were positive for class II molecule (Ia+) expression. Class II molecules are not usually expressed on murine T-lymphocytes, thus, this may indicate that the T-lymphocytes present in the insulitis were activated. Analysis of the T-lymphocyte subsets showed that Lyt-1 bearing cells (primarily T-helper/inducer cells) were more commonly present than those expressing the Lyt-2 antigen (primarily T-cytotoxic/suppressor cells). The T-lymphocytes were located close to the islets and also infiltrated deep within the islets. B-lymphocytes were also present at all stages of insulitis and gradually increased in numbers as the severity progressed. One-half or more of

PAGE 24

the infiltrating lymphocytes were B-lymphocytes in severe insulitis. The B-lymphocytes were located primarily near the blood vessels and at the periphery of the T-lymphocytes. Some natural killer (NK) cells were shown to be present close to pancreatic islet cells. Close examination of the ~-islet cells next to the infiltrating lymphocytes revealed evidence of cellular degeneration and necrosis as well as cell death (Fujino-Kurihara et al., 1985; Fujita and Yui, 1986; Fujita and Fujino-Kurihara, 1986). The histological and immunohistochemical studies described above suggest that the immune system plays an important role in the pathogenesis of diabetes in the NOD mouse, indicating an autoimmune etiology. Cellular autoimmunity. Data from several experimental systems support the hypothesis that diabetes observed in NOD mice has an autoimmune etiology. For example, generalized immunosuppression induced by sublethal irradiation (870 Rads) reduced the frequency of diabetes to 25% (2 of 8 NOD mice) within 29 weeks of age (Harada and Makino et al., 1976). Furthermore, a reduced frequency of diabetes was observed in neonatal thymectomized NOD mice (Hanafusa et al., 1986) suggesting a role for the cellular immunity since the thymus is involved in development and education of T-lymphocytes. Islet specific T-cell lines have been isolated from NOD pancreatic islets bearing mononuclear cell infiltrates 19

PAGE 25

(Hattori et al., 1986), indicating that T-lymphocytes are involved in the development of insulitis. The resulting cell lines were shown to be Thy-1.2+ and Lyt-2+ by flow cytometric analysis and to proliferate in response to NOD islet homogenates. Furthermore, intravenous injection of rabbit anti-mouse thymocyte antisera to deplete T-lyrnphocytes markedly decreased the incidence of overt diabetes in NOD female mice (Hanafusa et al., 1986) and reduced the severity of insulitis (Yokono et al., 1986) indicating an important role for thymus-derived lymphocytes in the development of diabetes. The development of diabetes in NOD mice was also shown to be dependent on the L3T4+ subpopulation of T-lymphocytes by intravenous injection of L3T4-specific antibody (GKl.5) for 12 weeks starting at 2 weeks of age (Koike et al., 1987). Monoclonal antibody GKl.5 binds to the cell surface antigen L3T4 which is primarily associated with T-helper cells reactive against antigen in the context of self class II molecules (Dialynas et al., 1983). Histological examination of pancreata from GKl.5 injected NOD mice showed no insulitis and the development of diabetes was prevented. In another study, diabetic NOD mice were engrafted with pancreatic islets from BALB/c mice and injected intravenously with GKl.5 for four consecutive days to prolong engraftment (Wang et al., 1987). The engraftment of pancreatic islets and the temporary depletion of L3T4+ 20

PAGE 26

cells resulted in a return to normoglycemia that persisted for 2-4 weeks. Histologic examination of the grafted pancreatic islet tissue, at the time when the animals again became hyperglycemic and the L3T4+ T-cells returned to normal numbers, showed that the islet tissue had been destroyed and replaced by mononuclear infiltrating cells. Considered together, these studies indicate an important role for T-helper cells in the pathogenesis of insulitis and diabetes in NOD mice. Theoretically, the binding of antigen by T-lymphocyte receptors requires the antigen to be presented on the surface of a cell in the context of a self MHC class I or class II molecule (Kronenberg et al., 1986). Therefore, initiation of the response by T-helper cells against the pancreatic ~-cells may require that an autoantigen be presented on the surface of a ~-cell in the context of a self class II molecule. Initially, cells of the macrophage lineage were considered to be responsible for antigen presentation to T-helper cells in the context of self class II MHC molecules (Unanue 1981; Ziegler and Unanue, 1981). However, other cell types have been hypothesized to be capable of antigen presentation in the context of self class II molecules including: 1) Langerhans cells {Tanaki et al., 1979), 2) B-lymphocytes {Chesnut and Grey, 1981; Chesnut et al., 1982), 3) thyroid cells {Hanafusa et al., 1983) and pancreatic ~-cells (Bottazzo et al., 1983; Bottazzo et al., 21

PAGE 27

22 1985; Londei et al., 1984). The first reports of aberrant class II molecule expression on endocrine cells associated with autoimmune disease were from patients with thyroid autoimmunity (Hanafusa et al., 1983) and type I IDDM (Bottazzo et al., 1983). It was suggested that the class II expressing p-cells may present auto-antigens in the context of self class II and thus initiate the autoimmune response (Bottazzo et al., 1983; Bottazzo et al., 1985; Londei et al., 1984). This theory would provide a possible explanation for the association of type I diabetes with the DR 3 and DR 4 histocompatibility phenotypes. Immunofluorescence studies have been performed to determine if I-A molecules might be expressed on the surface of pancreatic p-cells in NOD mice (Hanafusa et al., 1987) as described in human IDDM patients. Positive islet cells were seen in all 23 NOD mice examined. Double immunofluorescence staining showed that the I-A positive cells were all insulin producing cells but were negative for production of glucagon or somatostatin indicating that only the p-cells of the pancreatic islets expressed I-A. All of the islets with mononuclear cell infiltration and about half of the islets with no infiltration were positive for I-A molecule expression. Perhaps part of the genetic susceptibility in organ specific autoimmunity is related to a predisposition for aberrant expression of class II MHC molecules on the surface of the target cells resulting in sufficient

PAGE 28

presentation of autoantigens to L3T4+ lymphocytes. Further evidence for an autoimmune pathogenesis of diabetes in NOD mice comes from adoptive transfer experiments with which a pathological contribution by T-lyrnphocytes has been implied. In one experimental system, adoptive transfer procedures in NOD mice could induce rapid onset of diabetes at an age when spontaneous diabetes is rarely observed (Wicker et al., 1986a; Wicker et al., 1986b). NOD mice between 6 and 13 weeks of age were sublethally irradiated (775 Rads) and injected intravenously with spleen cells or enriched T-cells from overtly diabetic mice. In contrast to the very low percentage of female mice that would have become spontaneously diabetic by 13 weeks of age, 95% (79 of 82 mice) developed diabetes within 12-22 days after sublethal irradiation and injection of autoreactive cells from overtly diabetic NOD mice. To determine the subset(s) of T-cells responsible for the adoptive transfer of autoimmunity, Hanafusa and co-workers utilized T-cell depleted NOD mice which were injected with L3T4+ depleted spleen cells or Lyt-2+ depleted spleen cells from cyclophosphamide treated NOD mice (Hanafusa et al., 1988). The mice from the donor population were pre-treated with cyclophosphamide since this treatment has been shown to promote the onset of overt diabetes in NOD mice (Harada and Makino 1984). T-cell depletion in NOD mice was achieved by thyrnectomy at 4-5 weeks of age, followed by injection of 23

PAGE 29

antithymocyte antiserum 2 days later and sublethal irradiation (800 Rads) 3 weeks after thymectomization. T-cell depleted NOD mice had a reduced incidence of insulitis (23%) as compared to untreated NOD mice (80%) or to T-cell depleted NOD mice which were adoptively transferred with either whole spleen cell populations (69%) or L3T4+ enriched cell populations (81%). However, adoptive transfer of L3T4+ depleted spleen cells resulted in a reduced incidence of insulitis (20%). These data have been interpreted to mean that the L3T4+ T-lymphocytes are crucial in the pathogenesis of insulitis in NOD mice. Humeral autoimmunity. A possible role for humeral immunity in the development of diabetes has also been indicated from studies showing the presence of islet cell antibodies (ICA) and islet cell surface antibodies (ICSA) in NOD mice. The presence of ICA in NOD mice was detected by indirect immunofluorescence using serum samples from NOD mice reacted with pancreatic tissue sections from ICR mice followed by an incubation with a fluoresceinated rabbit anti-mouse antibody (Toyota et al., 1982; Takei et al., 1986). ICA was shown to be present in 75% of NOD mice at 4 weeks of age, a time prior to the presence of insulitis detectable by light microscopy~ A possible role for ICA in autoimmunity is not clear since the antigen is present in the cytoplasm rather then on the cell surface. The detection of ICSA in NOD mice was reported by two groups in 24

PAGE 30

Japan by indirect immunofluorescence using cultured islet cells (Kanazawa et al., 1984; Matsuba et al., 1986). Kanazawa and co-workers determined the presence of !CSA in NOD mice at various ages. !CSA was first present at 6 weeks of age in less than 10% (1/13) of the NOD mice and reached a peak prevalence of 50% (3/6) by 12 weeks of age. Subsequently, the detectable !CSA decreased as the mice became older. In contrast, Matsuba and co-workers reported a 69% (34/49) overall incidence of !CSA in NOD mice from all ages. Hybridoma cell lines secreting monoclonal antibody specific for !CSA have been produced by fusing lymphocytes from !CSA+ NOD mice with FO myeloma cells (Yokono et al., 1984). Prior to fusion the NOD lymphocytes had been transferred to sublethally irradiated (600 Rads) ICR mice by intravenous injection since this increased the frequency of !CSA reactive hybridoma production. The specificity of purified !CSA reactive monoclonal antibody (3A4) was determined in a radiolabelled protein-A binding assay using Syrian golden hamster insulinoma cells (In-111 cells). Immunoprecipitation of radiolabelled proteins from In-111 cells using monoclonal antibody 3A4 allowed the isolation of a 64 kilodalton cell surface protein and a 28 kilodalton cytoplasmic protein (Hari et al., 1986). The immunoprecipitated 64 kilodalton protein may represent the auto-antigen recognized by !CSA. Further evidence that !CSA may play a role in autoimmunity was indicated by positive killing in 25

PAGE 31

antibody-dependent cellular cytotoxicity (ADCC) assays using 3A4 antibody. Abnormal immune functions. Functional studies of the immune system from NOD mice have revealed an altered immune cell repertoire and a relatively diminished immune responsiveness. Experiments using 3 month old NOD mice showed possible abnormalities of cellular or humoral immunity when compared with results from ICR mice (Kataoka et al., 1983; Maruyama et al., 1984b; Toyota et al., 1986b). NOD mice appeared to have significantly lower numbers of mononuclear blood cells and mononuclear spleen cells as well as a significantly reduced number of T-lymphocytes as compared with ICR mice. A more specific examination of the T-lymphocyte subpopulations indicated a normal level of Lyt-1+ (T-helper cells) and a markedly reduced number of Lyt-2+ (T-cytotoxic/suppressor cells) (Toyota et al., 1986b). The number of B-lymphocytes in NOD mice was normal, which translates to a relative increase in the percentage of B-cells. However, there was a significantly reduced number of phagocytizing macrophages in NOD mouse spleens. The ability of NOD immune cells to mount a response in vitro and in vivo was also measured using various immunological assays. Natural killer (NK) cell activity against YAC-1 target cells, antibody-dependent cell-mediated cytotoxicity (ADCC) activity against chicken red blood cells, and the induction of cytotoxic T-lymphocytes (CTL) 26

PAGE 32

against EL-4 cells were all significantly reduced in NOD mice indicating a diminished ability to mount a cellular immune response. Furthermore, NOD spleen cells produced significantly less interleukin-2 (IL-2) in response to Concanavalin A stimulation than !CR spleen cells, and NOD mice showed a diminished resistance to a virulent herpes simplex virus following intraperitoneal inoculation (Toyota et al., 1986b). However, delayed type hypersensitivity (DTH) responsiveness to sheep erythrocytes was normal indicating normal cellular immunity for certain immune functions. Furthermore, NOD mice showed a significantly increased ability to produce antibodies against T-cell dependent antigens such as sheep erythrocytes, even though NOD mice possess normal numbers of B-cells, indicating a possible enhancement of T-helper cells that regulate B-cell responsiveness. These data indicate that there are alterations of immune response functions in NOD mice but that the specific defects are not well defined. Genetics of IDD in the NOD mouse. Breeding studies have been designed by several laboratories to determine the inheritance of genetic susceptibility for the development of autoimmune diabetes in NOD mice (Makino et al., 1985; Hattori et al., 1986; Prochazka et al., 1987; Wicker et al., 1987). Makino and co-workers examined the inheritance characteristics of insulitis in breeding studies between NOD and C57BL/6 mice. Insulitis was used as a marker since 27

PAGE 33

there is a 100% incidence of insulitis in NOD mice and the lesion is believed to be important in the development of diabetes (Makino et al., 1985). Insulitis was not observed in (NOD x C57BL/6)Fl or (Fl x C57BL/6) backcross progeny indicating insulitis had a recessive functional inheritance. However, insulitis was observed in 3.9% of (Fl x Fl)F2 and 23.7% of (Fl x NOD) backcross female mice as compared to 89.7% of (NOD x NOD)Pl mice. The 1/16 and 1/4 ratios for the incidence of insulitis in the F2 and backcross mice with respect to the parental (Pl) mice indicated to the authors that the development of insulitis was controlled by two independently segregated recessive genes. Hattori and co-workers specifically examined the possibility of MHC-linked susceptibility genes for the development of overt diabetes in NOD mice (Hattori et al., 1986). The MHC region of the NOD mouse was shown to be of a recombinant haplotype. Serological analysis indicated that the class I histocompatibility molecules were of the H-2Kd and H-2Db alleles. However, the initial screening with a panel of antisera to class II antigens indicated that the NOD I-A molecule was unique and that NOD did not express an I-E molecule. Assays to test the ability of NOD splenocytes to stimulate class II specific T-cell hybrid clones as well as mixed lymphocyte cultures using NOD mice and various C57BL/10 congenic mouse lines confirmed that the NOD class II I-A molecules were unique to the NOD mouse line. 28

PAGE 34

Analysis of mRNA from NOD mice showed no detectable Ea mRNA, thus confirming the serologic data which indicated a lack of I-E molecule expression. Out breeding studies were then carried out with NOD and C3H (H-2k) mice to examine a possible linkage of diabetes to the inheritance of the unique NOD I-A molecule (Hattori et al., 1986). A difference in the restriction fragment length polymorphisms (RFLP) of the Afi genes from NOD and C3H mice using southern blot analysis enabled the investigators to distinguish which I-A genes were inherited by the progeny. None of (C3H x NOD)Fl female mice developed diabetes as compared with 80% of NOD females. Three of 117 (Fl x Fl)F2 female progeny and 12 of 57 (Fl x NOD) backcross female mice developed diabetes. All of the diabetic mice that were analyzed by southern blot analysis for Afi RFLP were found to be homozygous for the NOD I-A molecule indicating a recessive nature for the NOD MHC-linked susceptibility gene. The authors suggested that there were probably two or more diabetes susceptibility genes in addition to the MHC-linked gene since only 15.6% of the (Fl x NOD) backcross females developed diabetes. Sequence analysis of cDNA clones from NOD A0 and Afi genes has been carried out to examine the possible role of NOD I-A molecules as diabetes susceptibility genes (AchaOrbea and McDevitt, 1987). Sequences from the NOD A0 cDNA clone were identical to that of the I-A0d gene. However, sequences from the first external domain of the NOD Afi cDNA 29

PAGE 35

clone were unique as compared to allelic sequences of I-A~k,b,d,s,u,f,q. Only one region, corresponding to amino acids 56 and 57, was completely unique to the NOD I-A~ sequence. Acha-Orbea and McDevitt suggested that the unique sequence of the NOD I-A~ gene could be responsible for the association of diabetes susceptibility and the I-A region of the NOD mouse. Prochazka and co-workers carried out breeding studies between NOD and the related non-obese normal (NON) strain of mice to more closely examine the polygenic nature of d iabetes susceptibility (Prochazka et al., 1987). None of the (NOD x NON)Fl progeny developed diabetes over a 12 month period. When 200 (Fl x NOD) backcross progeny were examined over a 12 month period, 19 (9.5%) developed overt diabetes. These data indicated to the authors that at least 3 unlinked autosomal recessive genes were involved in the development of autoimmune diabetes in the NOD mouse line. The investigators showed that all 19 of the diabetic backcross progeny were homozygous for the H-2Kd allele of the NOD haplotype indicating that one of the diabetogenic recessive genes was MHC-linked (chromosome 17). The investigators screened the backcross mice for a series of polymorphic genetic markers capable of distinguishing NOD from NON mice and showed that a second susceptibility gene mapped to chromosome 9 in the vicinity of the Thy-1 gene. However, the third susceptibility gene could not be mapped. 30

PAGE 36

A more complex role of susceptibility genes with respect to insulitis and the development of diabetes was inferred from breeding studies between NOD and BlO mice (Wicker et al, 1987). None of 200 (NOD x BlO)Fl mice developed diabetes but 1 of 54 examined histologically developed insulitis and 17 had small lymphoid infiltrations of the exocrine pancreas associated with vascular elements. These data confirmed a recessive inheritance for diabetes but suggested that insulitis may be controlled in part by an incompletely dominant gene that appears to have low penetrance in (NOD x BlO)Fl mice. None of the (NOD x BlO)F2 mice examined developed diabetes but most showed pancreatic lesions including mild to severe insulitis on histological examination. These data suggested to the authors that pancreatic inflammation and insulitis may be controlled by a single dominant or incompletely dominant gene and that fewer genes may be required for the inflammatory responses than for the full development of overt diabetes. There was a 12.5% (4 of 61 females) incidence of overt diabetes in the (Fl x NOD)BCl backcross progeny by 7 months of age and insulitis was observed in all the backcross mice examined. These data indicated that at least 3 functionally recessive genes or gene complexes control the development of diabetes in the NOD mice. The data discussed above on the genetic susceptibility for diabetes, suggests that inheritance of autoimmunity in NOD mice is of a complex polygenic nature. 31

PAGE 37

Experimental protocols for treatment and prevention of diabetes in the NOD mouse. Several experimental systems for the treatment and prevention of diabetes in NOD mice have recently been examined that have not yet been discussed. Takayoshi and co-workers gave 17 female NOD mice weekly intraperitoneal injections of a streptococcal preparation (OK-432) from 4-24 weeks of age (Toyota et al., 1986b). The OK-432 streptococcal preparation is a potent immunomodulator capable of activating both macrophages and killer T-cells and increasing interleukin-2 (IL-2) production. None of the 17 female NOD mice treated with OK-432 developed diabetes, whereas 14 of 17 NOD mice given only saline developed diabetes by 24 weeks of age. Histological examination of pancreatic sections from OK-432 treated NOD mice showed 98% of the islets were intact or only mildly infiltrated, whereas pancreatic sections from saline treated mice exhibited 79% of total islets affected with severe insulitis. Immunological characterization revealed that the number of mononuclear cells in the spleen and the reactivity of NK cells was significantly increased in OK-432 treated NOD mice. Oldstone examined the possibility that certain lymphotropic viruses may be capable of inducing a selective suppression of an autoimmune response (Oldstone, 1988). A lymphotropic variant of lymphocytic choriomeningitis virus (LCMV) was used that results in the infection of the 32

PAGE 38

T-helper subset. LCMV infection of NOD mice at birth or in adulthood prevented the development of overt diabetes. Adoptive transfer experiments were done to demonstrate the requirement for LCMV infection of lymphocytes to prevent diabetes. When non-diabetic NOD mice were sublethally irradiated (400 Rad) and injected intravenously with splenic lymphocytes from uninfected NOD mice, 80% developed diabetes within 30 days. In contrast, adoptive transfer of lymphocytes from LCMV infected NOD mice into sublethally irradiated NOD mice prevented the development of diabetes in all recipients. Lymphocytes from donors did not release infectious virus and no virus was detectable in the sera of recipients. These data indicated to Oldstone that the prevention of diabetes was caused by a virus-induced inactivation of potentially autoimmune reactive lymphocytes. Experiments have also been designed to examine the ability of preventing diabetes in NOD mice with large-dose treatment with nicotinamide (Yamada et al., 1982). Nicotinamide (0.5 mg/g body weight) was injected subcutaneously into 3 month old female NOD mice every day for 40 days. None of the nicotinamide treated NOD mice developed diabetes. In contrast, nearly 70% of untreated mice developed diabetes within the 40 day trial. Histological examination of pancreatic sections from treated mice showed normal numbers of islets and inflammatory cells were scanty and limited to the periphery of the islets. 33

PAGE 39

Bone marrow transplantation procedures have also been carried out to prevent and/or treat insulitis and overt diabetes in NOD mice. Ikehara and co-workers lethally irradiated NOD mice (greater than 5 months of age) and reconstituted them with bone marrow cells from BALB/cnu/nu mice (Ikehara et al., 1985; Ikehara et al., 1987). The allogeneic bone marrow reconstituted NOD mice showed a normal glucose tolerance profile at 3 months posttransplantation. In contrast, 8 month old untreated NOD mice exhibited impaired glucose tolerance similar to that seen in IDDM. Furthermore, insulitis was not present in the pancreatic islets of the allogeneic bone marrow reconstituted NOD mice. The allogeneic bone marrow transplantation was sufficient to treat NOD mice which still had sufficient numbers of pancreatic fi-cells to produce enough insulin for the prevention glucose intolerance and hyperglycemia. However, NOD mice with overt diabetes required treatment with concomitant allogeneic bone marrow transplantation plus pancreas transplantation (Yasumizu et al, 1987). Four of seven treated NOD mice survived over 90 days posttransplantation, as compared to, control NOD mice which all died within 2 weeks without insulin treatment. Furthermore, the transplanted NOD mice showed an improvement in glycosuria and a normal glucose tolerance profile. 34

PAGE 40

MATERIALS AND METHODS Animals. Inbred strains of mice used in this study were bred and maintained in the animal facility located in the Department of Pathology, University of Florida, Gainesville, Florida. They included AKR/J, BlO.BR/cd, B10.BUA16, BlO.Q, B10.RIII(71NS), BlO.S, B10.SAA48, C57BL/6, C57BL/10, CBA/J, DBA/2J, PL/J, non-obese diabetic (NOD) mice, (C57BL/10 x BlO.BR/cd)Fl, and (NOD X BlO.BR/cd)Fl. In bone marrow reconstitution experiments, female mice 8-12 weeks of age were used for donor bone marrow cell populations. All host mice receiving donor bone marrow were female mice 7 weeks of age. Antisera. Monoclonal antisera 15-3-ls (anti-H-2k), 31-3-4s (anti-H-2Dd), 28-13-3s (anti-H-2b), MK-D6 (anti-I-Ad), 31-3-4s (anti-H-2Kd), 10-2.16 (anti-I-Ak) and 20-10-5s (anti-Thy-1), obtained from American Type Culture Collection, Rockville, MD., were used to type for the histocompatibility antigens expressed on spleen cells of the chimeric animals. Alloantisera D.32 (anti-H-2Dk) and K.333 (anti-H-2Kb) provided by Dr. E.K. Wakeland (Department of Pathology, University of Florida), were also used. 35

PAGE 41

Spleen cells preparations. Single-cell suspensions of newborn and adult splenocytes were prepared by gently pressing freshly explanted tissues through wire mesh screens and washed with phosphate-buffeted saline (PBS). For mixed leukocyte culture reactions, the red blood cells were lysed with 0.84% ammonium chloride. The leukocytes were then washed once and resuspended in PBS to the appropriate concentrations. Bone marrow cells preparations. Adult bone marrow cells were prepared by cutting the epiphyses from freshly explanted femurs and tibias and then flushing the contents from the lumens with PBS using a 27 gauge needle and syringe. The tissue was dispersed to a single cell suspension using gentle pipetting. Bone marrow reconstitution of lethally irradiated hosts. Protocols used in bone marrow reconstitutions of lethally gamma-irradiated adult mice included a novel system utilizing newborn spleen cells to mediate engraftment. Newborn spleen-associated null suppressor-inducer cells (Peeler et al., 1983) were used to mediate bone marrow engraftment and to prevent graft-versus-host disease. Newborn spleen cell cultures were established by culturing 6 x 106 spleen cells/35cm2 dish for 24 hours in 3 ml eagle's high amino acid (EHAA) medium supplemented to 0.5% normal mouse serum (Click et al., 1972). This culture period 36

PAGE 42

permitted suppressor factors to be produced and secreted in the supernate. After 24 hrs, 60 x 106 freshly explanted adult donor bone marrow cells were added to each dish and co-cultured an additional 24 hrs. The cell mixtures were collected, washed and resuspended to a concentration of 100 X 106/ml. In a few experiments, as an alternative to the bone marrow/newborn spleen cell mixtures, animals were reconstituted with hematopoietic stem cells from long-term bone marrow cell cultures. Primary bone marrow cultures were established and maintained according to Dexter and co-workers (Dexter et al., 1977). Briefly, 10 x 106 freshly explanted bone marrow cells were cultured in 10 mls Fisher's medium for leukemic cells of mice supplemented to 20% horse serum and 10-6 molar hydrocortisone within a 75 mm2 culture flask. Weekly, 50% of the medium was discarded and replaced with fresh medium and on the third week the cultures were reseeded with an additional 10 x 106 bone marrow cells. At time of reconstitution, non-adherent cells were collected, washed and resuspended to 100 x 106/ml. These dexter cultures did not contain mature lymphocytes (Dexter and Spooncer, 1978; Spooncer and Dexter, 1983) and cells derived from these cultures failed to induce GVH reaction. Furthermore, the cells from these cultures would not be capable of passively transferring mature lymphocytes capable of inducing autoimmunity in the transplanted hosts. 37

PAGE 43

Each recipient mouse received 0.2 ml of reconstituting cells via the tail vein within 4 hrs after lethal irradiation (970-1050 Rads). Reconstituted animals were usually placed in laminar flow hoods and given water containing 10 mg/L polymyxin B plus 100 mg/L neomycin. In a few experiments, the mice were maintained under normal colony conditions and given acidified water to drink, but no differences in survival time/rates were observed. In any one experiment, all groups of animals were treated similarly. At various time points recipient mice were killed for histological and functional studies. Single cell suspensions of splenocytes were prepared as described above and tested for functional reactivities in mitogen stimulation and mixed leukocyte culture assays. Lymphocyte cultures. Mixed leukocyte cultures (MLCs) were performed according to Peck and Bach (Peck and Bach, 1973) and consisted of 0.5 x 106 splenic leukocytes co-cultured in flat bottom plates with an equal number of gamma-irradiated (2500 Rad) stimulating splenic leukocytes in 0.2 ml EHAA. At various time points, as indicated in the figures, cells were pulsed with 1.0 Ci of tritiated thymidine, harvested 8 hours later and 3H-uptake measured using standard scintillation procedures. Mitogenic responses were measured in a similar manner following stimulation with either 8 g concanavalin A (Con A) or 25 g 38

PAGE 44

lipopolysaccharide (LPS). Data were reported as the means of triplicate cultures minus background. Histocompatibility typing for chimerism. Leukocytes were serotyped using a two-step cytotoxicity assay. Spleen cells at 1.0 x 106 cells/ml were incubated with antisera against host specific or donor specific histocompatibility antigens for 45 min at 4C. The cells were then washed, resuspended in rabbit complement and incubated for 45 min at 37C. All tests were carried out in RPM! 1640 medium. Cell viability was assessed by trypan blue dye exclusion. Alternatively, when leukocytes differed only at class II loci, spleen cells were serotyped by immunofluorescence. The spleen cells were incubated with histocompatibility antigen specific antisera for 45 minutes at 4C. The cells were washed and then incubated an additional 45 minutes with fluoreseinated rabbit-anti-mouse antisera at 37C. The cells were then analyzed in a fluorescence activated cell sorter (FACS) to detect positive binding by specific antisera. Histological examinations. Freshly removed organs and skin tissue were fixed in 10% formalin for 18-24 hr, followed by dehydration in 80% ethanol. The tissues were embedded in paraffin, sectioned and stained with hematoxylin-eosin (H&E) dye. 39

PAGE 45

Measuring of blood glucose levels to determine the onset of overt diabetes. Blood glucose levels were determined using chemstrip bG test strips. A drop of blood drawn from the tail was placed on the test strip indicator, wiped off after one minute, and the glucose concentration was indicated by the color change in comparison with a calibrated scale. Glucose tolerance testing. Mice which had fasted for 12 hours were injected intraperitoneally with a glucose solution (1.5 grams per kilogram weight). Blood glucose levels were measured at the time of injection and subsequently at 15, 30, 60, 120 and 180 minutes as described above. 40

PAGE 46

THE USE OF NEONATAL SPLEEN CELLS TO INHIBIT GVH DISEASE FOLLOWING SEMI-ALLOGENEIC AND ALLOGENEIC BONE MARROW TRANSPLANTATION Introduction Bone marrow transplantation provides a practical means to manipulate the immune system. Successful engraftment with pluripotent stem cells from bone marrow provides a continuous source of differentiated leukocytes, platelets, erythrocytes, fixed macrophages of the reticuloendothelial system and osteoclasts (Desnick, 1987). Bone marrow can be readily ablated from the host animal by lethal irradiation (Johns, 1966) and then replaced by injection of donor bone marrow cells into the venous system. Replacement of the hematopoietic stem cell system provides a means to examine the effects of certain genes expressed by hematopoietic derived cells. The normal procedure for bone marrow transplantation is to ablate the hematopoietic progenitor cells of the host with a lethal dose of irradiation (Gale, 1982; O'Reilly, 1983). Donor bone marrow is then injected into the irradiated host to reconstitute the hematopoietic progenitor cell population. However, freshly explanted donor bone marrow also contains mature T-lymphocytes capable of 41

PAGE 47

inducing an immune response. Therefore, allogeneic bone marrow transplantation usually results in the induction of an immune response to allogeneic histocompatibility antigens resulting in graft-versus-host (GVH) disease and the eventual deat h of the animal if left untreated. A successful allogeneic bone marrow transplantation system must be capable of dealing with the normal occurrence of GVH disease. Spleens of newborn mice less than 3-4 days of age contain a naturally-occurring population of monocytes capable of suppressing T-dependent and T-independent immune responses of third-party adult cells both in vitro and in vivo. The presence of naturally occurring newborn spleen associated suppressor cells capable of inhibiting T-dependent and T-independent immune responses of third-party adult cells in vitro is now well established (Skownon-Cendrzak and Ptak, 1976; Ptak and Skownon-Cendrzak, 1977; Argyris, 1978). Despite earlier claims that this suppression was mediated by T suppressor cells (Argyris, 1978; Murgita et al., 1978; Ross and Pilarski, 1981), recent reports (Rodriguez et al., 1979; Hooper and Murgita, 1980; Peeler et al., 1983; Piguet et al., 1981) have identified the important cell population as a monocyte. These suppressor, or suppressor-inducer, monocytes elicit their suppressor activity in part through the secretion of soluble materials capable of activating the T suppressor 42

PAGE 48

limb of the immune response (Basset et al., 1977; Argyris, 1981; Peeler et al., 1983; Jadus and Peck, 1984). While no biological role for this phenomenon has yet been determined, it may be important in establishing maternal non-responsiveness to an allogeneic fetus (Peeler et al., 1983; Jadus and Peck, 1986) Recently, Jadus and Peck (Jadus and Peck, 1984) were able to utilize these naturally occurring suppressor-inducer monocytes to inhibit graft-versus-host (GVH) disease in sublethally gamma-irradiated hosts reconstituted with semi-allogeneic or allogeneic splenic T lymphocytes. Host animals which failed to develop GVH disease were shown to contain a mixture of host and donor leukocytes for 3-4 weeks but only host phenotype cells by 8-10 weeks. These data suggested that in sublethally irradiated animals, the newborn monocytes suppressed donor cell reactivity long enough for the host system to recover from the effects of irradiation. The ability to suppress allogeneic reactivity in the hosts unfortunately proved to be restricted by two genetic elements (Jadus and Peck, 1984): first, newborn monocytes and adult T cells had to be compatible at a H-2-linked region, and second, the newborn cells had to express a strongly stimulating non-H-2 phenotype. In the present section, I discuss the extension of this protocol to enhance the success of allogeneic bone marrow transplantation of lethally irradiated hosts. 43

PAGE 49

This protocol is based on the use of naturally occurring suppressor cells from the spleens of 2-3 day old mice (see methods). I have utilized this population of newborn suppressor cells to prevent acute GVH disease in lethally irradiated adult hosts reconstituted with semi-allogeneic or even fully allogeneic bone marrow cells. The advantage of this bone marrow reconstitution protocol is that it does not require any intervention with immunosuppressive drugs and does not require the elimination of T-cell subsets from the donor bone marrow. Spleen cells, rich in mature T-lymphocytes, were added to the donor cell population to enhance the development of GVH disease. The mice included B10.BR/cd, CBA/J, C57BL/6 and (C57BL/10 x B10.BR/cd)Fl. B10.BR/cd mice were the source of donor bone marrow cells and were co-cultured with newborn spleen cells from 2 day old CBA/J mice since this combination was previously shown to induce optimal suppression of GVH disease in sublethally irradiated animals. The bone marrow recipients utilized were C57BL/6 and (C57BL/10 x B10.BR/cd)Fl mice since they were readily available and provided sufficient host environment for the induction of GVH disease. The breeding efficiency of CBA/J mice was very good making the newborn suppressor cell mediated bone marrow transplantation procedure an efficient means for carrying out large numbers of allogeneic bone marrow reconstitutions. 44

PAGE 50

45 Pretreatment of the reconstituting cell populations with this newborn suppressor cells reduced the incidence of GVH disease from 100% to 20% in semi-allogeneic and from 100 to 40% in allogeneic combinations. Long-term surviving reconstituted hosts proved immunologically unresponsive to both donor and host histocompatibility antigens, yet possessed a fully chimeric lymphoid system responsive to T and B cell mitogens as well as unrelated third-party alloantigens. Results Newborn spleen cells prevent GVH disease in hosts engrafted with semi-allogeneic bone marrow. I first tested the ability to prevent GVH disease in semi-allogeneic bone marrow reconstituted hosts with the newborn suppressor cells since there would only be a 50% relative dosage of allogeneic histocompatibility antigens. In the first set of experiments, lethally gamma-irradiated (C57BL/10 x Bl0.BR/cd)F1 hosts were reconstituted with BlO.BR/cd bone marrow cells. As shown in Table 1, all irradiation control mice injected with PBS, but no reconstituting hematopoietic cells, died within 15 days. Likewise, control animals injected with BlO.BR/cd bone marrow plus spleen cells cultured 24 hours alone died within 16 days. However, 80% of hosts reconstituted with BlO.BR/cd bone marrow plus spleen cells co-cultured 24 hrs with CBA/J newborn suppressor cells were long-term survivors, living >60 days

PAGE 51

Table 1. Suppression of lethal GVH disease by newborn spleen cells in lethally irradiated (C57BL/10 x BlO.BR/cd)Fl hosts reconstituted with semiallogeneic BlO.BR/cd bone marrow cells. Reconstituting cell population None BlO.BR/cd adult bone marrow cells and spleen cells Adult BlO.BR/cd bone marrow cells and spleen cells co-cultured 24 hrs with CBA/J newborn spleen cells Survival times % Long-term (days) survivors 9 I 10 I 10 0 % 12, 14, 15 0 % 8, 8, 8, 8, 8, 8, 13 16 0 % 8, >60, >120, >170, >190 80 % 8, 8, >35, >55, >103, >103, >369, >375, >375, 535, 556 80 % > indicates the animal was sacrificed for characterization. 46

PAGE 52

(Figure 1). Long-term surviving hosts have been followed as long as 18 months after reconstitution before being sacrificed for functional studies. These data show that newborn suppressor splenocytes are capable of mediating successful development of semi-allogeneic bone marrow engrafted hosts. Newborn spleen cells prevent GVH disease in hosts engrafted with allogeneic bone marrow. The ability to suppress GVH disease in fully allogeneic bone marrow reconstituted hosts was then examined. In a second set of experiments, lethally gamma-irradiated C57BL/6 hosts were reconstituted with BlO.BR/cd bone marrow cells. As shown in Table 2, all irradiation control animals, injected with PBS but no reconstituting hematopoietic cells, died within 16 days. Similarly, animals injected with untreated BlO.BR/cd bone marrow plus spleen cells died within 19 days, with one animal dying at 27 days. Under the experimental and environmental conditions used 90-100% of C57BL/6 mice reconstituted with syngeneic bone marrow plus spleen cells survived long-term, whereas 55-60% of hosts reconstituted with fully allogeneic B10.BR/cd bone marrow plus spleen cells following co-culture with CBA/J newborn suppressor cells were long term survivors (Figure 1 and Table 1). These data indicate that the newborn spleen-associated suppressor cells are also capable of suppressing GVH disease in fully allogeneic engrafted hosts. 47

PAGE 53

Figure 1. Cumulative survival curves of reconstituted and non-reconstituted lethally irradiated hosts. a) Lethally irradiated (C57BL/10 x B10.BR/cd)F1 and C57BL/6 mice were reconstituted with syngeneic bone marrow (filled triangles) or injected with PBS but not reconstituted (open circles). b) Lethally irradiated (C57BL/10 x B10.BR/cd)F1 mice were reconstituted with untreated (open triangles) or newborn spleen cell pretreated (filled circles) semiallogeneic BlO.BR/cd bone marrow plus spleen cells. c) Lethally irradiated C57BL/6 mice were reconstituted with untreated (open triangles) or newborn spleen cell pretreated (closed circles) fully allogeneic BlO.BR/cd bone marrow plus spleen cells.

PAGE 54

SUMVAL RATES A. SYNGOEIC RECONSTITUT10N 100 o 0--------------------------,,.. .-tltuted o i 70 o .. 40 Ill ao ....-.-tltuted ,o 10 ,o 30 ,o 70 ,o ,o 100 110 120 TIME CDAYS) B. SEMI-ALLOGENEIC RECONSTITUTION ,o 70 ~-----~----~~'----------~*'-"*-------,,-f o en ,o 40 Ill ,o 10 newborn aplenic rnonocytOI ,o 30 ,o TIME (DAYS) C. ALLOGENEIC RECONSTITUTION 100 00 ,o 70 .J + newborn aplenic rnonocytea 100 110 ,o ----~*"---~*---"'*--*"---*"----'*'----------1.1,o 40 + newborn aplenic rnonocytu Ill 30 20 newborn aplenlc rnonocyt .. 10 ,o 30 ,o ,o ,o ,o 100 120 TIME CDAYS) 49

PAGE 55

Table 2. Suppression of lethal GVH disease by newborn spleen cells in lethally irradiated C57BL/6 hosts reconstituted with semi-allogeneic BlO.BR/cd bone marrow cells. Reconstituting cell population None BlO.BR/cd adult bone marrow cells and spleen cells Adult BlO.BR/cd bone marrow cells and spleen cells co-cultured 24 hrs with CBA/J newborn spleen cells Survival times (days) 3, 3, 4, 10, 10, 11, 11, 11, 12, 12, 13, 13, 13, 14 5, 6, 11, 12, 13, 16 9, 11, 11 7, 7, 9, 9 7, 11, 15, 19, 27 8, 9, 11, >21 >77, >87 7, 9, 9, 166, >323, >375 17, 20, >15, >33, >55 I 195 I 350 > indicates the animal was sacrificed for characterization. % Long-term survivors 0% 0% 0 % 0 % 0 % 50 % 50 % 71 % 50

PAGE 56

Histological examination of experimental mice. Physical examinations and histological studies were carried out to confirm that death of experimental animals following reconstitution resulted from GVH disease and not irradiation, per se. As depicted in figure 2, experimental mice which failed to thrive following bone marrow reconstitution showed signs of severe GVH disease: emaciation, exaggerated hunched appearance, patches of hair loss, sloughing of the tails, development of diarrhea, and hypothermia. Photomicrographs of histology sections from skin biopsies as well as liver and spleen sections are presented in Figures 3-5. Irradiation control mice showed fairly normal skin biopsies (Figure 3a) with the exception of noticeable loss of fat cells between connective tissue of the dermis and muscle. Similarly, liver histology appeared normal (Figure 3b). However, spleens from irradiation control mice showed severe generalized cytopenia in both white and red pulp, a collapsed architecture disrupting the white and red pulp areas, hemosiderin, and markedly visible connective tissue septae (Figure 3c & 3d). Experimental mice which where successfully reconstituted with bone marrow pretreated with newborn suppressor cells displayed normal skin (Figure 4a), liver (Figure 4b), and spleens (Figure 4c & 4d). Occasionally, slight leukocytic cell infiltration could be seen around portal tracts of the livers. 51

PAGE 57

Figure 2. Physical appearance of mice undergoing lethal GVH disease. Mice with GVH disease appear emaciated and maintain an exaggerated hunched position. Some mice also exhibit patches of hair loss and sloughing of tails.

PAGE 58

53 GVH CONTR O L

PAGE 59

Figure 3. Photomicrographs of H&E stained histology sections from irradiation control mice which were lethally irradiated and injected with PBS but not reconstituted with hematopoietic progenitor cells. a) Skin biopsies revealed fairly normal tissue morphology with the exception of a paucity of normal skin appendages and a loss of fat cells between the dermis connective tissue and muscle. (250 X magnification). b) Livers from irradiation control mice exhibited normal tissue morphology. (100 X magnification). c) Spleens from irradiation control mice exhibited severe cytopenia throughout the white and red pulp regions. (250 X magnification). d) The architecture of the spleens had a collapsed appearance and markedly visible connective tissue septae. (100 X magnification).

PAGE 60

55 A B

PAGE 61

56 C D Figure 3--continued.

PAGE 62

Figure 4. Photomicrographs of H&E stained histology sections from experimental mice which where successfully reconstituted with bone marrow plus spleen cells pretreated with newborn suppressor cells. a) Histologic sections of skin biopsies exhibited normal tissue morphology and skin appendages. {250 X magnification). b) Liver sections exhibited normal tissue morphology. {100 X magnification). c) Spleens sections from successfully reconstituted mice had re-established normal tissue morphology. (250 X magnification). d) The architecture of spleens from successfully reconstituted mice appeared normal with distinct red and white pulp areas.

PAGE 63

58 A B

PAGE 64

59 C D Figure 4--continued.

PAGE 65

Figure 5. Photomicrographs of H&E stained histology sections from experimental mice reconstituted with untreated bone marrow plus spleen cells. a) Skin biopsies from lethally irradiate mice reconstituted with untreated bone marrow plus spleen cells exhibited an atrophic epidermal layer with a hyperkeratotic surface and focal infiltrates within the epithelium. (250 X magnification). b) Liver sections exhibited many inflammatory cell infiltrates within the portal tracts. (250 X magnification). c) Fibrosis associated with inflammatory cell infiltrates and necrosis at the periphery of fibrotic regions was observed in a few liver sections. (250 X magnification) d) Spleen sections exhibited expanded and delineated white pulp regions as well as lymphocyte depletion. (250 X magnification). e) The architecture of the spleens appeared collapsed with no discernable red and white pulp regions. (100 X magnification).

PAGE 66

61 A

PAGE 67

62 B C Figure 5--continued.

PAGE 68

tinued. s--con Figure 63 D E

PAGE 69

In contrast, skin biopsies from mice reconstituted with untreated bone marrow plus spleen cells revealed an atrophic epidermal layer and a hyperkeratotic surface as well as a loss of skin appendages (Figure 5a). Upon closer examination, there were focal lymphocyte infiltrates within the surface epithelium and there were increased mononuclear cell numbers in and around blood vessels and the remains of skin appendages. The livers (Figure 5b & 5c) contained marked inflammatory lesions: inflammatory cell infiltrates were evident within the portal tracts whereas focal cell infiltrates were present in the sinusoids of the lobules and associated with patchy hepatocyte necrosis. A few livers showed fibrosis with inflammation and necrosis at the periphery of fibrotic regions. Spleens from these mice (Figure 5d & 5e) had expanded and delineated white pulp regions which contained cells with morphologic characteristics of blast transformation. At the time of severe cachexia, the white pulp showed lymphocyte depletion. The red pulp had collapsed, and Russell bodies, septae and hemosiderin were visible. The results from histological analysis indicated that lethally irradiated mice transplanted with untreated allogeneic bone marrow plus spleen cells had developed GVH disease. Furthermore, pretreatment with newborn spleen cells appeared to suppress the development of GVH disease resulting in successful engraftment and reconstitution of lethally irradiated hosts. 64

PAGE 70

Full chimerism in long-term survivors. Serological analysis of spleen cells from long-term surviving hosts was carried out to confirm that they were bone marrow chimeras. Specific antisera against donor and host histocompatibility antigens were used in complement mediated cytotoxicity assays to determine the origin of the hematopoietic cells within the reconstituted hosts. Data presented in Table 3 show that the surviving semi-allogenic and allogeneic bone marrow reconstituted hosts were fully chimeric. The leukocyte populations residing in the spleens of (C57BL/10 x B10.BR/cd)F1 hosts reconstituted with B10.BR/cd, as well as the C57BL/6 hosts reconstituted with B10.BR/cd, serotyped positive for H-2k but negative for H-2b indicating that the hematopoietic cells were of donor origin. Functional studies on splenocytes from long-term survivors. Semi-allogeneic and allogeneic bone marrow reconstituted hosts were killed at various times and their splenocytes assayed for immune reactivity to mitogenic stimulation and MLC reactivity (Figures 6 and 7). Cells from (C57BL/10 x B10.BR/cd)F1 recipients reconstituted with B10.BR/cd cells responded strongly to Con A and LPS stimulation, thus demonstrating immunocompetence of the T-cell and B-cell compartments at 60 days (Fig 6a). In addition, they responded to third-party alloantigens, e.g., DBA/2J (H-2d, Mlsa) and B10.SAA48 (H-2W3), while failing to 65

PAGE 71

66 Table 3. Serotyping of splenocytes from long-term surviving reconstituted hosts. Experimental group BlO.BR/cd C57BL/6 C57BL/6 reconstituted with syngeneic C57BL/6 bone marrow {C57BL/10 X BlO.BR/cd)Fl reconstituted with adult BlO.BR/cd bone marrow and spleen cells co-cultured with CBA/J newborn spleen cells C57BL/6 reconstituted with adult BlO.BR/cd bone marrow and spleen cells co-cultured with CBA/J newborn spleen cells % Cytotoxicity antiH-2k 85 % 16 % 15 % 96 % 80 % Normal antiH-2b mouse serum 21 % 25 % 99 % 80 % 15 % 13 % 15 % 15 %

PAGE 72

Figure 6. In vitro immune responses of spleen cells from lethally irradiated (C57BL/10 x B10.BR)F1 mice reconstituted with semiallogeneic B10.BR/cd bone marrow. a) Spleen cell responses following mitogenic stimulation with 25 gs/ml LPS (open circles) or 8 gs/ml Con A (filled squares). b) Mixed lymphocyte culture responses of spleen cells against gamma-irradiated B10.SAA48 (filled squares), DBA/2J (filled triangles), CBA/J (open triangles), C57BL/10 (filled circles), and B10.BR/cd (open circles) spleen cells.

PAGE 73

r() 'o )( a.. f(C57BL/IO Sn xBIO.BR/cd)F1 [810.BR/cd]--~) A. Mitogenic responses B. MLC responses 150 30 u 100 20 C 0 2 0 a. t.. 0 0 C Q) C 50 "O 10 E >. .i:: -I I rn 0 I I I I I I 0 24 48 72 96 Hours of culture 68

PAGE 74

Figure 7. In vitro immune responses of spleen cells from lethally irradiated C57BL/6 mice reconstituted with fully allogeneic B10.BR/cd bone marrow. a) Spleen cell responses following stimulation with 25 gs/ml LPS (filled circles, unbroken line) or in unstimulated culture (closed circle, dashed line). b) Mixed lymphocyte culture responses of spleen cells against gamma-irradiated DBA/2J (filled triangles), AKR/J (open triangles), BlO.Q (filled squares), C57BL/6J (filled circles), and CBA/H (open circles) spleen cells.

PAGE 75

70 C57BL/6J [BIO. BR/cd] A. Mitogenic responses B. MLC responses 60 60 r() 'o )( 40 40 a.. u C 0 C ... 0 a. ... 0 0 C Q) 20 20 C "O e >-.s::. -I I +medium r() -+-.t .. 0 0 I I I I I I 24 48 72 96 Hours of culture

PAGE 76

respond to BlO.BR/cd and C57BL/10 stimulator cells (Fig 6b). In addition, these cells responded against B10.RIII(71NS) (H-2r), B10.S (H-2s) and B10.BUA16 (H-2w22) stimulators (data not shown). These data indicate that the cells residing in the (C57BL/10 x B10.BR/cd)F1 chimeric hosts were tolerant to both donor and recipient histocompatibility antigens but not to third party alloantigens. This substantiates the immunocompetence of the T-cell compartment of these hosts. Similarly, C57BL/6 hosts reconstituted with B10.BR/cd bone marrow cells were assayed for mitogenic stimulation and MLC reactivity. Figure 7 shows data for mice sacrificed on day 90. The cells proved unresponsive to CBA/H and C57BL/6 cells indicating a tolerance to both donor and recipient haplotypes. In contrast, the cells were reactive to both LPS and histocompatibility alloantigens, e.g., expressed by BlO.Q (H-2q), AKR/J (H-2k, Mlsd) and DBA/2J (H-2d, Mlsa). Thus, these reconstituted hosts also exhibited immunocompetence of the T and B cell compartments. Discussion The results showed that lethally-irradiated hosts could be successfully reconstituted with semi-allogeneic (80%-90% long-term survival) or allogeneic (55%-60% long-term survival) bone marrow if the donor cells were first co-cultured 24 hours in the presence of newborn monocytes. 71

PAGE 77

Routine histological examination (Figure 8) of mice which had died or had been killed for functional testing revealed several points: 1) lethal irradiation caused complete destruction of the lymphoid compartment of the host, 2) hosts which had been reconstituted with histoincompatible bone marrow showed successful engraftment but subsequent development of severe organ and tissue lesions characteristic of classic graft-versus-host disease, and 3) hosts which had been successfully reconstituted with histoincompatible bone marrow pretreated with newborn suppressor cells showed normal lymphoid and tissue histology. Thus, pretreatment of histoincompatible donor bone marrow cells with newborn spleen-associated suppressor cells prevented the development of lethal GVH disease observed in host animals engrafted with untreated bone marrow cells. Functional studies using spleen cells from reconstituted chimeric hosts surviving >60 days showed that the cells were capable of proliferating in response to both T and B cell mitogens as well as histocompatibility alloantigens on third party cells. In contrast, these chimeric mice proved tolerant in their T cell responses to both donor and host histocompatibility antigens. Serotyping confirmed that the hosts were fully chimeric in that all the splenocytes expressed the donor histocompatibility antigens. These results are in marked contrast to those reported by 72

PAGE 78

Figure 8. Composite of photomicrographs for comparison of skin biopsies, liver and spleen sections from irradiation control mice (a-c), long-term surviving chimeras reconstituted with newborn suppressor cell treated allogeneic bone marrow (d-f), and allogeneic bone marrow-reconstituted animals undergoing GVH disease (g-i).

PAGE 79

74 SKIN LIVER SPLEEN

PAGE 80

Jadus and Peck (Jadus and Peck, 1984). In their study, cells from sublethally-irradiated, T cell engrafted hosts examined at day 60 post-engraftment remained tolerant to host cells but responded to donor and third-party cells. Furthermore, at day 60 all of the splenocytes serotyped as host cells. One interesting and unexpected result was the response of the chimeric (C57BL/10 x Bl0.BR/cd)F1 hosts toward adult CBA/J cells. Although syngeneic with the newborn suppressor-inducer population, CBA/J stimulated cells of the reconstituted hosts. This lack of tolerance may result from an insufficient quantity of CBA/J alloantigens present during the development of the chimeric immune system, or alternatively, tolerance toward non-MHC alloantigens, e.g., Mls, may never develop in such a protocol. This point needs further examination. Traditional protocols for bone marrow reconstitution in which bone marrow is "cleansed" the donor marrow of mature T-cells, primarily using anti-lymphocyte or anti-T cell antisera. Construction of semi-allogeneic parent~ F 1 hybrid chimeras through the pretreatment of donor bone marrow and spleen cells with anti-theta antiserum plus complement (Tyan, 1973; Sprent et al., 1975; Vallera et al., 1981; Thiefelder et al., 1983) has resulted in about 80-100% of the hosts surviving greater than 100 days. Such 75

PAGE 81

chimeric animals were totally populated with leukocytes of the donor strain and proved capable of either rejecting third-party skin allografts while remaining tolerant to skin grafts syngeneic with the donor mice or mounting an MLC response against third-party strains while showing no proliferative response against donor cells. Von Boehmer and co-workers (von Boehmer et al., 1975) obtained similar success rates in the production of tetra-parental chimeric mice. As would be predicted, construction of fully allogeneic chimeric mice using anti-lymphocyte antiserum has not proven as successful as in semi-allogeneic combinations: Theifelder and co-workers (Thiefelder et al., 1983) found only a 55% success rate while Vallera and co-workers (Vallera et al., 1981) had up to an 80% success rate. Successfully reconstituted hosts proved tolerant toward skin grafts of the donor while rejecting third-party unrelated grafts. In conclusion, bone marrow reconstitutions using either protocols in which the donor bone marrow is cleansed or pretreated with newborn suppressor-inducing cells appear similar. The major difference is that in the experimental protocol described herein, mature T cells have purposely been left in the reconstituting population, and have even been added without detrimental results. Thus, use of the 76

PAGE 82

newborn suppressor-inducer population in bone marrow reconstitution may provide us with a unique and natural method to control GVH disease even in allogeneic transplantations. 77

PAGE 83

PREVENTION AND TRANSFER OF AUTOIMMUNE DIABETES FOLLOWING ALLOGENEIC BONE MARROW TRANSPLANTATION Introduction Current evidence supports an autoimmune etiology for IDDM (Cahill and McDevitt, 1981; Eisenbarth, 1986) with the pancreatic beta (~) cell as the target for humoral and/or cell mediated responses. Since genes of the major histocompatibility complex (MHC) have been associated with susceptibility to IDDM (Platz et al., 1981; Cudworth and Wolf, 1982; Hattori et al., 1986) and have been shown to strongly influence immune responsiveness (Klein et al., 1981; Klein et al., 1983; Kaufman et al., 1984), it is possible that specific MHC molecules permit the development of aberrant immunity to autoantigens given the proper environmental conditions. NOD mice were utilized to more specifically examine the role of the immune system in the pathogenesis of autoimmune diabetes. NOD mice spontaneously develop insulitis (Figure 9) and a progressive destruction of the insulin producing~ cells within the islets of Langerhans leading to insulin dependent diabetes (Makino et al., 1980; Fujino-Kurihara et al., 1985). As depicted in Figure 10, there is a progressive lymphocytic infiltration starting with 78

PAGE 84

Figure 9. Photomicrograph of an H&E stained pancreatic histology section from an untreated NOD female mouse. Note the presence of insulitis within the pancreatic islet characteristic of autoimmune diabetes (400 X magnification). 79

PAGE 85

Figure 10. Photomicrographs of H&E stained pancreata from untreated NOD mice depicting the progression of lymphocytic infiltration (400 X magnification). a) Peri-islet lymphocytic infiltration abutting several clustered islets. b) Penetration of the pancreatic islet connective tissue sheath resulting in moderate insulitis. c) Increasing numbers of infiltrating cells resulting in the progression to severe insulitis. d) Advanced stage of insulitis is characterized by the destruction of the majority of the pancreatic islet cells and loss of normal islet architecture.

PAGE 86

81 A B

PAGE 87

82 C D Figure 10--continued.

PAGE 88

peri-islet infiltration, which advances to moderate insulitis as the infiltrating cells penetrate the pancreatic islet, and then to severe insulitis which results in the destruction of pancreatic islet. Several experimental protocols have recently been developed for the treatment and prevention of diabetes in NOD mice involving modulation of the immune system. Nonspecific immunosuppressive regimens such as sublethal irradiation (Harada and Makino, 1986), neonatal thymectomy (Hanafusa et al., 1986), and injection of anti-thymocyte antisera to deplete T-lymphocytes (Hanafusa et al., 1986) have all been shown to markedly reduce the incidence of diabetes in NOD mice. Bone marrow transplantation procedures have also been utilized to manipulate the immune system of NOD mice. Ikehara and co-workers carried out allogeneic bone marrow transplantations to prevent and/or treat insulitis and overt diabetes in NOD mice (Ikehara et al., 1985; Ikehara et al., 1987). In these studies, 5-6 month old NOD mice were lethally irradiated and reconstituted with bone marrow cells from BALB/cnu/nu mice. The allogeneic bone marrow reconstituted NOD mice showed normal glucose tolerance profiles at 3 months post-transplantation. Furthermore, histological examination showed that insulitis was not present in pancreatic islets of the allogeneic bone marrow reconstituted NOD mice. In contrast, 8 month old untreated 83

PAGE 89

NOD mice exhibited impaired glucose tolerance similar to that seen in IDDM. Thus, allogeneic bone marrow transplantation was sufficient to treat NOD mice which possessed sufficient numbers of pancreatic ~-cells to produce enough insulin for the prevention of glucose intolerance and hyperglycemia. However, NOD mice with overt diabetes required treatment with allogeneic bone marrow transplantation concomitant with pancreas transplantation (Yasumizu et al, 1987). In the present study, the newborn suppressor cell mediated bone marrow transplantation protocol discussed in the previous section was utilized to do reciprocal allogeneic bone marrow transplantations between diabetes susceptible NOD mice and non-susceptible strains of mice. The reconstitution of NOD mice with allogeneic bone marrow from diabetes non-susceptible strains of mice provided experimental animals with non-diabetogenic immune systems, within a diabetes prone host environment. Elements within the diabetes prone host environment include: the expression of potential MHC and non-MHC-linked diabetes susceptibility genes, appropriate auto-antigens capable of inducing an autoimmune response to pancreatic ~-cells, and a thymus that does not appear to delete or suppress the pertinent autoreactive clones during thymocyte development. Reconstitution of diabetes non-susceptible strains of mice 84

PAGE 90

with bone marrow from NOD mice provided experimental animals with diabetogenic immune systems in a normally nonsusceptible host environments. Bone marrow transplantation was carried out on female NOD mice at 7 weeks of age, prior to a large amount of pancreatic #-cell destruction, to determine if it would avert the progression of insulitis and prevent development of autoimmune diabetes. Likewise, the diabetes nonsusceptible strains of mice were reconstituted with NOD bone marrow at 7 weeks of age to determine if the diabetogenic immune system was sufficient for the development of autoimmunity between 2 months and 7 months post-transplantation. I have compared the newborn suppressor cell mediated protocol with one using a long-term bone marrow culture system initially developed by Dexter and co-workers (Dexter et al., 1977) as a source of hematopoietic stem cells for bone marrow reconstitutions. The long-term bone marrow culture conditions allow for the proliferation of hematopoietic progenitor cells to be maintained in vitro for up to 6 months. It takes approximately 6 weeks to establish a stable primary bone marrow culture during which time the cultured cells lose their immunoreactivity (Dexter and Spooncer 1978). The T-lymphocytes, which are responsible for the initiation of alloreactivity, do not survive in the long-term cultures. Therefore, it is possible to reconstitute a lethally 85

PAGE 91

irradiated mouse with allogeneic bone marrow culture cells without inducing lethal GVH disease (Spooncer and Dexter, 1983). Long-term bone marrow cultures, devoid of mature Tlymphocytes allowed us to control for the possibility that the transfer of autoimmunity to non-susceptible hosts was dependent on direct adoptive transfer of autoimmunity. Furthermore the use of an alternate source of hematopoietic progenitor cells allowed for us to control for the possible effects of newborn suppressor cells on autoimmunity. Results Normal onset of IDD in the NOD mouse following syngeneic bone marrow reconstitution. Female NOD mice, 7 weeks of age and prior to the onset of severe insulitis, were lethally irradiated and reconstituted with either freshly explanted NOD bone marrow cells that had been cocultured with newborn spleen cells or syngeneic NOD hematopoietic progenitor cells from long-term bone marrow cultures. Lethal irradiation was confirmed as all of the animals not reconstituted died within 15 days of irradiation. Of 13 NOD mice reconstituted with syngeneic bone marrow, all survived long-term ( > 2 months). 86 Starting 4 weeks post-transplantation, blood glucose levels were routinely screened using test-strips. Blood glucose levels greater than 200 mgs/dl, for 3 consecutive days, were considered to be diagnostic for overt diabetes. As

PAGE 92

shown in Figure 11, all 13 of the syngeneic reconstituted NOD mice developed overt diabetes within 26 weeks posttransplantation (blood glucose levels of 400-800 mgs/dl). Nine of the diabetic animals from this experimental group were sacrificed and their pancreata were examined histologically for the presence of insulitis. All nine of the syngeneic reconstituted NOD mice exhibited insulitis (Figure 12). As shown in Figure 13, there was a progression of lymphocytic infiltration from peri-islet infiltration, to moderate insulitis, and finally to severe insulitis resulting in the destruction of the pancreatic islets. The normal development of insulitis and progression to overt diabetes in the syngeneic reconstituted NOD mice indicated that the irradiation and bone marrow infusion procedures did not alter the pathogenesis of autoimmune diabetes. Furthermore, the same results were observed when NOD mice were reconstituted with newborn suppressor cell treated bone marrow or cultured bone marrow indicating that newborn spleen cells did not suppress the development of diabetes. Prevention of IDD in the NOD mouse with allogeneic bone marrow reconstitution. B10.BR/cd mice were chosen as bone marrow donors since they are fully allogeneic to NOD mice. Seven week old female NOD mice were lethally irradiated and reconstituted with B10.BR/cd bone marrow cells which had been co-cultured with newborn spleen cells to enhance engraftment and to suppress the development of GVH disease. 87

PAGE 93

Reconstitution: NOD NOD f BM reconstitution (7 wks of age) f TIME (wks post-transplantation) 0 4 8 12 16 20 24 Tested for elevated BGLs Time mice were killed: n n mm i i i ... Detection of insulitis ++ ++ +++ + + Onset of Diabetes: .A. .A. .A. .A. .A. .A. .A. .A. .A. .A. .A. .A. 88 28 Figure 11. Lethally irradiated NOD mice reconstituted with syngeneic hematopoietic stem cells at 7 weeks of age were routinely screened for elevated blood glucose levels (horizontal arrow) to detect overt diabetes. Individual mice were sacrificed at various time points during the experiment (vertical arrows) and examined histologically for the presence of insulitis (+ symbols represent individual animals in which insulitis was detected in pancreatic sections on H&E stained slides). The black triangles indicate the time point when mice developed overt diabetes.

PAGE 94

Figure 12. A photomicrograph showing a pancreatic islet from a lethally irradiated NOD mouse reconstituted with syngeneic NOD bone marrow. Note the cellular infiltrate characteristic of insulitis (H&E stained sections, 400 X magnification). 89

PAGE 95

Figure 13. Photomicrographs of H&E stained pancreata from NOD mice reconstituted with syngeneic NOD bone marrow mice depicting the progression of lymphocytic infiltration. a) Peri-islet lymphocytic infiltration abutting a pancreatic islet (400 X magnification). b) Penetration of the pancreatic islet connective tissue sheath resulting in moderate insulitis (250 X magnification. c) Increasing numbers of infiltrating cells resulting in the progression to severe insulitis (250 X magnification). d) Advanced stage of insulitis characterized by the infiltrating cell almost completely covering the pancreatic islet (400 X magnification).

PAGE 96

91 A B

PAGE 97

92 C D Figure 13--continued.

PAGE 98

Spleen cells from adult B10.BR/cd mice were not added to the reconstituting cell population since I wanted to increase the percentage long-term surviving reconstituted hosts. Of 14 mice reconstituted, 11 survived long-term (>2 months). As shown in Figure 14, all long-term survivors exhibited normal blood glucose levels (80-120 mg/dl) up to 6 months post-transplantation in a weekly screening with teststrips. At various times after 3 months posttransplantation, individual animals were killed for functional studies and histological examination. None of the pancreatic islets examined histologically showed the presence of insulitis (Figure 15). The use of young NOD mice as hosts for the reconstitution allowed us to avert the progression of insulitis before many pancreatic fi-cells were destroyed. Thus, bone marrow reconstitution of the NOD mouse using cells from the IDD non-susceptible B10.BR/cd mouse was sufficient to prevent the onset of insulitis, hyperglycemia and diabetes. Development of insulitis and diabetes in C57BL/6 and B10.BR/cd mice reconstituted with the NOD hematopoietic system. Diabetes non-susceptible C57BL/6 and B10.BR/cd mice were reconstituted with allogeneic bone marrow from NOD mice. Seven week old female C57BL/6 or B10.BR/cd mice were lethally irradiated and reconstituted with NOD bone marrow, which was previously co-cultured with newborn spleen cells. Animals which were irradiated but not reconstituted with 93

PAGE 99

Reconstitution: B 1 O.BR/ cd ... NOD f BM reconstitution (7 wks old) 0 4 Tested for elevated BGLs Time mice were killed: Detection of lnsulltls Onset of Diabetes: TIME (wks post-transplantation) 8 12 16 20 24 i i nt mi i i NONE 28 Figure 14. Lethally irradiated NOD mice reconstituted with allogeneic hematopoietic stem cells from BlO.BR/cd mice at 7 weeks of age were routinely screened for elevated blood glucose levels (horizontal arrow) to detect overt diabetes. Mice were sacrificed at various time points during the experiment (vertical arrows) and examined histologically for the presence of insulitis (-symbols mean insulitis was not detected in pancreatic sections on H & E stained slides), (* symbols indicate no histology). Note that none of the animals developed diabetes. 94

PAGE 100

Figure 15. Photomicrographs of H&E stained pancreatic islets from lethally irradiated NOD mice reconstituted with allogeneic BlO.BR/cd bone marrow. a) Examination of histological sections revealed healthy appearing pancreatic islets with no insulitis (400 X magnification). b) A 1000 X magnification of a second healthy appearing pancreatic islet with no signs of insulitis. 95 A

PAGE 101

96 B Figure 15--continued.

PAGE 102

bone marrow cells all died within 15 days of irradiation. Reconstituted mice were screened routinely for elevated blood glucose levels while at various time points individual animals were sacrificed for histological examination of pancreata to detect the presence of insulitis (Figure 16). Surprisingly, 100% of diabetes non-susceptible C57BL/6 and BlO.BR/cd mice reconstituted with a NOD hematopoietic system developed insulitis (Figures 16 and 17). Furthermore, one of the 9 reconstituted C57BL/6 mice progressed to overt diabetes as determined by a persistent elevated blood glucose level greater than 200 mg/dl. Characteristically, insulitis observed in these animals (Figure 18) involved a progressive infiltration of the pancreatic islets ranging from peri-islet infiltration to full-scale insulitis resulting in the destruction of the islet. However, the proportion of islets exhibiting infiltration was smaller and there were fewer islets with extensive insulitis compared with untreated NOD mice or NOD mice reconstituted with syngeneic bone marrow. The ability to induce insulitis and diabetes in C57BL/6 and BlO.BR/cd mice suggests that an autoantigen capable of inducing a response is present in the non-susceptible strains. These data also suggest that the expression of hematopoietic cell specific diabetogenic genes is sufficient to induce the autoimmune response, even within a diabetes resistant environment. 97

PAGE 103

NOD~ C57BL/6/ Reconstitution: and NOD~ B 1 O.BR/ cd f BM reconstitution (7 wks of age) f TIME (wks post-transplantation) 98 0 4 8 12 16 20 24 28 32 Tested for elevated BGLs Time mice were killed: i m mt i t Detection of lnsulitls ++ + ++ ++ + + + Onset of Diabetes: Figure 16. Lethally irradiated C57BL/6 and BlO.BR/cd mice reconstituted with allogeneic hematopoietic stem cells from NOD mice at 7 weeks of age were routinely screened for elevated blood glucose levels (horizontal arrow) to detect overt diabetes. Mice were sacrificed at various time points during the experiment (vertical arrows, long= C57BL/6 and short= BlO.BR/cd) and examined histologically for the presence of insulitis (+ symbols represent individual animals in which insulitis was detected in pancreatic sections on H&E stained slides),(* symbols indicate no histology). Black triangles indicate the time points when mice developed overt diabetes.

PAGE 104

Figure 17. Photomicrograph from a C57BL/6 mouse reconstituted with newborn suppressor cell pretreated NOD bone marrow. Note the lymphocytic infiltrate characteristic of insulitis present in the H&E stained pancreatic islet (400 X magnification). 99

PAGE 105

Figure 18. Photomicrographs of H&E stained pancreata from C57BL/6 mice reconstituted with allogeneic NOD bone marrow mice depicting the progression of lymphocytic infiltration. a) Peri-islet lymphocytic infiltration abutting two pancreatic islets (250 X magnification). b) Penetration of the pancreatic islet connective tissue sheath resulting in moderate insulitis (250 X magnification). c) Advanced stage of insulitis characterized by the infiltrating cell almost completely covering the pancreatic islet (400 X magnification).

PAGE 106

101 A B C

PAGE 107

102 Since T-lymphocytes were not depleted from the donor bone marrow prior to injection into lethally irradiated hosts, it was possible that the development of insulitis and diabetes in the reconstituted hosts was due to a direct passive transfer of autoreactive T-lymphocytes. To address this possibility, BlO.BR/cd mice were reconstituted with non-adherent cells from long-term bone marrow cultures derived from NOD mice. These modified Dexter cultures have been reported to be devoid of mature T-lymphocytes and do not induce lethal GVH disease when used as donors of hematopoietic progenitor cells for bone marrow transplantation (Dexter and Spooncer, 1978; Spooncer and Dexter, 1983). We have also confirmed the absence of mature T-lymphocytes in our cultures by indirect immunofluorescence using an anti-Thy-1 antisera and flow cytometric analysis, as shown in Figure 19. Therefore, donor derived T-lymphocytes would have to develop de novo within the long-term surviving reconstituted hosts. As shown in Figure 20, all 6 BlO.BR/cd mice reconstituted with long-term culture NOD bone marrow cells showed insulitis and one animal progressed to overt diabetes within the time frame of the experiment. Histological examination of pancreata (Figure 21) showed that the insulitis observed was similar to that seen in C57BL/6 and BlO.BR/cd mice reconstituted with NOD bone marrow pretreated with newborn spleen cells. The similar results obtained with the above

PAGE 108

103 A 300 Thymocytes 300 Bone marrow F"5 C 27. 251 SSC 6 254 F"LI 0 255 F"L2 0 255 101 FLI B 400 400 .. Thymocytes Cultured bone marrow tSC 69. t:'4 SSC 5. 65 F"Ll 0.255 tL.2 0 255 FL! FL! Figure 19. Flow cytometric analysis of the reconstituting cell populations using a fluorescent-activated cell sorter. Fluorescent staining intensity of (A) whole bone marrow cells and (B) long-term cultured bone marrow cells is compared with thymocytes. Measurements are of forward light scattering versus intensity of cell fluorescence following staining with an FITCconjugated anti-Thy-1 monoclonal antibody.

PAGE 109

Reconstitution: NOD_.. B 1 O.BR/ cd f BM reconstitution (7 wks of age) T TIME (wks post-transplantation) 0 4 8 12 16 20 24 28 Tested for elevated BGLs Time mice were killed: t t t t Tt Detection of lnsulitls + + + ++ Onset of Diabetes: 104 32 Figure 20. Lethally irradiated B10.BR/cd mice reconstituted with allogeneic hematopoietic stem cells from NOD long-term bone marrow cultures at 7 weeks of age were routinely screened for elevated blood glucose levels (horizontal arrow) to detect overt diabetes. Mice were sacrificed at various time points during the experiment (vertical arrows,) and examined histologically for the presence of insulitis (+ symbols represent individual animal in which insulitis was detected in pancreatic sections on H & E stained slides),(* symbols indicate no histology). Black triangles indicate the time points when mice developed overt diabetes.

PAGE 110

Figure 21. Photomicrographs of H&E stained pancreata from BlO.BR/cd mice reconstituted with allogeneic NOD long-tern bone marrow cultured cells. a) Penetration of the pancreatic islet connective tissue sheath resulting in moderate insulitis (400 X magnification. 105 A B b) more advanced penetration resulting in severe insulitis (400 X magnification).

PAGE 111

106 two experimental systems suggest that the development of the NOD immune system in a diabetes non-susceptible host environment still results in autoimmunity and that a direct adoptive transfer of autoimmunity apparently is not required. The frequency of progression to overt diabetes and the relative glucose tolerance in NOD reconstituted nonsusceptible strains of mice. The results above indicated that the progression to overt diabetes in C57BL/6 and B10.BR mice reconstituted with NOD bone marrow was less frequent than in NOD mice. However, we could not determine the actual frequency of the progression to overt diabetes in a time response study in which reconstituted mice were sacrificed at different times for characterization. Therefore, separate experiments were carried out in which C57BL/6 mice were reconstituted with an NOD hematopoietic system and all 10 mice were allowed to survive at least 30 weeks post-transplantation. Table 4 shows that only 1 of 10 NOD reconstituted C57BL/6 mice progressed to overt diabetes within 30 weeks post-transplantation. However, all ten mice showed the presence of insulitis on histological examination of the pancreata. In contrast, NOD animals reconstituted with a syngeneic NOD hematopoietic system all developed overt diabetes within 24 weeks post-transplantation. Since the lethal irradiation and bone marrow infusion

PAGE 112

107 Table 4. Determination of the frequency of progression to overt diabetes. Reconstitution % With Overt Diabetes NOD---> NOD 100% (13/13) NOD---> C57BL/6 10% (1/10) Time Elapsed Post-transplantation 12-24 weeks > 30 weeks Note: All animals examined histologically had insulitis.

PAGE 113

108 procedures do not prevent the development of insulitis and diabetes, these results show a dichotomy in the development of insulitis and the progression of overt diabetes in the NOD reconstituted diabetes non-susceptible hosts. Intraperitoneal glucose tolerance tests were carried out on randomly selected non-diabetic NOD reconstituted C57BL/6 mice at 30 weeks post-transplantation. Figure 22 shows that 3 of 3 control C57BL/6 mice reconstituted with B10.BR bone marrow had normal glucose tolerance profiles. In contrast, 1 of 3 C57BL/6 mice reconstituted with NOD bone marrow was glucose intolerant. These results indicated that while diabetes non-susceptible mice reconstituted with the NOD hematopoietic cell system are certain to develop moderate to severe insulitis, only a small portion of the animals actually progress to overt diabetes or show glucose intolerance. Serological analysis of the long-term surviving reconstituted hosts to confirm that they were bone marrow chimeras. Long-term surviving hosts, which were lethally irradiated and reconstituted with allogeneic bone marrow, were confirmed to be bone marrow chimeras in a complement mediated cytotoxicity assay using specific antisera against host and donor histocompatibility antigens. Serological typing of spleen cells revealed the hematopoietic system to be of donor origin in all of the experimental animals examined (Table 5).

PAGE 114

109 GLUCOSE TOLERANCE TEST B 10.BR/cd ... C57BL/6 f 450--~~~~~~~-NOD ... C57BL/6 f 400 Gi 350 > G) ...1300 G) 0 0 g250 CJ "tS g 200 co 150 100 4' .. . .. : . . . . . 50 ........ ~--........ ~-----i.~_.,_, initial 15 30 60 120 180 Time (min.) G) > G) ...I G) 0 0 0 3 CJ "tS 0 0 co 450 400 [iJ--.-. ' 350 ti 300 ' ' 250 ' r,;i 200 ' ' 150 ' : ... 100 .. 6 I 50--~--~--~--~-----initial 15 30 60 120 180 Time (min.) Figure 22. Intraperitoneal glucose tolerance tests on lethally irradiated C57BL/6 mice reconstituted with bone marrow from diabetes non-susceptible B10.BR/cd and diabetes prone NOD mice. Each line represents the glucose tolerance profile from individual animals chosen at random.

PAGE 115

110 Table 5. Serotyping of splenocytes from long-term surviving reconstituted hosts. Experimental group NOD BlO.BR/cd NOD mice reconstituted with syngeneic NOD hematopoietic cells BlO.BR/cd mice reconstituted with NOD hematopoietic cells NOD mice reconstituted with BlO.BR/cd hematopoietic cells Experimental group NOD C56BL/6 C57BL/6 mice reconstituted with NOD hematopoietic cells % Cytotoxicity antiH-2k 27 % 82 % 30 % 35 % 97 % Normal antiH-2b mouse serum 96 % 32 % 14 % 15 % 94 % 95 % 30 % 19 % 16 % Normal antiH-2b antiH-2Kd mouse serum 10 % 88 % 11 % 61 % 9 % 75 % 20 % 25 %

PAGE 116

111 Functional studies of splenocytes from long-term survivors. Long-term surviving bone marrow reconstituted hosts were sacrificed at various times and their splenocytes assayed for immune reactivity to mitogenic stimulation and MLC reactivity. Data from functional studies on NOD mice reconstituted with syngeneic or allogeneic bone marrow and from BlO.BR/cd mice reconstituted with NOD bone marrow are presented in Tables 6-8. Spleen cells from syngeneic reconstituted NOD mice responded strongly to mitogenic stimulation by Con A and LPS, thus demonstrating immunocompetence of the T-cell and B-cell compartments of the immune system (Table 6). Further studies indicated that these spleen cells were responsive to third-party histocompatibility alloantigens, i.e., BlO.BR/cd and BlO.RIII spleen cells, in mixed lymphocyte cultures but were tolerant to syngeneic (NOD) stimulator cells. Spleen cells from NOD mice reconstituted with allogeneic BlO.BR/cd bone marrow were capable of responding to both T-cell and B-cell mitogens as well as to thirdparty histocompatibility alloantigens, i.e., PL\J, in mixed lymphocyte cultures. However, the spleen cells were tolerant to host (NOD) and donor (BlO.BR/cd) alloantigens (Table 7). Similarly, BlO.BR/cd mice reconstituted with allogeneic NOD bone marrow were responsive to mitogenic stimulation and to third-party alloantigens in mixed lymphocyte cultures but were tolerant to both host and donor

PAGE 117

112 Table 6. Functional reactivities of spleen cells from longterm surviving bone marrow chimeras. NOD mice reconstituted with syngeneic NOD hematopoietic cells Stimulating agent Medium LPS (25 g/ml) Con A (8 g/ml) NOD spleen cells B10.BR Spleen cells B10.RIII SPLeen cells 3H-TdR Incorporation (Mean CPM S.D.) 5823 896 15681 6196 49096 7674 2185 276 17447 332 11606 0

PAGE 118

113 Table 7. Functional reactivities of spleen cells from longterm surviving bone marrow chimeras. NOD mice reconstituted with allogeneic BlO.BR/cd hematopoietic cells Stimulating agent Medium LPS (25 g/ml) Con A (8 g/ml) BlO.BR/cd spleen cells NOD spleen cells PL/J spleen cells 3H-TdR Incorporation (Mean CPM S.D.) 10924 779 41733 6339 73563 21529 4136 402 5197 1382 13647 208

PAGE 119

114 Table 8. Functional reactivities of spleen cells from longterm surviving bone marrow chimeras. BlO.BR/cd mice reconstituted with allogeneic NOD hematopoietic cells Stimulating agent Medium LPS (25 g/ml) Con A (8 g/ml) NOD spleen cells BlO.BR Spleen cells BlO.RIII SPLeen cells 3H-TdR Incorporation (Mean CPM S.D.) 3547 185 9635 3158 11037 567 3769 494 4840 99 12016 735

PAGE 120

alloantigens (Table 8). These results indicate that the immune systems of allogeneic bone marrow chimeras were immunocompetent and were tolerant to histocompatibility antigens of both recipient and reconstituting cell populations. 115 Immunogenetic studies to examine the role of the H-2 complex in the development of insulitis. Bone marrow transplantations were done using various strains of mice as donors and recipients to examine the role of the H-2 complex in the development of insulitis. In the first set of experiments, various strains of mice with different haplotypes were lethally irradiated and reconstituted with bone marrow from NOD mice (Figure 23). As expected, lethally irradiated NOD mice reconstituted with syngeneic bone marrow all developed insulitis. Similarly, lethally irradiated C57BL/6 mice of the b-haplotype, B10.BR/cd mice of the k-haplotype, (B10.BR/cd x NOD)Fl mice of the k/NOD haplotype and B10.GD mice of the g2-haplotype all developed insulitis when reconstituted with NOD bone marrow. These results indicate that the MHC haplotype of the bone marrow recipient does not alter the ability to transfer insulitis with the hematopoietic system of the NOD mouse. In the next set of experiments, lethally irradiated NOD mice were reconstituted with bone marrow from various strains of mice with different haplotypes (Figure 24). NOD mice reconstituted with bone marrow from B10.BR/cd of the

PAGE 121

116 Bone f Marrow ... -p' Recipient Donor Recipient haplotype lnsulitis K I-A 1-E D NOD ... NOD7 d nod b + NOD ... C57BL/6 f b b b + NOD ... B10.BR f k k k k + NOD ... (BRxNOD)F 1f k/d k/nod k k/b + NOD ... B10.GD 7 d d b + Figure 23. Immunogenetic studies to examine the role of the H-2 complex in the development of insulitis. Various strains of mice with different haplotype were lethally irradiated and reconstituted with NOD hematopoietic cells. Long-term surviving hosts were examined histologically for the presence of insulitis.

PAGE 122

117 Bone f Marrow .... Recipient Donor Recipient haplotype lnsulitis K I-A 1-E D NOD ... NODf d nod b + B 10.BR _. NoDf k k k k C57BL/6 ., NoDf b b b B10.GD .... NoDf d d b ( 1st 4 mo.) + (after 4 mo.) Figure 24. Immunogenetic studies to examine the role of the H-2 complex in the development of insulitis. NOD mice were lethally irradiated and reconstituted with hematopoietic cells from various strains of mice with different haplotypes. Long-term surviving hosts were examined histologically for the presence of insulitis.

PAGE 123

118 k-haplotype and C57BL/6 mice of the b-haplotype did not develop insulitis or diabetes. However, a different set of results was observed when NOD mice were reconstituted with hematopoietic cells from B10.GD mice, which have an MHC haplotype very similar to that of NOD mice (the B10.GD haplotype differs from that of NOD mice only at the I-A~ class II gene). There was a lack of insulitis during the first 4 months post-transplantation followed by a delayed response of insulitis in all animals examined thereafter (Figures 24 and 25). These results suggest that the overall MHC environment of the donor hematopoietic system affects the development of insulitis in the NOD mouse. Serotyping of spleen cells was carried out on the additional long-term surviving reconstituted hosts from the immunogenetic studies (Table 9 and Figures 26 & 27). Complement mediated cytotoxicity assays with allotypic antisera were used (Table 9) to distinguish donor and host hematopoietic derived cells in combinations with class I and class II disparity. However, long-term survivors from reciprocal bone marrow transplantations between NOD and B10.GD mice were serotyped using flow cytometric analysis since they differ only at class II alleles (Figures 26 & 27). The fluorescence activated cell sorter allowed the distinction of cells which did not bind with the specific antisera from those which do not express class II molecules.

PAGE 124

Figure 25. 119 A Photomicrographs of pancreata from lethally irradiated NOD mice reconstituted with BlO.GD bone marrow (400 X magnification, H&E stained). B a) No insulitis was observed in animals characterized during the first 4 months post-transplantation. b) Insulitis was observed in animals examined after 4 months post-transplantation indicating a delayed response.

PAGE 125

120 Table 9. Serotyping of splenocytes from long-term surviving reconstituted hosts. Experimental group NOD (BlO.BR/cd x NOD)Fl (BlO.BR/cd x NOD)Fl mice reconstituted with NOD hematopoietic cells Experimental group NOD C56BL/6 NOD mice reconstituted with NOD hernatopoietic cells % Cytotoxicity antiH-2k 14 % 66 % 20 % Normal antiH-2b mouse serum 86 % 20 % 64 % 64 % Normal antiH-2b antiH-2Kd mouse serum 20 % 82 % 100 % 7 % 16 % 99 % 7 % 26 %

PAGE 126

A 2 0 0 Serotyping by flow cytometry using TIB 93 antisera B10.GD Fluorescence Intensity C 200 c > 0 .0 E i B NOD_. B10.GD/ Fluorescence Intensit y 200 .I! C > 0 .0 E :, z NOD 011a0 101 102 Fluorescence Intensity Figure 26. Flow cytometric analysis of spleen cells to determine the origin of hematopoietic derived cells in bone marrow transplantation recipients. Fluorescent staining intensity of (A) recipient and (B) donor type cells were compared with spleen cells from BlO.GD mice reconstituted with NOD bone marrow. Measurements are of forward light scattering versus intensity of cell fluorescence following staining with a monoclonal antibody that binds NOD class II antigens. 121

PAGE 127

A 200 !! C .. > .. 0 .. .. -" E :, z Serotyping by flow cytometry using TIB 93 antisera Fluorescence Intensity fl i: .. > .. 0 .. .. -" E :, z B Fluorescence Intensity B10.GD~ NOD Fluorescence Intensity Figure 27. Flow cytometric analysis of spleen cells to determine the origin of hematopoietic derived cells in bone marrow transplantation recipients. Fluorescent staining intensity of (A) donor and (B) recipient type cells were compared with spleen cells from NOD mice reconstituted with BlO.GD bone marrow. Measurements are of forward light scattering versus intensity of cell fluorescence following staining with a monoclonal antibody that binds NOD class II antigens. 122

PAGE 128

123 The serotyping results indicated that the hematopoietic derived cells in all of the reconstituted host were of donor origin. The immunogenetic studies also enabled an examination of the possible role of I-E molecule expression in the development of diabetes in the NOD mouse. Nishimoto and coworkers (Nishimoto et al., 1987) suggested that the lack of I-E molecule expression in NOD mice may play a role in the development of autoimmune diabetes and that positive expression of I-E would play a protective role. Transgenic C57BL/6 mice harboring the I-E0d gene were bred with NOD mice and the resulting Fl progeny that expressed I-E were backcrossed to NOD. None of the backcross mice that expressed I-E developed insulitis. In contrast, 47% of backcross mice that expressed the I-Anod but lacked I-E molecule expression developed insulitis. The possibility that many other genes were transferred to the NOD genotype along with the I-E0 gene could not be accounted for. As shown in the immunogenetic transplantation studies (Figures 23 and 24), both C57BL/6 hosts which lack I-E molecule expression and BlO.BR/cd hosts which do express I-E developed insulitis when reconstitute with NOD bone marrow (Figure 23). Therefore, the expression of I-E molecules within the host environment was not sufficient to inhibit the development of insulitis. These observations prompted an examination of the development of autoimmunity in NOD

PAGE 129

mice reconstituted with C57BL/6 hematopoietic stem cells. In this set of chimeras, there was a hematopoietic system lacking I-E molecule expression in the NOD host environment which also lacked I-E molecule expression and contained all the background susceptibility genes for diabetes (Figure 24). No insulitis was observed in these mice. Together these results suggested that the expression of I-E or the lack of I-E molecule expression did not effect the development of autoimmunity. However, a full analysis of the possible effects of I-E molecule expression within the MHC environment of the NOD mouse would require the production of a transgenic NOD mouse harboring a functional I-Ea gene. Discussion The results from this study have confirmed the work by Ikehara and co-workers (Ikehara et al., 1985) which indicated that lethal irradiation of NOD mice and reconstitution with allogeneic bone marrow cells was sufficient to prevent diabetes. While the former study involved the treatment of 5-6 month old NOD mice, just prior to the onset of diabetes, the present experiments concerned the prevention of diabetes in young (7 week old) mice. The irradiation and bone marrow infusion procedures did not themselves affect the development of diabetes since all syngeneic bone marrow reconstituted NOD mice developed 124

PAGE 130

125 diabetes within 26 weeks post-transplantation. Collectively, the above results provide evidence that bone marrow replacement was enough to prevent the development of diabetes, suggesting an important role for the hematopoietic derived cells of the immune system. To determine if bone marrow from diabetes susceptible mice is sufficient to cause diabetes would require experiments in which non-susceptible strains of mice have been reconstituted with NOD bone marrow. Serreze and coworkers (Serreze et al., 1988) had recently carried out experiments to determine if adoptive transfer of (NOD x NON)Fl mice with NOD bone marrow could result in the transfer of diabetes to the normally resistant Fl animals. The (NOD x NON)Fl mice developed focal pancreatic lymphocytic infiltrates in perivascular and periductular regions as well as peri-islet infiltration; however the animals did not exhibit insulitis and remained diabetesresistant (Prochazka et al., 1987). In contrast, 4 week old (NOD x NON)Fl mice which had been lethally irradiated and injected intravenously with 50 x 106 untreated semiallogeneic NOD bone marrow cells developed diabetes in 6 of 8 NOD cases. These findings indicate that expression of certain diabetes susceptibility genes in hematopoietic derived cells can induce the development of diabetes in a host environment with 50% expression of background NOD susceptibility genes.

PAGE 131

In the present study, lethally irradiated C57BL/6 and BlO.BR/cd mice were used as diabetes non-susceptible recipients of allogeneic NOD bone marrow. C57BL/6 and BlO.BR/cd mice were utilize since they provide a host environment devoid of NOD susceptibility genes. Insulitis was observed in every C57BL/6 and BlO.BR/cd mouse reconstituted with NOD bone marrow but only 1 of 10 developed overt diabetes. Similar results were observed whether the source of donor hematopoietic progenitor cells contained or were devoid of mature T-lymphocytes. These results suggested that the development of NOD hematopoietic derived immune cells de novo in a diabetes non-susceptible host environment could result in the development of insulitis and occasionally diabetes. The results further suggest that a direct adoptive transfer by mature Tlymphocytes was not required. It is important to consider the possibility that the observed insulitis in non-susceptible strains of mice reconstituted with NOD hematopoietic cells was due to GVH reactivity since allogeneic bone marrow transplantation was involved. If the transfer of potential to develop 126 insulitis was actually due to an allospecific GVH response, then NOD mice reconstituted with BlO.BR bone marrow cells would exhibit insulitis as well. However, close examination of the pancreata from NOD mice reconstituted with BlO.BR/cd

PAGE 132

127 bone marrow cells mice indicated that mononuclear infiltration was minimal and, if present, restricted to the connective tissue regions supporting the microvasculature. Occasionally, a few focal infiltrating cells were observed in the acinar tissue. Rarely, an islet in close proximity to a connective tissue region, showed minor peri-islet infiltration. However, none of the infiltrating cells could be seen to have penetrated into the islet. Similar results were observed in pancreata from C57BL/6 mice which had been reconstituted with BlO.BR/cd bone marrow. These observations suggest that the insulitis observed in C57BL/6 and BlO.BR/cd mice reconstituted with hematopoietic cells from NOD mice was a pancreatic islet cell specific inflammatory response and not due to allospecific GVH reactivity. Such findings are in agreement with studies by Seemayer and co-workers (Seemayer et al., 1983) in which GVH disease was intentionally induced by intravenous injection of 50 x 106 semi-allogeneic lymphoid cells from spleen and lymph node. In their study, the majority of the pancreatic cellular infiltrates centered around microvasculature vessels in the connective tissue regions. Only pancreatic islets near invaded vascular regions showed infiltration of lymphocytes and usually only peri-islet infiltration was observed. Furthermore, no consistent changes in the number of pancreatic ~-cells were observed and the animals did not become diabetic.

PAGE 133

The ability to induce insulitis and occasionally diabetes in C57BL/6 and B10.BR/cd mice suggests that the hematopoietic derived immune cells from NOD mice were sufficient to induce anti-islet reactivity but may require the diabetogenic host environment to develop the frequency and severity of diabetes observed in NOD mice. The ability to transfer insulitis also suggests that if an autoantigen is involved in the inflammatory response it is present in diabetes non-susceptible strains of mice as well as in NOD mice. 128 The dependence for the prevention of diabetes and the transfer of diabetes on the source of hematopoietic progenitor cells suggests that the induction of tolerance to the diabetes autoantigen appears to be dependent on hematopoietic derived cells in the thymus. Thus, hematopoietic derived cells in the thymus of NOD mice and NOD bone marrow reconstituted mice may permit development of autoreactive clones specific for pancreatic antigens or fail to delete such clones. These results are consistent with previous studies which indicated that hematopoietic cells regulate tolerance induction by the thymus (von Boehmer and Schubiger, 1984; Marrack et al., 1988). The progression of insulitis to overt diabetes in C57BL/6 and B10.BR/cd mice reconstituted with an NOD immune system proved relatively low, about 10%. This result may be due to the lack of non-MHC-linked diabetes susceptibility

PAGE 134

genes expressed in the host environment of C57BL/6 and aaBlO.BR/cd mice. The low frequency of overt diabetes may also be due to different susceptibilities of the pancreatic islets to immunological damage or altered capacity of the immune cells to respond to antigen in the context of allogeneic histocompatibility antigens. 129

PAGE 135

SUMMARY AND FUTURE PERSPECTIVES Reciprocal allogeneic bone marrow transplantation between autoimmune diabetes prone NOD mice and diabetes non-susceptible strains provided a means for manipulating the immune system at the cellular level. Replacement of the hematopoietic progenitor cells for the immune system of NOD mice with bone marrow from non-susceptible strains of mice resulted in the prevention of autoimmunity and the development of diabetes. Conversely, replacement of the immune system from non-susceptible strains of mice with the hematopoietic system of NOD mice resulted in the development of autoimmune insulitis in every reconstituted mouse and diabetes in a relatively small percentage. These results emphasized the importance of hematopoietic derived cells in the development of autoimmune diabetes and suggested that genes expressed on immune cells play an important role in the pathogenesis. These results also suggest that hematopoietic derived cells within the thymus may play a more important role than epithelial cells for thymic education of tolerance for autoantigens. I propose two experimental systems to more specifically study these conclusions. 130

PAGE 136

A combination of transgenic methodology and allogeneic bone marrow transplantation may enable a more specific examination of which genes expressed by immune cells are involved in the pathogenesis of autoimmune diabetes. The production of founder transgenic mice that express a desired gene product could be utilized as a donor or recipient in bone marrow transplantation experiments. This combination of methods would allow an examination of specific genes expressed on hematopoietic derived cells on the pathogenesis of autoimmune diabetes. A good set of gene candidates for introduction into mouse embryos are the class II genes associated with diabetes susceptibility. 131 A combination of thymus transplantation and bone marrow transplantation may help elucidate the respective roles of hematopoietic derived cells and epithelial derived cells for education of tolerance induction in the thymus. Transplantation of thymus epithelium from diabetes susceptible animals or non-susceptible animals combined with bone marrow transplantation from the opposite source would allow the determination of which population is responsible for tolerance induction. The disadvantage of working with the above 2 experimental systems is that interpretation of results would be limited by the lack of a known antigenic target on pancreatic islet cells. The lack of a characterized autoantigen would limit the interpretation to the level of

PAGE 137

prevention or transfer of disease. The advantage of these systems is that there is an end-stage disease which can be measured and thus in vivo experimental results can be obtained. This is not readily available in experimental systems in which the antigens are well characterized. 132

PAGE 138

REFERENCES Acha-Orbea, H. and McDevitt, H. o. 1987. The first external domain of the nonobese diabetic mouse class II I-A fi chain is unique. Proceedings of the National Academy of Science 84:2435-2439. Argyris, B.F. neonatal mice. 1978. Suppressor activity in the spleen of Cellular Immunology 36:354-362. Argyris, B.F. 1981. Suppressor factor produced by neonatal mouse spleen cells. Cellular Immunology 62:412-424. Basset, M., Coons, T.A., Wallis, W., Goldberg, E.H., and Williams, R.C. 1977. Suppression of stimulation in mixed lymphocyte culture by newborn splenic lymphocyte in the mouse. Journal of Immunology 119:1855-1858. Bennett, P.H. 1983. Classification of Diabetes. In: Ellenberg, M. and Rifkin, H. (eds) Diabetes Mellitus, 3rd edition. Medical Examination Publishing Co., Inc. New York, pp 409-414. Boitard, C., Debray-Sachs, M., Pouplard, A., Assan, R., and Hamburger, J. 1981. Lymphocytes from diabetics suppress insulin release in vitro. Diabetologia 21:41-46. Bottazzo G.F., Dean, B.M., McNally J.M., MacKay E.H., Swift P.G.F. and Gamble, D.R. 1985. In Situ characterization of autoimmune phenomena and expression of HLA molecules in the pancreas in diabetic insulitis. New England Journal of Medicine 313:353-360. Bottazzo, G.F., Pujol-Borrell, R., and Hanafusa, T. 1983. Role of aberrant HLA-DR expression and antigen presentation in induction of endocrine autoimmunity. The Lancet 2:1115-1118. Cahill, G.F. and McDevitt, H.O. 1981. Insulin-dependent diabetes mellitus: the initial lesion. The New England Journal of Medicine 304:1454-1465. 133

PAGE 139

Chesnut, R.W., Colon, S.M., and Grey, H.M. 1982. Requirements for the processing of antigens by antigenpresenting B cells. The Journal of Immunology 129:2382-1079. Chesnut, R.W. and Grey, H.M. 1981. Studies on the capacity of B cells to serve as antigen-presenting cells. The Journal of Immunology 126:1075-1079. Click, R.E., Benck, L. and Alter., B.J. 1972. Immune responses in vitro. I. Culture conditions for antibody synthesis. Cellular Immunology 3:264-270. 134 Crabbe, M.F.C. 1987. The complications of diabetes. In: Crabbe, M.F.C. (ed) Diabetic Complications. Churchill Livingstone Inc., New York, pp 1-23. Cudworth, A.G., Wolf, E. 1982. The genetic susceptibility to Type I (insulin-dependent) diabetes mellitus. Clinics in Endocrinology and Metabolism 11:389-408. Deckert, T., Poulsen, J.E., and Larsen, M. 1978. Prognosis for diabetics with diabetes onset before the age of thirty-one. 1. Survival, Causes of Death, and Complications. Diabetologia 14:363-370. Desnick, R.J. 1987. Bone-marrow transplantation in genetic diseases: a status report. In: Frontiers in Genetic Medicine. Ross Laboratories, Columbus, Ohio, pp 71-87. Dexter, T.M., Allen, T.D. and Lajtha. L.C. 1977. Conditions controlling the proliferation of haemopoietic stem cells in vitro. Journal of Cell Physiology 91:335-344. Dexter, T.M. and Spooncer, E. 1978. Loss of immunoreactivity in long-term bone marrow culture. Nature 275:135-136. Dialynas, D.P., Wilde, D.B., Marrack, P., Pierres A., Wall K.A., Havran, W., Otten, G., Loken, M.R., Pierres, M., Kappler J., and Fitch, F.W. 1983. Characterization of the murine antigenic determinant, designated L3T4a, recognized by monoclonal antibody GKl.5: expression of L3T4a by functional T cell clones appears to correlate primarily with class II MHC antigen reactivity. Immunological Reviews 74:29-56. Dobersen, M.J., Scharff, J.E., Ginsberg-Fellner, F., and Notkins, A.L. 1980. cytotoxic autoantibodies to beta cells in the serum of patients with insulin-dependent diabetes

PAGE 140

mellitus. The New England Journal of Medicine 303:1493-1498. Dorman, J.S., Laporte, R.E., Kuller, L.H., Cruickshanks, K.J., Orchard, T.J., Wagener, D.K., Becker, D.J., and Cavender, D.E. 1984. The Pittsburgh insulin-dependent diabetes mellitus (IDDM) morbidity and mortality study, mortality results. Diabetes 33:271-276. 135 Eisenbarth, G.S. 1986. Type I diabetes mellitus, a chromic autoimmune disease. New England Journal of Medicine 314:1360-1368. Eisenbarth, G.S., Morris, M.A., and Scearce, R.M. 1981. Cytotoxic antibodies to cloned rat islet cells in serum of patients with diabetes mellitus. Journal of Clinical Investigation 67:403-408. Fein, F.S. and Scheuer, J. 1983. Heart disease in diabetes. In: Ellenberg, M. and Rifkin, H. (eds) Diabetes Mellitus, 3rd edition. Medical Examination Publishing Co., Inc. New York, pp 851-861. Foulis, A.K., Liddle, C.N., Farquharson, M.A., Richmond, J.A., and Weir, R.S. 1986. The histopathology of the pancreas in Type I (insulin-dependent) diabetes mellitus: a 25-year review of deaths in patients under 20 years of age in the United Kingdom. Diabetologia 29:267-274. Foulis, A.K. and Stewart, J.A. 1984. The pancreas in recent-onset Type I (insulin-dependent) diabetes mellitus: insulin content of islets, insulitis and associated changes in the exocrine acinar tissue. Diabetologia 26:456-461. Fujino-Kurihara, H., Fujita H., Hakura A., Nonaka, K., and Tarui, s. 1985. Morphological aspects on pancreatic islets of non-obese diabetic (NOD) mice. Virchows Archive Cell Pathology 49:107-120. Fujita, H. and Fujino-Kurihara, H. 1986. Fine structural aspects of the degenerating process of pancreatic islets of NOD mice. In: Tauri, S., Tochino, Y., and Nonaka, K. (eds) Insulitis and Type I Diabetes: Lessons From the NOD Mouse. Academic Press, Inc., Tokyo-Orlando-London, pp 51-60. Gale, R.P. 1982. Progress in bone marrow transplantation in man. Surveys in Immunological Research 1:40-66. Gepts, w. 1965. Pathologic anatomy of the pancreas in juvenile diabetes mellitus. Diabetes 14:619-633.

PAGE 141

136 Gepts, W. and Lecompte, P.M. 1981. The pancreatic islets in diabetes. The American Journal of Medicine 70:105-115. Gepts w. and Lecompte P.M. 1985. The pathology of Type I (Juvenile) diabetes. In: Volk B.W. and Arquilla, E.R. (eds) The Diabetic Pancreas. second edition. Plenum Medical Book Company, New York and London, pp 337-365. Hanafusa, T., Fujino-Kurihara, H., Miyazaki, ., Yamada, K., Nakajima, H., Miyagawa, J., Kono N., and Tarui, S. 1987. Expression of class II major histocompatibility complex antigens on pancreatic B cells in the NOD mouse. Diabetologia 30:104-108. Hanafusa, T., Pujol-Borrell, R., Chiovato, L., Russell, R.C.G., Doniach, D., Bottazzo, G.F. 1983. Aberrant expression of HLA-DR antigen on thyrocytes in Graves' disease: relevance for autoimmunity. The Lancet 2:1111-1115. Hanafusa, T., Nonaka, K., and Tarui, s. 1986. Immune pathogenesis of the NOD mouse: an overview. In: Tauri, S., Tochino, Y., and Nonaka, K. (eds) Insulitis and Type I Diabetes: Lessons From the NOD Mouse. Academic Press, Inc., Tokyo-Orlando-London, pp 75-82. Hanafusa, T., Sugihara, s., Fujino-Kurihara, H., Miyagawa J-I., Miyazaki, A., Yoshioka, T., Yamada K., Nakajima, H., Asakawa, H., Kono N., Fujiwara H., Hamaoka, T., and Tarui, S. 1988. Induction of insulitis by adoptive transfer with L3T4+ Lyt2T-lymphocytes in T-lymphocyte-depleted NOD mice. Diabetes 37:204-208. Harada M. and Makino, s. 1984. Promotion of spontaneous diabetes in non-obese diabetes-prone mice by cyclophosphamide. Diabetologia 27:604-606. Harada, M. and Makino, S. 1986. Immunological manipulation of diabetes production in NOD mice. In: Tauri, s., Tochino, Y., and Nonaka, K. (eds) Insulitis and Type I Diabetes: Lessons From the NOD Mouse. Academic Press, Inc., Tokyo-Orlando-London, pp 143-153. Hari, J., Yokono, K., Yonezawa, K., Amano, K., Yaso, S., Shii, K., Imamura, Y., and Baba, S. 1986. Immunochemical characterization of anti-islet cell surface monoclonal antibody from nonobese diabetic mice. Diabetes 35:517-522. Hattori, M., Buse, J.B., Jackson, R.A., Glimcher, L., Dorf, M.E., Minami, M., Makino, s., Moriwaki, K., Kuzuya, H., Imura, H., Strauss, W.M., Seidman, J.G., and Eisenbarth,

PAGE 142

137 G.S. 1986a. The NOD mouse: Recessive diabetogenic gene in the major histocompatibility complex. Science 231:733-735. Hattori, M., Ikegami, H., Adri M.N.S., and Watt, D.E. 1986b. Isolation of T lymphocyte lines from pancreatic islets of the NOD mouse. Diabetes 35(suppl.):76a. Henson, V., Maclaren, N., Winter, W., Riley, W., Rotter, J., and Wakeland, E.K. 1986. Molecular genetics of insulin-dependent diabetes mellitus. Molecular Biology and Medicine 3:129-136. Hooper, D.C. and Murgita, R.A. 1980. Evidence for two distinct neonatal suppressor cell populations. Federation Proceedings 39:354 A. Huang, S-H. and Maclaren, N.K. 1976. diabetes: a disease of autoaggression. Insulin-dependent Science 192:64-66. Ikehara, s., Nakamura, T., Sekita, K., Muso, E., Yasumizu, R., Ohtsuki, H., Hamashima, Y., and Good, R.A. 1987. Treatment of systemic and organ-specific autoimmune disease in mice by allogeneic bone marrow transplantation. Progress in Clinical and Biological Research 229:131-146. Ikehara, s., Ohtsuki, H., Good, R.A., Asamoto, H., Nakamura, T., Sekita, K., Muso, E., Tochino, Y., Ida, T., Kuzuya, H., Imura, H., and Hamashima, Y. 1985. Prevention of type I diabetes in nonobese diabetic mice by allogeneic bone marrow transplantation. Proceedings of the National Academy of Science 82:7743-7747. Ito, S. 1977. The Pancreas. In: Weiss, L. and Greep, R.O. (eds) Histology. fourth edition. McGraw-Hill Book Company, New York, pp 747-763. Jadus, M. and Peck, A.B. 1984. Naturally occurring spleen-associated suppressor activity of the newborn mouse. Requirement for two genetic restrictions in suppression of lethal graft-versus-host disease by newborn spleen cells. Scandinavian Journal of Immunology 20:81-91. Jadus, M. and Peck, A.B. 1986. Naturally occurring spleen associated suppressor activity of the newborn mouse. Biochemical and functional identification of three monokines secreted by newborn suppressor-inducer monocytes. Scandinavian Journal of Immunology 23:35-44. Johns, H.E. 1966. Radiobiology. In: The Physics for Radiology. Charles C. Thomas Publisher, Illinois, pp. 649-686.

PAGE 143

Kanazawa, K., Komeda, K., Sato, S., Mori, S., Akanuma K., and Takaku, F. 1984. Non-obese-diabetic mice: immune mechanisms of pancreatic ~-cell destruction. Diabetologia 27:113-115. 138 Kataoka, s., Satoh, J., Fujiya, H., Toyota, T., Suzuki, R., Itoh, K., and Kumagai, K. 1983. Immunologic aspects of the nonobese diabetic (NOD) mouse. Abnormalities of cellular immunity. Diabetes 32:247-253. Keens, H. and Ng Tang Fui, S. 1982. The definition and classification of diabetes mellitus. Clinics in Endocrinology and Metabolism 11:279-305. Koike, T., Itoh, Y., Ishii, T., Ito, I., Takabayashi, K., Maruyama, N., Tomioka, H., and Yosida, S. 1987. Preventive effect of monoclonal anti-L3T4 antibody on development of diabetes in NOD mice. Diabetes 36:539-541. Komeda, K. and Goto, N. 1986. Genetic monitoring of the NOD mouse. In: Tauri, s., Tochino, Y., and Nonaka, K. (eds) Insulitis and Type I Diabetes: Lessons From the NOD Mouse. Academic Press, Inc., Tokyo-Orlando-London, pp 11-22. Krolewski, A.S., Warram, J.H., Rand, L.I., and Kahn, C.R. 1987. Epidemiologic approach to the etiology of Type I diabetes mellitus and its complications. The New England Journal of Medicine 317:1390-1398. Kronenberg, M., Siu, G., Hood, 1986. The molecular genetics of and T-cell antigen recognition. 4: 529-91. L.E., and Shastri, N. the T-cell antigen receptor Annual Review of Immunology Lecompte, P.M. 1958. "Insulitis" in early juvenile diabetes. Archives of Pathology 66:450-457. Leiter, E.H., Prochazka, M., and Coleman, D.L. 1987. Animal model of human disease, the non-obese diabetic (NOD) mouse. American Journal of Pathology 128 (2):380-383. Lendrum, R., Walker, G., and Gamble, D.R. 1975. Isletcell antibodies in juvenile diabetes mellitus of recent onset. The Lancet 1:880-883. Lernmark, A., Freedman, Z.R., Hofmann, C., Rubenstein, A.H., Steiner, D.F., Jackson, R.L., Winter, R.J., and Traisman, H.S. 1978. Islet-cell-surface antibodies in juvenile diabetes mellitus. New England Journal of Medicine 299:375-380.

PAGE 144

139 Levin, M.E. and O'Neal, L.W. 1983. In: Ellenberg, M. and Rifkin, H. (eds) Diabetes Mellitus, 3rd edition. Medical Examination Publishing Co., Inc. New York, pp 803-828. Londei, M., Lamb, J.R., Bottazzo, G.F., and Feldmann, M. 1984. Epithelial cells expressing aberrant MHC class II determinants can present antigen to cloned human T cells. Nature 312:639-641. Maccuish, A.C., Barnes, E.W., Irvine, W.J., and Duncan, L.J.P. 1974. Antibodies to pancreatic islet cells in insulin-dependent diabetics with coexistent autoimmune disease. The Lancet 2:1529-1533. Maclaren, N.K., Huang, s-w., and Fogh, J. 1975. Antibody to cultured human insulinoma cells in insulin-dependent diabetes. The Lancet 1:997-1000. Makino, s., Kunimoto, K., Muraoka, Y., and Katagiri, K. 1981. Effect of castration on the appearance of diabetes in NOD mouse. Experimental Animals 30:137-140. Makino, S., Kunimoto, K., Muraoka, Y., Mizushima, Y., Katagiri, K. and Tochino, Y. 1980. Breeding of a nonobese, diabetic strain of mice. Experimental Animals 29 (1): 1-13. Makino, s., Muraoka, Y., Kishimoto, Y., and Hayashi, Y. 1985. Genetic analysis for insulitis in NOD mice. Experimental Animals 34:425-432. Marrack, P., Lo, D., Brinster, R., Palmiter, R., Burkly, L., Flavell, R.H. and Kappler, J. 1988. The effect of thymus environment on T cells development and tolerance. Cell 53:627-634. Maruyama, T., Takei, I., Matsuba, I., Tsuruoka, A., Taniyama, M., Ikeda, Y., Kataoka, K., Abe, M., and Matsuki, S. 1984a. Cell-mediated cytotoxic islet cell surface antibodies to human pancreatic beta cells. Diabetologia 26:30-33. Maruyama, T., Takei, I., Taniyama, K., Kataoka, K., and Matsuki, s. 1984b. Immunological aspect of non-obese diabetic mice: immune islet cell-killing mechanism and cell-mediated immunity. Diabetologia 27:121-123. Matsuba, I., Tsuruoka, A., Mori, Y., Sasaki, A., Ishii, K., Yamada, H., Tanese, T., Ikeda, Y., and Nishimura, M. 1986. Islet cell surface antibodies in the NOD mouse detected by indirect immunofluorescence assay using a fetal pancreatic p-cell line and rat islet cells. In: Tauri, s.,

PAGE 145

140 Tochino, Y., and Nonaka, K. (eds) Insulitis and Type I Diabetes: Lessons From the NOD Mouse. Academic Press, Inc., Tokyo-Orlando-London, pp 111-118. McMillan, D.E. 1983. Pathophysiology for diabetic macroand microvascular disease. In: Ellenberg, M. and Rifkin, H. (eds) Diabetes Mellitus, 3rd edition. Medical Examination Publishing Co., Inc. New York, pp 343-359. Miyazaki, A., Hanafusa, K., Yamada, J., Miyagawa, H., Fujino-Kurihara H., Nakajima, K., Nonaka and Tarui, s. 1985. Predominance of T lymphocytes in pancreatic islets and spleen of pre-diabetic non-obese diabetic (NOD mice: a longitudinal study. Clinical and Experimental Immunology 60:622-630. Miyazaki, A., Nonaka, K., and Tarui, s. 1986. Lymphocyte subsets in islets and spleen of the NOD mouse. In: Tauri, s., Tochino, Y., and Nonaka, K. (eds) Insulitis and Type I Diabetes: Lessons From the NOD Mouse. Academic Press, Inc., Tokyo-Orlando-London, pp 83-90. Murgita, R.A., Geidl, E.A., Kontiainin, s., Beverly, P.C.L., Wigzell, H. 1978. Adult murine T cells activated in vitro by alpha-fetoprotein and naturally occurring T cells in newborn mice. Identity in function and cell surface differentiation antigens. Proceedings of the National Academy of Science 75:2897-2901. National Diabetes Data Group. 1979. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 28:1039-1057. Nerup, J., Andersen, 0.0., Bendixen, G., Egeberg, J., Gunnarsson, R., Kromann, H., and Poulsen, J.E. 1974. Cellmediated immunity in diabetes mellitus. Proceedings of the Royal Society of Medicine 67:506-513. Nerup, J., Andersen, o.o., Bendixen, G., Egeberg, J., and Poulsen, J.E. 1971. Antipancreatic cellular hypersensitivity in diabetes mellitus. Diabetes 20:424-427. Nerup, J., Lernmark, A., and Scott, J. 1984. Autoimmunity. In: Gupta S. (ed) Immunology of Clinical and Experimental Diabetes. Plenum Medical Book Company, New York & London, pp 351-367. Neufeld, M., Maclaren, N.K., Riley, W.J., Lezotte, D., McLaughlin, J.V., Silverstein, J., and Rosenbloom, A.L. 1980. Islet cell and other organ-specific antibodies in

PAGE 146

U.S. caucasians and blacks with insulin-dependent diabetes mellitus. Diabetes 29:589-592. Norris, J.M., Dorman J.S., Rewers, M., and Porte, R.E. 1987. The epidemiology and genetics of insulin-dependent diabetes mellitus. Archives of Pathology and Laboratory Medicine 111:905-909. Notkins, A.L. 1979. The causes of Diabetes. Scientific American 241:62-72. Oldstone, M.B.A. 1988. Prevention of type I diabetes in nonobese diabetic mice by virus infection. Science 239:500-502. O'Reilly, R.J. 1983. Allogeneic bone marrow transplantation: current status and future directions. Blood 62:941-964. Peck, A.B. and Bach, F.H. 1973. a miniaturized mouse mixed leucocyte culture in serum-free and mouse serum supplemented media. Journal of Immunological Methods 3:147-164. 141 Peeler, K.L., Wigzell, H. and Peck, A.B. 1983. Isolation and identification of the naturally occurring, newborn spleen-associated suppressor cells. Scandinavian Journal of Immunology 17:443-453. Piguet, P.F., Irle, c., and Vassalli, P. 1981. Immunosuppressor cells from newborn mouse spleen are macrophages differentiating in vitro from monoblastic precursors. European Journal of Immunology 11:56-61. Platz, P., Jakobsen, B.K., Merling, N., Ryder, L.P., Svejgaard, A., Thomsen, M., Christy, M., Kromann, H., Benn, J., Nerup, J., Green, A., and Hauge, M. 1981. HLA-D and -DR antigens in genetic analysis of insulin dependent diabetes mellitus. Diabetologia 21:108-115. Prochazka M., Leiter, E.H., Serreze, D.V., and Coleman, D.L. (1987) Three recessive loci required for insulindependent diabetes in nonobese diabetic mice. Science 237:286-289. Ptak, W. and Skownon-Cendrzak, A. 1977. Fetal suppressor cells. Their influence on the cell-mediated immune responses. Transplantation 24:45-51. Rodriguez, G., Anderson, G., Wigzell, H., and Peck, A.B. 1979. Non T-cell nature of naturally occurring spleen-

PAGE 147

associated suppressor cells present in the newborn mouse. European Journal of Immunology 9:737-746. Rahier, J., Goebbels, R.M., and Henquin, J.C. 1983. Cellular composition of the human diabetic pancreas. Diabetologia 24:366-371. 142 Rosenbloom, A.L. 1983. Long-term complications of Type I (Insulin-dependent) diabetes mellitus. Pediatric Annals 12:665-685. Ross, C.A. and Pilarski, L.M. 1981. Suppressor T cells derived from early postnatal murine spleen inhibit cytotoxic T cell responses. Cellular Immunology 58:345-355. Rossini, A.A. and Chick, W.L. 1980. Microvascular pathology in diabetes mellitus. In: Kaley, G., and Altura, B.M. (eds) Microcirculation, Volume III. University Park Press, Baltimore, pp 245-271. Seemayer, T.A., Gartner, J.G., Colle, E. graft-versus host reaction in the pancreas. 35:72-77. 1983. Acute Transplantation Skowron-Cendrzak, A. and Ptak, W. 1976. Suppression of local graft-versus-host reactions by mouse fetal and newborn spleen cells. European Journal of Immunology 6:451-452. Spooncer, E., and Dexter, T.M. 1983. Transplantation of long-term cultured bone marrow cells. Transplantation 35:624-627. Sprent, J., von Boehmer, H. and Nabholz, M., 1975. Association of immunity and tolerance to host H-2 determinants in irradiated F 1 hybrid mice reconstituted with bone marrow cells from one parental strain. Journal of Experimental Medicine 142:321-331. Stewart-Brown, s., Haslum, M., and Butler, N. 1983. Evidence for increasing prevalence of diabetes mellitus in childhood. British Medical Journal 286:1855-1857. Takei, I., Maruyama, T., Taniyama, M., and Kataoka, K. 1986. Humoral immunity in the NOD mouse. In: Tauri, S., Tochino, Y., and Nonaka, K. (eds) Insulitis and Type I Diabetes: Lessons From the NOD Mouse. Academic Press, Inc., Tokyo-Orlando-London, pp 101-110. Tanaki, K., Stingl, G., Gullino, M., Sachs, D.H. and Katz,. S.I. 1979. Ia antigens in mouse skin are predominantly expressed on Langerhans cells. The Journal of Immunology 123:784-787.

PAGE 148

Tarui, S., Yamada, K., and Hanafusa, T. 1987. Animal models utilized in the research of diabetes mellitus: with special reference to insulitis-associated diabetes. Progress in Clinical and Biological Research 229:211-223. Thiefelder, s., Hoffmann-Fezer, G., Rodt, H., Doxiadis, I., Eulitz, M. and Kimmer, u. 1983. Antilymphocytic antibodies and marrow transplantation. VI. Absence of immune-suppression in vivo after injection of monoclonal antibodies blocking graft-versus-host reactions and humoral antibody formation in vitro. Transplantation 35:249-254. Tochino, Y. 1986. Discovery and breeding of the NOD mouse. In: Tauri, s., Tochino, Y., and Nonaka, K. (eds) Insulitis and Type I Diabetes: Lessons From the NOD Mouse. Academic Press, Inc., Tokyo-Orlando-London, pp 3-10. Toyota, T., Kataoka, s., Sato, J., Fujiya, H., Hayashida, Y., Goto, Y., and Kumagai, K. 1982. Islet-cell antibody and immunologic aspects of NOD mice. In: Mimura, G., Baba, S., Goto, Y., and Kobberling (eds) Clinico-Genetic Genesis of Diabetes Mellitus. Excerpta Medica, Amsterdam-OxfordPrinceton, pp 185-192. Toyota, T., Oya, K., Kataoka, S., Sato, J., Goto, Y., Suzuki, R., and Kumagai, K. 1986a. Characterization of abnormalities in immune response in the NOD mouse. In: Tauri, s., Tochino, Y., and Nonaka, K. (eds) Insulitis and Type I Diabetes: Lessons From the NOD Mouse. Academic Press, Inc., Tokyo-Orlando-London, pp 91-100. Toyota, T., Satoh, J., Oya, K., Shintani, s., and Okano, T. 1986b. Streptococcal preparation (OK-432) inhibits development of type I diabetes in NOD mice. Diabetes 35:496-499. 143 Tsuneo, F. and Yui, R. 1986. Histological changes in the pancreatic islets on NOD mice. In: Tauri, s., Tochino, Y., and Nonaka, K. (eds) Insulitis and Type I Diabetes: Lessons From the NOD Mouse. Academic Press, Inc., Tokyo-OrlandoLondon, pp 35-50. Tyan M.L. 1973. Prevention of graft-versus-host disease with antisera to the theta antigen or to whole serum. Transplantation 15:601-604. Unanue, E.R. 1981. The regulatory role of macrophages in antigenic stimulation. II. Symbiotic Relationship between lymphocytes and macrophages. Advances in Immunology 31:1-121.

PAGE 149

Volk, B.W. and Wellmann, K.F. 1985. Histology, cell types, and functional correlation of islets of Langerhans. In: Volk, B.W. and Arquilla, E.R. (eds) The Diabetic Pancreas. second edition. Plenum Medical Book Company, New York and London, pp 81-106. 144 Vallera, D.A., Soderling, C.C.B.V., Carlson, G.J., Kersey, J.H. 1981. Bone marrow transplantation across major histocompatibility barrier in mice. Transplantation 31:218-222. von Boehmer, H., Schubiger, K. 1984. Thymocytes appear to ignore class I major histocompatibility complex antigens expressed on thymus epithelial cells. European Journal of Immunology 14:1048-1052. von Boehmer, H., Sprent, J. and Nabholz, M. 1975. Tolerance to histocompatibility determinants in tetraparental bone marrow chimeras. Journal of Experimental Medicine 141:322-334. Wang, Y.I., Hao, L., Gill, R.G., and Lafferty, K.J. 1987. Autoimmune diabetes in NOD mouse is L3T4 T-lymphocyte dependent. Diabetes 36:535-538. Wicker, L.S., Miller, B.J., Coker, L.Z., McNally, S.E., Scott, S., Mullen, Y., and Appel, M.C. 1987. Genetic control of diabetes and insulitis in the nonobese diabetic (NOD) mouse. Journal of Experimental Medicine 165:1639-1654. Wicker, L.S., Miller, B.J., and Mullen, Y. 1986a. Transfer of autoimmune diabetes mellitus with splenocytes from nonobese diabetic (NOD) mice. Diabetes 35:855-860. Wicker, L.S., Miller, B.J., and Mullen, Y. 1986b. Transfer of autoimmune diabetes mellitus with T-cells from nonobese diabetic mice. Transplantation Proceedings 18:809-811. Wolf, E., Spencer, K.M., and Cudworth, A.G. 1983. The genetic susceptibility to Type I (insulin-dependent) diabetes: analysis of the HLA-DR association. Diabetologia 24:224-230. Yasumizu, R., Sugiura, K., Iwai, H., Inaba, M., Makino, S., Ida T., Imura, H., Hamashima, Y., Good, R.A., and Ikehara, s. 1987. Treatment of type 1 diabetes mellitus in non-obese diabetic mice by transplantation of allogeneic bone marrow and pancreatic tissue. Proceedings of the National Academy of Science 84:6555-6557.

PAGE 150

145 Yamada, K., Nonaka, K., Hanafusa, T., Miyazaki, A. Toyoshima, H., and Tarui, S. 1982. Preventive and therapeutic Effects of large-dose nicotinamide injections on diabetes associated with insulitis: An observation in nonobese diabetic (NOD) mice. Diabetes 31:749-753. Yokono, K., Amano, K., Suenaga, K., Hari, J., Shii, K., Yaso, s., Yonezawa, K., Imamura, Y., and Baba, s. 1986. Effect of antiserum to monoclonal anti-islet cell surface antibody on pancreatic insulitis in non-obese diabetic mice. Diabetes Research and Clinical Practice 1:315-321. Yokono, K., Shii, K., Hari, J., Yaso, s., Imamura, Y., Ejiri, K., Ishihara, K., Fujii, S., Kazumi, T., Taniguchi H., and Baba, S. 1984. Production of monoclonal antibodies to islet cell surface antigens using hybridization of spleen lymphocytes from non-obese diabetic mice. Diabetologia 26:379-385. Ziegler, K., and Unanue, E.R. 1981. Identification of a macrophage antigen-processing event required for I-regionrestricted antigen presentation to T lymphocytes. The Journal of Immunology 127:1869-1874.

PAGE 151

BIOGRAPHICAL SKETCH Drake Maurice La Face was born at the Naval Hospital in San Diego, California, on June 16, 1959. As a Navy dependent, Drake traveled with his parents to many different locations within the United States and overseas. Thus his pre-university education was quite varied and acquired from many different schools. Drake acquired his Bachelor of Science degree in Microbiology at the University of Central Florida in 1981. After one year of post-baccalaureate work at the University of South Florida, he enrolled in the Ph.D. program with the College of Medicine, Department of Pathology, at the University of Florida. He was admitted to Ph.D candidacy in the summer of 1984. The announcement was made at his wedding reception after his marriage to Marjorie Price. Drake then continued his dissertation research in the laboratory of Dr. Ammon B. Peck in the area of cellular immunology. In 1988, Drake accepted a post-doctoral position in the laboratory of Per Peterson at the Scripps Clinic and Research Foundation at La Jolla, California where he will do research in the area of molecular immunology. 146

PAGE 152

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Amm~._ Ee.ck-,--Cha-i.J;.m .... a ..... u..._ __ ~ociate Professor of Pathology and Laboratory Medicine I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Arthur K. Kimura Associate Professor of Pathology and Laboratory Medicine I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Noel K. Maclaren Professor of Pathology and Laboratory Medicine I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ~J'-~'t=r1-Sigurd J.ormann Professor of Pathology and Laboratory Medicine

PAGE 153

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. dtdwm.1~& I William E. Winter Assistant Professor of Pathology and Laboratory Medicine I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. mes Zuc ssociate Pr Medicine of This dissertation was submitted to the Graduate Faculty of the College of Medicine and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August 1988 ~1-rJ ~--k_";f--Dean, Graduate School


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ENSXEDJJX_AP8ESB INGEST_TIME 2017-05-22T19:16:33Z PACKAGE AA00054903_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES