<%BANNER%>

Reversal of Overt Type 1 Diabetes in the NOD Mouse through the Use of Combination Therapy

Permanent Link: http://ufdc.ufl.edu/UFE0022816/00001

Material Information

Title: Reversal of Overt Type 1 Diabetes in the NOD Mouse through the Use of Combination Therapy
Physical Description: 1 online resource (121 p.)
Language: english
Creator: Parker, Matthew
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: aat, atg, autoimmunity, beta, combination, diabetes, gcsf, hyperglycemia, insulin, januvia, lymphocyte, neupogen, nod, rapamycin, remission, reversal, scd26, sirolimus, sitagliptin, thymoglobulin
Immunology and Microbiology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The autoimmune destruction of the insulin-producing beta cells by autoreactive T lymphocytes results in a loss of blood glucose control and is known as type 1 diabetes (T1D). While exogenous insulin therapy has been available commercially since 1923, this treatment is not considered a cure as it is life-long and typically results in long-term complications including retinopathy, nephropathy, and various cardiovascular disorders. As such, there has been a desire to increase our understanding of disease pathogenesis in order to potentially prevent and/or cure overt diabetes. Many such studies have been carried out using animal models of T1D, perhaps most prolifically in the non-obese diabetic (NOD) mouse. After recent clinical trials in which prevention therapies have not proven efficacious, there has been an increasing desire for intervention therapies immediately post-onset. As such, the need to control the autoimmune response may be considered a critical component to these therapies. In these studies, anti-thymocyte globulin (ATG) was used immediately post-onset in NOD mice in order to halt the autoimmune destruction of the remaining beta cells as well as to attempt to induce a more tolerant immune phenotype. Only two doses of ATG were administered, resulting in a short-term T lymphocyte ablation. In addition, secondary compounds were added in combination with ATG in different stages of this project based upon the hypothesis that combination therapy targeting multiple pathways would prove more effective at inducing long-term disease remission than monotherapy. These studies will demonstrate that not only was ATG effective at inducing significant remission in new-onset NOD mice, but also that through combination therapy with granulocyte-colony stimulating factor (G-CSF) insulitis was diminished while beta cell area increased, as demonstrated through analysis of histology. In addition a more tolerant immune phenotype was induced as quantified through analysis of regulatory T cell populations.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Matthew Parker.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Atkinson, Mark A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022816:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022816/00001

Material Information

Title: Reversal of Overt Type 1 Diabetes in the NOD Mouse through the Use of Combination Therapy
Physical Description: 1 online resource (121 p.)
Language: english
Creator: Parker, Matthew
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: aat, atg, autoimmunity, beta, combination, diabetes, gcsf, hyperglycemia, insulin, januvia, lymphocyte, neupogen, nod, rapamycin, remission, reversal, scd26, sirolimus, sitagliptin, thymoglobulin
Immunology and Microbiology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The autoimmune destruction of the insulin-producing beta cells by autoreactive T lymphocytes results in a loss of blood glucose control and is known as type 1 diabetes (T1D). While exogenous insulin therapy has been available commercially since 1923, this treatment is not considered a cure as it is life-long and typically results in long-term complications including retinopathy, nephropathy, and various cardiovascular disorders. As such, there has been a desire to increase our understanding of disease pathogenesis in order to potentially prevent and/or cure overt diabetes. Many such studies have been carried out using animal models of T1D, perhaps most prolifically in the non-obese diabetic (NOD) mouse. After recent clinical trials in which prevention therapies have not proven efficacious, there has been an increasing desire for intervention therapies immediately post-onset. As such, the need to control the autoimmune response may be considered a critical component to these therapies. In these studies, anti-thymocyte globulin (ATG) was used immediately post-onset in NOD mice in order to halt the autoimmune destruction of the remaining beta cells as well as to attempt to induce a more tolerant immune phenotype. Only two doses of ATG were administered, resulting in a short-term T lymphocyte ablation. In addition, secondary compounds were added in combination with ATG in different stages of this project based upon the hypothesis that combination therapy targeting multiple pathways would prove more effective at inducing long-term disease remission than monotherapy. These studies will demonstrate that not only was ATG effective at inducing significant remission in new-onset NOD mice, but also that through combination therapy with granulocyte-colony stimulating factor (G-CSF) insulitis was diminished while beta cell area increased, as demonstrated through analysis of histology. In addition a more tolerant immune phenotype was induced as quantified through analysis of regulatory T cell populations.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Matthew Parker.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Atkinson, Mark A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022816:00001


This item has the following downloads:


Full Text

PAGE 1

REVERSAL OF OVERT TYPE I DIABETES IN THE NOD MOUSE THROUGH THE USE OF COMBINATION THERAPY By MATTHEW JOHN PARKER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Matthew John Parker 2

PAGE 3

To my parents, for their support and a ppreciation of education, and to my wife for her help while we both we nt through graduate school together. 3

PAGE 4

ACKNOWLEDGMENTS There are many people whom I wish to thank for making this research possible. Without their help, the quality or even the existence of this work w ould have been in question. I greatly appreciate the technical, intellectual, financial, and other help that has been provided for me over the years at the University of Florida. I must start by thanking my mentor, Dr. Mark Atkinson, for giving me the opportunity to perform research as an undergraduate, for hiring me for the eight months between my undergraduate and graduate programs, and for accepting me into his lab for my doctoral research. It has been an honor and a privilege to work with a giant in the field of type 1 diabetes research. His patience, unwavering support, and unique insight have provided me with an enjoyable graduate experience. Next, I must thank Clive Wasserfall for his cons tant help from day one in the lab. Working with someone who perhaps could be described as the 'coffee-drinking Yoda of immunology' to answer all of my questions has been invaluable. He has been a constant source of ideas and has greatly expanded my understandi ng of the field. I can't thank hi m enough for being effectively a second mentor to me. I also must thank many post-doctoral fellows who have worked with me over the years while in the Atkinson lab. Todd Brusko, currently a postdoc at UCSF, set the bar high as a graduate student and postdoc in the Atkinson lab and has been an outstanding role model to follow. His advice has stretched from my firs t days in the lab up to the writing of this dissertation. Brant Burkhardt, a former postdoc in the lab and current faculty member at Penn, was both patient and generous in allowing me to run his AAV experiment while I was just an undergrad. Clare Zhang provided excel lent guidance during my first da ys in the lab and aided my entrance in the IDP. Song Xue has worked side-by-side with me fo r a considerable time with her 4

PAGE 5

own animal studies. Comparing notes and excha nging assistance has made both of our projects that much easier. Finally, I must thank John Alexander who, despite only recently joining the lab, has already been a source of practical professional advice. My work in the Atkinson lab has been made much smoother due to the efforts of many other faculty members. I would like to extend my thanks to Dr. Campbell-Thompson and the Pathology Core for their endless work in pr oviding excellent histol ogy support. A practical clinical insight has been provide d by Drs. Des Schatz, Michael Haller, and Michael Stalvey that has kept the eventual goal of translational th erapy firmly grounded. Thanks to Dr. David Ostrov for allowing me to volunteer in his lab as an undergra duate and for providing an insight into the importance of x-ray crystallography and struct ure-based drug design. Dr. Clayton Mathews recently arrived at UF and has already helped out my project with hi s expertise in several experiments. My thanks also Neil Benson and the Flow Cytometry core for keeping the datagenerating FACScalibur running smoothly. Day-to-day work would have been much more di fficult and dull if it weren't for the help of countless administrators, students, volunteers, and technicians. Invaluable help with paperwork, animal work, cytokine analysis and other benchwork has been provided by Kim Young, Christy Popp, Lauren Steinberg, Joyce Connors, Fletcher Schwartz, Kieran McGrail, Marcus Moore, Peter Frances, Annie Song, Maigan Hulme, Cour tney Myhr, Sean McGrail, Jessica Greer, and Nathalie Anderson. Apart from the assistance within the lab, I was fortunate enough to have outstanding guidance from a premier committee whose members included Drs. Laurence Morel, Eric Sobel, Sihong Song, and Richard Johnson. Their insight, c onstructive criticism, and practical advice accelerated the pace of my research and ensured th at I was moving in the right direction. I would 5

PAGE 6

like to extend special thanks to Dr. Morel for allowing me to rotate th rough her lab during my first year of IDP where I was briefly able to experience research in the field of SLE. Last, but definitely not least, I would like to thank the members of my family for their lifelong support toward this goal. My dad, Martin, has led by example with his doctorate in physics and by eventually achieving the role of chairman over two departments. My mother, Sheila, has also led by example, acquiring two gr aduate degrees. My fraternal twin brother, David, has also been a constant source of s upport throughout my graduate work. His willingness and ability to suffer through long conversations about autoimmunity during his completion of law school has been astounding. I want to finish these acknowledgements by th anking my wife, Nicole, and her parents for their encouragement throughout my work in the IDP. Despite her own hectic schedule as a doctoral candidate in IDP, Nicole has always made the time to help me, both in and out of lab and has served as a true motivating force. Fo r her calming influence a nd understanding over the years, I can't thank her enough. Thanks also to her parents, Bill and Nihal, who as nuclear physicists have their own extensive experience with research and graduate education. Having their support of my work has been a tremendous psychological boost. 6

PAGE 7

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF FIGURES .......................................................................................................................10 LIST OF ABBREVIATIONS ........................................................................................................12 ABSTRACT ...................................................................................................................................14 CHAPTER 1 INTRODUCTION ................................................................................................................ ..16 Type 1 Diabetes ......................................................................................................................16 Worldwide Incidence .......................................................................................................17 Isolation of Insulin ...........................................................................................................18 Natural History of Type 1 Diabetes .................................................................................18 Genetic Susceptibility ......................................................................................................19 Long-Term Clinical Complications .................................................................................19 Non-Obese Mouse Model of Type 1 Diabetes .......................................................................20 Discovery and Development ...........................................................................................20 Natural History ................................................................................................................21 Prevention Therapy .........................................................................................................21 Reversal Therapies After Overt Onset ............................................................................22 Limitations of the NOD Model .......................................................................................23 Therapeutic Strategies for Type 1 Diabetes ............................................................................24 Insulin Replacement Therapy ..........................................................................................24 Pancreas or Islet Cell Transplantation .............................................................................24 Immunomodulatory Therapies ........................................................................................25 Combination Therapies ...................................................................................................26 2 GENERAL METHODS .........................................................................................................29 NOD Mouse Strain .................................................................................................................29 Determination of Hyperglycemia ...........................................................................................29 Standard Treatment at Onset of Hyperglycemia ....................................................................29 Anesthetization with Isoflurane .......................................................................................29 Implantation of Insulin Pellet ..........................................................................................30 Implantation of ID Chip ..................................................................................................30 Euthanasia ...............................................................................................................................30 Organ Harvesting for Histology .............................................................................................31 Splenocyte Purification ...........................................................................................................31 Regulatory T Cell Analysis via Flow Cytometry ...................................................................32 Multiplex Cytokine Analysis ..................................................................................................34 Statistical Analysis ..................................................................................................................35 7

PAGE 8

3 REVERSAL OF OVERT TYPE 1 DI ABETES IN THE NOD MOUSE USING COMBINATION IMMUNOSUPPRESSIVE THERAPIES .................................................36 Introduction: Two Agents of Therapeutic Efficacy ................................................................36 Methods ..................................................................................................................................37 Experimental Design .......................................................................................................37 Blood Glucose Criteria ....................................................................................................38 Implementation of Rapamycin Pellet ..............................................................................38 Analysis of C-peptide Serum Concentration ...................................................................39 Immunohistochemistry ....................................................................................................39 Insulitis Scoring ...............................................................................................................39 Results .....................................................................................................................................39 In Vivo Survival Curves of NOD Mice ...........................................................................39 C-peptide Analysis ..........................................................................................................40 Flow Cytometry Analysis of T Cell Subpopulations ......................................................41 Insulitis Scoring ...............................................................................................................41 Discussion ...............................................................................................................................42 4 AUGMENTATION OF IMMUNOSUPPRE SSION IN OVERTLY DIABETIC NOD MICE VIA AAT THERAPY ..................................................................................................47 Introduction: AAT For the Protection of Beta Cells ..............................................................47 Materials and Methods ...........................................................................................................48 Experimental Design .......................................................................................................48 Preparation of AAT .........................................................................................................49 Reagents for the Prevention of Anaphylaxis ...................................................................49 Results .....................................................................................................................................50 Induction of Fatal Anaphylaxis .......................................................................................50 Prevention of Anaphylaxis ..............................................................................................50 In Vivo Survival Curves of Treated NOD Mice ..............................................................51 Flow Cytometry Analysis of T Cell Subpopulations ......................................................52 Discussion ...............................................................................................................................52 5 ENHANCEMENT OF INSULIN SECRETION IN DIABETIC NOD MICE THROUGH INHIBITION OF DPPIV ...................................................................................58 Introduction: Enhancing Beta Cell Function ..........................................................................58 Materials and Methods ...........................................................................................................60 Experimental Design .......................................................................................................60 Preparation of Januvia .....................................................................................................61 DPPIV Activity Assay .....................................................................................................61 CD26 Analysis Via Flow Cytometry ..............................................................................61 Fasting IPGTT .................................................................................................................61 TELISA Analysis of sCD26 and SDF-1 ............................................................................62 Measurement of GLP-1, Glucagon, and Insulin ..............................................................62 Results .....................................................................................................................................63 Confirmation of DPPIV Inhibitory Activity ....................................................................63 8

PAGE 9

Januvia-mediated Effect Upon IPGTT ............................................................................63 Endocrine Analysis of IPGTT Sera .................................................................................64 Blood Glucose Values of Treated NOD Mice .................................................................64 In Vivo Survival Curves of Treated NOD Mice ..............................................................64 Endpoint IPGTT of Reversed NOD Mice .......................................................................65 sCD26 Concentrations Following In Vivo Therapy ........................................................65 Flow Cytometry Analysis of T Cell Subpopulations ......................................................66 SDF-1 Serum Concentrations and Surface CD26 Following Therapy ...........................66 Discussion ...............................................................................................................................67 6 INDUCTION OF A TOLEROGENIC PHE NOTYPE BY G-CSF ADMINISTRATION ....78 Introduction .............................................................................................................................78 Materials and Methods ...........................................................................................................79 Experimental Design .......................................................................................................79 Preparation of G-CSF ......................................................................................................80 Quantification of Leukocyte Depletion ...........................................................................80 NIT-1 Lysate Preparation ................................................................................................81 Analysis of T cells, DCs, Macrophages, and Neutrophils Via Flow Cytometry .............81 Immunoglobulin Isotyping ..............................................................................................82 Anti-G-CSF Antibody Measurement ..............................................................................82 Histology .........................................................................................................................83 Real Time PCR ................................................................................................................83 Results .....................................................................................................................................83 Depletion/Repopulation of Leukocytes ...........................................................................83 Flow Cytometry Analysis of Macrophages and Neutrophils ..........................................84 Alterations in B Lymphocyte Profile ..............................................................................84 Flow Cytometry Analysis of T Lymphocytes .................................................................85 Splenocyte Proliferation to NIT-1 Lysate In Vitro ..........................................................87 Splenocyte Cytokine Release in Response to Activation In Vitro ..................................87 Insulitis Scoring ...............................................................................................................87 Insulin Area .....................................................................................................................88 Conclusions .............................................................................................................................88 7 DISCUSSION AND CONCLUSIONS ................................................................................101 Discussion .............................................................................................................................101 Conclusions ...........................................................................................................................106 LIST OF REFERENCES .............................................................................................................108 BIOGRAPHICAL SKETCH .......................................................................................................120 9

PAGE 10

LIST OF FIGURES Figure page 3-1. ATG+RAPA reversal study experimental design describing the timing and duration of individual treatments. .........................................................................................................44 3-2. Kaplan-Meier survival curve of NOD mice following treatment at onset of hyperglycemia with ATG and/or RAPA. ...........................................................................44 3-3. Both ATG and RAPA lead to significant in creases in serum c-peptide levels at the time of sacrifice. .........................................................................................................................45 3-4. Flow cytometry analysis of the regulatory T cell population and CD4:CD8 ratios in splenocytes of treated mice at time of s acrifice revealed modulation by both ATG and RAPA. .........................................................................................................................45 3-5. Insulitis scoring of treated mice reveal s a marginal benefit of ATG therapy in the percentage of healthy islets. ...............................................................................................46 4-1. ATG+AAT reversal study experimental de sign describing the timing and duration of individual treatments. .........................................................................................................55 4-2. IP administration of either albumin or AAT leads to the inducti on of fatal anaphylaxis in the NOD mouse. ............................................................................................................55 4-3. Short-term protection from AAT-induced an aphylaxis is afforded by pretreatment with antihistamine and PAF antagonist .....................................................................................56 4-4. Kaplan-Meier survival curve of NOD mice following treatment at onset of hyperglycemia with ATG and/or AAT.. ............................................................................56 4-5. ATG increases the percentage of spleni c regulatory T cells versus control in mice sacrificed at trial endpoint ..................................................................................................57 5-1. ATG+Januvia reversal study experimental design describi ng the timing and duration of individual treatments. .........................................................................................................70 5-2. Purification of Januvia does not ablate DPPIV activity as measured by luminescence. .......70 5-3. A one-time dose of Januvia prior to IP GTT in non-diabetic 12 week old NOD mice does not significantly modulate bloo d glucose levels versus control. ...............................71 5-4. Januvia-pretreatment of non-diabetic 12 week old NOD mice mediates an increase in serum insulin levels during IPGTT. ...................................................................................71 5-5. Nonfasting blood glucose leve ls of NOD mice treated with control, ATG, Januvia, or ATG+JAN post-onset. .......................................................................................................72 10

PAGE 11

5-6. Kaplan-Meier survival of new-onset NOD mice treated with ATG and/or Januvia. ............73 5-7. Januvia does not augment ATG-mediated glucose control in NOD mice at endpoint as measured by an IPGTT, but does increase serum insulin levels. .......................................74 5-8. Januvia-treated diabetic NOD mice exhibit increased serum sCD26 versus controltreated mice. .......................................................................................................................75 5-9. ATG and ATG+JAN-treated mice exhib it altered regulatory T cell and CD4:CD8 profiles. ..............................................................................................................................76 5-10. Januvia has no significant impact upon the serum levels of SDF-1. ....................................77 5-11. Januvia has no significant impact upon th e surface expression of CD26 on CD4+ and CD8+ T cells. .....................................................................................................................77 6-1. ATG+G-CSF mechanistic study experimental design describing the timing and duration of individual treatments. ......................................................................................92 6-2. ATG+G-CSF mediates a significant increa se (p=0.0129) in peripheral blood leukocytes at 2 weeks vs. ATG monotherapy.. ....................................................................................92 6-3. G-CSF therapy, alone or in combination with ATG, induces a short-term increase in the percentage of both spleni c neutrophils and macrophages. ...........................................93 6-4. Total serum immunoglobulin is altered by treatment with G-CSF. ......................................94 6-5. Anti-G-CSF antibodies are produced in treated NOD mice ..................................................95 6-6. Both ATG and G-CSF altered the CD4:CD 8 ratio and/or the percentage of splenic regulatory T cells.. .............................................................................................................96 6-7. The percentages of CD4+ and CD8+ T cells are significantly reduced at 2 weeks with ATG and/or G-CSF therapies. ...........................................................................................97 6-8. The ratio of T to effector T cells (both CD4+ a nd CD8+) is increased with ATG and G-CSF therapies.reg ................................................................................................................98 6-9. G-CSF alters the cytokine re lease of stimulated splenocytes in vitro and appears to skew from a T 1 to a T 2 phenotype.H H ...............................................................................99 6-10. Insulitis scoring of pancreatic islets at the 8 week time point was improved in mice treated with a combination of ATG and G-CSF. .............................................................100 6-11. Combination therapy of ATG and G-CSF l eads to a robust, longlasting increase in the percentage of beta cell area. .......................................................................................100 7-1. Efficacy of ATG is independent of age at onset and dependent upon blood glucose at therapeutic onset. .............................................................................................................107 11

PAGE 12

LIST OF ABBREVIATIONS AAT Alpha-1-Antitrypsin ACS Animal Care Services AF Aldehyde Fuchsin AICD Activation-induced Cell Death APC Antigen presenting cell ATG Anti-Thymocyte Globulin AUC Area Under the Curve BB Bio-Breeding BrdU Bromo-Deoxy-Uridine CTLA-4 Cytotoxic T Lymphocyte Antigen-4 CD Cluster Of Differentiation DC Dendritic Cell DPPIV Dipeptidyl Peptidase IV ELISA Enzyme-Linked Immunosorbance Assay G-CSF Granulocyte Colony Stimulating Factor GLP-1 Glucagon-like Peptide-1 GVHD Graft Versus Host Disease H&E Hematoxylin & Eosin IACUC Institutional Animal Care & Use Committee IP Intraperitoneal IPGTT Intraperitoneal Gl ucose Tolerance Test mAb Monoclonal Antibody mATG Murine ATG MHC Major Histocompatibility Complex 12

PAGE 13

mTOR Mammalian Target Of Rapamycin NOD Non-Obese Diabetic PBS Phosphate-Buffered Saline PLN Pancreatic Lymph Node RAPA Rapamycin rIgG Rabbit Gamma Globulin RLU Relative Light Unit SDF-1 Stromal Cell-Derived Factor-1 SPF Specific-Pathogen Free SQ Subcutaneous T1D Type 1 Diabetes 13

PAGE 14

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy REVERSAL OF OVERT TYPE I DIABETES IN THE NOD MOUSE THROUGH THE USE OF COMBINATION THERAPY By Matthew John Parker December 2008 Chair: Mark A. Atkinson Major: Medical Sciences Immunology and Microbiology The autoimmune destruction of the insu lin-producing beta cells by autoreactive T lymphocytes results in a loss of blood glucose cont rol and is known as type 1 diabetes (T1D). While exogenous insulin therapy has been availabl e commercially since 1923, this treatment is not considered a cure as it is lif e-long and typically results in long-term complications including retinopathy, nephropathy, and variou s cardiovascular disorders. As such, there has been a desire to increase our understanding of disease pathogenesis in order to potentially prevent and/or cure overt diabetes. Many such studies have been carr ied out using animal mo dels of T1D, perhaps most prolifically in the non-obese diabetic (NOD) mouse. After r ecent clinical trials in which prevention therapies have not proven efficacious, there has been an increasing desire for intervention therapies immediately post-onset. As such, the need to control the autoimmune response may be considered a critical component to these therapies. In these studies, antithymocyte globulin (ATG) was used immediately post-onset in NOD mice in order to halt the autoimmune destruction of the re maining beta cells as well as to attempt to induce a more tolerant immune phenotype. Only two doses of ATG were administered, resu lting in a short-term T lymphocyte ablation. In addition, secondary co mpounds were added in combination with ATG in different stages of this project based upon the hypothesis that combin ation therapy targeting 14

PAGE 15

15 multiple pathways would prove more effective at inducing long-term disease remission than monotherapy. These studies will demonstrate that not only was ATG effective at inducing significant remission in new-onset NOD mice, but also that through combination therapy with granulocyte-colony stimulating f actor (G-CSF) insulitis was dimi nished while beta cell area increased, as demonstrated through analysis of histology. In addition a more tolerant immune phenotype was induced as quantified through an alysis of regulatory T cell populations.

PAGE 16

CHAPTER 1 INTRODUCTION This introduction serves to hi ghlight both the significance and the characteristics of type 1 diabetes. This will be accomplished through a disc ussion of worldwide disease incidence, natural history, insulin treatment, and the long-term clin ical complications. In addition, the NOD mouse, a popular mouse model of type 1 diabetes, will be di scussed in terms of its role in research into disease pathogenesis. Finally, current as well as preclinical rationales for the treatment of type 1 diabetes will be discussed. Type 1 Diabetes Diabetes consists of a multitude of metabolic disorders that are collectively linked by the common characteristic of elevat ed blood sugar, or hyperglycemia. This alteration in the control of blood sugar arises due to deficiencies either in the production of or uti lization of the endocrine hormone insulin. Insulin is produced and stored in the form of pr oinsulin in granules within beta cells in the Islets of Langerhans within the pancreas. The beta cells are responsive to changes in blood sugar, appropriately releasing proinsulin which is cleaved into insulin and c-peptide. Insulin enters into circulation where it aids in the u tilization and uptake of glucose. The two major forms of diabetes, type 1 and 2, differ in the manner of insulin deficiency. Type 1, formerly known as juvenile until the incidence of type 2 became prevalent in adolescents, diabetes results from the autoimmu ne destruction of insu lin-producing beta cells, which will be discussed in greater detail in subseq uent sections of this introduction, consequently reducing the availability of insulin to regulate blood sugar. Type 2 diabetes results from deficient action of insulin such that a resistance to insu lin develops and does not involve the autoimmune destruction of the beta cells. 16

PAGE 17

Worldwide Incidence As many as 246 million people worldwide as of 2007 were afflicted with a form of diabetes with type 1 diabetes accounting for as many as 10% of these cases (1). As of 2005, 20.8 million people in the United States were aff licted with diabetes, or 7% of the population according to the NIH online. Of these, 176,500 were under the age of 20. The incidence of type 1 diabetes worldwide depends upon racial and geographical factors. The incidence can vary dram atically depending upon location (2). Between 1990 and 1994 there was a 350-fold difference in the rate of onset of type 1 diabetes between 100 different worldwide populations in those under 14 years of age (3). Th e lowest incidence was s een in both China and Venezuela with roughly 0.1 out of every 100,000 bei ng diagnosed per year. At the other extreme, more than 40 out of every 100,000 were diagnosed in Finland. Worldwide, the overall incidence of both type 1 and type 2 diabetes has been increasing since the conclusion of World War II in regions with both high and low areas of prevalence (4). As of 2006, approximately 440,000 individuals under the age of 14 were afflicted with type 1 diabetes with 70,000 new cases each year. In add ition, this incidence has been increasing at roughly 3% per year. This trend of increasing frequency has only serv ed to highlight the need for improved diagnosis, treatment, and research. The rapid rate of increase worldwide indicates that genetics are not solely at play, but rather environmental factors are also involved. One indication of the role of environmental factors is present in the fact that fewer than half of identical twins become concordant for type 1 diabetes (5). Proposed environmen tal explanations range from ch anges in milk consumption to vaccinations. The fall in infectious diseases as well as intestinal parasites in developed countries has been proposed as a possible explanation fo r increasing rates of au toimmunity in what has become known as the hy giene hypothesis (6-9). 17

PAGE 18

Isolation of Insulin The groundwork for insulin therapy was laid in October 1920 by the University of Toronto researcher Frederick Banting when he devised an idea for taking th e pancreatic extracts of dogs in an attempt to treat diabetes based upon an article written by Moses Barron linking diabetes with the Islets of Langerhans (10). Upon presen ting his idea to Professor John Macleod, Banting was granted facilities the next summer. With the aid of Charles Best and biochemist J.B. Collip, the team at the University of Toronto successfully isolated sufficient quantities of insulin to treat their first patient, 14-year-old Leonard Th ompson, on January 23, 1922. By 1923, commercial quantities of insulin became available to thousands of patients. This accomplishment earned Banting and Macleod the 1923 Nobel Prize in Medicine, which was shared with Best and Collip, respectively, and effectively transforme d diabetes into a treatable disease. Natural History of Type 1 Diabetes While the Islets of Langerhans were linked with diabetes prior to th e isolation of insulin, the concept of an autoimmune basis of diabetes is more r ecent. Although autoimmunity was discovered during the 1950s (11), it wasn't un til the 1970s that evidence for an autoimmune lymphocytic infiltrate in the islets leading to diabetes onset was demonstrated. This was accomplished by the analysis of pancreatic histology from recent-onset juvenile diabetic patients by Dr. William Gepts (12). Autoantibodies have also become associated with type 1 diabetes since the 1970s. These autoantibodies exist to a variety of autoantigens including, but not limited to, insulin, IA-2, and glutamic acid decarboxylase (GAD). It is currently believed that although these autoantibodies may serve as effective markers of disease, they do not possess a significant role in the pathogenesis of type 1 diabetes. 18

PAGE 19

Our current understanding of type 1 diabetes implicates an abnormal, destructive T lymphocyte response in the destruction of beta cells This will be discussed in greater detail in a subsequent section of this intr oduction. The pathogenesis of type 1 diabetes is such that an individual begins with a "nor mal" beta cell mass which is reduced upon the initiation of a lymphocytic infiltration. During the initial destruction, the loss is de scribed as silent due to the lack of overt symptoms. Eventually sufficient beta cell mass is lost, estimates range from 5080%, such that overt symptoms appear. These include a fasting blood glucose of at least 126mg/dL as well as ketoacidosis. Following onse t, typical therapy involves daily insulin injections for the control of blood sugar. Despit e an initial reduction in the requirement of exogenous insulin, in what is known as the Honeymoon Phase, the requirement for insulin increases over time until any significant endogenous insulin production has been lost. Genetic Susceptibility In the course of more than 30 years of study, at least 18 genetic loci (IDDM1-IDDM18) have been identified as risk factors for type 1 diabetes (13). The firs t genetic region to be associated with type 1 diabetes was the HLA co mplex (14). This complex is highly associated with immune regulation and is comprised of at least 128 genes (15-17). While many risk factors for diabetes fall under HLA, there are several no n-HLA risk factors as well. The insulin gene VNTR region (IDDM2) and the cy totoxic t-lymphocyt e antigen-4 (CTLA-4 ) are two notable examples. Despite the discovery of these loci, they are able to predict only a modest risk. It is likely that rare, high risk variants also exist, but remain to be discovered. Long-Term Clinical Complications While the first use of insulin in 1922 was rightly hailed a watershed moment in the treatment of type 1 diabetes, it has proved to be an imperfect therapy. Despite changing diabetes from a fatal disorder into a treatable condition, in sulin is not a cure. Decades of insulin therapy 19

PAGE 20

have revealed a multitude of long-term complicat ions including significantly higher risks for nephropathy, retinopathy, amputation, and various cardiovascular diseases (18). The long-term treatment and complications of diabetes pose significant psychological and financial tolls upon those afflicte d. Constant blood glucose mon itoring and insulin injections result in significant stress, especially in childr en (19). The financial burde n is also substantial, with the total cost in the United Stat es estimated at $132 billion in 2002 (20). This has highlighted the need for an improved understanding of the path ogenesis of type 1 diabetes as well as the need to develop an effective long-term th erapy to eliminate the need for exogenous insulin therapy. Research into these issues has been sign ificantly aided by the use of animal models of type 1 diabetes. Non-Obese Mouse Model of Type 1 Diabetes Animal models are used in the study of virtua lly all diseases where pertinent models are available. There exist several animal models of di abetes for research purposes. Three of the most common models are streptozotoc in (STZ)-induced, the bio-breedi ng (BB) rat, and the non-obese diabetic (NOD) mouse. The STZ model was not used in this dissertation due to the fact that STZ administration causes beta cell a poptosis through chemical, rather than autoimmune, means. The BB rat was not used due to the difficulty in working with sufficient num bers of rats for the purposes of our experiments as well as due to our long-term familiarity with the NOD mouse model of type 1 diabetes. By working with th e NOD mouse model, we were able to study the pathogenesis and effects of our therapie s in a tightly controlled environment. Discovery and Development The NOD mouse was originated in Japan during the early 1970s during the development of the Cataract Shionogi strain, base d on outbred Jcl:ICR mice (21). Ce rtain cataract-free mice with elevated fasting blood sugar were separated a nd inbred. During subsequent generations, one 20

PAGE 21

normoglycemic female mouse developed spontaneous diabetes. This became the basis for future generations of NOD mice. In 1984 pairs of NOD mice were shipped to the Joslin Diabetes Center in the United States as well as to Australia. The NOD mouse has now been extensively characterized for more than 20 years. Natural History The pathogenesis of type 1 diabetes in the NOD mouse has much in common with the human form of disease. Both exhibit extensiv e lymphocytic infiltrati on into the Islets of Langerhans resulting in T cell-medi ated destruction of the beta cel ls. Both are polygenic diseases in which multiple susceptibility loci, which typi cally involve immune re gulation, are responsible for the induction of disease. Both involve def ective immune regulation and onset rates in both are likely affected by environmental factors. NOD mice begin to show signs of insulitis as ear ly as 5 weeks of age. By 10 weeks of age, a multitude of cell types are present in the infilt rate, consisting of CD4+ T cells, CD8+ T cells, B cells, and dendritic cells (DC). In a typical cohort, onset of hyperglycemia will begin in a percentage of NOD mice begi nning at roughly 12 weeks of ag e. By 30 weeks roughly 75% of female and 30% of male NOD mice will have become hyperglycemic. The shift from normoglycemia to extreme hyperglycemia (500+ mg/dL) is relatively rapid, often occurring within one week. Prevention Therapy Most therapeutic studies in the NOD mouse model have focused upon disease prevention. As previously described, the rate of onset is ma rkedly reduced if the mi ce are not housed in SPF conditions. Even in SPF conditions however, NOD mice appear to be susceptible to many forms of prevention therapy. Recent estimates place the number of successful prevention therapies at well over 200 (22). 21

PAGE 22

Part of the perceived problem with these prev ention trials is that early prevention at 4-6 weeks of age prior to insulitis allows for improve d rates of reversal comp ared to late prevention at 12 weeks after prolonged insu litis. Along these lines, appr oximately 70% of NOD prevention studies could be considered early prevention ( 23). Irrespective of when the prevention therapies initiated, there is a problem of translation into clinical efficacy. In recent years, several promising prevention therapies derived from an imal models have had no apparent success despite extensive testing in the clinic (24; 25). Given these apparent failures, recent studies have begun to focus more upon either late prevention and/or reversal studies following th e onset of overt hyperglycemia in the hope of improving translation into the clinic. Reversal Therapies After Overt Onset Few therapies have been successful for the revers al of overt diabetes in NOD mice. This is due, in part, to the focus upon prevention trials until relatively recently as well as due to the aggressive, rapid nature of di abetes progression following initial diagnosis of hyperglycemia. In spite of this, there are numerous published protoc ols that afford reversal of hyperglycemia in new-onset NOD mice. While these therapies rang e from microspheres (26) to islet-specific regulatory T cells (27), virtually all share some facet of immu nomodulation with the aim of halting autoimmune beta cell destruction. One such therapy, anti-cd3 m onoclonal antibody, has been extensively characterized and has been considered something of a benchmark for the reversal of overt di abetes since it was first described in 1992 (28; 29). Since it's introducti on, the use of anti-cd3 mAb therapy has been shown to induce a tolerogenic state through the induction of TGFbeta dependent regulatory T cells (30). In addition, the use of this therapy typically yields a reversal rate gr eater than 50%. Given the success of anti-cd3, limited intervention trials have been performed for the treatment 22

PAGE 23

of new-onset patients. These have demonstrated an ability to preserve c-peptide for 18-24 months after therapy (31). More extensive trials are also currently in the formative stages. Other published therapies have indicated significa nt levels of diabetes reversal in the NOD mouse, yet many are impractical for translation in to clinical use. Examples of such protocols may involve the use of Complete Freund's Adjuvant in combination with cellular-based therapies (32). Consequently, while therapeutic efficacy is cl early a crucial factor, th e ability to translate such a therapy into clinical use is of paramount importance. As su ch, the studies laid out in this dissertation describe the use of FDA-approved reagents with a dem onstrated safety record in the clinic. Limitations of the NOD Model Although the NOD mouse model has much in co mmon with the human form of type 1 diabetes and serves as a convenient tool for th e study of type 1 diabetes, there are limitations. These limitations are a consequence of the differences between mouse and man. The NOD mouse is an inbred, i.e. syngeneic strain. Consequently, the variet y of disease pathogenesis that may exist in the clinic may not be represented by the more uniform NOD mouse. In addition, the rate of diabetes onset in th e NOD mouse is extremely reliant upon the conditions in which the mice are housed. Without specific-pathogen free conditions, the onset rate in the NOD mouse drops dramatically. In addition, NOD mouse lack hemolytic complement C5 and have defective natural killer (NK) cell function. Finally, as me ntioned above, there are hundreds of ways to reduce or prevent onset in the NOD mouse, yet n one of these has any demonstrated effect in humans to date. Ultimately, ther e is no guarantee that a preventi on or reversal therapy that is effective in the NOD mouse will prove clinically efficacious. 23

PAGE 24

Therapeutic Strategies for Type 1 Diabetes There are several potential strategies for the treatment of type 1 diabetes. Of these, two comprise the overwhelming majority of trea tments. Insulin replacement therapy may be considered the standard approach, while islet cell or pancreas tran splantation is also used albeit to a lesser degree. This section of the introduction will serve to discuss not only these two strategies, but also more recent, experimental approaches. Insulin Replacement Therapy Insulin replacement therapy has served as th e conventional management therapy for type 1 diabetes since its commercial de but in 1923. While initially percei ved as an effective solution, the emergence of long-term complications su ch as retinopathy, nephropathy, amputation, and other cardiovascular-related conditi ons have appeared in subsequent decades (33-35). The rate of retinopathy, for instance, is reported as greater than 90% over th e lifetime of type 1 diabetes patients (35; 36). Recent technical developments may aid in the reduction of these complications in subsequent generations of type 1 diabetes patients by improving blood glucose control. This may be achieved through a variety of formulations of insulin that are tailore d to provide different durations of insulin activity, such as the rapid-acting insulin glulisine or the l ong-acting insulin glargine. In addition, real-time blood glucose monitor implants enable patients to make more informed decisions on their dosing, providing tig hter blood glucose with fewer excursions. Ultimately, insulin replacement therapy necessitates a lifetime of constant monitoring and insulin injections without any guarantee of eliminating long-term complications. Pancreas or Islet Cell Transplantation Insulin replacement therapy was the only option fo r type 1 diabetes patients until the first pancreas transplant was performed in 1966 at the University of Minnesota (37). Since then, more 24

PAGE 25

than 20,000 such transplants have been performed w ith increasing rates of success with pancreas rejection rates of less than 25% at 48 months. This is largely attributable to an improved understanding of immunosuppressive therapies. More recently, islet cell transpla ntation has also become an attractive method for attaining insulin independence. The Edmonton Protocol, which utilizes a regimen of humanized IL-2R antibody (Daclizumab), sirolimus, and tacrolimus, has led to insulin independence rates of 80% 12 months into therapy (38). Unfortunately, long-term (5+ years) success remains elusive with failure at 90% or greater. The advantage over panc reas transplantation, howe ver, is that islet cell transplantation does not require major surgery an d may be considered an outpatient procedure. Although the rate of insulin independence is grea ter with pancreas tran splantation, there is the requirement for more extensive surgery as we ll as considerable exocri ne drainage from the transplant. Islet cell tr ansplantation using the Edmonton pr otocol is not without its own drawbacks. Apart from the low long-term su ccess rate, two to three donor pancreases are required for sufficient islet numbers. In addition, both of these therapies require lifelong immune suppression and suffer from the ever present s hortage of organ donors. Consequently, these therapies are available only to a selected subgroup of type 1 diabetes patients. Immunomodulatory Therapies While the mainstream strategies for the treatm ent of type 1 diabetes consist of insulin replacement or transplantation, there has been a limited effort to treat overt disease through the use of immunomodulatory agents in order to halt the autoimmune destruction of beta cells. Due to the relative safety of insulin replacement ther apy as well as age consid erations regarding the treatment of children, im munomodulatory therapy has been approached cautiously in the clinic. The objective of such studies is to alter the immune phenotype such that the autoreactivity to the beta cells is brought under control th rough a more regulatory, suppressive milieu. This 25

PAGE 26

may be accomplished through a number of ways, such as ablation of the cytotoxic T cell component, inducing an increase in the CD4+CD25+Foxp3+ regulatory T cell (Treg) population, and/or through the induction of tolerogenic antigen presenting cells (APCs) such as dendritic cells (DC) (39; 40). Other methods, such as the induction of antigen-specific Treg have also been described (27). One notable clinical trial involving multiple i mmune modulating agents is currently taking place in Brazil. In this trial, 15 recent-onset (< 6 weeks from diagnosis) patients age 14-31 were treated with autologous non-myeloablative hemat opoietic stem cell transp lantation (AHST). In this protocol, cyclophosphamide and granulocytecolony stimulating factor (G-CSF) to mobilize the hematopoietic stem cells followed by c onditioning with cyclophosphamide and ATG. The cells were then injected intrav enously (IV) and the patients were followed for morbidity as well as insulin requirements (41). Insulin independenc e was achieved as far out as 35 months and was achieved in 14 out of the 15 patients. Combination Therapies In recent years, following the lack of success in clinical prevention trials, there has been an increased interest in the use of combination th erapy for the purpose of reversing overt type 1 diabetes. The above trial in Brazil is but one ex ample of the potential benefit of the use of combination therapy for this purpose. It has be en proposed that given the success of multiple therapies over monotherapy for the treatment of HIV and many forms of cancer, such as malignant glioma and breast cancer, that treati ng multiple aspects of diabetes pathogenesis may result in greater therapeutic efficacy, provided that such therapies have a demonstrated degree of safety (42-47). While monotherapies have yielded the successf ul prevention of type 1 diabetes in NOD mice, their use for purposes of reversing overt disease remain limited. Few clinically relevant 26

PAGE 27

agents have been demonstrated to possess the cap ability of robust diabetes reversal with ATG and anti-CD3 being among the exceptions (48; 49 ). Even in these cases, the reversal rate typically hovers close to 50%. The potential for diabetes reversal appears to be significantly greater when such therapies are considered for us e in combination with one another rather than solely as monotherapies. Previous studies demonstrate the complementar y effect of using combination therapy for the reversal of type 1 diabet es in the NOD mouse. One study de monstrated that while neither lisofylline nor exendin-4 were capable of i nducing long-term remission from hyperglycemia, the combination of the two yielded a reversal rate of 100% (6/6) for up to 145 days (50). In a separate study, the use of anti-lymphocyte se rum alone remitted 40% of the new onset NOD mice treated, while the addition of exendin-4 to this therapy booste d the reversal rate close to 100% (23/26) for up to 200 days (51). In other instances, such as the addition of exendin-4 to anti-cd3, the reversal rate was not improved, but the quality of the remission may have been improved as measured by a reduction in beta cell apoptosis (48). Th ese and other published reports repeatedly demonstrate the increased efficacy of combination therapies for the purposes of inducing long-term remission from hyperglycemia in NOD mice (27; 32; 52-54) with rates approaching or reaching 100%. This dissertation will descri be the use of such an approach by testing combination therapies in the setting of the NOD mouse for the purpose of rever ting overt disease in new onset mice. Specifically, we have identified three ther apeutic targets upon which to act. The first, and arguably most critical, is to ha lt ongoing autoimmunity at time of diabetes onset. The second is to protect the beta cell mass rema ining at the time of diagnosis. Finally, the third facet is to increase the beta cell mass eith er through neogenesis or replicat ion of existing beta cell. FDA27

PAGE 28

28 approved agents with an establishe d safety record were utilized in all studies carri ed out herein with the potential goal of rapid translation into the clinic.

PAGE 29

CHAPTER 2 GENERAL METHODS NOD Mouse Strain These studies used female NOD/ShiLtJ mice to test various combination therapies as outlined in this dissertation. The animals were or dered at 8 weeks of age from Jackson Labs Inc (Bar Harbor, ME, USA) and housed in SPF cond itions in the Animal Care Services (ACS) Communicore basement facility at the University of Florida. All procedur es were carried out as described in approved Institutional Animal Care & Use Committee (IACUC) protocols. Determination of Hyperglycemia During the course of these studies, it wa s necessary to determine the onset of hyperglycemia in order to begin therapy or to de termine failure of a therapy. The threshold of 240mg/dL blood glucose for two consecutive days was used as classification of diabetes onset. A tail bleed was performed three times per week beginning at 11 weeks of age. A number 11 scalpel was used to obtain a drop of blood that was measured using a One Touch Ultra testing meter. Standard Treatment at Onset of Hyperglycemia Despite using different therapies for each stage of this project, there were certain procedures that remained uniform throughout th e course of the study. Following determination of diabetes onset, the mice were subject to standard processing in cluding implantation of subcutaneous insulin pellets and electronic identity chips as described below. Anesthetization with Isoflurane Prior to the implantation of subcutaneous pell ets, the animals were anaesthetized using an isoflurane assembly located in a sterile laminar flow hood. Animals were initially placed in an acrylic chamber. Once the animals were anaesthetiz ed as tested by toe pinch, they were moved to 29

PAGE 30

a nose cone apparatus and a patch approximately one inch in diam eter was shaved in the dorsal hair and swabbed with an iodine wipe. The anaesthetization was maintained while the implantations were performed, as described belo w. Following the procedure, the isoflurane was turned off and the animals were observed until they were able to right themselves. Implantation of Insulin Pellet Ideally, insulin injections would be given to e ach mouse in order to tailor the insulin dose to the appropriate blood sugar. Due to logistical issues, we used a sustai ned release subcutaneous insulin pellet. We purchased LinBit insulin pe llets from LinShin Canada (Toronto, Ontario, Canada) to deliver an approximate dose of 0.1U in sulin per day for up to 30 days. The pellet was implanted under the mid dorsal skin using a ster ilized trocar and stylet, provided by LinShin. Suturing was not required, a nd the pellet was degraded in vivo obviating the need for surgical removal. Implantation of ID Chip In order to ensure the identity of each mous e treated, animals were implanted with an identity microchip. We purchased FriendChip (AVID, Norco, CA, USA) pellets which we implanted subcutaneously in the mid dorsal region using the all-in-one sy ringe containing the ID chip. The pellet was placed into the same ope ning created during implantation of the LinBit pellet and consequently did not require suturi ng. Animals could be identified using a handheld scanner yielding a unique 9 digit identity code afforded by the microchip. Euthanasia Once at the endpoint of each study, the mice were sacrificed. The method of sacrifice for the experiments in Chapters 3 and 4 began w ith asphyxiation using a carbon dioxide chamber. For the experiments in Chapters 5 and 6 the an imals were anaesthetized using an isoflurane assembly. To ensure death in both instances, th e animals were exsanguina ted using a 1cc insulin 30

PAGE 31

syringe (Becton Dickinson, Franklin Lakes, NJ, USA) followed by cervical dislocation as well as a bilateral thoracotomy. These pr ocedures were approved by the IACUC in each phase of the study. Organ Harvesting for Histology Several organs were harvested following sacrif ice for histological analysis. The mice were sprayed with 70% ethanol and sterile instruments were used to remove the organs. The specific organs collected depended upon the phase of th e project and are desc ribed in subsequent chapters. Their colle ction as outlined here remains the same throughout. Tissues were placed in cassettes and fixed overnight in a 10% formalin solution (Fisher Scientific, Pittsburgh, PA, USA). Cassettes were then placed in 1x PBS and sent to the University of Florida Pathology Core for processing. All tissues were stained with hematoxylin and eosin (H&E). The pancreas was also stained for insulin. Specific pancreatic co-stains are described in subsequent chapters. Splenocyte Purification Following sacrifice of the animal, the spleen was acquired using sterile instruments and placed in a 50mL conical tube containing ~5mL of Hank's Buffered Salt Solution (CellGro, Herndon, VA, USA). The tube was sprayed with ethanol and placed in a laminar flow hood for the remainder of the isolation. The spleen was poured over a 100um nylon filter (BD Falcon, Bedford, MA, USA) on top of a new 50mL conical tube and perfused wi th 1.5mL Hank's Buffer using a 3mL BD syringe with attached 27 gauge needle. The plunger was then removed from the syringe and used to grind the spleen. The filter was washed with Hank's buffer during grinding until the total volume reached 25mL. The tube was centrifuged at 300g for 10 minutes at 4 degrees. The supernatant was then decanted and the pellet was resuspended by flic king the bottom of the tube. 25mL of Hank's buffer was added to the cells. The centrifugation was repeated at the same settings. Following the 31

PAGE 32

next decanting the pellet was resuspended in 3m L sterile ammonium chloride solution (StemCell Technologies, Vancouver BC, Canada) and placed on i ce for 5 minutes in order to lyse the red blood cells. Two washes and spins with 25mL Hank's buffer were then performed. At this point, the resuspension depended upon the use of the splenocytes. Cells were resuspended in 5mL of Hank's Buffer if flow cy tometry was to be performed or in 5mL of RPMI, 10% FBS (described in subsequent methodol ogy) if cells were going to be cultured. Cells were then counted using a hematocytometer. A vol ume of 10uL of cells was mixed with 90uL of Trypan Blue dye. 10uL of this mixture was pipette d under the cover slip of the hematocytometer and the cells in the 5x5 grid were counted. The number obtained was multiplied by 105 to determine the number of cells per mL. Regulatory T Cell Analysis via Flow Cytometry Splenocytes were isolated and counted as described above. 1x106 cells from each sample were placed in individual, 5mL Falcon flow t ubes (Becton Dickinson, Franklin Lakes, NJ, USA). The same number of cells was also placed into tubes for the isotype control as well as for compensation, as described below. With each tube containing 1x106 cells in 1m L of Hank's Buffered Salt Solution, 2mL of cold BD Pharmingen Stain Buffer (San Diego, CA USA) was added to each tube. The tubes were then centrifuged at 300g for 10 minutes at 4 degrees. The supernatant was then aspirated from each tube, and the tubes were vortexed at setting 7 to resuspend th e pellet. 100uL of stain buffer was added to each tube and the overhead lights were turned off for subsequent lightsensitive steps in the protocol. The appropriate antibody cocktails were th en added to the corresponding tubes. The surface "test" antibodies consisted of 1ug anti-mouse CD4-PerCP, clone RM4-5 (BD Biosciences, San Jose, CA, USA) and 1ug an ti-mouse CD25-APC, clone PC61.5 (eBioscience, 32

PAGE 33

San Diego, CA, USA). The corresponding isotype c ontrol antibodies for these test antibodies were Rat IgG2a PerCP (BD Biosciences, San Jose, CA, USA) and Rat IgG1 APC (eBioscience, San Diego, CA, USA) respectively. For the compen sation tubes, a clear, strong positive signal was desired, so anti-mouse CD4 an tibodies were used conjugated to PerCP, PE, and APC. The cells were placed in a rack, covere d with foil, and placed in a refrigerator for 30 minutes. Following this incubation, 2mL of stain buffer was added to each tube. The tubes were then centrifuged at 300g for 10 minutes followed by another aspiration and resuspension in 2mL stain buffer. The centrifugation and aspiration was then repeated. At this point, the protocol for the compensation and test/isotype tubes became distinct. The compensation tubes were resuspended in 250uL BD Cytofix (San Di ego, CA, USA), containing formaldehyde, and incubated for 15 minutes in the dark at 4 de grees. Following this incubation, 2mL stain buffer was added to each compensation tube which were then centrifuged at 300g for 10 minutes. One more wash with 2mL stain buffer was performed, followed by another centrifugation. Finally, the compensation tubes were resusp ended in 250uL stain buffer and ke pt in the dark at 4 degrees until time of analysis. The isotype and test tubes were resuspended in 1mL eBioscience fixation/permeabilization buffer overnight in the dark at 4 degrees. Following this overnight incubation, 2mL of eBioscience 1x permeabilization buffer was added to each tube. The tubes were centrifuged at 300g for 10 minutes, aspirated, and resuspended in another 2mL 1x permeabilization buffer. Following one more spin and aspiration, the cells were resuspended in 100uL permeabilization buffer. 1ug eBioscience purified anti-mouse CD16/CD32 (clone 93) was added to block Fc binding. The tubes were incubated for 15 minutes at 4 degrees in the dark. 1ug of eBioscience 33

PAGE 34

anti-mouse Foxp3 PE (clone FJK-16a) or 1ug eBio science Rat IgG2ak were then added to the test and isotype tubes, respectivel y. The tubes were incubated in th e dark at 4 degrees for another 30 minutes. 2mL permeabilization buffer was added to each tube, which were then centrifuged for 10 minutes at 300g. The cells were aspirated, resuspended in another 2mL permeabilization buffer, and centrifuged again. The cells were aspi rated and finally resuspended in 250uL stain buffer. All tubes were stored in the da rk at 4 degrees until time of analysis. Cells were analyzed on a BD FACScalibur flow cytometer within 24 hours of staining using Cell Quest software on the Apple OS. Co mpensation tubes were used to acquire the appropriate instrument settings, which were th en used throughout each study (i.e. compensation tubes were rerun for each phase of the study to account for instrument "drift."). 1x105 cells were run per tube. The data were analyzed usi ng FCS Expression v.3 software (DeNovo Software, Thornhill, Ontario, Canada). Isotype controls were used to determ ine positive staining such that 99%+ of the isotype was gated as negative. Gati ng for regulatory T cells began with forward and side scatter to isolate the lymphocyte populati on. Cells were then ga ted for CD4, CD25, and Foxp3. Regulatory T cells were cons idered as triple positive. Multiplex Cytokine Analysis A collection of both serum and cell culture supern atants were analyzed for their levels of cytokines using the Luminex xMAP System (Austin, TX, USA). This instrument has several advantages over the use of tradit ional ELISAs, such as a diminish ed required sample volume and the ability to detect up to 100 different analytes in the same sample with an improved range of detection. Several kits specific to this inst rument were used throughout the studies. The particular panels used are de scribed in subsequent methodology. 34

PAGE 35

35 Statistical Analysis GraphPad Prism v4.0 software (S an Diego, CA, USA) was used for statistical analysis of the data. Significance was defined as P<0.05. Kapl an-Meier survival curves were used for analysis of diabetes remission. Additional analyses of cytokines, flow cy tometry, and other data were analyzed with either two-tailed unpaired t tests or by one way ANOVA, depending upon the particular data being analyzed. This is de scribed in greater detail in subsequent methodology.

PAGE 36

CHAPTER 3 REVERSAL OF OVERT TYPE 1 DI ABETES IN THE NOD MOUSE USING COMBINATION IMMUNOSUPPRESSIVE THERAPIES This chapter continues to address the rationa le behind the use of co mbination therapy and subsequently carries one such therapy out for the purpose of reversal of ove rt type 1 diabetes in the NOD mouse. In particular, these experime nts focus upon the augmentation of mATG-therapy with the addition of the immunomodulator rapamy cin. These experiments will demonstrate that the therapeutic criteria for success, most notab ly the control of normal blood glucose and an improvement in pancreatic histology, did not reve al any apparent benefit through the addition of Rapamycin. Introduction: Two Agents of Therapeutic Efficacy The prevention of T1D in the NOD mouse has been achieved through a variety of immunomodulatory agents (5561). Although the reversal of overt T1D in the NOD has proven to be more challenging than prevention, this too has been achieved by both anti-cd3 (28-30; 62) and, in our lab, by mATG (49). It is the goal of this study to improve upon the reversal achieved by mATG monotherapy by administering a seco nd immunomodulator for the purpose of enhancing and/or maintaining the imm unoregulatory benefit achieved by mATG. The mATG is produced by inoculating New Zealand White rabbits with a mixture of mouse thymocytes. The rabbits incur a polyc lonal antibody response against the foreign thymocytes. The IgG portion of this response is collected and purified in the preparation of mATG. ATG is currently used clinically for a variety of purposes including the treatment of acute rejection, graft versus hos t disease, and conditioning for stem cell transplantation (63-65). It appears to target more than 40 epitopes acts partially through lymphocyt e depletion as well as the induction of regulatory T cells (66-72). 36

PAGE 37

Our criteria for selecting the second agent included a demons trated efficacy in the NOD model, namely prevention, as well as FDA approval for the sake of easing c linical translation of any benefit we might demonstrat e in this preclinical setting. Given these considerations, we chose the immunomodulator rapamy cin. Rapamycin acts by halting cell cycle progression at the G1 to S phase transition through the inhibition of mTOR. Previous work has demonstrated that rapamycin is capable of inducing regulatory T cells (73) and is capable of late T1D prevention in the NOD mouse both as a monotherapy and when combined with IL-10 therapy (74; 75). This chapter presents and discusses results of a T1D reversal study in the NOD mouse using both mATG and rapamycin. The impact of these therapies upon blood glucose, Tlymphocyte populations, and pancreatic histology will be examined. Subsequent to these findings, the concept of combination therapy in re gards to this combination and others will be discussed in detail. Methods These studies dealt with the diagnosis and attempted reversal of overt hyperglycemia in NOD mice using ATG and/or Rapamycin. As such, the methods described here provide details on the administration of these compounds as well as the analyses performed, including but not limited to immunohistochemistry and c-peptide analysis. Experimental Design In this study, female NOD mice were mon itored for the onset of overt hyperglycemia beginning at 11 weeks of age as determined th rough blood glucose. Once confirmed as diabetic, the animals were placed into one of four treatment groups: cont rol (rIgG + control pellet), ATG, Rapamycin, or ATG combined with rapaymcin. All animals received two subcutaneous LinBit insulin pellets while anaesthetized with isoflu rane as well an AVID Fr iendChip ID microchip. 37

PAGE 38

Rabbit immunoglobulin or ATG were administered at day 0 and 3 at a dose of 500ug/dose. Rapamycin was administered continuously for up to six weeks (Figure 3-1). All animals were monitored for the potential return to hyperglycemia (upon exhaustion of the implanted insulin pellets) on a weekly basi s. Upon confirmation of either hyperglycemia or upon reaching the 120 day endpoint, all animals were sacrificed and various analyses performed upon selected tissues. Blood Glucose Criteria As described in general methods, blood glucose was determined using the One Touch Ultra testing system. Onset of diabetes was de fined as a non-fasting blood glucose reading of 240mg/dL or higher on two consecutive days. Blood glucose was measured once per week. Failure of therapy was determined in the same fa shion as the original di agnosis of hyperglycemia such that two consecutive elevated (240+mg/dL) readings were grounds for the sacrifice of the animals. Implementation of Rapamycin Pellet The administration of rapamycin posed a moderate challenge given the number of mice and the duration of therapy. Due to the limited so lubility of rapamycin, the typical formulation involves a suspension in a mix of carboxymethylce llulose and must be given orally. Due to safety concerns over repeated or al gavages, we opted to purchase custom-made subcutaneous sustained-release rapamycin pellets (Innovative Re search of America, Sarasota, FL, USA). The pellets were formulated for a 60 day release with first order kinetic s providing a dose of 24 ug/day. Placebo pellets were also provided for the control group. The pellets were implanted using the same method as the implantation of the LinBit insulin pe llets using the same perforation in order to minimize st ress and injury to the animals. 38

PAGE 39

Analysis of C-peptide Serum Concentration C-peptide I and c-peptide II levels in sera were an alyzed via enzyme-linked immunosorbant assay (ELISA). The kits were bo th purchased from Alpc o Diagnostics (Salem, NH, USA). Immunohistochemistry Several organs were harvested following sacrif ice for histological analysis. The mice were sprayed with 70% ethanol and sterile instruments were used to remove the organs. The organs collected included spleen, pancreas, salivary gl ands, pancreatic draining lymph node (PLN), thymus, and kidney. Tissues were placed in cassettes and fixed overnight in a 10% formalin solution (Fisher Scientific, Pittsburgh, PA, USA). Cassettes were then placed in 1x PBS and sent to the University of Florida Pathology Core for processing. All tissues were stained with hematoxylin and eosin (H&E). The pancreas was also co-stained for insulin and bromodeoxyuridine (BrdU). Insulitis Scoring Insulitis scoring was performed on H&E pancreat ic sections from mice sacrificed either at the time of therapy failure or the predetermine d successful endpoint of 133 days. The islets of langerhans were scored from 0 to 3. A score of 0 was considered an islet free of any infiltrate. A score of 1 indicated peri-insulitis, while a score of 2 was considered infiltration in up to 50% of the islet area. Finally, a score of 3 was given to islets with infiltration in greater than 50% of the islet area. Results In Vivo Survival Curves of NOD Mice The most critical measurement of success for therapy was the ability to maintain euglycemia. Beginning therapy at onset of diab etes, all animals received two subcutaneous 39

PAGE 40

insulin pellets capable of controlling blood gluc ose for approximately three weeks. Once these pellets were spent, any unsuccessful therapy wo uld become evident by a spike in blood glucose toward hyperglycemic (i.e. glucose > 240mg/dL) le vels, whereas a successful therapy would be capable of maintaining euglycemia without th e presence of exogenous insulin. As such, the blood glucose values were translated in success and failure by time with failure defined as consecutive measurements in the hyperglycemia range as previous ly described. The survival of mice treated at diabetes onset with control, AT G, RAPA, or a combination of ATG and RAPA is shown as a Kaplan-Meier survival curve out to as many as 133 days post-onset (Figure 3-2). There was not a significant difference between the control and RAPA-treated mice. There was also no significant difference between ATG and ATG+RAPA-treated mice. Both ATG and ATG+RAPA exhibited a significant (p<0.05) diff erence between both control and RAPA-treated mice. C-peptide Analysis In order to differentiate any remaining e xogenous insulin-derived glucose control and endogenous control, a marker other than insulin was chosen. Mice express two insulin genes. As a consequence, we measured levels of both C-pe ptide I (Figure 3-3a) and C-peptide II (Figure 33b) in sera collected at time of sacrifice. For c-peptide II, the only stat istical significance was observed in an increase in ATGtreated mice versus control-treated mice (p< 0.05). No other significant differences were observed. For c-peptide I measurements, both AT G and RAPA-treated mi ce exhibited significant increases versus control-treated mice (p< 0.01 and p<0.02, respectively). ATG+RAPA-treated mice trended higher, but, surprisingly, were not significantly differe nt from controls. 40

PAGE 41

Flow Cytometry Analysis of T Cell Subpopulations In order to explain the immunological mechanis m(s) behind the metabolic differences seen with each therapy, we chose to analyze the T-ly mphocyte profiles of all animals at the time of sacrifice. Splenocytes obtained at time of sacrifice were stained for the analysis of T cell subpopulations. The percentage of splenic regulatory T cells (C D4+CD25+Foxp3+) was analyzed (Figure 3-4a). The rati o of splenic CD4:CD8 T lymphocyt es was also analyzed (Figure 3-4b). Due to a limited sample size, the percentage of regulatory T cells was not statistically significant when comparing diffe rences between therapies, although ATG and RAPA-treated mice trended higher than controls. The splenic CD4:CD8 ratio was significantly higher in the combination-treated mice versus control (p=0.01) as well as in ATG-treated mice versus control (p<0.01) indicating a long-term (up to 133 days) impact of ATG upon the T-lymphocyte profile. The ratios also trended higher in RAPA -treated mice versus controls (p=0.14). Insulitis Scoring Apart from indirect metabolic analyses, we perf ormed an analysis of the pancreas itself, in order to observe the extent of protection afforded the islets by the given therapies. Insulitis scoring was performed on H&E stained pancre atic sections from mice obtained at time of sacrifice. Scoring was represented as a percentage of a given score out of the total number of islets observed (Figure 3-5). Although the scoring did not reveal statistically significant diffe rences, several trends were observed. The percentage of healthy (i.e. score of 0) islets trended (p= 0.07) toward a greater percentage in ATG-treated mice versus controltreated mice. In additi on, the percentage of heavily infiltrated islets (i.e. sc ore of 3) was reduced in ATG-treated mice versus control-treated to the brink of significance (p=0.0585). 41

PAGE 42

Discussion This study confirms the efficacy of ATG for th e purpose of reversing overt diabetes in the NOD mouse by achieving reversal rates approaching 25%. These data indicate that ATG is capable of inducing long-term glucose control in these animals in conjunc tion with increases in c-peptide I and c-peptide II seru m levels. In addition, a long-term impact was seen on the T cell profile with a trend toward in creasing regulatory T cells as we ll as a significant long-term increase in the CD4:CD8 ratio. Th is increased ratio is curious as it contrasts with the reported decreased ratio seen w ith a similar therapy, an ti-cd3 mAb (28; 29). While that study implicated the induction of CD8 regulatory cells, we observe d a trend toward increasing CD4 regulatory T cells instead. Rapamycin monotherapy yielded contrasting re sults. The measured blood glucose values, as observed by survival in Figure 3-2 clearly dem onstrate the inability of rapamycin to control diabetes in this setting. In sp ite of this, c-peptide measuremen ts were substantially greater (significantly so for c-peptide I) than in control-treated mice. In addition, the percentage of regulatory T cells trended higher when compared with control-treated animals, as would have been predicted from previous studies in the NOD mouse (73). Insulitis scoring also demonstrated a trend toward a greater percentage of healthy islets than in controls. This may indicate that success and failure are divided by a knife edge based on the similar data between ATG-treated (success) and rapamycin-treated (failure) mice. It may also indicate the presence of a secondary complicating factor induced, but not measured, by this therapy. Curiously, despite achieving the same reve rsal rate as ATG monotherapy, ATG and rapamycin combination therapy yielded different immunological and metabolic data. Combination therapy yielded a lo wer percentage of regulatory T cells and c-peptide (I and II) 42

PAGE 43

levels indicating that th e effect of these two therapies was not additive, and that they may, in fact, interfere with one another in an as-yet undetermined manner. The ultimate result is that rapamycin therapy could be considered a failure when combined with ATG despite its ability to improve the immunological and metabolic markers chosen for this study. This raises the possi bility that while rapamycin ma y be acting as intended, an unintended secondary effect may be taking place. Previous clinic al studies have demonstrated that rapamycin therapy yields an unintended consequence of pe ripheral insulin resistance and modulates insulin receptor substrate phosphorylation ( 76-79). If this were to occur in the setting of new-onset diabetes, any negative impact upon beta cell function would be immensely detrimental given the greatly reduced beta ce ll mass at this time. Any future study using rapamycin in this setting will therefore be best served by analyzing th e potential induction of peripheral insulin resistance. In summary, while both ATG and rapamyci n exhibit immunological and metabolic benefits in the setting of new-onset T1D in the NOD mouse, only ATG wa s capable of reversing hyperglycemia for a prolonged period of time (up to 133 days). Despite the rationale behind combining these two therapies and the apparent strength of each, achieving long-term remission of diabetes remission remains a challenging task Future studies may be best served by complementing ATG with a second agent that target s a second pathway other than the control of autoimmunity, such as inducti on of beta cell regeneration. 43

PAGE 44

Figure 3-1. ATG+RAPA reversal study experime ntal design describing the timing and duration of individual treatments. New onset NOD mi ce received either ATG or rIgG as well as either RAPA or saline. All animals r eceived 2 insulin pellets at onset. Arrows indicate timing of a given treatment. Figure 3-2. Kaplan-Meier surv ival curve of NOD mice follo wing treatment at onset of hyperglycemia with ATG and/or RAPA. Bo th ATG (n=11) and ATG+RAPA (n=11) treatments led to signifi cant (p<0.05, Log-Rank Test) im provement in survival vs. control (n=12). 44

PAGE 45

Figure 3-3. Both ATG and RAPA lead to signifi cant increases in serum c-peptide levels at the time of sacrifice. C-peptide II is significan tly increased in A) ATG-treated (n=8) mice (p<0.05, unpaired t test) and c-peptide I is significantly increased in B) both ATG and RAPA-treated (n=5) mice (p<0.01, p<0.02 resp ectively, unpaired t test) vs. control (n=7). Figure 3-4. Flow cytometry anal ysis of the regulator y T cell population and CD4:CD8 ratios in splenocytes of treated mice at time of sacr ifice revealed modulation by both ATG and RAPA. A) Both ATG and RAPA therapies tr ended toward greater Treg levels than controls (p=NS, unpaired t test). B) ATG (n=4) therapy (p<0.01, unpaired t test) and ATG+RAPA (n=6) therapy (p=0.01, unpaired t test) yield a greater CD4:CD8 ratio than controls (n=4). 45

PAGE 46

Figure 3-5. Insulitis scoring of treated mice reveals a marginal benefit of ATG therapy in the percentage of healthy islets. ATG (n=4) i nduced a trend toward a higher percentage of healthy islets (p=0.07, unpa ired t test) and toward a lo wer percentage of heavily infiltrated islets (p= 0.0585, unpaired t test) ve rsus control (n=4). 46

PAGE 47

CHAPTER 4 AUGMENTATION OF IMMUNOSUPPRESSION IN OVERTLY DIABETIC NOD MICE VIA AAT THERAPY We have achieved successful reversal of overt T1D in the NOD mouse with ATG monotherapy, but did not experience efficacy e nhancement when rapamycin was added as a combination therapy. Consequently we decided to change our tact such that we would maintain our desire to reverse T1D using combination therapy by using ATG as our base therapy, but that we would choose a different pathwa y of action for our second agent. If one were to consider the various pathways capable of being target ed for the purpose of T1D reversal, three might come to mind. The firs t, which appears to be addressed successfully by ATG, is the control of autoimmunity. If the au toimmune destruction of beta cells is achieved, that leaves two other pathways to consider. The fi rst is the protection of remaining beta cells at onset of diabetes. There are severa l reports indicating that a substa ntial percentage of beta cells remain at onset (80), therefore the protection of these cells may allow for the eventual recovery of glucose control in the absence of exogenous insulin. The second potential pathway to target would be the initiation or acceleration of beta cell regeneration or ne ogenesis. In this chapter, the use of alpha-1-antitrypsin (AAT) will be discussed as a secondary agent used in combination with ATG for the purposes of beta cell protection. Introduction: AAT For the Protection of Beta Cells AAT is a naturally occurring glycoprotein and a member of the serine protease inhibitor (SERPIN) family with an array of physiologi cal functions (81-83). Perhaps most notably, AAT is produced by hepatocytes for the purpose of inhibiting damage to lung tissue caused by neutrophil elastase and prot einase 3 (84). In a subset of emphys ema patients, the disease is due to the production of a mutant form of AAT such that lung damage occurs. Consequently, exogenous AAT therapy has been used for thes e individuals since th e 1980s (85). Other 47

PAGE 48

functions include the inhibition of trypsin, thrombin, and cathepsin G (84). Apart from acting as an antioxidant, AAT has been shown to exhibit increased expre ssion in response to inflammation, with levels rising as much as 4-fold in the acute phase (84). Of particular interest to the field of autoimmunity, AAT has been shown to inhibit inflammatory cytokines and alter effector T ce ll trafficking. More spec ifically, AAT has shown promise for the treatment of diabetes with seve ral studies demonstrating the ability of AAT to prolong islet allograft survival (86), protect against streptozotocin (STZ)-mediated beta cell destruction in mice (87), and th e prevention of T1D onset in the NOD mouse (88). The activity of AAT in sera from type 1 diab etic patients has also been show n to be reduced versus healthy controls (85; 88). Based on these data, we believe that ATG in co mbination with AAT at the time of diabetes onset in the NOD mouse would serve to cont rol autoimmunity, induce a regulatory T cell population, and protect the remaining beta cells fr om further destruction. In this chapter, the results of this study will be pr esented and discussed, and an une xpected caveat will emerge that will severely impact the me thod of AAT administration. Materials and Methods These studies once again dealt with the diagnosis and attempted reversal of overt hyperglycemia in NOD mice. The reagents used in this project were ATG and/or AAT. As such, the methods described here provide details on th e administration of these compounds as well as the attempts to prevent anaphylaxis. Experimental Design In this study, female NOD mice were mon itored for the onset of overt hyperglycemia beginning at 11 weeks of age as determined th rough blood glucose. Once confirmed as diabetic, the animals were placed into one of five treatment groups: control (rIgG + saline), ATG, 48

PAGE 49

Prolastin, or ATG combined with Prolastin with the Prolastin eith er at onset or delayed for one week (Figure 4-1). All animal s received two subcutaneous LinBit insulin pellets while anaesthetized with isoflurane as well an AVID FriendChip ID microchip. Rabbit immunoglobulin or ATG were administered at day 0 and 3 at a dose of 500ug/dose. Prolastin was originally sche duled for two weeks of daily subcutaneous injections. For reasons that will become evident in the results, the dose timing was altered to four doses every other day either beginning at ons et or delayed for one week, then administered every other day for four injections. All animals were monitored for the potential return to hyperglycemia (upon exhaustion of the implanted insulin pellets) on a weekly basi s. Upon confirmation of either hyperglycemia or upon reaching the 120 day endpoint, all animals were sacrificed and various analyses performed upon selected tissues. Preparation of AAT Alpha-1-antitrypsin is commercially available in several formulations. For this experiment, we used Prolastin (Talecris Biotherapeutics, Research Triangle Park, NC, USA). The Prolastin was reconstituted according the manufacturer's guidelines and was administered subcutaneously at a dose of 2mg per injection. Reagents for the Prevention of Anaphylaxis In the course of performing this experiment, we encountered sudden, fatal anaphylaxis that necessitated an attempt to preven t the anaphylaxis. In order to accomplish th is, a combination of a PAF antagonist and an antihistamine were us ed. The H1 antihistamine, triprolidine, was purchased from Sigma-Aldrich (St. Louis, MO, US A) and injected IP 45 minutes prior to the fourth (and any future) injection of AAT at a dose of 200ug in 100uL of sterile saline. The PAF antagonist, CV-3988, was purchased from Wako Chemicals USA Inc. (Richmond, VA, USA) 49

PAGE 50

and administered at a dose of 66ug in 100uL sterile saline at pH 7.4 via in travenous tail injection 5 minutes prior to tr eatment with the AAT. Results Induction of Fatal Anaphylaxis Beginning this study, we intended to administer AAT via daily IP injections for a duration of two weeks. Quite quickly, however, we enco untered unexpected deaths in the animals receiving this regimen of AAT as well as in an imals receiving the albumin control treatments. Subsequent monitoring of the animals for the peri od following injections revealed acute deaths occurring within approximately 30 minutes typica lly after 1 week of th erapy. We decided to repeat the injection cycle, this time providi ng saline, albumin, or AAT on Mondays and Fridays to the three groups of non-diabetic NOD mice. The survival curve of this small trial was plotted (Figure 4-2). What we discovered was that during the sec ond week of therapy, animals receiving AAT began to experience acute death beginning as earl y as the third injection and completely by the fourth injection. Albumin administration resulted in a similar event, while saline-receiving mice did not experience any deaths by th e fifth injection. As is evident from this figure, the survival of albumin and AAT treated mice was significantly re duced (p<0.05, log-rank test for trend) when compared with saline-treated animals. Prevention of Anaphylaxis In order to use either albumin or AAT for the full duration as originally intended, we would be required to develop a method for preven ting the anaphylactic deat hs we had observed. Using a regimen of an H1 antihistamine, tr iprolidine, and a PAF antagonist, CV-3988, we performed pre-treatment of NOD mice receivi ng AAT, successfully prev enting anaphylaxis for up to 5 injections (Figure 4-3). We decided to re peat this experiment for a greater number of 50

PAGE 51

injections to determine whether this protection from anaphylaxis could be used long-term in a diabetes reversal study. What quickly became evident was that th e protection was not 100% effective with prolonged injections of AAT -possi bly for a variety of factors disc ussed later-, wi th only 2 of 5 mice surviving for up to 7 injections at which point the experiment was terminated (data not shown). In Vivo Survival Curves of Treated NOD Mice Despite the potential benefit of preventing anaphylax is using the above pre-treatment, the logistics and potential conflicting mechanisms behind such an approach proved to be incompatible with our desire to provide a safe defined mechanistic method for reversing overt T1D in the NOD mouse. Consequently we decided to administer a maximum of just four doses of either albumin or AAT, give n every other day over a period of 7 days, in order to avoid the onset of anaphylaxis. In order to discriminate any potential conflicting effects between ATG and AAT, two schedules (and as a result two cohorts of mice) were used such that one cohort received AAT at onset, while a second cohort re ceived AAT one week post onset following the initial dosing of ATG at onset. The survival curv es of all treatment groups were plotted (Figure 4-4). As expected, control treatments did not lead to long-term reversal of disease. AAT monotherapy at onset significantl y delayed the return of hypergly cemia versus control (p<0.03), but did not lead to long-term remission. AAT monotherapy delayed one week was far less effective and not significantly different from control. Surprisingly, ATG monotherapy yielded the greatest rate of remission (57%) and was si gnificantly improved (p<0.0001) versus control. ATG and AAT combination therapy at onset yielded a statistically similar result (50%) versus 51

PAGE 52

control (p<0.0001), while AAT administered one week post onset in conjunction with ATG actually reduced the remission rate ( 25%) versus ATG monotherapy (p<0.0001). Flow Cytometry Analysis of T Cell Subpopulations In order to determine the immunological im pact of each therapy on the T lymphocyte population, flow cytometry was performed on splenocyt es obtained at the time of sacrifice. The percentage of splenic regulatory T ly mphocytes was determined (Figure 4-5). ATG monotherapy, the most successful at mainta ining long-term euglycemia, also yielded the greatest percentage of re gulatory T cells and was improved significantly versus control treatment (p<0.05). While none were as significa ntly increased at ATG monotherapy, all other therapies yielded increased percentages of regulat ory T cells versus control with, surprisingly, albumin at onset (p<0.02), AAT at onset (p<0.02), and ATG+AAT at onset (p<0.03) significantly so. Discussion The viability of ATG to induce long-term e uglycemia in overtly diabetic NOD mice was once again demonstrated in this study. The abilit y of a second therapy to augment ATG's effect, however, was beset with complications. The induction of fatal an aphylaxis by either albumin or AAT proved to be an immense hurdle to overcome and severely dampened any benefit that may have been induced by the AAT therapy. This a pparent anaphylaxis in the NOD mouse is not unprecedented however. Previous work in the NOD has shown that administration of the insulin B:9-23 peptide is capable of inducing fatal anaphylaxis (89). As with our study, the use of a combination of triprolidine and a PAF antagonist was capable of preventing the anaphylaxis from occurring. A follow up study from the same group has implicated the release rate of th e peptide as correlating with the induction of anaphylaxis. In this se cond study, the peptide's isoelectric point was 52

PAGE 53

modified and subsequently offered protection from the previously observed anaphylaxis (90). More recently, our group in conjunction with a co llaborator has shown that the anaphylaxis that has been observed appears to be NOD-specific, with other strains such as C57Bl/6J not experiencing any deaths despite repe ated administration of protein (91). In spite of the anaphylaxis, it is important to note that the reversal study was performed, albeit with the albumin or AAT doses limited to f our injections. As was clear from the survival curves, the only benefit with AAT was observed when it was admini stered at onset, either alone or in combination with ATG. Delaying the dose until one week post-onset ablated the effect of the monotherapy and actually diminished the therapeutic potential of ATG when used in combination. Whether a longer dosing timetable (e .g. 2 weeks or greater) would have afforded greater protection is uncer tain. Given the increase in the percentage of regulatory T cells in virtually all groups when compared with control, the question of a nonspecific response is raised. Given the intense anaphylactic response observe d from both albumin or AAT, it is reasonable to presume that anti-albumin or anti-AAT antibodies are produced even with just four doses given that any additional doses induces death. Whether the upregulation in the percentage of regulatory T cells is a natural response against anaphylaxis is speculatory and should be addressed in future studies. Given the obstacle of anaphylaxis in the NOD, there are several ways that AAT might be administered successfully for a prolonged period of time. Apart from directly preventing anaphylaxis using antihistamines and PAF antagoni sts, altering the rate of AAT release, as implied through the aforementioned isoelectric po int study, should amelio rate this response. Therefore an alternate route of injection, such as subcutaneous or the use of a slow-release osmotic pump may provide an accept able alternative. Ultimately it is important to note that the 53

PAGE 54

anaphylactic response appears to be NOD-specific. While this may complicate studies using AAT in this excellent mouse model is diabetes, it may be unreasonable to translate this negative side effect into future clinical use of AAT. This is especially true gi ven the clinical use of exogenous AAT since the 1980s with no reported cases of anaphylaxis (85). In summary, the use of ATG and AAT combin ation therapy for the purpose of reversing overt T1D in the NOD mouse was complicated by the induction of anaphylaxis following repeated injection of foreign protein. While ATG monotherapy was once again successful at reversing a substantial percentage (57%) of new-onset mice and AAT monotherapy at onset delayed hyperglycemia for greate r than 70 days post-onset, the combination of the two in the reduced dosing schedule did not result in any augm ented benefit. As a result, any future use of this combination will likely requ ire an alternative method of AAT administration in combination with a prolonged dosing schedule. In a ddition, this NOD-specifi c phenomenon should not dissuade the use of AAT in the clinic, given AAT's safety reco rd of approximately 20 years. 54

PAGE 55

Figure 4-1. ATG+AAT reversal st udy experimental design descri bing the timing and duration of individual treatments. New onset NOD mice received either ATG or rIgG as well as either AAT or Albumin. All animals receiv ed 2 insulin pellets at onset. Arrows indicate timing of a given treatment. Figure 4-2. IP administration of either albumin or AAT leads to the induction of fatal anaphylaxis in the NOD mouse. Either tr eatment (n=4) significantly diminished survival versus saline (p<0.05, Log-Rank Te st). Mice were injected on Mondays and Fridays for up to 5 injections. 55

PAGE 56

Figure 4-3. Short-term protecti on from AAT-induced anaphylaxis is afforded by pretreatment with antihistamine and PAF antagonist. n=4 Figure 4-4. Kaplan-Meier surv ival curve of NOD mice follo wing treatment at onset of hyperglycemia with ATG and/or AAT. AAT monotherapy (n=9, p<0.03, Log-Rank Test), ATG+AAT (n=8, p<0.0001, Log-Ra nk Test), ATG+AAT(onset) (n=8, p<0.0001, Log-Rank Test), and ATG (n =7, p<0.0001, Log-Rank Test) yielded significant increases in surviv al versus control therapy. 56

PAGE 57

57 Figure 4-5. ATG increases the percentage of sple nic regulatory T cells versus control in mice sacrificed at trial endpoin t (n=4, p<0.05, unpaired t test ). Albumin (onset), AAT (onset), and ATG+AAT (onset) also yi elded significant increases (n=6, p<0.02, p<0.02, p<0.03 respectively, unpaired t test) versus c ontrol (n=6).

PAGE 58

CHAPTER 5 ENHANCEMENT OF INSULIN SECRETION IN DIABETIC NOD MICE THROUGH INHIBITION OF DPPIV In our previous T1D reversal st udies, we have repeatedly de monstrated the ability of ATG to induce long-term remission of hyperglycemia in NOD mice. Despite this success, previous attempts to augment ATG therapy through the use of a second immunomodulator, rapamycin, and through the use of an anti-i nflammatory, alpha-1-antitrypsin, have proven unsuccessful via our methods of administration. Given these setbac ks, a third as-yet-untested pathway remains to be tested using the NOD model of diabetes. To date we have successfully addressed th e issue of autoimmun ity through the use of ATG. We have not achieved success through our attempt to preserve beta cell mass using AAT. The third pathway to address is whether the beta cell mass remaining at onset, of which there is believed to be a substantial amount (80), may be functionally enhanced and potentially be induced to regenerate. In this study, we will be examining the ability of an FDA-approved drug, Januvia, currently in use clinical ly for the treatment of type II diabetes, to address this third pathway. Introduction: Enhancing Beta Cell Function In both mouse and man, the body possesses th e ability to regulat e glucose control following intake of a meal. One such method to accomplish this relie s upon the release of glucagon-like peptide I (GLP-1). Upon nutrients entering the small intestine, L cells within the intestinal lumen respond by releasing GLP-1 into circulation (92). GLP1 enters the pancreas where it acts by stimulating insulin secretion as well as by supp ressing glucagon secretion (93). GLP-1 binds to its receptor, a heterotrimeric G protein on the beta cell (94). Upon binding the subunit is released leading to a chain of events culminating in the release of calcium by the sarcoplasmic reticulum which in turn enhances the release of insulin. GLP-1, however, has a 58

PAGE 59

very short half life of two minutes, as it is rapidly degraded by Dipeptidyl peptidase IV (DPPIV); in fact, less than fifty percent of the GLP-1 released in this manner reaches the pancreas. DPPIV is not only present in circulat ion but also membrane-bound on a variety of tissues throughout the body such as the thymus, spleen, pancreas, kidneys, and the intestine (95). The function of GLP-1 has been of particular interest in the fi eld of type II diabetes for the purpose of counteracting insulin resistance by using this mechanism to enhance endogenous insulin secretion. One approach to achieving this is through the us e of a GLP-1 analogue, such as exendin-4, that is resistant to DPPIV-mediated degradation while yielding the ability to stimulate insulin secretion (96-99). This analogue, however carries with it the caveat of altered gastric emptying, the risk of hypoglycemia, and se vere complications, albeit rare (100). Instead of using a synthetic GLP-1 analogue, su ch as exendin-4, we chose to use a DPPIV inhibitor, Januvia (Sitagliptin) to prolong the half -life of endogenous GLP-1. Unlike exendin-4, Januvia does not alter gastric emptying and, to date, has not been associated with the same, rare complications as exendin-4 nor with hypoglycemia (101-104). Sitagliptin exhibits first order kinetics and has a half life of approximately twelve hours (95). This inhibitor has already demonstrated potential for the treatment of T1D in NOD mice based upon previously published reports. In one such study, a sitag liptin analogue has been reported to preserve beta cell mass in streptozotocin-diabetic mice fed a high fat diet (103). Based upon the documented safety and efficacy record of Sitagliptin, it was our belief that it would e ffectively enhance the activity of the beta cell mass remaining in NOD mice at the time of diabetes onset. Apart from the potential metabolic benefit a fforded by Sitagliptin, we also planned to study the potential impact upon th e immune profile based upon th e expression of DPPIV on the surface of T lymphocytes (CD26) as well as the potential immunomodulator y function of soluble 59

PAGE 60

CD26. CD26 has a demonstrated role in T ly mphocyte activation via binding to adenosine deaminase (ADA) on the surface of antigen presen ting cells (105), while sCD26 has been shown to enhance transendothelial migration via the ma nnose-6-phosphate/insulin-lik e growth factor II receptor (106). While it is unlikely to have a dramatic immunological effect, given the extensive safety record, we believed it was important to pe rform relevant analyses considering the altered state of immune regulation in thes e mice at time of diabetes onset. Materials and Methods These studies once again dealt with the diagnosis and attempted reversal of overt hyperglycemia in NOD mice. The reagents used in this project were ATG and/or Januvia. As such, the methods described here provide details on the administration of these compounds as well as the metabolic and immunological experiments that were performed. Experimental Design In this study, female NOD mice were mon itored for the onset of overt hyperglycemia beginning at 11 weeks of age as determined th rough blood glucose. Once confirmed as diabetic, the animals were placed into one of four treatment groups : control (rIgG + saline), ATG, Januvia, or ATG combined with Ja nuvia. All therapies began at onset. All animals received one subcutaneous LinBit insulin pellet while anaesthetized with isoflurane as well an AVID FriendChip ID microchip. Rabbit immunoglobulin or ATG were administered at day 0 and 3 at a dose of 500ug/dose. Januvia was administered orally at 1mg/dose in 100uL total volume daily for up to six weeks (barring failu re). 100uL total volume of saline was administered orally as a control for Januvia (Figure 5-1). All animals were monitored for the potential return to hyperglycemia (upon exhaustion of the implanted insulin pellets) three times per week. Confirmation of hyperglycemia was considered as a blood glucose greater than 300 mg/dL for one week. Upon confirmation of either 60

PAGE 61

hyperglycemia or upon reaching the 120 day endpoint, all animals were sacrificed and various analyses performed upon selected tissues. Preparation of Januvia Januvia is sold commercially in 100mg tablets. In order to convert the Januvia into a viable format to treat mice, each tablet was dissolved in 10mL sterile saline. The suspension was filtered twice. The first filtration was performe d using a 100um nylon filter. The second filtration was performed using a 0.45um pore size PES filter (Whatman, Kent, UK) to yield a translucent solution. DPPIV Activity Assay In order to determine whether the filtration me thod or choice of saline suspension affected the viability of Januvia's DPPIV inhibitory activity, it was necessary to perform a DPPIV activity assay. A DPPIV-Glo Protease Assay was purchas ed from Promega (Madison, WI, USA) that measured inhibitory activity of the sample, in this case Januvia. CD26 Analysis Via Flow Cytometry CD26 was measured via flow cytometry in co mbination with stains for CD4 and CD8. All three stains were purchased from eBioscie nce. CD26-FITC (clone H194-112), CD4-PerCP (clone RM4-5), and CD8-APC (clone 53-6.7) were used. 1ug of each was added to 1x106 splenocytes collected, as previously described, at either failure or endpoint of the trial. The data obtained were analyzed using FCS Express v3.0. CD26 MFI was measured as expressed on ungated lymphocytes, CD4+ gated lymphocytes as well as CD8+ gated lymphocytes. Fasting IPGTT Intraperitoneal glucose tolerance tests (IPGTT) were performed in order to determine the robustness of the metabolic control of long-term (i.e. 120 day) reversed mice. In order to achieve 61

PAGE 62

a stable baseline blood glucose reading, animals were fasted overnight (12 hours) prior to IPGTT. Water remained available to the animals throughout the experiment. 30 minutes prior to injection of d-glucose, a baseline blood glucose reading was taken as previously described through a tail bleed. D-glucose was prepared by dilution of 800uL 50% dextrose solution in 1200uL sterile saline. Animal s were injected IP wi th 200uL total volume of the diluted solution at time 0 minutes. Blood glucose readings were taken at -30, 0, 20, 40, 60, and 120 minutes relative to dextrose injection. Blood samples were also taken for serum collection at -30, 10, and 60 minutes. Protease inhibitor cocktail, previously described, was a dded to each serum collection tube prior to each bleed. The effect of a one-time dose of Januvia (or sa line) was also tested via IPGTT. 12 week old non-diabetic, female NOD mice were fasted and treated as described above. Mice in this cohort (n=4 per group) were also treated orally at -30 minutes w ith either 1mg Januvia or saline in 100uL total volume. ELISA Analysis of sCD26 and SDF-1 The levels of soluble CD26 and stromal cellderived factor-1 (SDF-1 ) were measured from serum collected at the endpoint for all mi ce, either at failure or 120 days. sCD26 was measured using an R&D Systems DPPIV/ CD26 ELISA (catalogue DY954). SDF-1 was measured using an R&D Systems M ouse SDF-1 ELISA (catalogue MCX120). Measurement of GLP-1, Glucagon, and Insulin Due to the limited availability of serum collected during IPGTTs, GLP-1, glucagon, and insulin were measured via multiplex analysis. A Li nco Mouse Endocrine 3-plex kit (St. Charles, MO, USA) was used to measure these 3 analyt es using a Luminex xMap 100 system. Data was exported to and analyzed usi ng SoftMax Pro 4.8 software. 62

PAGE 63

Results Confirmation of DPPIV Inhibitory Activity In our study, we sought to administer Januvia to overtly diabetic NOD mice. Januvia, however, was only available in th e commercial pill form. Consequently we were obliged to dissolve the pellet in sterile saline followed by filt ration of the particulate in order to acquire a solution suitable for oral delivery. Given this devi ation from the clinical administration of the drug, there were concerns that our purification method would so mehow affect the ability of Januvia to inhibit DPPIV activity. To address this concern, we performed a DPPIV activity luminescence assay in which the impact of varyin g doses of Januvia could be determined (Figure 5-2). We determined that the filtered Januvia reta ins a robust ability to inhibit DPPIV activity, such that as little as 100ng/mL of Januvia was capable of ablating virt ually 100% of the DPPIV activity. Januvia-mediated Effect Upon IPGTT Given that our preparation of Januvia app eared suitable for administration to NOD mice and with our understanding of the known mechanism of action, we sought to test whether Januvia would have an impact in the in vivo setting. We chose to perform an intraperitoneal glucose tolerance test (IPGTT) with pretreatment of Januvia or saline 30 minutes prior to injection of dextrose. We then measured blood glucose values at predetermined time points (Figure 5-3). With a limited cohort (n=4 per group, 12 w eek old non-diabetic female NOD mice), we observed a non-significant trend toward highe r blood glucose values in, surprisingly, Januviatreated mice versus control. Possible reasons for this result, albeit non-sig nificant, are presented in the discussion. 63

PAGE 64

Endocrine Analysis of IPGTT Sera The results of the initial IPGTT were unexpect ed, but the blood glucose is only one aspect that was measured. Serum was collected from these same mice during the IPGTT at -30, 10, and 60 minutes. The level of insulin in each serum sample was then measured (Figure 5-4). Insulin levels at the -30 minute time point were not significantly different between treatments. Insulin levels 10 minutes after injection of dextrose, not surprisingly, rose, with Januvia-treated mice trending highe r. Finally, at the 60 minute tim e point, Januvia-treated mice exhibited significantly higher leve ls of serum insulin when comp ared with control-treated mice (p=0.0015). Blood Glucose Values of Treated NOD Mice Following our examination of Januvia's ability to modulate aspects of an IPGTT in prediabetic NOD mice, we sought to treat overtly di abetic NOD mice with control, ATG, Januvia, or a combination of ATG and Ja nuvia (ATG+JAN). The blood glucose values, measured 3 times per week (Figure 5-5). Control and Januvia-treated mice all returned to hyperglycemi a within 40 days of onset, with Januvia-treated mice exhibiting a slight delay in this failure versus control. Both ATG and ATG+JAN-treated mice experienced a significant percentage of euglycem ia out to the 120 day study endpoint. In Vivo Survival Curves of Treated NOD Mice In order to more accurately co mpare therapies, the blood glucose values were plotted in terms of percentage of euglycemia animals by treatment by time post-onset. The Kaplan-Meier curve of these results was plotted (Figure 5-6). As we have observed in previous studie s, ATG monotherapy yielded significant enhancement of long-term euglycemia versus control (p<0.0001) with animals successfully 64

PAGE 65

treated until the 120 day study e ndpoint. Januvia-treated mice, in the absence of any immunosuppression, all returned to hyperglycemia, but did so at a significantly later time point versus control-treated mice (p=0.0035). The comb ination of these two monotherapies, however, did not appear to be synergistic with a result nearly identical to that of ATG monotherapy when compared with control (p<0.0001). Endpoint IPGTT of Reversed NOD Mice An IPGTT was performed on all mice that successfully reached the 120 day endpoint while remaining euglycemic. Both blood glucose (Figure 5-7a) and serum insulin (Figure 5-7b) was measured during the IPGTT. These analyses were limited largely by the number of mice reaching the 120 day endpoint successfully. Blood glucose cont rol during IPGTT for both ATGtreated and ATG+JAN-treated mice was diminished versus non-diabetic 12 week old control NOD mice. Serum insulin measurements, however, appeared to replicate the results from Figur e 5-4 in that the addition of Januvia to ATG appeared to augment the serum in sulin content (p=0.0571) despite the lack of overt differences in glucose control. Reasons for this discrepancy will be discussed. sCD26 Concentrations Following In Vivo Therapy To address the question of whether Januvia, a DPPIV inhibitor, would impact the levels of sCD26, either by inducing the shedding of CD26 fr om the surface of cells or by affected sCD26 directly, we measured the levels of sCD26 in serum from contro l and Januvia-treate d mice at the time of sacrifice (Figure 5-8). Analysis of the sera revealed a Januvia-i nduced significant increas e (p=0.0223) in serum sCD26 levels. The potential source of this increase and the subsequent physiological impact will be discussed later in this chapter. 65

PAGE 66

Flow Cytometry Analysis of T Cell Subpopulations In order to determine the immunomodulatory im pact of each therapy, a series of analyses were performed on splenocytes obtained at time of sacrifice via flow cytometry. The Treg:CD8 ratio (Figure 5-9a), the Treg:CD4eff ratio (Figure 5-9b), and CD4:CD8 ratio (Figure 5-9c) were plotted. In order to determine the effect in the regulatory profile, the ratio of regulatory T cells to CD4+ and CD8+ effector T cells was analy zed. Both ATG and ATG+JAN, by virtue of preferential depletion of CD8+ T cells by ATG, exhibited si gnificantly increased CD4:CD8 ratios as well as si gnificantly increased Treg:CD8 ratios. The ratio of Treg:CD4 effector T cells, which would not be impacted by differential CD 4 versus CD8 ATG-mediated depletion, was not significantly higher in any treatment versus control, but tren ded heavily toward significance in the ATG+JAN combin ation group (p=0.0653). Given the limited sample volumes obtained at th e time of sacrifice, a determination of the functional capacity of these regula tory T cells, such as through an in vitro suppression assay, was not determined. SDF-1 Serum Concentrations and Surface CD26 Following Therapy One of the goals of this study was to observe whether any immunomodulatory, rather than solely metabolic, changes were evident thr ough the use of this Januvia-mediated DPPIV inhibitory therapy. Previous re ports have indicated the potential impact of DPPIV inhibitors upon the levels of stromal-cell derived factor 1 (SDF-1). As a substrate for DPPIV/CD26, an alteration in the levels of SDF-1 could have a significant impact upon th e immune profile given its role in the tissue invasion and chemotaxis. We measured serum levels of SDF-1 from samples obtained at the time of sacrifice in both sa line and Januvia-treated mice (Figure 5-10). 66

PAGE 67

Clearly, Januvia appears to have no significant impact upon the levels of SDF-1 in the serum compared with control-treated mice. Anothe r potential effect was the alteration of surface CD26. Given the documented role of CD26 as a costimulatory marker on the surface of T lymphocytes, we measured the levels of CD26 on both CD4+ and CD8+ T lymphocytes from splenocytes of mice at time of s acrifice from mice treated with either control or Januvia (Figure 5-11). Discussion As we have demonstrated previously, ATG was capable of inducing long-term remission from hyperglycemia when administered at di abetes onset to NOD mice. Januvia monotherapy yielded a significant increase in the time require d before euglycemia was lost when compared with control-treated mice. Combination ther apy of ATG and Januvia, however, was not significantly different from ATG monotherapy for the purpos e of prolonging euglycemia. Despite the apparent lack of reversal enhancement when combined with ATG, Januvia did appear to have a beneficial metabolic impact a nd potentially a slight impact upon certain aspects of immune regulation. In both the treatment of prediabetic 12 week old and long-term post-onset euglycemic mice treated with Januvia and ATG+J AN, respectively, no significant change was observed in the glucose response curve of an IPGTT when compared with saline or ATG-treated mice, respectively. Despite this, significant increases in serum insulin were observed in these mice during the same IPGTT. This curious, but perh aps explains the results of the Kaplan-Meier survival curve when comparing controls and Januvia-treated mice. While both do not induce long-term remission from hyperglycemia, Januvia th erapy significantly prolongs the euglycemic duration. Since we have no i ndication of any immunoregulat ory shift following Januvia 67

PAGE 68

administration, it appears that Januvia is simply ab le to increase the insulin released from beta cells through the previously described GLP-1 pathway. Soluble CD26 was analyzed from mice receivi ng either saline or Januvia therapies. Previous reports have noted the potential role of sCD26 as a marker for disease activity in systemic lupus erythematosus (SLE) as well as possessing the potential fo r uptake in monocytes leading to increased T ce ll proliferation (107; 108). sCD26 serum levels from the Januvia-treated mice were increased compared with control-trea ted diabetic mice. While we can state that Januvia did not reduce the levels of sCD26 in spite of its DPPIV i nhibitory capacit y, it is unclear whether the therapy itself is res ponsible for this increase. We observed significantly lower levels of sCD26 in pre-diabetic 12 w eek old NOD mice, leading to th e possibility that sCD26 may simply increase with the duration of disease (thus supporting the similar notion observed in SLE). We did not observe any overt increase in T cell proliferation, therefore we are unable at this time to state whether ch anges in sCD26 have any impact upon the T cell phenotype. Future studies will be required. A comparison of splenic regulator y T cells revealed significant increases in the ratio of Treg to CD8 effectors in both ATG and AT G+JAN-treated mice. When comparing Treg to C effectors, only the combination therapy of ATG and Januvia trende d toward significance (p=0.0653). The CD4 to CD8 ratios were also significantly increased in both ATG and combination therapy groups. Ultimately any enhancement of the regulatory T cell profile induced by the addition of Januvia wa s slight, at best. This is s upported by the lack of any longterm enhancement observed in the previously-discussed survival curves. D4 Two peripheral immunomodulatory avenues of interest were analyzed following Januvia therapy based upon previously reported effects of DPPIV inhibition. SDF-1, which plays an 68

PAGE 69

important role in chemotaxis, is degraded by D PPIV, therefore inhibiti on by Januvia may affect its levels (109). In addition, CD26 is present on the surface of T lymphocytes and is reported to play a role in costimulation (110). High surface leve ls of CD26 are also reported to be present in cases of allergic asthma and multiple sclerosis (111; 112). We did not observe any significant impact of Januvia upon either the levels of SDF-1 or upon the surface expression of CD26 on T cells. It is conceivable that any effect of Januvia upon SDF-1 or surface CD26 was site-specific and would not show up in the serum or spleen, re spectively, but given the relatively large dose of Januvia it is likely there is no impact of consequen ce in this regard. In spite of this, any future studies using other DPPIV inhibi tors that are approaching the market, such as Vildagliptin, should be analyzed for any potential impact upon these pathways (113). Ultimately, the greatest impact of Januvia therap y appeared to be metabolic rather than immunomodulatory. Januvia increased serum insulin content in the setting of an IPGTT and also prolonged the duration of euglycemia in new-onset diabetic mice when co mpared with control therapy. While no significant difference was observe d by combining Januvia with ATG, this may be due to our 6 week window of therapy. Observi ng the survival curves, there appeared to be a delay in the return to hyperglycemia in mice that failed versus those that failed using ATG alone. This difference appeared to disappear following cessation of Januvia therapy. Given the apparent lack of immunomodulatory impact and failure to induce long-term diabetes remission, Januvia is unlikely suitable for further study for purpose of T1D intervention therapy. 69

PAGE 70

Figure 5-1. ATG+Januvia reversal study experimental design describing the timing and duration of individual treatments. New onset NOD mi ce received either ATG or rIgG as well as either Januvia or saline. All animals re ceived one insulin pellet at onset. Arrows indicate timing of a given treatment. Figure 5-2. Purification of Januvia does not ablate DPPIV activity as measured by luminescence. 70

PAGE 71

Figure 5-3. A one-time dose of Januvia prior to IPGTT in non-diabetic 12 week old NOD mice does not significantly modulate blood gl ucose levels versus control. n=4 Figure 5-4. Januvia-pretr eatment of non-diabetic 12 week old NOD mice mediates an increase in serum insulin levels during IPGTT. This increase was significant at 60 minutes (n=4, p=0.0015, unpaired t te st) versus control. 71

PAGE 72

A B C D Figure 5-5. Nonfasting blood gluc ose levels of NOD mice treated with control, ATG, Januvia, or ATG+JAN post-onset. Blood glucose levels in A) control-treated (n=8), B) ATGtreated (n=9), C) Januvia-tr eated (n=10), and D) ATG+JAN-treated (n=8) mice are shown following diabetes onset. One subcutan eous insulin pellet was administered at onset to control blood glucose for up to 3 weeks. 72

PAGE 73

Figure 5-6. Kaplan-Meier surv ival curves of new-onset NOD mice treated with ATG and/or Januvia. Januvia (n=10, p=0.0035, Log-Ra nk Test), ATG (n=9, p<0.0001, Log-Rank Test), and ATG+JAN (n=8, p<0.0001, LogRank Test) all exhibit significant improvement in survival versus control-tr eated (n=8) mice. Januvia was administered for up to 42 days. 73

PAGE 74

Figure 5-7. Januvia does not augment ATG-mediat ed glucose control in NOD mice at endpoint as measured by an IPGTT, but does increase serum insulin levels. A) Both ATG and ATG+JAN exhibit inferior glucose control vs. non-diabetic 12 week old NOD mice as measured by AUC (n=4, p<0.05, unpaired t test). B) ATG+JAN (n=3) trends toward greater serum insulin vs. ATG (n=4) monotherapy at 10 minutes (p=0.0666, unpaired t test). 74

PAGE 75

Figure 5-8. Januvia-treated diabe tic NOD mice exhibit increased serum sCD26 versus controltreated mice (n=6, p=0.0223, unpaired t test). 75

PAGE 76

A B C Figure 5-9. ATG and ATG+JAN-treated mice exhi bit altered regulatory T cell and CD4:CD8 profiles. A) Both ATG (n=8) and ATG+JAN (n=6) treatment increased the ratio of Treg to CD8 effector T cells (p=0.0149, p=0.0002 respectively, unpaired t test). B) ATG+JAN therapy yielded the greatest trended increase in the Treg to CD4 effector ratio (n=7, p=0.0653, unpaired t test). C) The CD4:CD8 ratio was significantly increased in ATG (n=8) and ATG+JAN-treat ed (n=6) mice versus control (p=0.0034, p<0.0001 respectively, u npaired t test). 76

PAGE 77

77 Figure 5-10. Januvia has no significant imp act upon the serum levels of SDF-1. n=6 Figure 5-11. Januvia has no significant impact upon the surface expression of CD26 on CD4+ and CD8+ T cells.

PAGE 78

CHAPTER 6 INDUCTION OF A TOLEROGENIC PHE NOTYPE BY G-CSF ADMINISTRATION Introduction We have previously demonstrated the abil ity of ATG to induce significant long-term euglycemia when administered to new-onset, di abetic NOD mice. We have also demonstrated the positive short-term impact of both alpha-1-antitrypsin and the DPPIV inhibitor Sitagliptin upon blood glucose values following diabetes onset. The ability of a second agent to significantly enhance the diabetes remission rate of ATG remain s elusive, however. With the assumption that the control of autoimmunity is the most critical factor to control, as demonstrated by the efficacy of ATG, we sought to expand upon this therapeutic pathway. In particular, in this study it was our intention to use a second immunomodulatory, non-depleting agent with a substantial clinical safety record. Granulocyte-colony stimulating factor (G-CSF) was initially developed as a means of repopulating neutrophils in cancer patients w ho had received immune ablative chemoand radiation therapies (114). To th is end, G-CSF has demonstrated a robust ability to repopulate neutrophils with a safe clinical record (115). Recent studies, how ever, have indicated the ability of G-CSF to induce an immunor egulatory shift from a TH1 to a TH2 cytokine phenotype (116). In addition, studies have shown GCSF-mediated induction of tolerogenic dendritic cells as well as the mobilization of regulatory T cells (117; 118). More pertinently to our study, G-CSF ha successfully prevented the onset of T1D in th e NOD mouse via the indu ction of tolerogenic dendritic and regulatory T cells (119; 120). s Given this information, G-CSF appeared to be suitable for the use of augmenting ATG therapy in new-onset diabetic NOD mice. In order to more accurately char acterize the effects of G-CSF during the course of therapy, we have split our trial into two f acets: reversal and pre78

PAGE 79

diabetic timecourse. The results of the reversal trial will not be presented in this chapter in lieu of presenting the pre-diabetic time course data. In this study, it was our in tention to analyze the change in leukocyte populations over time, the alteration in cyt okine secretion, and the histology of the pancreatic islets, both in terms of insulitis and insulin c ontent. Sections of spleen and pancreas were also saved in a c ohort prior to therapy as well as fr om mice in all treatment groups at the end point of the trial for RT-PCR analysis. G-CSF is available commercially with a daily (Neupogen) and weekly (Neulasta) formulations (121). Since previous st udies were performed using Neupogen, we chose it for our study. Based upon previously published literature and unpublished resu lts indicating the ability of G-CSF to augment the ability of ATG to indu ce long-term diabetes remission, we hypothesized that G-CSF would mask ATG-induced leukoc yte depletion through the mobilization of neutrophils and macrophages. We also expected to observe an induction of regulatory T cells and a reduction in TH1 cytokines that we anticip ated would translate into favorable improvements in both insulitis scoring and insulin co ntent from pancreatic histology. Materials and Methods This study was different in scope from previ ous studies in that it focused upon mechanism analysis rather than reversal of overt diabetes. In spite of this many of the protocols remained the same. The reagents used in this project were ATG and/or G-CSF. As such, the methods described here provide details on the admini stration of these compounds as well as new procedures such real time PCR and the use of a NIT-1 lysate for the purpose of in vitro splenocyte stimulation. Experimental Design In this study, 12 week old, non-diabetic fe male NOD mice were used for a mechanistic analysis of ATG and G-CSF comb ination therapy. The animals were placed into one of four 79

PAGE 80

treatment groups: control (rIgG + saline), ATG, G-CSF, or ATG combined with G-CSF. All therapies began at day 1. Rabb it immunoglobulin or ATG were admi nistered at day 0 and 3 at a dose of 500ug/dose. G-CSF was administered IP at 6ug/dose in 100uL total volume daily for up to eight weeks (barring failure, th e eventuality of which is desc ribed below). 100uL total volume of saline was administered daily IP as a control for G-CSF. Timed sacrifices were performed at 4 time points: week 0, 2, 4, and 8 post-initiation of therapy (Figure 6-1). A total of 5 animals per group were used for the 2, 4, and 8 week time poi nts, while 5 non-diabetic 12 week-old female NOD control mice were sacrificed as the single 0 week reference group. All animals were monitored for the potent ial onset of hyperglycemia once per week. Confirmation of hyperglycemia was considered as a blood glucose value greater than 240 mg/dL for two consecutive days. Upon conf irmation of hyperglycemia, the animal was prioritized into the next available sacrifice window. Otherwise, upon reaching the each predetermined endpoint, the designated animals were sacrificed and va rious analyses performe d upon selected tissues. A second phase, repeating the fi rst cohort, was also performe d replacing timed sacrifices with serial tail bleeds. This was performed in or der to acquire peripheral blood T cell analysis through flow cytometry. Preparation of G-CSF G-CSF is sold commercially as Neupogen, F ilgrastim (Amgen, Thousand Oaks, CA, USA) in 1mL vials of 300ug per vial. The G-CSF was d iluted in 4mL 5% dextro se solution providing a total volume of 5mL. Animals were dosed with 6ug/dose IP on a daily basis in 100uL total volume. Due to concerns over storage, samp les were prepared fresh on a daily basis. Quantification of Leukocyte Depletion To measure the leukocyte concentration, w hole blood was collected from each mouse at each timed sacrifice. Blood was collected into EDTA tubes to prevent clotting, followed by 80

PAGE 81

analysis using a Coulter Ac T Diff Hematology Analyzer (B eckman Coulter, Fullerton, CA, USA) which requires as little as 12uL of blood. Ve terinary Applications Software was used to analyze and output data for each sample. NIT-1 Lysate Preparation A confluent vented flask of cultured NIT-1 in sulinoma cells was obtained as a gift from the lab of Dr. Clayton Mathews. The cells were lysed using M-Per Mammalian Protein Extraction Buffer (Pierce Biotechnology, Rockford, IL, USA). Cells were incubated with 1mL buffer for 5 minutes while rocking. The lysate was collected and protease inhibito r cocktail was added (Roche, Basel, Switzerland). The protein content was then measured by Bradford protein assay. The lysate was then stored in al iquots at -20 degrees until use. Analysis of T cells, DCs, Macrophages, and Neutrophils Via Flow Cytometry Splenocytes taken at each time point were processed as previously described. Flow cytometry was performed in order to quantif y the populations of T cells, dendritic cells, macrophages, and neutrophils at each time point. All antibodies were purchased from eBioscience with the single exception of CD4-PerCP which was purchased from BD Biosciences. T cells were stained for CD8-FITC (clone 53-6.7), CD4-PerCP (clone RM4-5), Foxp3-PE (clone FJK-16a), and CD25-APC (clone PC61). De ndritic cells were stained for CD11c-FITC (clone N418), CD86-PE (clone GL1), and MHC class II (clone M5 /114.15.2). Macrophages were stained with CD11b-FITC (clone M1/70) CD14-PE (Sa2-8), and CD16/CD32-APC (clone 93). Neutrophils were stained w ith F4/80-PE (clone BM8) and Gr-1-APC (clone RB6-8C5). All were added at a concentration of 1ug per 1x106 cells per tube. Respective to the above, the isotype controls us ed for T cells were Rat IgG2a-FITC (clone eBR2a), Rat IgG2a-PE (clone eBR2a), Rat IgG2ak-PerCP, and Rat IgG2b-APC (clone 81

PAGE 82

eB149/10H5). For dendritic cells, Armenian Hamster IgG-FITC (clone eBio299Arm), Rat IgG2a-PE (clone eBR2a), and Rat IgG2bAPC (clone eB149/10H5) were used. For macrophages, rat IgG2b-FITC (clone eB149/10H 5), Rat IgG2a-PE (clone eBR2a), and Rat IgG2a-APC (clone eBR2a) were used. For neut rophils, Rat IgG2a-PE (c lone eBR2a) and Rat IgG2b-APC (clone eB 149/10H5) were used. Immunoglobulin Isotyping Immunoglobulin isotyping was performed on sera obtained at each sacrifice time point using a Mouse Immunoglobulin Isotyping kit (Mil lipore, Billerica, MA USA) in order to measure IgA, IgM, IgG1, IgG2a, IgG2b, and IgG3. Mouse isotyping serum diluent and mouse immunoglobulin isotyping standard were ordered separately (Millipore). Anti-G-CSF Antibody Measurement Sera was collected from mice at 0, 2, 4, and 8 week time points in the prediabetic study. To determine whether the immunoglob ulin increases seen were G-CSF-specific, Nunc-Immuno 96well plates were coated with 50uL/well of 2ug/mL GCSF (Amgen, Thousand Oaks, CA, USA) overnight at 4o. Plates were blocked for 2 hours w ith 300uL/well 5%BSA/PBS and washed 5x with PBS/Tween. Sera were diluted 1:10000 and was incubated for 2 hours on a plate shaker. The plates were washed 5x as before, and were then coated with either 50uL/well 1:2500 Rat anti-mouse IgM-HRP or 1:5000 Donkey anti-mouse IgG-HRP (Southern Biotech, Birmingham, AL, USA) for 1 hour on a plate shaker. The plates were once again washed 5x followed by addition of 50uL TMB protected from light. After 5 minutes, the reaction was stopped with 50uL stop solution (Mercodia, Uppsala, Sweden) and the plate read at 450nm wavelength on a Spectramax Plate Reader (Molecular Devices). 82

PAGE 83

Histology The beta cell area was calculated using Me taMorph software (Molecu lar Devices) analysis with insulin stained with fast red (FR) on pa ncreatic sections. The insulin positive area was divided by the total acin ar area to yield a final percentage Insulitis scoring was performed on hematoxylin and eosin stained pancreatic sections as described in previously. Real Time PCR Real time PCR was performed on pancreatic lymph nodes and spleen collected from 12 week control mice as well as all mice at the 8 week endpoint. Tissues were frozen in RNAlater (Ambion, Inc., Austin, TX, USA) at -20 degrees overnight, then placed at -80 degrees for longterm storage until use. mRNA was extracted from the tissues using RNAqueous kits (Ambion). cDNA was produced from the mRNA using SuperScr ipt III Reverse Transc riptase (Invitrogen). cDNA samples were analyzed with a 384-pane l Mouse Immunology 384 StellArray qPCR array (Bar Harbor Biotechnolo gy, Bar Harbor, ME, USA). Results Depletion/Repopulatio n of Leukocytes We have shown previously that ATG induces a short-term decrea se in the number of leukocytes within the first mont h of treatment post-onset. We pr edicted that the addition of GCSF, due to its ability to mobili ze neutrophils, would ablate this depletion and possibly lead to a significant increase in the number of leukocytes measured during th e course of therapy. Using a Coulter Hematology Analyzer, we measured pe ripheral blood at 0, 2, and 4 weeks during the course of therapy (Figure 6-2). We encountered difficulties due to clotting, most likely due to the small volumes of blood involved during collection. As such, more vari ation than anticipated was observed in the measurements. What is clear, however, is the presence of a near-significant trend toward an 83

PAGE 84

increase in leukocyte concentrations in both the G-CSF and ATG+G-CSF treatment groups when compared with control and ATG-treated mice at the two week time point. A decrease in counts (p=NS), as anticipated, was observed in the ATG-tr eated group versus contro l at this same time point. Surprisingly, all groups appeared to return to baseline levels by the four week time point with no significant differences between treatments Measurements (not shown) were also taken from the final eight week time point that recapitula ted the data observed at four weeks, such that no groups were significant different from one another. Flow Cytometry Analysis of Macrophages and Neutrophils In order to more accurately characterize the change in leukocyte counts over time, we performed flow cytometric analysis of splenocyt es at zero, two, four, and eight weeks following onset of therapy. Both the pe rcentage of splenic neutroph ils (Figure 6-3a) and splenic macrophages (Figure 6-3b) were plotted. Clearly, the spike in leukocyt e counts (Figure 6-2) correla tes with a robust, significant increase in splenic neutrophils and macrophage s at the two week time point in both G-CSF (p<0.05) and ATG+G-CSF-treated (p<0.05) mice. As with the leukocyte counts, however, this increase in neutrophils and macrophages was shor t-lived. The four and ei ght week time points revealed no differences in these cell popul ations despite ongoi ng G-CSF therapy. Alterations in B Lymphocyte Profile The short-term effect of G-CSF upon neutrophils was surprising considering administration continued on a daily basis for the entire 8 week experimental period. RT-PCR analysis of pancreatic lymph nodes and sections of spleen from mice sacrificed at 8 weeks revealed significant fold changes in mRNA expression for multiple B lymphocyte markers. GCSF, both alone and in combination with ATG, i nduced significant increases in IgM, IgG1, and the B cell maturation factor btk. 84

PAGE 85

Immunoglobulin isotyping was performed on se ra obtained throughout the 8 week time course. Notable alterations in both IgG1 (Figur e 6-4a) and IgM (Figure 6-4b) were observed throughout the time course. In particular, combination therapy yielded the greatest concentrations of both IgG1 and IgM, with bot h significantly raised compared with control therapy at 2, 4, and 8 week time points. Given this dramatic increase in total seru m immunoglobulin and the lack of long-term mobilization of neutrophils despite prolonged G-CSF therapy, we believed it possible that antibodies were being produced against G-CSF, thus neutralizing its effect beyond 2 weeks. Anti-G-CSF IgM (Figure 6-5a) a nd anti-G-CSF IgG (Figure 6-6b) ELISAs were performed using sera from each time point. In all mice treated with G-CSF, alone or in combination with ATG, anti-G-CSF antibodies were observed as early as 2 weeks in the IgM assay and as early as 4 weeks in the IgG assay. These anti-G-CSF an tibodies remained evident throughout the study endpoint at 8 weeks. Flow Cytometry Analysis of T Lymphocytes Apart from neutrophils and macrophages, the sp lenic T cell populations were also analyzed through flow cytometry. The splenic ratio of CD4:CD8 (Figure 6-6a) and the percentage of regulatory T cells (Figure 6-6b) was measured. As with our previous studies, ATG therapy i nduced a signifi cant increase (p<0.05) in the CD4:CD8 ratios of splenocytes. While G-CSF did not augment this effect, it did not ablate it either. Analysis of the regulatory T cell population revealed a decrease in the percentage of Treg in all groups versus control at 2 weeks. This co rrelated with an ATG-mediated, short-term T cell ablation or a relative increase in the percentage of macropha ges and neutrophils by G-CSF, effectively lowering the percentage of Treg in these groups. By the final 8 week time point, at which point the leukocyte decreases and incr eases of ATG and G-CSF, respectively, had 85

PAGE 86

diminished, G-CSF treatment trended toward a higher percentage of Treg while ATG+G-CSF combination therapy yielded a significant incr ease (p<0.05) when compared with control therapy. In order to explain the altera tion in the CD4:CD8 ratio, we examined the percentage of CD4+ and CD8+ splenic T cells at each time point in mice treated with control (Figure 6-7a), GCSF (Figure 6-7b), ATG (Figure 6-7c), or ATG+G-CSF (Figure 6-7d) therapies. Treatment with control led to small, non-signi ficant fluctuations in the percentages of CD4+ and CD8+ T lymphocytes. ATG therapy led to a reduction of ~50% CD4+ T cells while reducing CD8+ T cells by almost 80%. While the decrease in CD4+ cells had returned to baseline levels at 8 weeks, the percentage of CD8+ T cells remained significantly (p<0.05) reduced. Corresponding with an increase of splenic macrophages and neutrophils (Figure 6-3), the percentage of CD4+ and CD8+ T cells was reduced at 2 weeks in G-CSF treated mice. By 8 weeks, both populations had retu rned to, or increased beyond, starting levels. Combination therapy also yielded a decrease in both CD4+ and CD8+ T cells beyond that which seen in either monotherapy indicating an addi tive effect. By 8 weeks, while CD8+ T cells remained significantly reduced versus the st arting time point, the percentage of CD4+ T cells had actually increased (p=0.043). To obviate the consequence of the alteration of different T cell subpopulations the ratios of regulatory T cells and effect or CD8+ (Figure 6-8a) or CD8+ (Figure 6-8b) T cells were compared. Corresponding with the ATG-induced increa se in CD4:CD8, both ATG and ATG+G-CSF therapies resulted in a significant increase in the Treg:CD8 ratio. Only combination therapy, 86

PAGE 87

however, resulted in a signi ficant increase in the Treg:CD4 effector ratio at 8 weeks when compared with control therapy. Splenocyte Proliferation to NIT-1 Lysate In Vitro The splenocyte prolifera tive response to NIT-1 lysate was measured from cells obtained at the 0 and 8 week sacrifice time points. No significant change in prolif eration was observed in any treatment group to this lysate (data not shown). Splenocyte Cytokine Release in Response to Activation In Vitro Based upon previous reports indicating that G-CSF therapy induces a shift from a TH1 to a TH2 cytokine profile, we cultured splenocytes fr om mice sacrificed at the 0 and 8 week time points and stimulated with anti-CD3 and anti -CD28 monoclonal antibodies. Supernatants were collected at 24 and 72 hours post-stimulation an d measured for cytokines (Figure 6-9). As was expected based on previously reporte d results, G-CSF therapy in combination with ATG significantly reduced the levels of TH1 cytokines TNF, IFN, and IL-2 at 72 hours following stimulation in vitro Levels of the TH2 cytokines IL-4 and IL-5 both trended higher in combination therapy-treated mice versus control mice. Insulitis Scoring Since one of the primary goals of any diabetes intervention therapy is the protection of beta cells from ongoing autoimmune attack, the degree of lymphocytic infiltration in the islets, as measured through insulitis scoring, is critical to any assessment of therapeutic efficacy. Mice from each therapy were sacrificed at 0, 2, 4, and 8 weeks. Using H&E stained pancreatic sections, insulitis scoring was performed (Figure 6-10). Since the goal of this experiment was to sust ain the health of the islets for as long as possible following treatment, only the 8 week ti me point data are presented in Figure 6-10. Combination therapy yielded a trend towards a greater average number of healthy islets. The 87

PAGE 88

number of near-healthy islets (with only peri-insulitis) was signif icantly increased versus control only in the combination-treated mice (p=0.005). In fact, the number of healthy or near-healthy islets trended higher in the combination-treate d mice even when compared with non-diabetic 12 week old NOD mice. Insulin Area Apart from measuring the degree of lymphocytic in filtration in th e islets, we also sought to quantify the percentage of beta cel l area relative to the acinar area at several time points in order to elucidate the extent of in sulin content in the pancreas of mice from each treatment group (Figure 6-11). Staining for insulin with fast red allowed fo r clear differentiation be tween beta cells and acinar tissue. The ratio of the beta cell area to acinar area was compared to yield the results observed (Figure 6-11b). While a serial measurem ent was obviously not possible as sacrifice was required for each time point, the analysis rev ealed the ability of AT G and G-CSF combination therapy to provide long-term prot ection to the beta cell mass with significantly greater beta cell area compared with control therapy (p<0.05). Conclusions While the focus of this dissertation has, until this chapter, focuse d upon the reversal of overt T1D in the NOD mouse, this study was perfor med with the intention of characterizing the effects of ATG and G-CSF combination therapy over the course of an 8 week pre-diabetic treatment window, the results of which will be combined with an adjunct reversal study also carried out by our group. It should be noted that since the data are obtaine d from timed sacrifice points, the collecti on is not serial. Perhaps the greatest surprise resulted from the leukocyte measurements taken at 0, 2, 4, and 8 weeks. It was anticipate d that G-CSF would induce the pr oliferation and mobilization of 88

PAGE 89

neutrophils, as well as possibly reverse the depleting effect seen with ATG monotherapy. What we did not expect was that this leukocytic increase mediated by G-CSF would only be evident at the two week time point. Counts from the 4 and 8 week time points revealed no differences between G-CSF-treated and cont rol-treated mice. The depletion induced by ATG was, as we have shown previously (49), limite d to approximately 4 weeks before levels returned to values similar to baseline. In order to examine which leukocytes were al tered at these time points, flow cytometry was performed using stains for both neutrophils and macrophages. The levels of both of these populations correlated closely with the aforementioned leukocyte c ounts. One possibility for this short-term increase is that th e relatively high dose of G-CSF administered modulated the GCSFR (CD114) on the surface of neutrophils. Previ ous reports have indicated the ability of GCSF to downregulate its own recepto r (122). In fact, several studies have also noted the critical nature of CD114 to mobilize ne utrophils into peripheral blo od (123; 124). RT-PCR analysis revealed a downregulation of CD114 mRNA in G-CSF treated mice at 8 weeks. Overwhelming evidence, however, points to the generation of neut ralizing antibodies are the cause of the lack of long-term mobilization. Future efforts may be best spent using pegylated G-CSF, Neulasta, in an attempt to minimize the anti-G-C SF antibody response. The silver lin ing of this result is that short-term therapy was effective. The therapeutic impact upon T lymphocytes wa s also examined through flow cytometry. The levels of CD4+ and CD8+ T cells remained stable in the control-tr eated mice, whereas ATG induced a significant drop in both subpopulations. Th is depletion preferentia lly targeted CD8+ T cells, reducing the percentage in the spleen by roughly 80%, whereas CD4+ T cells were reduced by approximately 50%. G-CSF induced a relative reduction in both subpopulations at 2 weeks 89

PAGE 90

due to the relative increase in neutrophils and macrophages. This effect was not seen at 4 or 8 weeks. Combination therapy appeared to intens ity the initial reduction in both CD4+ and CD8+ T cells. While the percentage of CD8+ T cells had only recovered to r oughly 50% of baseline, the percentage of CD4+ T cells had actually incr eased versus baseline. Th e consequence of these changes was a dramatic increase in the CD4:CD 8 ratios in mice receiv ing ATG monotherapy or combination therapy, which is actually the opposite of what has been reported with anti-CD3 mAb therapy. An analysis of the regulatory T cell population revealed that while neither ATG nor G-CSF monotherapies yielded significantly higher percentages of splenic Treg, combination therapy did induce a significant increase. A comparison of the Treg:Teff ratios revealed that combination therapy yielded the greatest increases in the Treg:CD4eff ratio as well as the greatest increase in the Treg:CD8 ratio. Given that ATG's putative mech anism was reported as the induction of regulatory T cells, this apparent benefit provided by augmentation with G-CSF is of therapeutic significance. Given the improvement in the immunoregulatory profile in combination therapy-treated mice, the question remained whether this improve ment would translate into beta cell protection. Analysis of both insulitis scoring and insulin st aining from pancreatic se ctions revealed that combination therapy had a significantly great er average number of non-infiltrated islets compared with all other treatment groups. At the final 8 week time point, only combination therapy appeared to have preser ved insulin content within the is lets relative to control therapy. This is encouraging should it translate in the setting of T1D reversal given previous studies correlating the maintenance of beta cell ma ss with protection from overt diabetes. 90

PAGE 91

Given these results, apart from testing this therapy in the setting of diabetes reversal, the use of Neulasta should be considered in a comparison to Neupogen. Given the short-term mobilization of macrophages and neut rophils, it is possible that two or even one dose of Neulasta may provide a similar effect. Give n that the ultimate goal of these therapies is translation into clinical use, this may provide a practical alteration in the treatment protocol. 91

PAGE 92

Figure 6-1. ATG+G-CSF mechan istic study experimental desi gn describing the timing and duration of individual treatments. 12-week-o ld pre-diabetic NOD mice received either ATG or rIgG as well as either G-CSF or saline. Arrows indicate timing of a given treatment and timed sacrifices. Figure 6-2. ATG+G-CSF mediat es a significant increase (n= 5, p=0.0129, unpaired t test) in peripheral blood leukocytes at 2 weeks vs ATG monotherapy (n=4). G-CSF (n=5) induces a near significant increase at 2 w eeks, while ATG mediates a near significant decrease at 2 weeks. 92

PAGE 93

Figure 6-3. G-CSF therapy, alone or in combination with ATG, i nduces a short-term increase in the percentage of both splenic neutroph ils and macrophages. This increase was significant (p<0.05, unpaired t test) for both A) neutrophils and B) macrophages at the 2 week timepoint (n=5). This effect is not evident at later time points. 93

PAGE 94

Figure 6-4. Total serum immunogl obulin is altered by treatment with G-CSF. (a) Total serum IgG1 (n=5) was significantly increased vs. control at 2 weeks by combination therapy (p=0.0024, unpaired t test). At 4 weeks, AT G, G-CSF, and combination exhibited significantly greater levels vs. c ontrol (p=0.028, p=0.0154, p=0.008, respectively, unpaired t test). At 8 weeks, both G-C SF (p=0.0016, unpaired t test) and combination (p=0.0007, unpaired t test) were greater than control. (b) Total serum IgM (n=5) was significantly increased vs. control at 2 weeks by combination (p=0.0022, unpaired t test). At 4 weeks, both G-CSF (p=0.0189, unpaired t test) and combination (p=0.0111, unpaired t test) were increased vs. control. At 8 weeks, both G-CSF (p=0.0013, unpaired t test) and combination (p=0.0014, unpaired t te st) increased IgM vs. controls. 94

PAGE 95

Figure 6-5. Anti-G-CSF antibodies are produced in treated NOD mice. (a) Serum levels of antiG-CSF IgM (n=5) were significantly incr eased in both G-CSF (4 weeks: p=0.0141, 8 weeks: p=0.001, unpaired t test) and mATG + G-CSF (4 weeks: p<0.0001, 8 weeks: p<0.0001, unpaired t test) groups versus contro l, but were not raised significantly with mATG therapy. (b) Serum levels of anti-G-CSF IgG (n=5) were also significantly raised with both G-C SF (4 weeks: p=0.0039, 8 weeks: p=0.001, unpaired t test) and mATG + G-CSF (4 weeks: p=0.0016, 8 weeks: p<0.0001, unpaired t test) groups versus control, while mATG did not induce a significant increase. 95

PAGE 96

Figure 6-6. Both ATG and G-CSF altered the CD4: CD8 ratio and/or the pe rcentage of splenic regulatory T cells. A) Both ATG and ATG+ G-CSF (n=5) led to significant increases (p<0.05, unpaired t test) in th e CD4:CD8 ratio at 2, 4, and 8 week time points. B) By the final time point, combination therapy yielded a significant increase in the percentage of splenic Treg (p<0.05, unpaired t test). 96

PAGE 97

A B C D Figure 6-7. The percentages of CD4+ and CD8+ T cells are significan tly reduced (n=5, p<0.05, unpaired t test) at 2 weeks with A) ATG and/ or B) G-CSF therapies. C) and D) By the 8 week time point, the CD4+ population ha s recovered while the CD8+ population remains significantly reduced (n=5, p<0.05, unpa ired t test) compared with the initial, 0 week time point. 97

PAGE 98

Figure 6-8. The ratio of Treg to effector T cells (both CD4+ and CD8+) is increased with ATG and G-CSF therapies. A) Both ATG and ATG+G-CSF induced a significant increase (n=5, p<0.05, unpaired t test) in Treg:CD8 at 2, 4, and 8 week time points versus control. B) ATG+G-CSG combination thera py yielded a significant increase (n=5, p<0.05, unpaired t test) in Treg:CD4eff versus both control and ATG at the final 8 week time point. 98

PAGE 99

Figure 6-9. G-CSF alters the cytokine release of stimulated splenocytes in vitro and appears to skew from a TH1 to a TH2 phenotype. Combination therapy trended toward increases in A) IL-4 and B) IL-5 versus control wh ile significantly decreasi ng the levels of C) TNF, D) IFN, and E) IL-2 (n=5, p<0.05, unpaired t test) versus control at 72 hours. 99

PAGE 100

100 Figure 6-10. Insulitis scoring of pancreatic islets at the 8 week time point was improved in mice treated with a combination of ATG and G-C SF. The average number of healthy islets (score=0) from combination-treated mice trended higher than control while the number of islets with peri-insulitis (score=1) was significantly higher than any other treatment (n=5, p=0.005, unpaired t test). Figure 6-11. Combination therapy of ATG and G-CSF leads to a r obust, long-lasting increase in the percentage of beta cell area. A) Repres entative islet stained for insulin with FR. B) ATG+G-CSF significantly in creased beta cell area versus control at 8 weeks (n=5, p<0.05, unpaired t test).

PAGE 101

CHAPTER 7 DISCUSSION AND CONCLUSIONS Discussion In spite of the enormous success of insulin therapy for most th an 85 years, the presence of severe long-term complications associated w ith T1D, such as retinopathy, nephropathy, and cardiovascular disease, fuel the need for more effec tive treatments in order to restore endogenous insulin production (22; 23; 42; 45). Apart from me dical complications, there is also a substantial financial burden associated with T1D. The cost of insulin therapy over the average lifetime of a type 1 diabetic patient is approximately $70,000 as of 2007 (125). Clearly, the development of a long-lasting therapy for T1D woul d curtail not only medical but also financial complications. In order to develop therapies to address these concerns, we have tu rned to the use of a mouse model of disease. We have used the NOD mouse as our model of T1D given its incidence rate of up to 80% in 30 week old females as well as its close resemblance to human disease as it is polygenetic and presents immunoregulatory def ects (21; 126). In this dissertation, several studies have been presented in which the reversal of overt T1D was achieved via combination therapies. These were all performed using compounds that are currently used clinically and offer rapid translation into clinical use for T1D. Sp ecifically each was based upon a core therapy of ATG, which serves to target the ongoing autoimmune attack at time of onset. As we have previously reported, ATG induces a regulatory T cell population, protects the remaining beta cell mass, and is capable of both prevention a nd reversal of T1D in the NOD mouse (49). In our first attempt to augment the efficacy of ATG, we utilized a second agent to target the pathway of autoimmune at tack through the use of rapamy cin. Unfortunately, this study yielded mixed results. While rapamycin mono therapy or combination therapy induced an increase in the percentage of sp lenic regulatory T cells as well as an increase in serum c-peptide 101

PAGE 102

levels, it did not reduce the degree of insulitis pr esent in the pancreas, nor did it induce long-term euglycemia in new-onset mice. Therapeutic efficacy, as measured by blood glucose, revealed rapamycin to be not significantly different from control therapy. While th e ability of rapamycin to prevent T1D in the NOD mouse through the induction of regulator y T cells has been documented (74; 75), reports of an unanticipated increase in peripheral insulin resistance and alteration of IRS signaling has also been report ed (76; 77). It is lik ely that while immune regulation may have taken place, with the success of reversal therapy eff ectively on a knife edge, these deleterious complications may have prevente d any therapeutic success by rapamycin in this setting. After the rapamycin trial, we sought to targ et a different pathway. Specifically, we aimed to enhance the survival of beta cells by reduci ng apoptosis through the use of alpha-1-antitrypsin therapy. AAT has a long safety record in clinic al use for emphysema and via gene therapy has yielded successful diabetes prevention in the NOD mouse (84; 88). Unfortunately, the use of recombinant AAT in the NOD mouse triggered an unexpected anaphylactic event. While prevention of anaphylaxis could be achieved for up to 5 doses using triprolidine and a PAF antagonist, repeating this prevention for a longer duration revealed imperfect protection. Consequently, our dosing of AAT was suboptimal, li mited to just 4 injections. As a result, we observed only a partial impact upon the remission from hyperglycemia, with one AAT-treated mouse lasting 70 days before hyperglycemia return ed. Future trials in the NOD must consider this anaphylactic predisposition, potentially av oiding this response by slowing the release of AAT via subcutaneous injecti ons or through osmotic pumps. Our third reversal trial in the NOD mouse targeted the path way of beta cell regeneration. While the GLP-1 mediated increases in insulin se cretion have been well characterized during the 102

PAGE 103

clinical use of Januvia, there were also reports indicating the potential regeneration in mice induced via DPPIV inhibition. The presence of DPPIV on the surface of T cells (CD26) acting as a costimulatory marker offered yet another face t for investigation whil e treating NOD mice with Januvia. Implementation of this therapy yielded only partial benefit, however. While Januvia monotherapy provided significant improvement in remission from hyperglycemia versus control, no Januvia-treated mice were afforded long-term remission. Combination therapy hinted at a similar improvement, but due to the 6 week treatmen t window, this effect appeared to be ablated with the withdrawal of Januvia. Immunological analyses revealed few differences following Januvia administration with the ex ception of sCD26, which is likely simply a marker of disease duration. The limited viability of both AAT and Januvia prompted a return to arguably the most critical pathway: the targeting of autoimmun ity. To this end, we chose the immunomodulatory, non-depleting G-CSF. Specifica lly, we performed a mechanistic time course in which combination therapy was administered to pre-di abetic 12 week old NOD mice for up to 8 weeks. Timed sacrifices at 0, 2, 4, and 8 weeks provided a timeline that al lowed for a description of the metabolic and immunological changes associated with these therapies. Analysis revealed a robust mobilization induced by G-CSF upon bot h macrophages and neutrophils that was ephemeral, lasting only at the 2 week time point. In spite of this short-term effect, the impact upon regulatory T cells appeared to increase w ith each subsequent measurement such that combination therapy yielded the greatest percenta ge of splenic regulatory T lymphocytes at the final 8 week time point. This increase was asso ciated with favorable insulitis scoring and significantly greater insulin area in combina tion-treated mice versus controls. Given the augmentation of ATG's immunological and metabolic benefits, it is likely th at this therapy would 103

PAGE 104

prove efficacious if administ ered to new-onset diabetic NOD mice. The success of ATG monotherapy has demonstrated an upper limit for re versal that is independ ent of age at onset, but rather dependent upon blood glucose levels at ons et. The efficacy of ATG appears to all but disappear with starting blood glucose values of 400mg/dL or great er (Figure 7-1). One crucial analysis of any ATG+G-CSF comb ination trial would be to examine whether this threshold is increased beyond 400mg/dL. In spite of several published reports of successful T1D reversal therapies in the NOD mouse, the translation into clinic al use is not clear. As has al ready been discussed, hundreds of therapies are capable of preventing T1D in the NOD mouse, with several capable of diabetes remission, yet translation into clinical use has be en limited (23; 41). Conc erns remain regarding differences between mouse and man in regards to the ability of the beta cell mass to regenerate or replenish via hematopoietic stem cells (127). Previous work in mouse models indicates that the timing, dosage, and protocols for a given therapy can have a dramatic effect upon the outcome (128). Several previous reports have supported this notion, arguing that previous clinical failures may be predominantly due to in sufficient dosing, a problem they argue could be solved through greater optimization in the NOD mouse model (126; 129). Other considerations regarding the utility of the NOD mouse pertain to other abnormalities of this strain. Apart from islet autoreactiv ity, the NOD mouse e xhibits no known C5 complement-pathway, has defective cytokine ex pression from macrophages, and has aberrant MAdCAM-1 and VCAM-1 expression, which may result in abnormal NK cell function and/or homing (130-132). In addition, the NOD mouse model, apart from spontaneous diabetes, also experiences a Sjgren's-like syndrome (133). Thus, considering the alterati on of several immune 104

PAGE 105

phenotypes seen in this syngeneic model of dise ase, the NOD mouse is unlikely to be a perfect representation of c linical disease. Diabetes prevention and reversal trials have already or are in the process of taking place worldwide. Two notable diabetes prevention tria ls are the Diabetes Prevention Trial Type 1 (DPT-1) and the European Nicotinamide Diabetes Intervention Trial (ENDIT). DPT-1 initially appeared to show no indication of successful prevention of T1D using oral or subcutaneous insulin (24; 25), but a breakdown of the patients revealed a possibl y protective effect in high-risk individuals treated with oral in sulin. The ENDIT study thus far has failed to provide evidence for successful prevention of T1D (134). More recently, more aggressive therapies have been used for the purpose of T1D intervention following disease onset. Anti-CD3 therapy has yielded positive results, with just six doses yielding enhanced beta cell function for 18 months following onset (135). In a separate tria l in Brazil, combination therapy including the use of ATG, G-CSF, hematopoietic stem cells, and cyclophosphami de yielded improved c-peptide levels and diminished insulin requirements, with several becoming insulin-free, up to 24 months following disease onset (41). Given the relatively low number of successful diabetes revers al therapies compared with the number of successful preven tion therapies in the NOD mouse, in addition to the starting blood glucose limits seen at onset in several reve rsal trials in the NOD, the success of therapy appears likely to hinge upon the beta cell mass re maining at time of onset. As such, therapies aimed at immunoregulatory modulation prior or immediately following onset are most likely to experience success. While there are reports of be ta cells present in individuals decades after diabetes onset (136; 137), the beta cell mass is unlikely to su pport glucose control even if autoimmunity is controlled. 105

PAGE 106

Conclusions The diabetes intervention therapies in the NOD mouse presented in this dissertation focused upon the projected benefit of combination therapy versus conventional monotherapy. Specifically, it was hypothesized that by targetin g multiple pathways of autoimmune beta cell destruction, combination therapy would be more effective than monotherapy. As the limited number of published successful T1D reversal therapies implies (27; 32; 48; 50-54), successful reversal of diabetes in the NOD mouse is challe nging given the aggressive nature of beta cell destruction in this model. The studies in this dissertation demonstrate bot h the difficulty and the benefit in this pursuit, with al pha-1-antitrypsin and Januvia providing a partial, yet insufficient, benefit for the reversal of overt disease. Our self-imposed requirement of "off-the-shelf" FDAapproved drugs most likely made this effort even more challenging as we could not tailor therapies, such as the use of ex vivo generated Treg, to the model of disease. The combination of ATG and G-CSF, however, provide s compelling evidence for the ability of two therapies to complement one another through the enhancemen t of the regulatory T cell phenotype and the preservation of beta cells in the NOD mouse. This, in turn, may also promote the focus of reversal therapies upon the regula tion of autoimmunity, rather th an beta cell regeneration, given the success of ATG and G-CSF and the minimal impact of Januvia. Further studies will undoubtedly require care ful modulation of both the timing and dose of each of these therapies, w ith careful consideration given to the eventual translation into clinical use. Optimization of a given combination therap y, such as ATG and G-CSF, will necessitate a detailed analysis of multiple immune cell types as well as a long-term study to ensure prolonged protection from autoimmunity. 106

PAGE 107

107 Figure 7-1. Efficacy of ATG is i ndependent of age at onset and dependent upon blood glucose at therapeutic onset.

PAGE 108

LIST OF REFERENCES 1. Apelqvist J, Bakker K, van Houtum WH, Schaper NC: The development of global consensus guidelines on the management of the diabetic foot. Diabetes Metab Res Rev 24 Suppl 1:S116-118, 2008 2. Ehehalt S, Popovic P, Muntoni S, Muntoni S, Willasch A, Hub R, Ranke MB, Neu A: Incidence of diabetes mellitus among children of Italian migrants substantiates the role of genetic factors in the pathoge nesis of type 1 diabetes. Eur J Pediatr 2008 3. Karvonen M, Viik-Kajander M, Moltchanova E, Libman I, LaPorte R, Tuomilehto J: Incidence of childhood type 1 diabetes worldw ide. Diabetes Mondiale (DiaMond) Project Group. Diabetes Care 23:1516-1526, 2000 4. Zhang H, Xia W, Yu Q, Wang B, Chen S, Wa ng Z, Love EJ: Increasi ng incidence of type 1 diabetes in children aged 0-14 years in Harbin, China (1990-2000). Prim Care Diabetes 2:121-126, 2008 5. Petrovsky N, Kyvik KO, Bonnevie-Nielsen V, Beck-Nielsen H, Green A, Harrison LC: Evidence from twins for acquired cellular im mune hyperactivity in type 1 diabetes. Immunology 106:584-589, 2002 6. Cardwell CR, Carson DJ, Patterson CC: No a ssociation between routinely recorded infections in early lif e and subsequent risk of childhood-onset Type 1 diabetes: a matched case-control study using the UK Gene ral Practice Research Database. Diabet Med 25:261-267, 2008 7. Cardwell CR, Carson DJ, Yarnell J, Shields MD, Patterson CC: Atopy, home environment and the risk of childhood-onset type 1 diabetes: a population-based casecontrol study. Pediatr Diabetes 9:191-196, 2008 8. Friedrich N, Kocher T, Wallaschofski H, Schw ahn C, Ludemann J, Kerner W, Volzke H: Inverse association between peri odontitis and respiratory allerg ies in patients with type 1 diabetes mellitus. J Clin Periodontol 35:305-310, 2008 9. Kukreja A, Maclaren NK: NKT cells and type -1 diabetes and the "hygiene hypothesis" to explain the rising incidence rates. Diabetes Technol Ther 4:323-333, 2002 10. Rafuse J: Seventy-five years later, insu lin remains Canada's major medical-research coup. Cmaj 155:1306-1308, 1996 11. Gale EA: The discovery of type 1 diabetes. Diabetes 50:217-226, 2001 12. Gepts W: Islet changes suggesting a possi ble immune aetiology of human diabetes mellitus. Acta Endocrinol Suppl (Copenh) 205:95-106, 1976 13. Kim MS, Polychronakos C: Immunogenetics of type 1 diabetes. Horm Res 64:180-188, 2005 108

PAGE 109

14. Alizadeh BZ, Koeleman BP: Genetic polym orphisms in susceptibility to Type 1 Diabetes. Clin Chim Acta 387:9-17, 2008 15. Barker JM, Triolo TM, Aly TA, Baschal EE, Babu SR, Kretowski A, Rewers MJ, Eisenbarth GS: Two Single Nucleotide Polymo rphisms Identify Highest-Risk Diabetes Human Leukocyte Antigen Genotype: Potential for Rapid Screening. Diabetes 2008 16. Morran MP, Omenn GS, Pietropaolo M: Imm unology and genetics of type 1 diabetes. Mt Sinai J Med 75:314-327, 2008 17. Wallden J, Ilonen J, Roivainen M, Ludvigsson J, Vaarala O: Effect of HLA Genotype or CTLA-4 Polymorphism on Cytokine Response in Healthy Children. Scand J Immunol 68:345-350, 2008 18. Salsali A, Nathan M: A review of types 1 a nd 2 diabetes mellitus and their treatment with insulin. Am J Ther 13:349-361, 2006 19. Scheidegger U, Allemann S, Scheidegger K, Diem P: Continuous subcutaneous insulin infusion therapy: effects on quality of life. Swiss Med Wkly 137:476-482, 2007 20. Urbanski P, Wolf A, Herman WH: Cost -effectiveness of diabetes education. J Am Diet Assoc 108:S6-11, 2008 21. Makino S, Kunimoto K, Muraoka Y, Mizushim a Y, Katagiri K, Tochino Y: Breeding of a non-obese, diabetic strain of mice. Jikken Dobutsu 29:1-13, 1980 22. Roep BO, Atkinson M, von Herrath M: Satisf action (not) guaranteed : re-evaluating the use of animal models of type 1 diabetes. Nat Rev Immunol 4:989-997, 2004 23. Roep BO: Are insights gained from NOD mice su fficient to guide clinical translation? Another inconvenient truth. Ann N Y Acad Sci 1103:1-10, 2007 24. Singh N, Palmer JP: Therapeutic targets for the prevention of type 1 diabetes mellitus. Curr Drug Targets Immune Endocr Metabol Disord 5:227-236, 2005 25. Vialettes B, Valero R, Mattei C, Nervi S: [P revention of type 1 diabetes: what have we achieved? What should we say and propose next to families?]. Diabetes Metab 29:650654, 2003 26. Phillips B, Nylander K, Harnaha J, Machen J, Lakomy R, Styche A, Gillis K, Brown L, Lafreniere D, Gallo M, Knox J, Hogeland K, Trucco M, Giannoukakis N: A microspherebased vaccine prevents and reverses new-onset autoimmune diabetes. Diabetes 57:15441555, 2008 27. Tarbell KV, Petit L, Zuo X, Toy P, Luo X, Mqadmi A, Yang H, Suthanthiran M, Mojsov S, Steinman RM: Dendritic cell-expa nded, islet-specific CD4+ CD25+ CD62L+ regulatory T cells restore normogl ycemia in diabetic NOD mice. J Exp Med 204:191-201, 2007 109

PAGE 110

28. Chatenoud L, Primo J, Bach JF: CD3 antibody-i nduced dominant self tolerance in overtly diabetic NOD mice. J Immunol 158:2947-2954, 1997 29. Chatenoud L, Thervet E, Primo J, Bach JF: [R emission of established disease in diabetic NOD mice induced by anti-CD3 monoclonal antibody]. C R Acad Sci III 315:225-228, 1992 30. You S, Leforban B, Garcia C, Bach JF, Bluestone JA, Chatenoud L: Adaptive TGF-betadependent regulatory T cel ls control autoimmune diabetes and are a privile ged target of anti-CD3 antibody treatment. Proc Natl Acad Sci U S A 104:6335-6340, 2007 31. Haller MJ, Gottlieb PA, Schatz DA: Type 1 di abetes intervention trials 2007: where are we and where are we going? Curr Opin Endocrinol Diabetes Obes 14:283-287, 2007 32. Suri A, Calderon B, Esparza TJ, Frederic k K, Bittner P, Unanue ER: Immunological reversal of autoimmune diab etes without hematopoietic replacement of beta cells. Science 311:1778-1780, 2006 33. Beran D, Yudkin JS: Diabetes care in sub-Saharan Africa. Lancet 368:1689-1695, 2006 34. Bhatia E, Aggarwal A: Insulin therapy for patients with type 1 diabetes. J Assoc Physicians India 55 Suppl:29-34, 39-40, 2007 35. King GL: The role of inflammatory cytoki nes in diabetes and its complications. J Periodontol 79:1527-1534, 2008 36. Keenan HA, Costacou T, Sun JK, Doria A, Ca vellerano J, Coney J, Orchard TJ, Aiello LP, King GL: Clinical factors associated with resistance to microvascular complications in diabetic patients of extreme diseas e duration: the 50-year medalist study. Diabetes Care 30:1995-1997, 2007 37. Sutherland DE, Gruessner RW, Dunn DL, Ma tas AJ, Humar A, Kandaswamy R, Mauer SM, Kennedy WR, Goetz FC, Robertson RP, Gruessner AC, Najarian JS: Lessons learned from more than 1,000 pancreas transplants at a si ngle institution. Ann Surg 233:463-501, 2001 38. Balamurugan AN, Bottino R, Giannoukakis N, Smetanka C: Prospective and challenges of islet transplantation for the therapy of autoimmune diabetes. Pancreas 32:231-243, 2006 39. Bertry-Coussot L, Lucas B, Danel C, Halb wachs-Mecarelli L, Bach JF, Chatenoud L, Lemarchand P: Long-term reversal of establ ished autoimmunity upon transient blockade of the LFA-1/intercellular adhesion molecule-1 pathway. J Immunol 168:3641-3648, 2002 40. Maki T, Gottschalk R, Ogawa N, Monaco AP: Prevention and cure of autoimmune diabetes in nonobese diabetic mice by continuous administration of FTY720. Transplantation 79:1051-1055, 2005 110

PAGE 111

41. Voltarelli JC, Couri CE, Stracieri AB, Olivei ra MC, Moraes DA, Pieroni F, Coutinho M, Malmegrim KC, Foss-Freitas MC, Simoes BP, Foss MC, Squiers E, Burt RK: Autologous nonmyeloablative hematopoietic stem cell transplantation in newly diagnosed type 1 diabetes mellitus. Jama 297:1568-1576, 2007 42. Goudy KS, Tisch R: Immunotherapy for the prev ention and treatment of type 1 diabetes. Int Rev Immunol 24:307-326, 2005 43. Hurwitz JL, Slobod KS, Lockey TD, Wang S, Chou TH, Lu S: Application of the polyvalent approach to HIV-1 vaccine development. Curr Drug Targets Infect Disord 5:143-156, 2005 44. Jones SE: Metastatic breast cancer: the treatment challenge. Clin Breast Cancer 8:224233, 2008 45. Schatz D, Gale EA, Atkinson MA: Why can't we prevent type 1 diabetes?: maybe it's time to try a different combination. Diabetes Care 26:3326-3328, 2003 46. Slobod KS, Bonsignori M, Brown SA, Zhan X, Stambas J, Hurwitz JL: HIV vaccines: brief review and discussion of future directions. Expert Rev Vaccines 4:305-313, 2005 47. Wen PY, Kesari S: Malignant gliomas in adults. N Engl J Med 359:492-507, 2008 48. Sherry NA, Chen W, Kushner JA, Glandt M, Tang Q, Tsai S, Santamaria P, Bluestone JA, Brillantes AM, Herold KC: Exendin-4 im proves reversal of di abetes in NOD mice treated with anti-CD3 monoclonal antibody by enhancing recovery of beta-cells. Endocrinology 148:5136-5144, 2007 49. Simon G, Parker M, Ramiya V, Wasserfa ll C, Huang Y, Bresson D, Schwartz RF, Campbell-Thompson M, Tenace L, Brusko T, Xue S, Scaria A, Lukason M, Eisenbeis S, Williams J, Clare-Salzler M, Schatz D, Kapl an B, Von Herrath M, Womer K, Atkinson MA: Murine antithymocyte gl obulin therapy alters diseas e progression in NOD mice by a time-dependent induction of immunoregulation. Diabetes 57:405-414, 2008 50. Yang Z, Chen M, Carter JD, Nunemaker CS, Garmey JC, Kimble SD, Nadler JL: Combined treatment with lisofylline and exendin-4 reverses autoimmune diabetes. Biochem Biophys Res Commun 344:1017-1022, 2006 51. Ogawa N, List JF, Habener JF, Maki T: Cure of overt diabetes in NOD mice by transient treatment with anti-lymphocyte serum and exendin-4. Diabetes 53:1700-1705, 2004 52. Suarez-Pinzon WL, Yan Y, Power R, Brand SJ, Rabinovitch A: Combination therapy with epidermal growth factor and gastrin increases beta-cell mass and reverses hyperglycemia in diabetic NOD mice. Diabetes 54:2596-2601, 2005 53. Tian C, Ansari MJ, Paez-Cortez J, Bagley J, Godwin J, Donnarumma M, Sayegh MH, Iacomini J: Induction of robust diabetes resi stance and prevention of recurrent type 1 diabetes following islet transplantation by gene therapy. J Immunol 179:6762-6769, 2007 111

PAGE 112

54. Zhang C, Todorov I, Lin CL, Atkinson M, Kand eel F, Forman S, Zeng D: Elimination of insulitis and augmentation of islet beta cel l regeneration via induction of chimerism in overtly diabetic NOD mice. Proc Natl Acad Sci U S A 104:2337-2342, 2007 55. Gaudreau S, Guindi C, Menard M, Besi n G, Dupuis G, Amrani A: Granulocytemacrophage colony-stimulating factor preven ts diabetes development in NOD mice by inducing tolerogenic dendritic cells that sustain the suppres sive function of CD4+CD25+ regulatory T cells. J Immunol 179:3638-3647, 2007 56. Hadjiyanni I, Baggio LL, Poussier P, Drucker DJ: Exendin-4 modulates diabetes onset in nonobese diabetic mice. Endocrinology 149:1338-1349, 2008 57. Jain R, Tartar DM, Gregg RK, Divekar RD, Bell JJ, Lee HH, Yu P, Ellis JS, Hoeman CM, Franklin CL, Zaghouani H: Innocuous IFNgamma induced by adjuvant-free antigen restores normoglycemia in NOD mice th rough inhibition of IL-17 production. J Exp Med 205:207-218, 2008 58. Kobayashi M, Abiru N, Arakawa T, Fukushima K, Zhou H, Kawasaki E, Yamasaki H, Liu E, Miao D, Wong FS, Eisenbarth GS, Eguchi K: Altered B:9-23 insulin, when administered intranasally with cholera to xin adjuvant, suppresses the expression of insulin autoantibodies and prevents diabetes. J Immunol 179:2082-2088, 2007 59. Verda L, Kim DA, Ikehara S, Statkute L, Br onesky D, Petrenko Y, Oyama Y, He X, Link C, Vahanian NN, Burt RK: Hematopoietic mixed chimerism derived from allogeneic embryonic stem cells prevents auto immune diabetes mellitus in NOD mice. Stem Cells 26:381-386, 2008 60. Xiu Y, Wong CP, Bouaziz JD, Hamaguchi Y, Wang Y, Pop SM, Tisch RM, Tedder TF: B lymphocyte depletion by CD20 monoclonal antibody prevents diabetes in nonobese diabetic mice despite isotype-specific differe nces in Fc gamma R effector functions. J Immunol 180:2863-2875, 2008 61. Zhang J, Tokui Y, Yamagata K, Kozawa J, Sa yama K, Iwahashi H, Okita K, Miuchi M, Konya H, Hamaguchi T, Namba M, Shimomura I, Miyagawa JI: Continuous stimulation of human glucagon-like peptide-1 (7-36) amid e in a mouse model (NOD) delays onset of autoimmune type 1 diabetes. Diabetologia 50:1900-1909, 2007 62. Chen G, Han G, Wang J, Wang R, Xu R, Shen B, Qian J, Li Y: E ssential roles of TGFbeta in anti-CD3 antibody therapy: reversal of diabetes in no nobese diabetic mice independent of Foxp3+CD4+ regulatory T cells. J Leukoc Biol 83:280-287, 2008 63. Esposito L, Kamar N, Tkaczuk J, Abbal M, Durand D, Rostaing L: Long-term evolution of lymphocytes subsets after induction thera py based on continuous versus discontinuous administration of anti-thymocyte globu lins in renal-transplant patients. Transplant Proc 37:785-787, 2005 64. Seidel MG, Fritsch G, Matthes-Martin S, Lawitschka A, Lion T, Potschger U, Rosenmayr A, Fischer G, Gadner H, Peters C: In vitro and in vivo T-cell depletion with 112

PAGE 113

myeloablative or reduced-intensity conditio ning in pediatric hematopoietic stem cell transplantation. Haematologica 90:1405-1414, 2005 65. Toor A, Rodriguez T, Bauml M, Mathews H, Shanti S, Senitzer D, Kini A, Norton J, Parthasarathy M, Mohideen N, Petrowsky C, Bo nilla B, Smith S, Stiff P: Feasibility of conditioning with thymoglobulin and redu ced intensity TBI to reduce acute GVHD in recipients of allogeneic SCT. Bone Marrow Transplant 2008 66. Ayuk FA, Fang L, Fehse B, Zander AR, Kr oger N: Antithymocyte globulin induces complement-dependent cell lysis and caspase -dependent apoptosis in myeloma cells. Exp Hematol 33:1531-1536, 2005 67. Dubey S, Nityanand S: Involvement of Fa s and TNF pathways in the induction of apoptosis of T cells by antithymocyte globulin. Ann Hematol 82:496-499, 2003 68. Fang L, Fehse B, Engel M, Zander A, Kroge r N: Antithymocyte globulin induces ex vivo and in vivo depletion of myeloid and plasmacytoid dendritic cells. Transplantation 79:369-371, 2005 69. Gillet-Hladky S, de Carvalho CM, Bernaud J, Bendahou C, Bloy C, Rigal D: Rabbit antithymocyte globulin inhibits monocyte-deri ved dendritic cells maturation in vitro and polarizes monocyte-derived dendritic cells to wards tolerogenic dendr itic cells expressing indoleamine 2,3-dioxygenase. Transplantation 82:965-974, 2006 70. Haidinger M, Geyeregger R, Poglitsch M, Weichhart T, Zeyda M, Vodenik B, Stulnig TM, Bohmig GA, Horl WH, Saemann MD: Antithymocyte globulin impairs Tcell/antigen-presenting cell interaction: disruption of immunological synapse and conjugate formation. Transplantation 84:117-121, 2007 71. Naujokat C, Berges C, Fuchs D, Sadeghi M, Opelz G, Daniel V: Antithymocyte globulins suppress dendritic cell function by multiple mechanisms. Transplantation 83:485-497, 2007 72. Zand MS, Vo T, Huggins J, Felgar R, Liesveld J, Pellegrin T, Bozorgzadeh A, Sanz I, Briggs BJ: Polyclonal rabbit antithymocyte globulin triggers B-cell and plasma cell apoptosis by multiple pathways. Transplantation 79:1507-1515, 2005 73. Battaglia M, Stabilini A, Roncarol o MG: Rapamycin selectively expands CD4+CD25+FoxP3+ regulatory T cells. Blood 105:4743-4748, 2005 74. Battaglia M, Stabilini A, Dra ghici E, Migliavacca B, Gregori S, Bonifacio E, Roncarolo MG: Induction of tolerance in type 1 diab etes via both CD4+CD25+ T regulatory cells and T regulatory type 1 cells. Diabetes 55:1571-1580, 2006 75. Battaglia M, Stabilini A, Migliavacca B, Horejs-Hoeck J, Kaupper T, Roncarolo MG: Rapamycin promotes expansion of functiona l CD4+CD25+FOXP3+ regulatory T cells of both healthy subjects and type 1 diabetic patients. J Immunol 177:8338-8347, 2006 113

PAGE 114

76. Di Paolo S, Teutonico A, Leogrande D, Capobi anco C, Schena PF: Chronic inhibition of mammalian target of rapamycin signaling downregulates insulin receptor substrates 1 and 2 and AKT activation: A crossroad between cancer and diabetes? J Am Soc Nephrol 17:2236-2244, 2006 77. Fraenkel M, Ketzinel-Gilad M, Ariav Y, Pa ppo O, Karaca M, Castel J, Berthault MF, Magnan C, Cerasi E, Kaiser N, Leibowitz G: mTOR inhibition by rapamycin prevents beta-cell adaptation to hyperglycemia and ex acerbates the metabolic state in type 2 diabetes. Diabetes 57:945-957, 2008 78. Kasiske BL, de Mattos A, Flechner SM, Gallon L, Meier-Kriesche HU, Weir MR, Wilkinson A: Mammalian target of rapamycin inhibitor dyslipidemia in kidney transplant recipients. Am J Transplant 8:1384-1392, 2008 79. Morath C, Arns W, Schwenger V, Mehrabi A, Fonouni H, Schmidt J, Zeier M: Sirolimus in renal transplantation. Nephrol Dial Transplant 22 Suppl 8:viii61-viii65, 2007 80. Klinke DJ, 2nd: Extent of beta cell destructi on is important but insufficient to predict the onset of type 1 diabetes mellitus. PLoS ONE 3:e1374, 2008 81. Janciauskiene SM, Nita IM, Stevens T: Alpha1-antitrypsin, old dog, new tricks. Alpha1antitrypsin exerts in vitro anti-inflammat ory activity in human monocytes by elevating cAMP. J Biol Chem 282:8573-8582, 2007 82. Nita I, Hollander C, Westin U, Janciauski ene SM: Prolastin, a pharmaceutical preparation of purified human alpha1-antitrypsin, blocks endotoxin-mediated cytokine release. Respir Res 6:12, 2005 83. Nita IM, Serapinas D, Janciauskiene SM: alpha1-Antitrypsin regulates CD14 expression and soluble CD14 levels in human monocytes in vitro. Int J Biochem Cell Biol 39:11651176, 2007 84. Song S, Goudy K, Campbell-Thompson M, Wa sserfall C, Scott-Jorgensen M, Wang J, Tang Q, Crawford JM, Ellis TM, Atkinson MA, Flotte TR: Recombinant adenoassociated virus-mediated alpha-1 antitrypsin gene therapy prevents type I diabetes in NOD mice. Gene Ther 11:181-186, 2004 85. Fregonese L, Stolk J: Hereditary alpha-1antitrypsin deficiency and its clinical consequences. Orphanet J Rare Dis 3:16, 2008 86. Pileggi A, Molano RD, Song S, Zahr E, SanJos e S, Villate S, Wasse rfall C, Ricordi C, Atkinson MA, Inverardi L: Al pha-1 antitrypsin treatment of spontaneously diabetic nonobese diabetic mice receiving islet allografts. Transplant Proc 40:457-458, 2008 87. Zhang B, Lu Y, Campbell-Thompson M, Spen cer T, Wasserfall C, Atkinson M, Song S: Alpha1-antitrypsin protects beta-cells from apoptosis. Diabetes 56:1316-1323, 2007 114

PAGE 115

88. Lu Y, Tang M, Wasserfall C, Kou Z, Campbe ll-Thompson M, Gardemann T, Crawford J, Atkinson M, Song S: Alpha1-antitrypsin gene therapy modul ates cellular immunity and efficiently prevents type 1 diab etes in nonobese diabetic mice. Hum Gene Ther 17:625634, 2006 89. Liu E, Moriyama H, Abiru N, Miao D, Yu L, Taylor RM, Finkelman FD, Eisenbarth GS: Anti-peptide autoantibodies and fatal anaphylax is in NOD mice in response to insulin self-peptides B:9-23 and B:13-23. J Clin Invest 110:1021-1027, 2002 90. Liu E, Moriyama H, Abiru N, Paronen J, Devendra D, Finkelman FD, Eisenbarth GS: Preventing peptide-induced anaphylaxis: additi on of C-terminal amino acids to produce a neutral isoelectric point. J Allergy Clin Immunol 114:607-613, 2004 91. Lu Y, Parker M, Pileggi A, Zhang B, C hoi YK, Molano RD, Wasserfall C, Ricordi C, Inverardi L, Brantly M, Schatz D, Atki nson M, Song S: Human alpha 1-antitrypsin therapy induces fatal anaphylaxis in non-obese diabetic mice. Clin Exp Immunol, 2008 92. Buteau J: GLP-1 receptor signaling: effects on pancreatic beta-cell proliferation and survival. Diabetes Metab 34 Suppl 2:S73-77, 2008 93. Meier JJ, Nauck MA: Glucagon-like peptide 1(GLP-1) in biology and pathology. Diabetes Metab Res Rev 21:91-117, 2005 94. Langley AK, Suffoletta TJ, Jennings HR: Dipe ptidyl peptidase IV inhibitors and the incretin system in type 2 diabetes mellitus. Pharmacotherapy 27:1163-1180, 2007 95. Choy M, Lam S: Sitagliptin: a novel drug fo r the treatment of type 2 diabetes. Cardiol Rev 15:264-271, 2007 96. Drucker DJ, Buse JB, Taylor K, Kendall DM, Trautmann M, Zhuang D, Porter L: Exenatide once weekly versus twice daily for the treatment of type 2 diabetes: a randomised, open-label, non-inferiority study. Lancet 2008 97. Forti A, Garcia EG, Yu MB, Jimenez MC, Brodows RG, Oliveira JH: Efficacy and safety of exenatide administered before the two larg est daily meals of Latin American patients with type 2 diabetes. Curr Med Res Opin 2008 98. Pratley RE: Overview of glucagon-like pep tide-1 analogs and dipeptidyl peptidase-4 inhibitors for type 2 diabetes. Medscape J Med 10:171, 2008 99. Thongtang N, Sriwijitkamol A: Incretins: the novel therapy of type 2 diabetes. J Med Assoc Thai 91:943-954, 2008 100. Ahmad SR, Swann J: Exenatid e and rare adverse events. N Engl J Med 358:1970-1971; discussion 1971-1972, 2008 101. Ahren B: Novel combination treatment of t ype 2 diabetes DPP-4 inhibition + metformin. Vasc Health Risk Manag 4:383-394, 2008 115

PAGE 116

102. Barnett AH: New treatments in t ype 2 diabetes a focus on th e incretin-based therapies. Clin Endocrinol (Oxf) 2008 103. Mu J, Woods J, Zhou YP, Roy RS, Li Z, Zycband E, Feng Y, Zhu L, Li C, Howard AD, Moller DE, Thornberry NA, Zhang BB: Chronic i nhibition of dipeptidyl peptidase-4 with a sitagliptin analog preserves pa ncreatic beta-cell mass and function in a rodent model of type 2 diabetes. Diabetes 55:1695-1704, 2006 104. Pei Z: From the bench to the bedside: dipept idyl peptidase IV inhibitors, a new class of oral antihyperglycemic agents. Curr Opin Drug Discov Devel 11:512-532, 2008 105. Dong RP, Tachibana K, Hegen M, Munakata Y, Cho D, Schlossman SF, Morimoto C: Determination of adenosine deaminase binding domain on CD26 and its immunoregulatory effect on T cell activation. J Immunol 159:6070-6076, 1997 106. Ikushima H, Munakata Y, Iw ata S, Ohnuma K, Kobayashi S, Dang NH, Morimoto C: Soluble CD26/dipeptidyl peptidase IV enha nces transendothelial migration via its interaction with mannose 6phosphate/insulin-like grow th factor II receptor. Cell Immunol 215:106-110, 2002 107. Kobayashi H, Hosono O, Mimori T, Kawasa ki H, Dang NH, Tanaka H, Morimoto C: Reduction of serum soluble CD26/dipeptidyl peptidase IV enzyme activity and its correlation with disease activity in systemic lupus erythematosus. J Rheumatol 29:18581866, 2002 108. Ohnuma K, Munakata Y, Ishii T, Iwata S, Kobayashi S, Hosono O, Kawasaki H, Dang NH, Morimoto C: Soluble CD26/dipeptidyl peptidase IV induces T cell proliferation through CD86 up-regulation on APCs. J Immunol 167:6745-6755, 2001 109. Hildebrandt M, Schabath R: SDF-1 (CXCL12) in haematopoiesis and leukaemia: impact of DPP IV/CD26. Front Biosci 13:1774-1779, 2008 110. Pacheco R, Martinez-Navio JM, Lejeune M, Climent N, Oliva H, Gatell JM, Gallart T, Mallol J, Lluis C, Franco R: CD26, adenosine deaminase, and adenosine receptors mediate costimulatory signals in the immunological synapse. Proc Natl Acad Sci U S A 102:9583-9588, 2005 111. Krakauer M, Sorensen PS, Sellebjerg F: CD4( +) memory T cells with high CD26 surface expression are enriched for Th1 markers and co rrelate with clinical severity of multiple sclerosis. J Neuroimmunol 181:157-164, 2006 112. Lun SW, Wong CK, Ko FW, Hui DS, Lam CW : Increased expression of plasma and CD4+ T lymphocyte costimulatory molecule CD26 in adult patients wi th allergic asthma. J Clin Immunol 27:430-437, 2007 113. Kleppinger EL, Helms K: The role of vildaglip tin in the management of type 2 diabetes mellitus. Ann Pharmacother 41:824-832, 2007 116

PAGE 117

114. Morishita M, Leonard RC: Pegfilgrastim; a neutrophil mediated granulocyte colony stimulating factor-expanding uses in cancer chemotherapy. Expert Opin Biol Ther 8:9931001, 2008 115. Abdallah FK: Single dose filgastrim in cy totoxic-induced neutropaenia in children. East Afr Med J 85:30-35, 2008 116. Hartung T, Docke WD, Gantner F, Krieger G, Sauer A, Stevens P, Volk HD, Wendel A: Effect of granulocyte colony-stimulating f actor treatment on ex vivo blood cytokine response in human volunteers. Blood 85:2482-2489, 1995 117. Rutella S: Granulocyte colony-stimulating f actor for the induction of T-cell tolerance. Transplantation 84:S26-30, 2007 118. Rutella S, Bonanno G, Pierelli L, Mariotti A, Capoluongo E, Contemi AM, Ameglio F, Curti A, De Ritis DG, Voso MT, Perillo A, Mancuso S, Scambia G, Lemoli RM, Leone G: Granulocyte colony-stimulating factor pr omotes the generation of regulatory DC through induction of IL-10 and IFN-alpha. Eur J Immunol 34:1291-1302, 2004 119. Hadaya K, Kared H, Masson A, Chatenoud L, Zavala F: G-CSF treatment prevents cyclophosphamide acceleration of autoim mune diabetes in the NOD mouse. J Autoimmun 24:125-134, 2005 120. Kared H, Masson A, Adle-Biassette H, Bach JF, Chatenoud L, Zavala F: Treatment with granulocyte colony-stimulati ng factor prevents diabetes in NOD mice by recruiting plasmacytoid dendritic cells and functi onal CD4(+)CD25(+) regulatory T-cells. Diabetes 54:78-84, 2005 121. Sierra J, Szer J, Kassis J, Herrmann R, Lazzarino M, Thomas X, Noga SJ, Baker N, Dansey R, Bosi A: A single dose of pegfilg rastim compared with daily filgrastim for supporting neutrophil recovery in patients tr eated for low-to-inter mediate risk acute myeloid leukemia: results from a randomized, double-blind, phase 2 trial. BMC Cancer 8:195, 2008 122. Jilma B, Hergovich N, Homoncik M, Jilma-St ohlawetz P, Kreuzer C, Eichler HG, Zellner M, Pugin J: Granulocyte col ony-stimulating factor (G-CSF) downregulates its receptor (CD114) on neutrophils and induces gelatinase B release in humans. Br J Haematol 111:314-320, 2000 123. Hermans MH, Ward AC, Antonissen C, Kari s A, Lowenberg B, Touw IP: Perturbed granulopoiesis in mice with a targeted mutation in the granulocyte colony-stimulating factor receptor gene associated w ith severe chronic neutropenia. Blood 92:32-39, 1998 124. Ward AC, van Aesch YM, Gits J, Schelen AM, de Koning JP, van Leeuwen D, Freedman MH, Touw IP: Novel point mutation in th e extracellular domain of the granulocyte colony-stimulating factor (G-C SF) receptor in a case of se vere congenital neutropenia hyporesponsive to G-CSF treatment. J Exp Med 190:497-507, 1999 117

PAGE 118

125. Palmer AJ, Valentine WJ, Ray JA, Foos V, Lu rati F, Smith I, Lammert M, Roze S: An economic assessment of analogue basal-bolus in sulin versus human basal-bolus insulin in subjects with type 1 diabetes in the UK. Curr Med Res Opin 23:895-901, 2007 126. Harrison LC: Vaccination against self to preven t autoimmune disease: the type 1 diabetes model. Immunol Cell Biol 86:139-145, 2008 127. Butler AE, Huang A, Rao PN, Bhushan A, Hogan WJ, Rizza RA, Butler PC: Hematopoietic stem cells derived from adult donors are not a source of pancreatic betacells in adult nondiabetic humans. Diabetes 56:1810-1816, 2007 128. Shoda LK, Young DL, Ramanujan S, Wh iting CC, Atkinson MA, Bluestone JA, Eisenbarth GS, Mathis D, Rossini AA, Campbell SE, Kahn R, Kreuwel HT: A comprehensive review of interventions in the NOD mouse and implications for translation. Immunity 23:115-126, 2005 129. Harrison LC, Honeyman MC, Morahan G, Wentworth JM, Elkassaby S, Colman PG, Fourlanos S: Type 1 diabetes: Lesso ns for other autoimmune diseases? J Autoimmun 2008 130. Baxter AG, Cooke A: Complement lytic activ ity has no role in the pathogenesis of autoimmune diabetes in NOD mice. Diabetes 42:1574-1578, 1993 131. Burke SD, Dong H, Hazan AD, Croy BA: Aberrant endometrial features of pregnancy in diabetic NOD mice. Diabetes 56:2919-2926, 2007 132. Fan H, Longacre A, Meng F, Patel V, Hsiao K, Koh JS, Levine JS: Cytokine dysregulation induced by apoptotic cells is a shared characteristic of macrophages from nonobese diabetic and systemic l upus erythematosus-prone mice. J Immunol 172:48344843, 2004 133. Rosignoli F, Roca V, Meiss R, Leceta J, Gomariz RP, Perez Leiros C: Defective signalling in salivary glands precedes the auto immune response in the non-obese diabetic mouse model of sialadenitis. Clin Exp Immunol 142:411-418, 2005 134. Gale EA, Bingley PJ, Emmett CL, Collier T: European Nicotinamide Diabetes Intervention Trial (ENDIT): a ra ndomised controlled trial of in tervention before the onset of type 1 diabetes. Lancet 363:925-931, 2004 135. Keymeulen B, Vandemeulebroucke E, Ziegle r AG, Mathieu C, Kaufman L, Hale G, Gorus F, Goldman M, Walter M, Candon S, Schandene L, Crenier L, De Block C, Seigneurin JM, De Pauw P, Pierard D, Weets I, Rebello P, Bird P, Berrie E, Frewin M, Waldmann H, Bach JF, Pipeleers D, Chat enoud L: Insulin needs after CD3-antibody therapy in new-onset type 1 diabetes. N Engl J Med 352:2598-2608, 2005 136. Meier JJ, Bhushan A, Butler AE, Rizza RA, Butle r PC: Sustained beta cell apoptosis in patients with long-standing type 1 diabetes: indirect evidence for islet regeneration? Diabetologia 48:2221-2228, 2005 118

PAGE 119

119 137. Meier JJ, Lin JC, Butler AE, Galasso R, Ma rtinez DS, Butler PC: Direct evidence of attempted beta cell regeneration in an 89-y ear-old patient with recent-onset type 1 diabetes. Diabetologia 49:1838-1844, 2006

PAGE 120

BIOGRAPHICAL SKETCH Matthew John Parker was born to Sheila and Martin Parker in Stockport, England. In 1986 he moved with his parents and twin brother, David, to Mobile, Alabama. In May 2000, he graduated from the Alabama School of Mathem atics and Science. Following high school, he went out-of-state in fall 2000 in or der to enroll at the Un iversity of Florida as a National Merit Scholar. While pursuing a Bachelor of Scien ce degree in Microbiology and Cell Science, he completed the University Honors Program and was inducted into the Phi Eta Sigma Honors Society. In the summer following his freshman year, Matthew volunteered in the lab of Dr. Charles Martin in order to study the processe s of gold and carbon nanotube synthesis in the department of analytical chemistry. During his sophomore year, he continued working with Dr. Martin part-time with the assistance of Dr. Mark Wirtz. In the summer following his sophomore year, Matthew began working in the lab of Dr. Mark Atkinson and was introduced to the realm of type 1 diabetes research and autoimmunity. During this time, he also volunteered in the la b of Dr. David Ostrov, where he was exposed to the field of x-ray crystallogra phy and computer-aided structure-based drug screening. During his final year of college, he used his research with Dr. Atkinson in order to participate in the University Scholar's Program. He continued this work part-time until his graduation in fall 2003. Matthew continued working with Dr. Atkinson full-tim e as a research laboratory technician until beginning graduate studies in fall 2004 in the In terdisciplinary Program (IDP) in Biomedical Sciences at the College of Medicine at the Un iversity of Florida, where he was awarded an Alumni Fellowship. While in the IDP, Matthew joined the lab of Dr. Mark Atkinson as part of the immunology concentration. There he investigated the use of combination therapy for the purposes of reversing type 1 diabetes in the non-obese diabetic mouse model, with the eventual goal of such therapy

PAGE 121

translating into clinical use. During his graduate studi es, he married fellow IDP student Nicole Teoh in November 2006. After graduation, Matthew plan s to continue his rese arch in the field of autoimmunity during a po st-doctoral fellowship.