<%BANNER%>

Biomaterial-Based Modulation of Dendritic Cells

Permanent Link: http://ufdc.ufl.edu/UFE0041515/00001

Material Information

Title: Biomaterial-Based Modulation of Dendritic Cells Adhesion Based Modulation and Highthroughput Particle Vaccine Generation, Optimization and Delivery.
Physical Description: 1 online resource (156 p.)
Language: english
Creator: Acharya, Abhinav
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: activation, adhesion, arrays, delivery, dendritic, diabetes, drug, highthroughput, nodmouse, optimization, particles, plga, surface, tcells, vaccine
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Modulation of immune-cell responses using biomaterials based cues is an exciting field of research that holds great potential to help solve disorders such as autoimmune diseases, cancer and infections. Medical devices used in numerous applications such as tissue-engineered constructs, combination products (e.g., drug-eluting stents) and therapeutic vaccines are excellent tools for modulating the immune system. Since dendritic cells (DCs) are the key regulators of the immune system, DCs can be targeted to generate a desired immune response. Dendritic cell functions can be altered using several different signals such as adhesion signaling, targeting extracellular/intracellular toll like receptors for activation and mechanical cues. Specifically, we are interested in modulating DC behavior using adhesion cues and delivery of agents that target extracellular and intracellular receptors. Exogenously generated DCs have been suggested as a potential solution to diseases such as type 1 diabetes (T1D). Ex vivo expansion of cells require isolation from the patient and culturing them on tissue culture treated plates where they come in contact with several different proteins. Adhesion of such cells on different extracellular matrix proteins can modulate cellular responses. While it is well-known that adsorbed proteins on biomaterials modulate inflammatory responses in vivo, modulation of dendritic cells (DCs), a key regulator of immune system, via adhesion-dependent signaling has only been begun to be characterized. Currently, DC-based immunotherapy approaches for diseases such as cancer and autoimmune diseases like type-I diabetes rely on ex vivo culture and expansion of patient-derived DCs onto tissue culture-treated polystyrene plates. The adhesive substrate provided for DCs in this ex vivo approach is typically tissue culture-treated polystyrene presenting serum proteins adsorbed from the culture media. We therefore chose to examine serum-coated tissue culture-treated polystyrene as a relevant benchmark to compare the effect of adhesion-dependent modulation of DC function when cultured on several of the extracellular matrix proteins. Furthermore, DCs isolated from non-obese diabetic mice and its background control of wild type mice were cultured on the extracellular matrix proteins and compared for optimal protein substrate for generating DC based vaccines. In addition to modulating DCs ex vivo, DCs can be targeted in vivo using particle-based vaccines. Currently there are scores of known antigenic epitopes and adjuvants, and numerous synthetic delivery systems accessible for formulation of vaccines. However, the lack of an efficient means to test immune cell responses to the abundant combinations available represents a significant blockade on the development of new vaccines. In order to overcome this barrier, we report fabrication of a new class of microarray consisting of antigen/adjuvant-loadable poly(D,L lactide-co-glycolide) microparticles (PLGA MPs), identified as a promising carrier for immunotherapeutics, that are cultured with DCs. Furthermore, a technique was generated to manufacture scores of particle-based vaccines in a highthroughput manner. The intention is to utilize this high-throughput platform to optimize particle-based vaccines designed to target DCs in vivo for immune system-related disorders, such as autoimmune diseases, cancer and infection.
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 Abhinav Acharya.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Keselowsky, Ben.
Local: Co-adviser: Gower, Laurie B.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041515:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041515/00001

Material Information

Title: Biomaterial-Based Modulation of Dendritic Cells Adhesion Based Modulation and Highthroughput Particle Vaccine Generation, Optimization and Delivery.
Physical Description: 1 online resource (156 p.)
Language: english
Creator: Acharya, Abhinav
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: activation, adhesion, arrays, delivery, dendritic, diabetes, drug, highthroughput, nodmouse, optimization, particles, plga, surface, tcells, vaccine
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Modulation of immune-cell responses using biomaterials based cues is an exciting field of research that holds great potential to help solve disorders such as autoimmune diseases, cancer and infections. Medical devices used in numerous applications such as tissue-engineered constructs, combination products (e.g., drug-eluting stents) and therapeutic vaccines are excellent tools for modulating the immune system. Since dendritic cells (DCs) are the key regulators of the immune system, DCs can be targeted to generate a desired immune response. Dendritic cell functions can be altered using several different signals such as adhesion signaling, targeting extracellular/intracellular toll like receptors for activation and mechanical cues. Specifically, we are interested in modulating DC behavior using adhesion cues and delivery of agents that target extracellular and intracellular receptors. Exogenously generated DCs have been suggested as a potential solution to diseases such as type 1 diabetes (T1D). Ex vivo expansion of cells require isolation from the patient and culturing them on tissue culture treated plates where they come in contact with several different proteins. Adhesion of such cells on different extracellular matrix proteins can modulate cellular responses. While it is well-known that adsorbed proteins on biomaterials modulate inflammatory responses in vivo, modulation of dendritic cells (DCs), a key regulator of immune system, via adhesion-dependent signaling has only been begun to be characterized. Currently, DC-based immunotherapy approaches for diseases such as cancer and autoimmune diseases like type-I diabetes rely on ex vivo culture and expansion of patient-derived DCs onto tissue culture-treated polystyrene plates. The adhesive substrate provided for DCs in this ex vivo approach is typically tissue culture-treated polystyrene presenting serum proteins adsorbed from the culture media. We therefore chose to examine serum-coated tissue culture-treated polystyrene as a relevant benchmark to compare the effect of adhesion-dependent modulation of DC function when cultured on several of the extracellular matrix proteins. Furthermore, DCs isolated from non-obese diabetic mice and its background control of wild type mice were cultured on the extracellular matrix proteins and compared for optimal protein substrate for generating DC based vaccines. In addition to modulating DCs ex vivo, DCs can be targeted in vivo using particle-based vaccines. Currently there are scores of known antigenic epitopes and adjuvants, and numerous synthetic delivery systems accessible for formulation of vaccines. However, the lack of an efficient means to test immune cell responses to the abundant combinations available represents a significant blockade on the development of new vaccines. In order to overcome this barrier, we report fabrication of a new class of microarray consisting of antigen/adjuvant-loadable poly(D,L lactide-co-glycolide) microparticles (PLGA MPs), identified as a promising carrier for immunotherapeutics, that are cultured with DCs. Furthermore, a technique was generated to manufacture scores of particle-based vaccines in a highthroughput manner. The intention is to utilize this high-throughput platform to optimize particle-based vaccines designed to target DCs in vivo for immune system-related disorders, such as autoimmune diseases, cancer and infection.
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 Abhinav Acharya.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Keselowsky, Ben.
Local: Co-adviser: Gower, Laurie B.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041515:00001


This item has the following downloads:


Full Text

PAGE 1

1 BIOMATERIAL-BASED MODULATION OF DENDRITIC CELLS: ADHESION BASED MODULATION AND HIGHTHROUGHPUT PARTICLE VACCINE GENERATION, EVALUATION AND DELIVERY By ABHINAV P. ACHARYA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Abhinav P. Acharya

PAGE 3

3 To my parents

PAGE 4

4 ACKNOWLEDGMENTS There are a lot of people who have guided and supported me throughout my Ph.D. and I would like to thank them. Firstly, I w ould like to thank Dr. Benjamin Keselowsky, who has been my mentor and guide for the past five years. I also thank my committee members, Dr. Laurie Gower, Dr. Christopher Batich, Dr. Eugene Goldberg, Dr Anthony Brennan, and Dr. Michael Clare-Salzler for their teachings, guidance and support. I have always been blessed with excellent teachers and mentors throughout my educational career and I would like to extend a special thanks to Dr. Henry Hess, Dr. William Ogle and Dr. Brain Sorg. I thank all my colleagues in the Keselo wsky group Natalia Dolgova, Jerome Karpiak, Toral Zaveri, Jamal Lewis, Ma tt Carston and Matt Sines for helping and supporting me in my work. I would like to spec ially thank Natalia for helping me with cell biology and being such a great friend. Furthermore, I woul d like to thank colleagues outside the lab, Ashutosh and Amit for driving me to deliver two most productive years of my Ph.D. work; Phil and Mamta for shapi ng my ideas, making the work enjoyable and for the memorable discussions over l ong lunches; Sewoon Choe for keeping me company in the lab through the long busy nigh ts of research; Ying for being so nice and entertaining and introducing me to the hot pot; Chong, and Shan for keeping the research light and fun. Further, I would like to thank Jerome and Elona for their invaluable advice both out side and in the lab.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF FIGURES .......................................................................................................... 8 ABSTRACT ................................................................................................................... 16 CHA PTER 1 INTRODUC TION .................................................................................................... 19 Adhesion-Dependent Modulat ion of DC-Function ................................................... 20 Activation of Dendritic Cells upon Adhesion Type 1 Di abetes vs. Wild Type .............................................................................................................. 21 Integrin-Peptide Based Controlled Activation of D endritic Cells ....................... 25 High-Throughput Production and Biological Evaluation of Antigen Presenting Cell-Directed Va ccine Part icles ........................................................................... 26 High-throughput Microparticle Microa rray for Dendritic Cell Targeted Vaccines ........................................................................................................ 27 Dendritic Cell Arrays for Biological Evaluation of Va ccine Particles ................. 30 Parallel Vaccine Particle Production ................................................................. 31 2 ADHESIVE SUBSTRATE-MODULAT ION OF ADAPTI VE IMMUNE RESPONSES IN C57BL6/J MICE .......................................................................... 32 Introduc tion ............................................................................................................. 32 Generation of Murine B one Marrow-Deri ved DCs .................................................. 32 Isolation of T-Cells .................................................................................................. 34 Protein Coating and DC Culture ............................................................................. 34 Dendritic Cell Adhesion and Proliferation ................................................................ 36 Quantification of DC Surface Maturati on Mark ers ................................................... 36 Quantification of DC Cytokine Pr oduction ............................................................... 37 Mixed Lymphocyt e Reaction ................................................................................... 37 Statistical Analysis .................................................................................................. 38 Results .................................................................................................................... 38 Dendritic Cell Mor phology and Ad hesion.......................................................... 38 Dendritic Cell Phenotype .................................................................................. 41 Dendritic Cell Cyto kine Secretion ..................................................................... 46 Mixed Lymphocyt e Reaction ............................................................................ 50 Impact of t he Study ................................................................................................. 53 3 ADHESIVE SUBSTRATE-MODULAT ION OF ADAPTI VE IMMUNE RESPONSES IN NOD MICE .................................................................................. 56 Introduc tion ............................................................................................................. 56

PAGE 6

6 Generation of Murine B one Marrow-deriv ed DCs ................................................... 56 Isolation of T-Cells .................................................................................................. 57 Protein Coating and DC Culture ............................................................................. 57 Protocols for Studying DC-Func tions ...................................................................... 57 Mixed Lymphocyt e Reaction ................................................................................... 57 Statistical Analysis .................................................................................................. 58 Results .................................................................................................................... 58 Endotoxin Quantification, DC Morphology an d Adhesion ................................. 58 Dendritic Cell Phenotype .................................................................................. 62 Dendritic Cell Cyto kine Secretion ..................................................................... 72 Mixed Lymphocyt e Reaction ............................................................................ 75 Impact of t he Study ................................................................................................. 80 4 INTEGRIN-MEDIATED PEPTIDE ADHESI ON BASED CONTROLLED ACTIVATION OF DENDRITIC CELLS .................................................................... 88 Introduc tion ............................................................................................................. 88 Chip Manuf acture ................................................................................................... 88 Dendritic Cell Isolation ............................................................................................ 89 Dendritic Cell Cu lture on Chip ................................................................................. 90 Immunostaining ...................................................................................................... 90 MHC-II and CD86 ............................................................................................. 90 IL-10 and IL -12p40 ........................................................................................... 91 V Staini ng ....................................................................................................... 91 Imaging and Analysis .............................................................................................. 92 Statistical Analysis .................................................................................................. 94 Results .................................................................................................................... 94 Impact of t he Study ............................................................................................... 105 5 A HIGH-THROUGHPUT MICROPARTICL E MICROARRAY PL ATFORM FOR DENDRITIC CELL-TARG ETING VACCINES ....................................................... 107 Introduc tion ........................................................................................................... 107 Preparation of PLGA Microparti cles ...................................................................... 107 Characterization of Microparti cles ......................................................................... 109 Particle Size Measurements ........................................................................... 109 Scanning Electron Microscopy ....................................................................... 109 Efficiency of Protei n Encapsul ation ....................................................................... 109 Degradation of Mi crop arti cles ............................................................................... 110 Particle Array Fabrication ...................................................................................... 110 Dendritic Cell Cult ure and Stai ning ....................................................................... 114 Statistical Analysis ................................................................................................ 115 Results .................................................................................................................. 115 Array Fabrication and Micropar ticle Characte rization ..................................... 115 Particle Array Valida tion ................................................................................. 120 Dendritic Cell A rray Fabric ation ...................................................................... 126 Impact of t he Study ............................................................................................... 128

PAGE 7

7 6 DENDRITIC CELL ARRAYS FOR BIOLOGICAL EVALUATION OF VACCINE PARTICLES .......................................................................................................... 130 Introduc tion ........................................................................................................... 130 Materials and Methods .......................................................................................... 130 Statistical Analysis ................................................................................................ 131 Results .................................................................................................................. 131 Impact of t he Study ............................................................................................... 135 7 HIGHTHROUGHPUT PARTICLE-BAS ED VACCINE GENERATION .................. 137 Introduc tion ........................................................................................................... 137 Materials Ut ilized .................................................................................................. 138 Parallel Partic le Produc tion ................................................................................... 138 Particle Analysis ................................................................................................... 140 Statistical Analysis ................................................................................................ 140 Results .................................................................................................................. 141 Impact of t he Study ............................................................................................... 145 8 CONCLUSION AND FU TURE OUTLOOK ........................................................... 146 LIST OF REFERENCES ............................................................................................. 149 BIOGRAPHICAL SKETCH .......................................................................................... 156

PAGE 8

8 LIST OF FIGURES Figure page 1-1 Schematic of immunotherapy approac h with the introduction of engineered adhesiv e substrates to direct cell maturation that can be potential used for Diabetes Type I. ................................................................................................. 21 2-1. Schematic of adsorption of prot eins on tissue cult ure treated polystyrene surfaces and culturing DCs on the modified surfaces. ........................................ 36 2-2 Initial adhesion of DCs is statisti cally not different for different adhesiv e substrates with an overall ANOVA p-value of less than 1. Data represents average and standard error of at l east 6 data points (replicates)........................ 39 2-3 Murine C57BL6/j dendritic cell ( DC) morphology is modulated by adhesive substrate. Inset micrographs represent a typical zoomed-in morphology of DCs on the given substrate. ............................................................................... 40 2-4 Adhesive substrates activate muri ne C57BL6/j bone marrow-derived dendritic cells (DCs) as evidenced through phenotypic surface presentation of major histocompatibility (MHC). Data re present average and standard error of at least 6 data points (replicates). DCs we re obtained from at least 3 separate mice and each mouse handled as an independent experiment repeat. A) Dendritic cells positive for su rface expression of MHC-II. B) Dendritic cells surface expression of MHC-II. C) Dendritic cells positive for surface expression of MHC-II when cult ured in the presence of LPS. D) Dendritic cells surface expression of MHC-II w hen cultured in the presence of LPS. The significant pair symbols are described in the text. ....................................... 43 2-5 Adhesive substrates activate muri ne C57BL6/j bone marrow-derived dendritic cells (DCs) as evidenced through phenot ypic surface presentation costimulatory molecules. Data represent average and standard error of at least 6 data points (replicates). DCs were obtained from at least 3 separate mice and each mouse handled as an independent experiment repeat. A) Dendritic cells positive for surface expression of CD86. B) Dendritic cells surface expression of CD86. C) Dendritic cells positive for surface expression of CD86 when cultured in the presence of LPS. D) Dendritic cells surface expression of CD86 when cultured in t he presence of LPS. The significant pair symbols are descri bed in the text. ............................................................... 45 2-6 Representative phenotypic density plots of data summarized in Figs. 3 and 4 for murine C57BL6/j-derived DCs cultured on protein-coated tissue culturetreated polystyrene 24 h. Shown also are iDC purity and immaturity (lack of MHC-II, CD80 and CD86). .................................................................................. 46

PAGE 9

9 2-7 Adhesive substrates differentiall y modulate murine C57BL6/j bone marrowderived dendritic cell (DC) cytokine production. Data represent average and standard error of at least 9 data points (replic ates). DCs were obtained from at least 3 separate mice and eac h mouse handled as an independent experiment repeat. A) Dendritic cells producti on of IL-12 upon culture on adhesive substrates. B) Dendritic cells production of IL-12 upon culture on adhesive substrates in the presence of LPS. C) Dendritic cells production of IL-10 when cultured on adhesive substrates. D) Dendritic cells production of IL-10 when cultured on adhesive substrat es in the presence of LPS. The significant pair symbols are described in the text. .............................................. 49 2-8. Adhesive substrates differentially modulate dendritic cell-mediated T-cell proliferation. A total of 20,000 CD4+ T-cells were analyzed for each run. Each experiment was independently repeated at least 4 times (C57Bl6/j for DCs and BALB/cbyj for CD4+ T-cells). A) T-cell proliferation at 48 h B) T-cell proliferation at 96 h. T he significant pair symbols are described in the text. ....... 53 3-1 Adhesive substrates differentially modulate NOD-DC mor phology. A pre-coated tissue culture treated plate was utiliz ed to culture NOD-DCs for 24 hrs and phase contrast micrographs were acquired (scale bar =100 m). In set micrographs are shown to clearly demonstrate morphology of DCs on the given substrate (scale bar = 50 m). ................................................................... 59 3-2 Non-obese diabetic mouse derived DCs can be divided into different groups according to the typical activated mo rphological manifestations of veil, dendrites and clusters is demonstrat ed as a Venn diagram upon culture on different adhesive protein substrates : bovine serum albumin (BSA), colla gen (COL), fibrinogen (FG), fibronectin (F N), laminin (LN), serum (SER) and vitronectin (VN) while DCs cultured in the presence of lipopolysaccharide (LPS) was included as a cont rol. ........................................................................ 61 3-3 Initial-adhesion of DCs cultured on tissue culture treated 96-well plates precoated with following proteins: bovine serum albumin (BSA), collagen (COL), fibrinogen (FG), fibronectin (FN), lami nin (LN), serum (SER) and vitronectin (VN) has a differential profile. Immatu re DCs (No Pre-coat) and culture with lipopolysaccharide (LPS, 1 g/mL) were included as controls. Data represents mean and standard error of at least 12 data points (replicates) consolidated from two mice. The signi ficant pair symbols are described in the text. ..................................................................................................................... 62 3-4 Adhesive substrates differentially activate non-obes e diabetic (NOD) bone marrow derived-dendritic cells (DCs) as evidenced through surface presentation of major hist ocompatibility (MHC-II). Dendritic cells were isolated from at least 2 separate mice and each mouse handled as an independent experiment repeat. It is im portant to note that all the DCs adhered or floating were included in the analysis. A) Dendritic cells positive for surface expression of MHC-II. B) Dendritic cells surface expression of

PAGE 10

10 MHC-II. C) Dendritic cells positive for su rface expression of MHC-II when cultured in the presence of LPS. D) Dendritic cells surface expression of MHC-II when cultured in the presence of LPS. The significant pair symbols are described in the text. .................................................................................... 66 3-5 Adhesive substrates differentially activate non-obes e diabetic (NOD) bone marrow derived-dendritic cells (DCs) as evidenced through surface presentation of co-stimulatory molecule, CD86. Dendritic cells were isolated from at least 2 separate mice and each mouse handled as an independent experiment repeat. It is important to note that al l the DCs adhered or floating were included in the analysis. A) Dendritic cells positive for surface expression of CD86. B) Dendritic cells surface expression of CD86. C) Dendritic cells positive for surface ex pression of CD86 when cultured in the presence of LPS. D) Dendritic cells surfac e expression of CD86 when cultured in the presence of LPS. The significant pair symbols are described in the text. ........................................................................................................... 68 3-6 Adhesive substrates differentially activate non-obes e diabetic (NOD) bone marrow derived-dendritic cells (DCs) as evidenced through surface presentation of co-stimulatory molecule, CD80. Dendritic cells were isolated from at least 2 separate mice and each mouse handled as an independent experiment repeat. It is important to note that al l the DCs adhered or floating were included in the analysis. A) Percentage of DC population expressing CD80. B) Dendritic cell expression of CD80. The significant pair symbols are described in the text. .......................................................................................... 69 3-7 Representative phenotypic density plots of data summarized in Figure 3-5, 3-6 for murine NOD-derived DCs cultured on protein-coated tissue culturetreated polystyrene substrate for 24 h. Shown also is iDC purity and maturation ma rkers. ............................................................................................ 71 3-8 Soluble proteins do not activate DCs, quantified via percentage of cells express ing stimulatory (M HC-II) and co-stimulatory (CD86) molecules. Density plots of MHC-II vs. CD 86 are generated through FCS Express version 3 software and the data was obt ained using flow cytometry. The percentage of DCs positive for these mo lecules is reported in the insets. .......... 71 3-9 Adhesive substrate modulates non-obes e diabetic (NOD) mouse dendritic cells (DCs) cytokine production. Dendritic ce lls were isolated from at least 3 separate mice and each mouse handled as an independent experiment repeat. It is important to note that all the DCs adhered or floating were included in the analysis. A) Dendritic cells production of IL-12 upon culture on adhesive substrates. B) Dendritic cells production of IL-12 upon culture on adhesive substrates in the presence of LPS. C) Dendritic cells production of IL-10 when cultured on adhesive substrates. D) Dendritic cells production of IL-10 when cultured on adhesive substrat es in the presence of LPS. The significant pair symbols are described in the text. .............................................. 74

PAGE 11

11 3-10 In a 96-hour mixed-lymphocyte reaction (1:6, DC to T-cell ratio) DCs modulate T-cell prolif eration differentially. Immature DCs were cultured on these modified surfaces for 24 h before adding T-cells isolated from the spleen. Each experiment was independently repeat ed at least 3 times (NOD/LtJ for DCs and BALB/cbyj for CD4+ T-cells). At least 20,000 CD4+ T-cells were analyzed for each run. The si gnificant pair symbols are described in the text. ... 75 3-11 In a 96-hour mixed-lymphocyte reaction (1:6, DC to T-cell ratio) DCs modulate CD4+ T-cell cytokine production and expres sion of these cytokines. Immature DCs were cultured on these modified surfaces for 24 h before adding T-cells isolated from the spleen. Each experiment was independently repeated at least 3 times (NOD/LtJ for DCs and BALB/cbyj for CD4+ T-cells). At least 20,000 CD4+ T-cells were analyzed for each run. A) Percentage T cells expressing IFN. B) T cells expressing IFN. C) Percentage T cells expressing IL-4. D) T cells expressing IL-4. T he significant pair symbols are described in the text. .......................................................................................... 76 3-12 Representative phenotypic density plots of data summarized in Figure 3-10, 311 for murine BALB/c byj CD4+ T-cells co-cultured with NOD-DCs on protein-coated tissue culture-treated polystyrene substrate for 96 hours. .......... 79 3-13 In a 96-hour mixed-lymphocyte reaction (1:6, DC to T-cell ratio) B6-DCs do not modulate T-cell proliferation differentiall y when cultured with soluble protein. ... 80 3-14 Dendritic cell surface stimulatory and co-stimulatory molecules and cytokine produced may modulate T-cell response, in a mixed ly mphocyte reaction, determined using Pearsons correlation valu es (0.1 to 0.4 low correlation, 0.4 to 0.7 moderate correlation and 0.7 to 1.0 high correlation). .............................. 85 3-15 A 3-factor bubble graph between expression of CD80 by DCs (gMFI-CD80) on the x-axis and DC-cytoki ne production (IL-10 and IL -12p40) on the y-axis is plotted with the diamet er of the bubble repres ent ing the percentage of CD4+ T-cells producing IL-4 or IFN. These plots have been constructed to understand the effect of cumulative effe cts of expression of co-stimulatory molecule and cytokine produced by DCs on T-helper type responses. A) Dependence of percent of T-cells producing IL-4 on DC IL-10 production and CD80 expression. B) Dependence of percent of T-cells producing IFNon DC IL-12p40 production and CD80 expr ession. ................................................. 86 4-1 Gradient SAM substrates can be conv erted into biomolecule functionaliz ed gradient surfaces with defined concentration, s patial orientation and complete chemical specificity by coupling an alkyne-terminated linker through the UV-oxidation generated ca rboxyl groups. The steps A to D show different process done to crosslink pept ide on the s ubstrate. ........................................... 89 4-2 Dendritic cells cultured and immunof luorescence stained on the RGD-gradient can be quantified for the express ion of surface molecules such as MHC-II-

PAGE 12

12 stimulatory molecule; CD86-co-stimulato ry molecule and intra-cellular antiinflammatory cytokine IL-10; pr o-inflammatory cytokine IL-12p40. A) The schematic of the chip utilized for making the RGD-gradient is shown. B) A representative micrograph of a chip obtained using fluorescent microscope is shown with increasing gradient of RGD from left to right and DCs stained for surface expression of CD86 with FITC (s hown in green) and nuclei with DAPI (shown in blue). C) Image analysis was performed on the micrographs of stained DCs for nuclei and surface mark ers or intracellu lar cytokine. ................ 93 4-3 Dendritic cells cultured on RGD-gradien t chips demonstrate that the number of DCs adhering does not have a correlation wit h the RGD-gradient or control gradient. A) Number of DCs on RGD gradient B) Number of DCs on control gradien t. ............................................................................................................. 95 4-4 Dendritic cells cultured on RGD-gr adient chips demonstrate that surface express ion of CD86 co-stimulatory mo lecule is proportional to the RGDgradient. A) CD86 expression by DCs cultured on RGD gradient and control gradient. B) Pearsonss Correlation coefficient for DCs cultured on RGD gradient. C) Pearsonss Correlation coefficient for DCs cultured on control gradien t. ............................................................................................................. 97 4-5 Dendritic cells cultured on RGD-gr adient chips demonstrate that surface express ion of MHC-II st imulatory molecule is proportional to the RGDgradient. A) MHC-II expression by DCs cultured on RGD gradient and control gradient. B) Pearsonss Correlation coefficient for DCs cultured on RGD gradient. C) Pearsonss Correlation coefficient for DCs cultured on control gradien t. ............................................................................................................. 99 4-6 Dendritic cells cultured on the RGD-gr adient chips and the control chips for 24 h, were stained for the intracellular cytokine IL-10 using immunofluorescence staining and image analysi s was performed to quantify the fluorescence intensity at varying RGD surface density present on the chip. A) IL-10 expression by DCs cultured on RG D gradient and control gradient. B) Pearsonss Correlation coefficien t for DCs cultured on RGD gradient. C) Pearsonss Correlation coefficient for DCs cultured on control gradient. .......... 100 4-7 Dendritic cells cultured on RGD-gradien t chips demonstrate that intracellular pro-inflammatory cytokine IL-12p40 is proportional to the RGD-gradient. A) IL-12p40 expression by DCs cultured on RGD gradient and control gradient. B) Pearsonss Correlation coefficient for DCs cultured on RGD gradient. C) Pearsonss Correlation coefficient for DCs cultured on control gradient. .......... 103 4-8 Dendritic cells demons trate i ncreased integrin V expression with increase in RGD-peptide surface density. A) RGD bound V integrin expression by DCs cultured on RGD gradient and control gradient. Arrow shows the threshold value upon which the RGD bound V expression is different from the control.

PAGE 13

13 B) Pearsonss Correlation coefficient for DCs cultured on RGD gradient. C) Pearsonss Correlation coefficient for DCs cultured on control gradient. .......... 105 5-1 Microparticle/dendritic ce ll (MP/DC) array fabrication incorpor ates miniarraying solid pin contact printing equipmen t, silane and alkanethiol surface chemistry, and physisorption of MPs to provide MP/DC co-localization on isolated spots. A.) Surface-engineering MP array. B.) Cross-section of a single spot in a MP-array illustrating physisorbed MPs on NH2-terminated spots with a polyethylene glycol-based non-adhesive background surface chemistry (not to scale). C.) Dendritic cells are seeded on MP arrays, selectively adherent to NH2-terminated spots providi ng co-localized DCs/MPs arrays. .............................................................................................................. 112 5-2 A typical size distribution curve of PLGA-microparticles quantified via dynamic light scattering analysis (based on vo lume estimation) demonstrates an average siz e of particles of 1.08 m. A bimodal poly-disper sity in the particle size is observed; with the smaller popul ation-set has an average diameter of 0.76 m, whereas the larger size population-set has 1.97 m average diameter. .......................................................................................................... 116 5-3 Surface-adsorbed PLGA-microparticles degrade in a pH-dependent fashion and are quantified in situ by image analysis. ........................................................... 118 5-4 Scanning electron micrographs demonstrat e constructed arrays of surfaceadsorbed PLGA MPs (microparticles) on NH2-terminated silane spots (visible as circular regions devoid of metal deposition). A.) Overspotting pin diameter is optimized for aligned delivery of MPs. B.) Representative micrograph of MPs printed on the adhesive-islands indicating MP smooth surface morphology and spherical shape (scale bar = 5 m). ....................................... 119 5-5 Delivery of microparticles (MPs) by so lid pin microarray printing is controlled by source plate particle suspension concentration to deliver a lower limit of 16 2 surface-adsorbed MPs per spot. A.) Representative fluorescenc e micrograph of serial 1:2 dilutions of rhodamine dye-loaded PLGA MPs printed in quadruplet in a 4 x 8 array format is shown (for 4 separate preparations; scale bar = 200 m.). B.) Surface-adsorbed MP numbers are quantified by image analysis, average del ivered MP values with standard deviations are plotted as a function of source plate concentration, and linear fit parameters are shown. The significant pair symbols are described in the text. ................................................................................................................... 122 5-6 Fluorescence micrograph of different microparticle (MP) formulations quantitatively printed using solid pin c ontact printing in dif ferent MP-dose combinations with minima l cross-contam ination. .............................................. 124 5-7 Particles encapsulating rhodamine (RHOD, red) and 9anthracenecarboxylic acid (ACA, blue) are printed and surface-adsorbed microparticle (MP)

PAGE 14

14 numbers are quantified by fluorescenc e image analysis on a spot-by-spot basis. Data from 4 separate array preparations were pooled and average and standard deviations for each arrayed spot are presented in 3-D bar plots. Values on xand y-axes represent the RHOD or ACA source plate MP concentration while z-axis values represent the number of surface-adsorbed MPs of either ACA ( A. ) or RHOD (B. ). C.) Average delivered MP values with standard deviations are plotted as a func tion of source plate concentration, and linear fit paramet ers are shown. ................................................................ 125 5-8 Dendritic cell (DC) adhesion is restrict ed to adhesive islands, and DCs are colocalized with printed microparticles (MPs) on isolated islands against a nonfouling PEG-based background. Inter-island DC migration is absent for up to at least 24 h. A.) Dendritic cells were seeded onto arrays of printed rhodamine-encapsulated MPs. Fluoresc ence micrograph overlay of nuclear staining, shown in blue, and MPs, shown in red, is shown (scale bar = 500 m). B.) Representative phase-contrast mi crograph of DCs cultured 24 hr on a 12 x 12 array, demonstrating that DCs are restricted to adhesive islands and are adherent and moderately spread on adhesiv e islands. ....................... 128 6-1 Dendritic cells can uptake MPs encapsulated with CyPher5E dye and fluoresce in the Cy5 filter thus, suggesting that the particles have been phagocytosed. Here green-cytosol, blue-parti cles, pink-phagocytosed MPs. ........................... 132 6-2 Dendritic cells selectively adhere to NH2-terminated spots (presenting ads orbed MPs) and do not adhere to the PEG ba ckground. Quantitation is obtained through image analysis (e.g., 150 MPs : 50 DCs per spot). Physi-sorbed particles are able to lifted from the substrate (seen in SEM image) and are phagocytosable (seen in DIC image and pH-s ensitive dye). Phagocytosis is quantifiable using MP loading of pH-sensitive dye, CyPHer5E dye. MP to cell ratio can be optimized so all MPs are taken up in 16 h. ................................... 132 6-3 As proof-of-principle, DCs were arra yed, co-localiz ed with MPs loaded with an increasing dose of activating factor (L PS), with equal number of particles and cells per spot. Dendritic cells were incubated 4 hrs and then immuno-stained for markers of activation, MHC-II & CD86. Quantification reveals over a 4-fold increase compared to unloaded MPs for the highest LPS dose. Furthermore, after 24 h, cytokine production of IL-10 & IL-12 was quantified in situ through golgi-stop treatment, followed by immuno-staining for intracellular cytokines. Comparison of pooled IL10 data with 3 different randomized DC arrays revealed the lack of positi onal dependence, suggesting limited crosstalk across islands. The symbols represents that the condition is significantly different from si milar sym bols. .......................................................................... 133 6-4 MPs (10:1 ratio MPs : DCs) with factors surfacetethered were incubated 24 h on MP/DC arrays and analyzed for production of IL-10, an antiinflammatory cytokine. Notably, MPs with tethered TGF1 and RGD peptide induced elevated DC production of IL-10 compar ed to other factors (peptides: 4N1K,

PAGE 15

15 P2, P-D2, VIP), and a 2-fold increas e over untreated MPs (PLGA-blank). Two-component mixtures of (1:1) MP formulations (surface-tethered: P2, PD2, 4N1K, CS-1, RGD; encapsulated: VIP, IL-4, poly I:C; surface-adsorbed: PEG-Pluronic) were investigated on MP/ DC arrays for the ability to induce IL10 cytokin e. ...................................................................................................... 134 7-1 Particles were generated using a wate r-oil-water bas ed emul sificationdiffusion method in a multi-well polystyrene plate with multiple fluorescent dyes as representative drug formulat ions. ..................................................................... 139 7-2 Combination of fluorescent dyes at diff erent dilutions can be printed in the same well accurately. The coefficient of determination was obt ained by simple regression analysis with the decreasing concentration of dye in the source plate: R2 Coumarin = 0.99; R2 Rhodamine = 0.99; R2 Cyanine = 0.94. ................................ 141 7-3 Combination of fluorescent dyes at different dilutions can be encapsulat ed in PLGA particles with uniform distribut ion of dyes within the individual population of particles. ...................................................................................... 142 7-4 Particles were produced via parallel particle production method and the particles were analy zed using SEM and dynamic li ght scattering. A typical size distribution curve of PLGA-particles quantified via dynamic light scattering analysis (based on volume estimation) demonstrates an average size of particles of 1.04 m. .......................................................................................... 143 7-5 The fluorescence intensities of t he particles generated via PPP method can be visualized by quick scan imaging. The particles encaps ulating 3 different fluorescent dyes in 6 dilutions, thus forming 216 different combinations, were printed onto a glass slide and scanned using typhoon 9410 and fluorescent microscope. This method is useful for image quick scan and visualization of the amount encapsulated factors. ..................................................................... 144

PAGE 16

16 Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy BIOMATERIAL-BASED MODULATION OF DENDRITIC CELLS: ADHESION BASED MODULATION AND HIGHTHROUGHPUT PARTICLE VACCINE GENERATION, EVALUATION AND DELIVERY By Abhinav P. Acharya May 2010 Chair: Benjamin Keselowsky Cochair: Laurie Gower Major: Materials Science and Engineering Modulation of immune-cell responses usi ng biomaterials based cues is an exciting field of research that holds great potentia l to help solve disorders such as autoimmune diseases, cancer and infections. Medical devices used in numerous applications such as tissue-engineered constructs, combinatio n products (e.g., drug-eluting stents) and therapeutic vaccines are excellent tools for modulating the immune system. Since dendritic cells (DCs) are the key regulators of the immune system, DCs can be targeted to generate a desired immune response. Dendritic cell functions can be altered using several different signals such as adhesion signaling, targeting extracellular/intracellular toll like receptors for activation and mechanica l cues. Specifically, we are interested in modulating DC behavior using adhesion cues and delivery of agents that target extracellular and intrac ellular receptors. Exogenously generated DCs have been sug gested as a potential solution to diseases such as type 1 diabetes (T1D). Ex vivo expansion of cells require isolation from the patient and culturing them on tissue culture treated plates where they come in

PAGE 17

17 contact with several different proteins. Adhesion of such cells on different extracellular matrix proteins can modulate cellular responses.While it is well-known that adsorbed proteins on biomaterials modulate inflammatory responses in vivo, modulation of dendritic cells (DCs), a key regulator of immune system, via adhesion-dependent signaling has only been begun to be c haracterized. Currently, DC-based immunotherapy approaches for diseases such as cancer and autoimmune diseases like type-I diabetes rely on ex vivo culture and expansion of patientderived DCs onto tissue culture-treated polystyrene plates. The adhesive substrate provided for DCs in this ex vivo approach is typically tissue culture-treated polystyrene presen ting serum proteins adsorbed from the culture medi a. We therefore chose to examine serum-coated tissue culture-treated polystyrene as a relevant benc hmark to compare the effect of adhesiondependent modulation of DC function when cult ured on several of the extracellular matrix proteins. Furthermore, DCs isolated from non-obese diabetic mice and its background control of wild type mice were cu ltured on the extracellular matrix proteins and compared for optimal protein substrate for generating DC based vaccines. In addition to modulating DCs ex vivo, DCs can be targeted in vivo using particle-based vaccines. Currently there are scores of known antigenic epitopes and adjuvants, and numerous synthetic delivery systems accessible for formulation of vaccines. However, the lack of an efficient means to test immune cell responses to the abundant combinations available represents a significant blockade on the development of new vaccines. In order to overcome this barrier we report fabrication of a new class of microarray consisting of antigen/adjuvant-loadable poly(D,L lactide-co-glycolide) microparticles (PLGA MPs), identified as a promising carrier for immunotherapeutics,

PAGE 18

18 that are cultured with DCs. Furthermore, a technique was generated to manufacture scores of particle-based vaccines in a highthroughput manner. The intention is to utilize this high-throughput platform to optimize particle-based vaccines designed to target DCs in vivo for immune system-related di sorders, such as autoimmune diseases, cancer and infection.

PAGE 19

19 CHAPTER 1 INTRODUCTION Active interaction of immune system with foreign body such as biomaterials provides an opportunity to rationally design medical devices incorporating synthetic and biological components that c an generate desired immune responses. Dendritic cells (DCs) are the most efficient antigen presenting cells of the immune system that can generate an effective immune response. Hence targeting DCs to modulate the immune system is an attractive strategy. Implant able biomaterials, particle based targeting devices and biomaterials used to culture DCs are the availabl e tools that can be modified to modulate DC-functions. Subsequently, these modified DCs can generate an effective and desired immune response. We ar e interested in generating biomaterials influenced vaccines live DC vaccines and sy nthetic particle based vaccines targeting autoimmune diseases, cancer and infection. Immunogenomic approaches, functional insight into pattern recognition receptors and progress in tolerance-inducing strategi es, have aided in the rational design of vaccine strategies targeting antigen presenti ng cells, and in particular, dendritic cells (DCs) for different immune related disorders [1-6]. Specifically, type 1 diabetes (T1D) is an autoimmune disease characterized by Tcell mediated destruction of insulinproducing -cells. The prevalence of T1D in US children is 1.7 to 2.5 cases per 1000 individuals, and the incidence is between 15 and 17 per 100,000/year. In the United States, 10,000 to 15,000 new cases of T1D are diagnosed each year. The associated cost of diabetes treatment and care in the US in 2002 was estimated at $132 billion [7]. Clinical trials are underway for generating immunotherapeutic solutions to T1D [8]. Notably, clinical safety studies are being pursued for the use of antisense

PAGE 20

20 oligonucleotide-treated DCs dow n-regulating co-stimulatory molecules which have been shown to confer diabetes protection in NOD mice [9]. In this body of work, a systematic analysis of the interaction of exogenously generated DCs with several extracellular matrix proteins and their derivative was per formed. The motivation for this research stems from the concept that adhesion of DCs to extracellular matrix proteins and their derivatives might induce modulation in their functions. There are several limitations to cell-based vaccines, for e.g. dissemination of exogenously delivered DCs is inefficient and treatment involves isolation and storage of DCs over a period of time, which amounts to high treatment costs that will prohibit widespread application. An alternative strate gy involves the development of a synthetic particle-encapsulated vaccine, or vaccine parti cle that can be easily administered with delivery of both prime & boost doses using time-release mate rials (in particular, poly lactide-co-glycolide, PLGA) [10-12]. Fu rthermore, a technique was developed to generate particle-based vaccines in a hight hroughput format and characterize their effects on DC-functions using in situ high fi delity highthroughput microarrays. This work leverages the versatility and customizability of synthetic microparticles as vaccine carriers and develops new strategies for the high-throughput production and immunologic assessment of combinatorially-loaded and surface-modified synthetic vaccine particles. Adhesion-Dependent M odulation of DC-Function Interactions of DCs with biomaterials have been demonstrat ed to modulate DC functions. Ex vivo culture and expansion of DCs is an imm unotherapeutic approach being pursued for treating diseases such as type 1 diabetes ( Figure 1-1 ). Hence, it is important to study the modulation of DC-function before generation of DC-based

PAGE 21

21 vaccines. 1. Immune cell isolation/expansion 2. Disease-specific antigen presentation & maturation on engineered adhesive substrates 3. Autologous immune cells delivered as a vaccine against disease Figure 1-1. Schematic of immunotherapy approach with the introduction of engineered adhesive substrates to direct cell maturation that can be potential used for Diabetes Type I. Activation of Dendritic Cells upon Adhesion Type 1 Diabetes vs. Wild Type Upon implantation, numerous proteins are quickly adsorbed onto biomaterials, including extracellular matrix proteins. Some of these (e .g., fibronectin, fibrinogen, vitronectin) have been shown to modulate inflammatory responses. DC adhesion to extracellular matrix proteins is therefore an important c onsideration in biomaterials. Additionally, DC adhesion to extracellular matrix proteins is of interest physiologically because DCs reside for much of their lifetime in connective tissues comprised largely of extracellular matrix proteins in both lymphoid and non-lymphoid organs, which may influence immune responses in the wake of injury, disease or tissue transplantation. However, despite its significance, modul ation of innate and adaptive immune responses by DCs upon adhesion to extracellular ma trix proteins has only been begun to be characterized. DCs are critical for both immunity and tolerance and are involved in guiding innate and adaptive immune responses [13,14]. Dendritic cells act as sentinels,

PAGE 22

22 constantly patrolling the body and presenting both self and nonself antigens to lymphocytes such as B-cells and T-cells [ 15,16]. Immature DCs (iDCs) mature/activate following interaction with pathogen associated molecular patterns or danger signals as well as self molecules (e.g. uric acid) [17-19]. DCs subsequently up-regulate antigen presenting molecules, co-stimulatory mole cules, cytokines, and chemokine receptors. The latter mediate migration to secondary ly mphoid tissues where they initiate adaptive immune responses and direct the development of T cell responses. DCs are the principal antigen-presenting cell involved in ac tivation of nave T-cells, as they provide three requisite signals: antigen presented in the context of major histocompatibility molecules (MHC), co-stimulatory molecu les (e.g., CD80, CD86 and CD83), cytokines (e.g., interleukin-12; IL-12) and other factor s that direct T cell functional development. Through these factors, DCs direct the different iation of T-cells into different functional groups: interferon(IFN) producing effector Th1 cells IL-4/IL-5 producing effector Th2 cells, TGFproducing T-cells, and Th17 cells [20]. However, if DCs coming in contact with antigens remain in a resting or quiescent state, they are thought to promote tolerance through induction of regulatory T cells including CD4+/CD25+/FoxP3+ Treg and IL-10 producing Tr1 subtypes [21]. In addition, DCs activate natural killer (NK) cell and invariant NK T-cell responses as well as B-cell responses [22]. Dysregulation of DC function, therefore, can have enormous consequence and a role for DCs has been implicated in numerous pathologies such as type 1 diabetes, atherosclerosis, allergy and graft versus host disease [23,24]. Fo llowing trans-endothelia l migration, DCs interact with tissue-specific extracellular ma trix proteins present in connective tissues. The integrin family of cell-surface receptors is the primary rec eptor responsible for

PAGE 23

23 mediating adhesion to extracellular matrix proteins, which has been shown to modulate numerous cell functions including proliferation and different iation [25]. While it has been shown that DCs express multiple integrins, t here are surprisingly few investigations into the effects of integrin bind ing to extracellular matrix proteins on DC maturation. Interventional immunotherapies and tiss ue engineering constr ucts are being investigated as strategies to alleviate and/or ameliorate symptoms of type 1 diabetes (T1D). Tissue engineered constructs may incorporate scaffold components of both synthetic and biological origin, for e.g. extrac ellular matrix (ECM) proteins. Furthermore, upon implantation, numerous adhesive proteins adsorb onto synthetic constructs (e.g., vitronectin, fibrinogen, fibronectin) [26-31]. Adsorption of such proteins onto biomaterials might generate an unwanted immune response. The culture of DCs onto different ECM proteins might differentially modulate adaptive immune responses [32],[33]. Hence, it is important to investigate ECM proteins-mediated modulation of immune responses via DCs isolated from a type 1 diabetes mouse model. Furthermore, interventional immunotherapies involve modulating DCs in vitro, where culture of ex vivo expanded DCs is performed in the presence of serum proteins which absorb onto culture substr ates [34,35]. There is precedent that DCs present in a diabetic pathology themselves might be ma lfunctioning [36-39]. It has been reported that the non-obese diabetic (NOD) mice derived DCs have a defective maturation [40]. These defective DCs can potentially affect immune responses in the diabetic patient to tissue engineered constructs. Interestingly, EC M-protein fibronectin (FN) is up-regulated in targeted organs of diabetic angiopathy [41]. Additionally, it has been demonstrated that muscle capillary basement membrane containing numerous proteins namely,

PAGE 24

24 laminin (LN) and collagen (COL) is abnormally enlarged in diabetic and pre-diabetic patients [42,43]. Thus, the up-regulated EC M-proteins might cause an increased migration and/or retention of antigen present ing cells in inflammatory-prone tissues. Furthermore, the adhesion of leukocytes to ECM-proteins that are up-regulated in prediabetic patients might initiate and propagate pro-inflammatory immune responses. Additionally, there is a potential for pathology-associat ed altered ECM-protein production in specific tissues to prov ide adhesion cues which may exacerbate pathogenesis. Dendritic cells can direct either pro-inflammatory or tolerogenic immune response [44] and defects in these functi ons may be linked to autoimmune disorders like T1D. Interestingly, adoptively transfe rred DCs presenting auto-reactive antigens induced acute autoimmune diabetes, thus suggesting a dominant role of DCs in inducing and propagating the di sease [44,45]. Notably, immunotherapies such as selective activation of Th2 cell subset by DCs is one of the mechanisms that might induce antigen-specific toleranc e, other mechanism that are worth investigating include DC-mediated induction of regulatory T-cells and IL-10 producing Tr1 cells, T-cell anergy and promotion of antigen-specific T-cell apoptosis [46,47]. Despite the growing interest in DC-based immunotherapies for T1D, ther e are very few studies to understand the effect of adhesive substrates on DC-matu ration and adaptive i mmune responses. In this work, we studied the effect of se veral adhesive substrates on activation and maturation of DCs derived from non-obese di abetic (NOD) mice and ability of such adhesive substrates to modulate an adaptive immune response via DCs. We are the first to report modulation of DC-responses to culture on adhesive substrates where the DCs were isolated from an animal predisposed toward T1D.

PAGE 25

25 Integrin-Peptide Based Controlled Activation of Dendritic Cells Growing use of biomaterials as targeted therapeutics and tissue constructs has generated the need to understand bi omaterials induced imm une responses and utilize the potential of biomaterials to modul ate the immune system. Integrins are transmembrane cell-adhesion receptors comprised of two distinct subunits called (alpha) and (beta) with small cytoplasmic domains [48]. Several integrins are present on DCs that are involved in various cellular functions. Integrins can recognize different peptide sequences, for example, 8 of the known 24 heterodimeric integrins can recognize RGD peptide sequence and severa l of these are expressed by DCs. Peptides are short amino-acid polymeric chains present in the protei ns providing special adhesion sites [49]. A short-peptide GRGDSPC (RGD) (glycine arginine glycine asparatic acid serine praline cystei n) has been investigated extensively as an adhesion molecule for modulating cell function. The V subunit of the in tegrins present on the cell surface has been shown to be co-localized with the RGD-peptide [50]. The RGD-peptide is present in severa l of the extracellular matrix proteins such as fibronectin and fibrinogen [51]. It has been demonstrated that modulation in cell-function is dependent on surface density of RGD [52]. Several studies have been done to modify the surface of a substrate to generate RGD-peptide gradient s [53]. In this study, RGD peptide gradient developed by Gallant et al were utilized to quantify the adhesion based activation of DCs, via major histocom patibility complexII (MHC-II), CD86 cell surface molecule expression and intracellula r IL (interleukin) -10 and IL-12p40 cytokine production. Furthermore, DC-expression of V integrin was quantified. The DC surface molecules, MHC-II a stimulatory molecule th at interacts with T-cell receptors and CD86

PAGE 26

26 a co-stimulatory molecule that interacts with CD28 molecules on T-cells along with the soluble cytokines are the three signals required for directing T-cell-functions. Additionally, the level of expression of these signals generated by DCs, determine the type of immune response i.e. anti-inflammato ry or pro-inflammatory. Furthermore, different stages of physiologically relev ant DC-activation stages of immature, semimature and matured state have been observed. Hence, it is interesting to study the level of activation of DCs upon controlled presentat ion of such adhesive ligands. This study will help understand the extent of adhesion signaling required via biomaterials to generate effective immune responses and potent ially modulate function of ex vivo cultured DCs in a highly controlled manner. High-Throughput Production and Biological Evaluation of Antigen Presenting Cell-Directed Vaccine Particles Advances in vaccine technologies promise solutions to some of todays most pressing medical problems incl uding the induction of immune tolerance for applications toward autoimmune disease and organ transpl antation. An attractive approach in vaccine technology involves the developmen t of a synthetic particle-encapsulated vaccine, or vaccine particle that can be easily administered with simultaneous delivery of both prime & boost doses using time-rel ease materials. This approach greatly simplifies issues related to manufactu ring, storage and shipping, as biomaterial encapsulation provides vaccine stability and improved shelf-life. Furthermore, vaccine particles can be engineered to be multifunctional and modular. Features of particular interest are: control over phagocytosability, targeting to dendritic cells (DCs; a prime target cell for vaccines), and providing a depot for antigens, adjuvants, immunosuppressants, chemokines and growth factors. This system can thereby be

PAGE 27

27 designed to attract DCs and precursors into a vaccination site, provide signals to drive differentiation into tolerogenic DCs, prom ote uptake of antigen and induce specific tolerance. However, the problem lies in t he fact that although there are now scores of known antigenic epitopes and adjuvants, there has not em erged a systematic examination of the f unctional responses of immune cells in a combinatorial, highthroughput manner. The lack of an efficient means to produce and test numerous combinations of potential components represents a significant blockade on the development of new vaccines. In order to overcome this barrier, we are developing novel core technologies for high-throughput microparticle synthesis and evaluation. High-throughput Microparticle Microarray fo r Dendritic Cell Targeted Vaccines Immunotherapeutic strategies utilizing biomaterials involve modulation of immune responses by targeted delivery of immuno-m odulatory molecules to leukocytes via a synthetic carrier. Microparti cles (MPs), nanoparticles, micelles, vesicles, dendrimers and microchips have been all investigated as drug delivery vehicles designed to effect immunomodulation [54-59]. Of these, MPs fabricated using poly (d,l lactide-co-glycolide) (PLGA) have been the most investigated vehicle for delivering immunotherapeutics [60,61]. Poly (d,l lactide-co-glycolide) (PLGA), approved by the U.S. Food and Drug Administration (FDA) for biodegradable surgical sutures and drug delivery products, is degraded in the body via bulk erosion and hydrol ysis. By altering the lactide/glycolide ratio, PLGA MPs can be designed to prov ide an initial burst of the encapsulated immuno-modulatory molecule followed by su stained release, permitting the design of a one-time drug administration with prime and boost doses [62]. Furthermore, PLGA MPs of appropriate size are phagocytosed efficien tly by antigen presenting cells providing direct delivery of antigens for immune recogni tion [63]. Critically, following phagocytosis

PAGE 28

28 by antigen-presenting cells, phagolysomal re lease of encapsulated antigens from PLGA microparticles can generate bot h MHC-II-directed, as well as MHC-I-directed immune response through cross-presentation [64,65]. Po ly (d,l lactide-co-glycolide) MPs are therefore an excellent candidate as a carri er vehicle for vaccines utilizing encapsulated antigenic proteins or pept ides along with immunomodulatory molecules such as adjuvants. Furthermore, PLGA MPs can be surface-modified to modulate activation, uptake and targeting of a key subset of antigen presenting cell, the dendritic cell. Dendritic cells (DCs) are the most effici ent antigen presenting cell [66-68]. Moreover, DCs are central regulators of the imm une system, processing and presenting antigen along with expression of an array of molecule s (i.e., stimulatory, co-stimulatory and cytokines), directing T-cell subsets (e.g., Th1, Th2, Treg) providing either antigenspecific tolerance or immunity [69,70]. Therefore, modulation of these cells is critical, and a number of groups are investigati ng vaccines consisting of antigen-loaded particles targeting DCs [71-74]. Several established adjuvants exist and num erous more are under investigation. For example, molecules binding to pattern recognition receptors, such as toll-like receptors (TLRs) and c-type lectins, stimul ate DC-activation [75]. Recently, Scholosser et al. demonstrated that the co-encapsulation of antigens and TLR 9 ligand, CpG oligonucleotide, or TLR 3 ligand, polyI:C encapsulated in PLGA MPs provided DC activation and generated potent cytotoxic Tlymphocyte responses [76]. Similarly, Elamanchili et al. reported that PLGA nanopa rticles encapsulated with TLR 4 ligand, mono-phosphoryl lipid A, resu lted in increased expression of stimulatory and costimulatory molecules on DCs and result ed in a robust Th1 type response [77,78].

PAGE 29

29 Additionally, antisense oligonucle otides specific for either co-stimulatory molecules or IL-10, loaded into PLGA MPs have been shown to be able to either induce immune suppression or direct specific Th-1 helper-t ype response respectively [79,80]. Finally, in addition to directly loading antigenic proteins or peptides into particles, it has been demonstrated that delivery of DNA encoding for antigenic protein via PLGA MPs is a viable immunotherapeutic option [81]. Whil e results have been promising, translation into new effective vaccines has stalled. We believe the problem lies in the fact that although there are now scores of known antigenic epitopes and adjuv ants, there has not emerged a platform for the systematic examin ation of immune cell responses in a highthroughput manner. The lack of an efficient means to test numerous combinations represents a significant blockade on the dev elopment of new vaccines. In order to overcome this barrier, we set out to devel op a system to efficiently evaluate a large number of multi-parameter combinations of particle-based vaccine formulations simultaneously. Miniarraying technology has revolutionized the fields of genomics and proteomics. The success of high-throughput arrays of DNA and other biomolecules fabricated on substrates through contact pin printing has led the way to the recent development of cell-based arrays [82]. For instance, arrays have been constructed consisting of various types of stem ce lls adherent on combinations of printed extracellular matrix proteins, cell adhesion molecules and growth factors, in order to systematically investigate the effects of molecular microenvironment on stem cell differentiation [83-85]. Miniaturization would have the added benefit of requiring minimal patient sample, allowing the use of patient-isolated DCs to be screened for responses to thousands of

PAGE 30

30 DC-targeting MP-vaccine formulations on a single chip. Patient-optimized MP-vaccine formulations could then be administered to the patient. This work describes construction of a new class of microarray arrays of co-localized particles and cells. Utilizing standard miniarraying equipment in conjuncti on with surface chemistry derivatization techniques, we have generated arrays of PL GA MPs co-localized with DCs onto adhesive islands against a non-fo uling background. Our intent ion is to employ this platform as a high-throughput technique to test multi-parameter combinations of microparticle-based vaccines targeting DCs. Dendritic Cell Arrays for Biological Evaluation of Vaccine Particles The large number of antigen/adjuvant stra tegies available in the design of new vaccines, selection of an optimal anti gen/adjuvant approach has become a challenge further complicated by patient-specific responses. Combinatorial approaches of delivering multiple immuno-modul atory signals to improve the vaccine efficacy will likely be required to overcome this problem. Therefore we have developed a system to efficiently evaluate a large number of multi-parameter combinati ons of particle-based vaccine formulations simultaneously. Th is high-throughput small-volume method consequently provides multiple advantages ov er traditional trial-an d-error methods to test efficacy and effectiveness of vaccine fo rmulations. Additionally, with a minimal patient sample required, optimization through a small-volume high-throughput screening technique may lead toward developing personalized vaccines. Micro-arraying technology has revolutionized the fields of genomics and prot eomics [85-87]. The success of these high-throughput arrays of co mplimentary DNA and other biomolecules on glass using contact pin printing and analyz ing them by fluorescent probes, has led the way to the recent development of cell-based arrays. For instance, stem cell arrays

PAGE 31

31 have been constructed by culturing various types of stem cells adherent on different combinations of printed ex tracellular matrix proteins cell adhesion molecules and morphogens/growth factors, in order to syst ematically investigate the effects of molecular microenvironment on stem cell differ entiation [88]. In this work, we describe the construction of a co-localized particle/cell array. Ut ilizing standard microarraying equipment in conjunction with surface chem istry derivatization techniques, we have generated microarrays of PLGA MPs co-loc alized with DCs onto adhesive islands against a non-fouling background. Our intention is to employ this platform as a highthroughput technique to test multi-parameter combinations of microparticle-based vaccines for immunotherapies targeting DCs. Parallel Vaccine Particle Production Currently, standard methodology for the generation of polymeric particles consists of single-batch processing using the doubleemulsion/solvent-evaporation method. This process can take 4-5 hours and skilled hands used to handling multiple samples may be limited to producing less than a dozen MP form ulations in one day. As a result of our development of high-throughput approaches fo r the design and optimization of particlebased vaccines, we have identified a unique, critical need to generate large numbers (hundreds to thousands) of diff erent MPs formulated with mu lti-parameter combinations of immunomodulatory molecules. To meet this need, we hav e developed a parallel particle production technology, utilizing solid-pin miniarraying equipm ent for the robotic loading of pre-particle solutions with combinatorial formulations into 384-well plate wells as loading/particle-ge neration chambers.

PAGE 32

32 CHAPTER 2 ADHESIVE SUBSTRATE-MODULATION OF ADAPTIVE IMMUNE RESPONSES IN C57BL6/J MICE Introduction Dendritic Cells (DC) are key regulator s of the innate and adaptive immune system hence, modulation of DC responses resulting from interactions with biomaterials is critical. Additionally, interactions of DCs with biomaterials have been demonstrated to modulate DC functions. Since, C57BL6/j (B-6) mice represent the wild type mice; it is physiologically relevant to assess the immune response from DCs isolated from B-6 mice. Dendritic cells may interact with the proteins adsorbed onto the implanted biomaterials and generate an adaptive immune response. Furthermore, DCs reside in connective tissues and interact with the extr acellular matrix for t he majority of their lifetime. Despite its significance, modul ation of innate and adaptive immune responses by DCs upon adhesion to extracellular ma trix proteins has only been begun to be characterized. In the present study, DCs we re cultured on extracellular matrix proteins and substrate-mediated modulation of DC maturation and DC-directed adaptive immune responses (T-cell proliferation and T-helper responses) were quantified. Generation of Murine Bone Marrow-Derived DCs Immature bone marrow-deriv ed DCs were generated from 7-week-old female C57BL6/j mice in accordance with protoc ol approved by the University of Florida (protocol number E751) using a modified 10-day protocol. Briefly, femur and tibia from mice were isolated and kept in wash m edia composed of DMEM/F-12 (1:1) with Lglutamine (Cellgro, Herndon, VA) and 10% fetal bovine serum (Bio-Whittaker). The ends of the bones were cut and bone marrow was flushed out with 10 ml wash media using a 25 G needle and mixed to make a homogeneous suspension. The suspension

PAGE 33

33 was then strained using 70 m cell strainers (Becton Dickinson) and cells were collected by centrifugation at 330xg for 6 min. Precursor cells were isolated by centrifuging NycoPrep gradient (10 ml) and cell suspension (25 ml) at 670xg for 20 min at 22 0C. Leukocytes were isolated by pipetting out the layer of cells that forms at the interface of wash media and gradient. The precursor cell suspension was then washed twice with wash media and re-suspended in DMEM/F-12 with L-glutamine (Cellgro, Herndon, VA), 10% fetal bovine serum, 1% s odium pyruvate (Lonza, Walkersville, MD), 1% non-essential amino acids (Lonza, Walk ersville, MD), 1% penicillinstreptomycin (Hyclone) and 20 ng/ml GM-CSF (R&D systems) (DC media). This cell suspension was then seeded in a tissue culture treated T-flask (day 0). After 48 h (day 2), floating cells were collected, re-suspended in fresh media and seeded on low attachment plates for 6 additional days. Half of the media was c hanged every alternate day. At the end of 6 days (day 8), cells were lifted from the lo w attachment wells by gentle pipetting, resuspended and seeded on tissue culture-treat ed polystyrene plates for 2 more days. Cells were then lifted (at day 10) using 5 mM Na2EDTA solution in phosphate-buffered saline (PBS, Hyclone) and used for all the ex periments. Purity, yield and immaturity of DCs (CD11c+ and MHC-II) were verified via imm unofluorescence staining and flow cytometry, whereas cell viab ility (>99% viable) was det ermined using Trypan blue. Marrow derived DC stimulatory capacity in terms of up-regulation of cell-surface markers MHC-II, CD80 and CD86, when cultured in the presence of LPS, was verified in comparison to immature DCs. Dendritic ce lls were isolated from at least 3 separate mice for each type of experiment.

PAGE 34

34 Isolation of T-Cells Spleens were isolated from 6-w eek-old BALB/cbyj mice. Single cell suspensions were prepared by mincing the sp leen through a cell strainer. The effluent was centrifuged for 10 min at 300xg. This su spension was then strained again using cell strainer to separate debris and cells were counted using a hemoc ytometer. The cells were then spun down at 300xg for 10 min an d the pellet was re-suspended in 4 ml of buffer (0.5% BSA and 2 mM EDTA in PBS) per million cells. Negative selection of CD4+ T-cells was performed. A biotin-labeled antib ody cocktail (CD8a (Ly-2) (rat IgG2a), CD11b (Mac-1) (rat IgG2b), CD45R (B220) (rat IgG2a), DX5 (rat IgM) and Ter-119 (rat IgG2b); Miltenyi) was added (10 ml per 10 million cells) and incubated for 10 min at 40C. Buffer (30 l) and anti-biotin microbeads (20 l) were added to the mixture per 10 million cells. After 15 min incubation at 4 0C, cells were centrifuged at 300xg for 10 min and re-suspended in 500 l of buffer per 100 million cells. A MiniMACS magnetic column was pre-washed with 500 l of buffer solution. Cell suspension was added to the column and the effluent comprised of CD4+ T-cells was collected. The column was then washed thrice with buffer soluti on and the effluents were mixed. The CD4+ T-cells were centrifuged at 300xg for 10 min and used in mixed lymphocyte reaction. Protein Coating and DC Culture Extracellular matrix proteins were coated onto 12-well tissue cult ure-treated polystyrene plates by overnight incubation of 20 g/ml protein solution in PBS ( Figure 2-1). For these single-component coating conditions, substrates are expected to be fully saturated with respect to protein surface densities. The wells were then washed with 1 M PBS with calcium and magnesium. Immature DCs were seeded (1 x 106 cells/well)

PAGE 35

35 with or without lipopolysaccharide (1 mg/m l LPS; maturation signal) on the following protein-coated substrates: human plasmaderived fibronectin (FN) (BD Bioscience), EngelbrethHolmSwarm mous e tumor-derived laminin (LN) (BD Bioscience), bovine dermis-derived collagen type I (COL) (BD Bioscience), human plasma-derived vitronectin (VN) (BD Bioscience) and bovine plasma fibrinogen (FG) (Mp Biomedicals). Bovine serum albumin (BSA) (Fisher Bi oreagents) and fetal bovine serum (SER) (Hyclone) protein-coated substrates were included as reference substrates. Speciesspecific protein sequence homologies, as co mpared to murine, are as follows: FN 92%, COL 89%, VN 76%, FG 81% and BSA 70%; determined by HomoloGene, an online resource made available through the National Center for Biotechnology Information. Note that t he SER substrate represents a standard culture condition, as quite often, DCs are cultured on tissue-culture polystyrene which has no pre-coated protein, but which allows uncontrolled protein adsorption from the serum-containing culture medium. One million DCs were cultured on each substrate for 24 h, supernatants were collected for cytokine ana lysis and cells were lifted using Na2EDTA and immunofluorescently stained for maturation markers. Dendritic cells used for positive (+LPS) and negative (iDCs) controls re mained in the plates in which they were generated and were not reseeded onto new plates Dendritic cell viability was tested using Trypan blue staining. Phase-contrast microscopy images were taken at 100X magnification. In order to assess the amount of endotoxin present on the adsorbed substrates, the chromo-Limulus Amebocyte Lysate (chromo-LAL) assay was performed as per the manufacturers instructions (C ape Cod) using a 50 ml reaction volume per substrate with a 20 min incubation time.

PAGE 36

36 Figure 2-1. Schematic of adsorption of proteins on tissue culture treated polystyrene surfaces and culturing DCs on the modified surfaces. Dendritic Cell Adhesion and Proliferation Dendritic cells were seeded on adhesive substr ates for 2 h. Dendritic cells were then washed once with 1x PBS, fixed with 3. 7% paraformaldehyde (USB Corp.) and nuclei were stained with Hoechst (Invitrogen) according to the manufacturers directions. The number of DCs was counted using Axiovi sion software 4.6.3 (Carl Zeiss Imaging Solution). Immature DCs and DCs +LPS were in cluded as controls, where cells were replated onto tissue culture polystyrene without pre-coating. Additionally, DC proliferation on adhesive substrates was quantified. Dend ritic cells were cultured on protein substrates for 24 h, and bromodeoxyuridine (B rdU) (kit from Beckton Dickinson) was added to the culture for the last 16 h. Dendr itic cells were then immunofluorescently stained for BrdU according to manufacturer s specifications. Fluorescence output was quantified in a fluorescence plate reader (PerkinElmer). Quantification of DC Surface Maturation Markers Dendritic cell maturation was quantified by measuring cell-surface marker levels by flow cytometry. Briefly, DCs were lifted by incubating with 5 mM Na2EDTA solution in 1 M PBS solution at 37 0C for 20 min. Dendritic cells were then washed with 1% fetal bovine serum in PBS and incubated with antibodies against CD16/CD32 (Fcg III/II Receptor) (clone 2.4G2, IgG2b,k (BD Pharmingen)) for 40 min at 40C to block Fcg receptors on DCs. Cells were washed and then stained with antibodies against CD80 Protein coated tissue culture treated polystyrene plates Dendritic Cell Seeding

PAGE 37

37 (clone 16-10A1, IgG2, k), CD86 (clone GL1, IgG2a, k), I-A/I-E (clone M5/114.15.2 IgG2b, k), CD11c (clone HL3, IgG1, l2) (BD Pharmingen) for 40 min at 4 0C. Appropriate isotypes were used for each antibody species as negative controls. Data acquisition was performed using (FACScalibur, Bect on Dickinson) flow cytometry and the geometric fluorescent intensities determi ned. More than 20,000 events were acquired for each sample and data analysis was perfo rmed using FCS Express version 3 (De Novo Software, Los Angeles, CA). Quantification of DC Cytokine Production Cell culture supernatants were collected after 24 h of cell culture, centrifuged to remove any cell debris and stored at 20 0C until analysis. The IL-12 cytokine subunit, IL12p40, and IL-10 cytokine production was analyzed using sandwich enzyme-linked immunosorbant assay (ELISA) kits (Becton Dickinson) according to manufacturers directions. Mixed Lymphocyte Reaction Immature C57BL6/j muri ne bone marrow-derived DCs (40,000 cells/well) were cultured on protein-coated (adsorbed overnight, 20 g/ml coating concentration) Ubottom tissue culture-treated 96 well plates for 24 h. CD4+ T-cells purified from spleen of BALB/cbyj mice in the ratio of 1:6 we re then added to the wells and co-cultured with adherent DCs for an additional 48 or 96 h. BrdU (Beckton Dickinson) (final concentration 10 mM) was added to the co-culture along with protein transport inhibitor (0.7 l for every 1 ml of media) (Beckton Dickinson) 5 h before labeling T-cells with CD4 fluorescently tagged antibodies. At t he end of 48 or 96 h, T-cells were immunofluorescently stained fo r BrdU (BD Pharmingen) to quantify proliferation rates.

PAGE 38

38 Additionally, cells were immunofluorescently stained for cell-surface markers CD4 (BD Pharmingen) and intracellula r cytokine staining of IFN(BD Pharmingen) and IL-4 (BD Pharmingen). Flow cytometry was utilized fo r data acquisition of 20,000 cells. Analysis was performed on CD4+ gated cells and was further classified for the presence of IFN, IL-4 and BrdU. T-cells were stimulated wi th phorbol 12-myristate 13-acetate/Ionomycin (10 ng/ml for 2 days) to verify T-cell prol iferation potential (dat a not shown), while a mixed lymphocyte reaction conducted with iDCs without pre-culturing on adhesive substrates comprised the negative control. Statistical Analysis Statistical analyses were performed using general linear nested model ANOVA, linear regression analysis and/ or Pearsons correlation, as appropriate, using Systat (Version 12, Systat Software, Inc., San Jose CA). Pair-wise comparisons were made, with p-values of less than or equal to 0.05 considered to be significant. Results Dendritic Cell Morphology and Adhesion Quantification of endotoxin levels by ch romo-LAL revealed that the substrate preparation yielded negligible endotoxin levels (<0.050 endot oxin units/ml). Immature DCs were seeded on substrates, cultured for 24 h and phase-contrast microscopy images were acquired. Different stages in maturation of pure DC cultures have been described as having characteristic morphologi es. For example, the presence of dendritic processes is widely considered to represent a mature state. Likewise, the formation of clusters of rounded cells, li kely mediated through E-cadher in, has also been attributed to a mature state. Morpholog ical differences in B6-derived DCs were demonstrated to

PAGE 39

39 be modulated by the adhesive substrate ( Figure 2-2). Figure 2-2. Initial adhesion of DCs is stat istically not different for different adhesive substrates with an overall ANOVA p-value of less than 1. Data represents average and standard error of at l east 6 data points (replicates).

PAGE 40

40 Figure 2-3. Murine C57BL6/j dendritic cell (DC) morphology is modulated by adhesive substrate. Inset micrographs represent a typical zoomed-in morphology of DCs on the given substrate. Dendritic cells cultured on COL and LN s ubstrates, and in the presence of LPS, evidenced both dendritic processes and cluste rs of rounded cells. In contrast, DCs cultured on substrates pre-coated with VN formed dendritic processes but not clusters, while DCs cultured on substrat es pre-coated with FG formed clusters but very few wellformed dendritic processes. In contrast, DCs cultured on substrates pre-coated with FN

PAGE 41

41 or SER demonstrated neither the presenc e of dendritic processes nor clusters. Interestingly, DCs cultured on BSA and FG substrates showed fibroblast-like morphology. Overall, t hese data demonstrate differ ential modulation of DC morphologies in a substrate dependent manner. In terms of a tradi tional view of DC morphology, COL, LN, VN and FG substrates potentially support increased levels of DC maturation, whereas DCs cultured on FN and SER demonstrate potentially lower levels of maturation. However, although there has been some tradition in linking DC morphology to maturation state, this is clearly not a sufficient indicator of maturation and further investigation was carried out. Additionally, in order to determine if substrate modulated DC adhesion, we quantified the number of adherent DCs at 2 h ( Figure 2-3) 24 h (data not shown) and found an equiva lent number of DCs adherent to all substrates. Dendritic Cell Phenotype In order to more clearly define subs trate-dependent maturation of DC expression levels of surface molecules (stimulatory : MHC-II; co-stimulatory: CD80, CD86) and secreted cytokines (IL-10, IL-12p40) were quant ified. Collectively, these metrics are descriptive of the extent and quality of matu ration. DC expression of stimulatory and costimulatory molecules was quantified by flow cytometry and data were pooled for statistical analysis. Results were plotted as bar graphs ( Figure 2-4, Figure 2-5 ), with representative density plots included ( Figure 2-6). The percentage of DCs expressing MHC-II molecule (~80% of L PS-stimulated cultur es/positive control) was equivalent across the substrate, indicating a degr ee of maturation on all substrates ( Figure 2-4A all other conditions; all conditions except VN.). A much lower percentage of

PAGE 42

42 iDCs expressed stimulatory molecule MHCII (negative control). Overall significance was determined by ANOVA, p-value < 1x10-10. The trend in significant pairs is summarized by the inequality expressi on: +LPS >LN=COL=FG=FN =SER= BSA=VN> iDCs. These results indicate that all the s ubstrates induced equivalent levels of MHC-II positive DCs, in a substrate-independent manner. It is important to note we found that when iDCs were lifted and re-plated onto clean tissue culture polystyrene, they demonstrated only a small increase in the level of phenotypic maturation at 24 h, compared to non-lifted iDCs (~8% increase in MHC-II positive cells, ~3% increase in CD86 positive cells and no increase in CD80 positive cells), thus indicating that the DC maturation observed is indeed attributable to substrate-dependent signals. Furthermore, the level of MHC-II expression was quantified by calculating the geometric mean fluorescence intensity (gMFI) ( Figure 2-4B all other conditions; SER. ). Dendritic cells cultured on different adhesive substrates expressed equivalent level of MHC-II (2-fold less than posit ive control, +LPS). Overall significance was determined by ANOVA, p-value < 1x10-10. The trend in significant pairs is summarized by the inequality expressions: +LPS >BSA=COL = FG= FN= LN= SER=VN, and SER>iDCs. These results indicate that all the substrates are capable of supporting DC maturation. In order to investigate the combined effe cts of adhesive substrates with a soluble maturation signal, we also quantified DC res ponses in the presence of LPS. With the addition of LPS to the cultur e, DC MHC-II expression levels were found to be slightly elevated for all substrates ( Figure 2-4B). MHC-II levels of LPS-stimulated DCs cultured on BSA, FG and VN was relatively lower as compared to LPS-stimulated DCs cultured on COL, LN and SER. Although LPS-stimulated DCs cultured on FN had higher MHC-II

PAGE 43

43 levels than iDCs, these values were not signifi cantly different from the other substrates. Overall significance was determi ned by ANOVA, p-value < 1 x 10-10. In the presence of LPS, the trend in salient significant pairs is summarized by: SER = LN = COL > BSA = FG = VN > iDCs. These results indicate that MHC-II expression was slightly elevated for DCs cultured on SER, LN and COL substrates in the presence of LPS compared to DCs cultured on BSA, FG and VN substrates. Figure 2-4. Adhesive substrates activate murine C57BL6/j bone marrow-derived dendritic cells (DCs) as evidenced th rough phenotypic surface presentation of major histocompatibility (MHC). Data represent average and standard error of at least 6 data points (replicates). DCs were obtained from at least 3 separate mice and each mouse handled as an independent experiment repeat. A) Dendritic cells positive for su rface expression of MHC-II. B) Dendritic cells surface expression of MHC-II. C) Dendritic cells positive for surface expression of MHC-II when cult ured in the presence of LPS. D) Dendritic cells surface expression of MHC-II when cult ured in the presence of LPS. The significant pair symbols are described in the text.

PAGE 44

44 Similar levels of co-stimulatory surf ace molecule, CD80, were found on DCs cultured on all substrates, comparable to the LPS-stimulated cultur es and significantly higher than iDCs (overall ANOVA, p-value < 1 x 10-10), suggesting a degree of maturation ( Figure 2-4C COL + LPS, LN + LPS, +LPS, iDCs; BSA + LPS, FG + LPS, VN + LPS, iDCs; iDCs. ). DC expression of co-stimulatory surface molecule, CD80, was not remarkably elevated in the presence of LPS ( Figure 2-4D COL + LPS, FN + LPS, LN + LPS, iDCs; SER + LPS, VN + LPS and iDCs; LN + LPS and iDCs; FG + LPS, FN + LPS, SER + LPS, VN + LPS and iDCs; iDCs; ** VN + LPS and iDCs. ) as compared to DCs cultured on proteins alone. However, DCs cultured in the presence of LPS on COL, FN, LN and SER substrates had slightly higher (~10%) expression of CD80 as compared to DCs cultured on BSA and VN (Figure 2-4D). Overall significance was determined by ANOVA, p-value < 1 x 10-10. In the presence of LPS, the trend in salient significant pairs is summarized by: COL = FN = LN = SER > BSA = VN > iDCs. Final ly, the expression of the co-stimulatory surface molecule, CD86, was found to be elevated for DCs cultured in the presence of LPS as compared to DCs cultured on adhesive protein substrates alone ( Figure 2-5A all other conditions; 2.5B all other conditions; 2.5C BSA + LPS, VN + LPS and iDCs; COL + LPS, FN + LPS, +LPS, LN + LPS and iDCs; iDCs; 2.5D

PAGE 45

45 with all other conditions ). Figure 2-5. Adhesive substrates activate murine C57BL6/j bone marrow-derived dendritic cells (DCs) as evidenced through phenotypic surface presentation co-stimulatory molecules. Data re present average and standard error of at least 6 data points (replicates). DCs we re obtained from at least 3 separate mice and each mouse handled as an independent experiment repeat. A) Dendritic cells positive for surface expression of CD86. B) Dendritic cells surface expression of CD86. C) Dendritic cells positive for surface expression of CD86 when cultured in the presence of LPS. D) Dendritic cells surface expression of CD86 when cultured in t he presence of LPS. The significant pair symbols are described in the text. However, statistical analysis of pooled data revealed that the expression of CD86 was statistically equivalent on DCs cultured on all substrates, with or without LPS (data not shown). Collectively, this phenotypic c haracterization indicates that the adhesive substrates examined support DC maturation, but do not give rise to large differences in DC expression of stimulatory (MHC-II) and co -stimulatory (CD80, CD86) molecules either in the presence or absence of LPS stimulation.

PAGE 46

46 Figure 2-6. Representative phen otypic density plots of data summarized in Figs. 3 and 4 for murine C57BL6/j-derived DCs cultured on protein-coated tissue culturetreated polystyrene 24 h. Shown also are iDC purity and immaturity (lack of MHC-II, CD80 and CD86). Dendritic Cell Cytokine Secretion DCs cultured on adhesive substrates demonstrated a differential cytokine production profile of the pro-in flammatory cytokine, IL-12p40 ( Figure 2-7A LN, COL, +LPS, VN and iDCs; ** FN, COL, +LPS, VN and iDCs BSA, FG, FN, LN, SER and iDCs; COL, +LPS and VN; COL, +LPS, VN and iDCs; COL, FG, FN, LN, +LPS, SER and VN. ). DCs cultured on VN and COL coated substrates

PAGE 47

47 secreted high levels (equivalent to positiv e control) of IL12p40, compared to DCs cultured on FN, LN, FG and SER coated subs trates. Notably, DC IL-12p40 secretion on VN and COL showed a greater than 2-fold increase over SER, representing standard culture conditions. The BSA coated substrate elicited intermediate levels of IL-12p40 production, lower than VN, COL, but not di fferent from FG, FN, LN and SER coated substrates. DCs cultured on LN coated substr ates also produced intermediate levels of IL-12p40, lower than COL, VN and positiv e control, and higher than FN coated substrates. Overall significance was determined by ANOVA, p-value < 1 x 10-10. The trend in salient significant pairs is summari zed by: +LPS = COL = VN > FG = FN = SER > iDCs. These results indicate that COL and VN induce the highest production of IL12p40 cytokine from DCs. Interestingly, IL-12p40 cytokine trend corresponds to the presence of either the clustered or dendritic morphologies observed on these substrates, however, DCs cultured on LN did not show this trend. We then quantified IL-12p40 production of DCs cultured on substrates in the presence of LPS. As expected, overall le vels of IL-12p40 production were elevated when cultured with LPS ( Figure 2-7B SER + LPS, VN + LPS, LN + LPS and iDCs; VN + LPS and iDCs; for FN + LPS, VN + LPS, BSA + LPS and iDCs; all the other conditions. ) in comparison to substrates without LPS. LPS-stimulated DCs cultured on VN produced the highest le vel of IL-12p40 (~45% increase over SER+LPS, representing standard positive control culture c onditions), while the other conditions induced moderate levels of IL -12p40 cytokine production. LPS-stimulated DCs cultured on BSA and FN coated substrates produced slightly higher levels of IL12p40 than on SER and LN coated substrates Overall significance was obtained by

PAGE 48

48 ANOVA, p-value < 1 x 10-10. In the presence of LPS, the tr end in salient significant pairs is summarized by: VN > BSA = FN > SER = LN > iDCs. These results emphasize that adhesive substrates induce a differential proinflammatory IL-12p40 response in DCs in the presence of LPS, with VN inducing t he elevated production of IL-12p40 cytokine. Adhesive proteins mediated a differentia l DC cytokine producti on profile of antiinflammatory cytokine, IL-10 (Figure 2-7C COL, FG, FN, LN, VN and iDCs. ). IL-10 cytokine production is most pr onounced (equivalent to positive control, roughly 4-fold higher than all other substrates) w hen DCs were cultured on BSA and SER coated substrates. In contrast, levels of IL-10 production by DCs cultured on COL, FG, FN, LN and VN coated substrates were relatively low, equivalent to the negat ive control. Overall significance was obtained by ANOVA, p-value < 1 x 10-10. The trend in significant pairs is summarized by: +LPS = BSA = SER > COL = FG = FN = LN = VN = iDCs. These results indicate that DC culture on SER and BSA substrates results in high production levels of anti-inflammatory cytokine IL -10 compared to other substrates. As expected, overall levels of IL-10 production were elevated when cultured with LPS ( Figure 2-7D COL + LPS, FG + LPS, LN + LPS and iDCs; FG + LPS, SER + LPS, BSA + LPS and iDCs; FN + LPS, VN + LPS, SER + LPS, BSA + LPS and COL + LPS; FG + LPS, LN + LPS and iDCs; FN + LPS, VN + LPS, SER + LPS and BSA + LPS; SER + LPS, BSA + LPS, COL + LPS, FN + LPS and VN + LPS. ) in comparison to substrates without L PS. It is notable that DCs cultured on FG and LN substrates apparently resisted LPS st imulation, in term s of IL-10 cytokine production, with levels that were no higher than the negative control (iDC). LPSstimulated DCs cultured on BSA, SER, FN and VN coated substrates stimulated the

PAGE 49

49 highest levels of IL-10, higher than FG and LN coated substrates LPS-stimulated DCs cultured on COL coated substr ates produced intermediate le vels of IL-10, less than BSA and SER and more than FG coated substr ates. Overall significance was obtained by ANOVA, p-value < 1 x 10-10. The trend in salient significant pairs is summarized by: BSA = FN = SER = VN > FG = LN = iDCs. T hese results indicate substrate-directed modulation in LPS-stimulated DC production of anti-inflammatory IL-10 cytokine. Figure 2-7. Adhesive substrates differentially modulate murine C57BL6/j bone marrowderived dendritic cell (DC) cytokine production. Data represent average and standard error of at least 9 data points (replicates). DCs were obtained from at least 3 separate mice and eac h mouse handled as an independent experiment repeat. A) Dendritic cells producti on of IL-12 upon culture on adhesive substrates. B) Dendritic cells production of IL-12 upon culture on adhesive substrates in the presence of LPS. C) Dendritic cells production of IL-10 when cultured on adhesive substrates. D) Dendritic cells production of

PAGE 50

50 IL-10 when cultured on adhesive substrat es in the presence of LPS. The significant pair symbols ar e described in the text. It is remarkable that DCs cultured on COL and VN adhesive substrates upregulate IL-12p40 cytokine production while maintaining relatively low IL-10 cytokine production levels, which is skewed toward the promotion of a Th1 type response. Furthermore, it is interesting that DCs cultured on FG and LN limit production the immunomodulatory cytokine, IL-10, even after stimulating with LPS, suggesting that these substrates suppress anti-inflammato ry responses. Taken together, these data indicate that adhesive substrates differentia lly direct DC production of cytokines known to direct specific T-helper cell type res ponses and can thereby skew adaptive immunity. Mixed Lymphocyte Reaction DC-mediated priming of T-cells is of par ticular interest to understand the influence of substrate-mediated modulation of immune responses. CD4+ T-cell proliferation and T-helper cell type responses were examined at two different time points. Immature DCs were seeded and cultured on protein-c oated substrates for 24 hours and CD4+ T-cells were then subsequently co-cultured for ei ther an additional 48 or 96 hours. T-cell proliferation was measured by the quantification of BrdU incorporation and intracellular cytokine production of IL-4 and IFNby CD4+ cells was quantified by flow cytometry. Adhesive substrates were found to modulate DC-directed CD4+ T-cell function in a mixed lymphocyte reaction afte r 48 hours of co-culture ( Figure 2-8A). The highest Tcell proliferation at 48 hours was observed in DCs cultured on BSA and LN, at levels 2 3 -fold higher than T-cells incubated with iDCs (negative control) On the other hand, DCs cultured on COL stimulated in termediate levels of T-cell proliferation, lower than BSA and LN coated substrates and higher t han iDCs. All other substrates produced

PAGE 51

51 relatively low T-cell proliferative responses, equivalent to iDCs. Overall significance was obtained by ANOVA, p-value < 2.2 x 10-5. The trend in salient significant pairs is summarized by: BSA = LN > FG = FN = SER = VN = iDCs. These results indicate that BSA and LN adhesive substrates induce the highest DC-mediated T-cell proliferative response at 48 hours. Critically, CD4+ T-cells cultured on adhesive substrates in the absence of DCs did not proliferate, indica ting that substrate-dependent modulation of T cell function was mediated through the adherent DC culture (data not shown). As expected, overall T-cell proliferati on quantified at 96 hours was higher (~8-fold) compared to 48 hours ( Figure 2-8B). Adhesive substrates were found to modulate DCdirected CD4+ T-cell function in the mixed lymphocyte reaction after 96 hours of coculture. T-cell proliferat ion at 96 hours was found to be the maximum when co-cultured with DCs seeded on VN compared to FN and LN (approximately 2-fold higher). BSA, COL, FG, and SER substrates produced intermediate levels of T-cell proliferation, indistinguishable from LN, FN, VN, and iDCs. Interestingly, at 96 hours of co-culture, DCs cultured on LN induced lower proliferation of T-cells t han iDCs. Overall significance was obtained by ANOVA, p-value < 1.5 x 10-3. The trend in salient significant pairs is summarized by: VN > FN = LN. T hese results demonstrate that CD4+ T-cell proliferation quantified at 96 hours is adhesive substrate-dependent. Interestingly, the mixed lymphocyte r eaction demonstrated a dynamic modulation of substrate-dependent differences in T-cell proliferation over time. DCs cultured on BSA produced the highest T-cell proliferat ive response at 48 hours, but demonstrated only a modest ~2-fold increase in T-cell prolif eration at 96 hours. Similarly, DCs cultured on LN mediated a high T-cell proliferative response at 48 hours, however, this response

PAGE 52

52 was exhausted by 96 hours, producing an even lower T-cell proliferation than iDCs ( Figure 2-8A COL, FG, FN, SER, VN, iDCs; BSA, LN, iDCs; ** BSA, FG, FN, SER, VN, iDCs; 2-8B all other conditions except FN; ** FN and LN. ). On the other hand, the largest increase (~10-fold ) in T-cell proliferation from 48 to 96 hours was found in T-cells cultured with DCs on the VN substrate. All other substrates showed a 6 -fold increase in Tcell proliferation. T-helper cell responses also demons trated DC-directed substrate-dependent differences. Quantification of IL-4 and IFNcytokine production was performed using flow cytometry, and average values of the percentage of T-cells producing either IL-4 or IFNcytokines. Compared to iDCs, T-cells co-cultured with DCs on COL and LN substrates produced a balanced respons e, with high levels of both Th2 (IL-4) and Th1 type (IFN) cytokines at both 48 and 96 hours. T-cells co-cultured with DCs on VN, on the other hand, produced a balanced response, with IL-4 and IFNlevels similar to iDCs at 48 hours, which increased to ~4-f old higher than iDCs at 96 hours. Overall, these data demonstrate that adhesive envir onments modulate DC-directed adaptive immune responses, inducing T-cell proliferat ion and specific helper T-cell responses.

PAGE 53

53 Figure 2-8. Adhesive substrates differentia lly modulate dendritic cell-mediated T-cell proliferation. A total of 20,000 CD4+ T-cells were analyzed for each run. Each experiment was independently repeated at least 4 times (C57Bl6/j for DCs and BALB/cbyj for CD4+ T-cells). A) T-cell proliferation at 48 h B) T-cell proliferation at 96 h. The significant pair symbols are described in the text. The significant pair symbols are described in the text. Impact of the Study We demonstrate that DC cult ure on extracellular matrix proteins modulates DC maturation and T-cell activation. Specifical ly, we found that cult ure on extracellular matrix proteins supported different DC mo rphologies, but equivalent levels of DC adhesion and phenotypic maturation, as characterized by high expression levels of stimulatory and co-stimulatory molecule s. Interestingly, substrate-dependent presentation of stimulatory MHCantigen complexes and co-stimulatory molecules, along with substrate-dependent modulation of DC cytokine production, correlates with differential T-cell proliferati on. Although the cytokine IL-2 is the most notable cytokine linked to T-cell proliferation, interesti ngly, we found that s ubstrate-dependent T-cell proliferative responses at 96 h correspond with the levels of IL-12p40 cytokine produced by DCs. This finding is supported by previ ous work that has also demonstrated correlations of IL-12 levels with Tcell proliferation. However, our data are the first to link adhesive substrate-dependent differences in DC IL-12p40 cytokine production with Tcell proliferation. For ex ample, adhesive substrates VN (high IL-12p40) and COL (high IL-12p40) induced high T-cell proliferation, whereas LN (moderate IL-12p40) and FN (moderate IL-12p40) induced low and moderate T-cell proliferation responses, respectively, at 96 h. overall, we quant itatively demonstrate that adaptive immune responses can be directed by the adhesive substrate on which DCs are cultured. Specifically, our findings suggest that substrate-dependent modul ation of DC IL-12p40

PAGE 54

54 cytokine production correlates with substrate-dependent CD4+ T-cell proliferation and Th1 type response in terms of IFN-g produc ing T-helper cells. On the contrary, our results indicate that the Th2 type response of IL-4 production does not correlate with DC-produced IL-10 cytokine, but do suggest a substrate-dependent trend. It is wellestablished that integrin receptors are the primary mediator of cell adhesion to extracellular matrix proteins and integrin binding to extracellular matrix proteins adsorbed onto synthetic materials has been de monstrated to dire ct cell adhesion and differentiation [89]. We theref ore hypothesize that the subs trate-mediated modulation of DC response demonstrated in this study is d ue to differential integrin binding to the adhesive substrates. Although it is known that DCs express multiple integrins [90,91], there are surprisingly little data on the effect of integrin binding to extracellular matrix proteins on DC response. This work impacts the fields of bi omaterials and DC-dire cted immunotherapy and begins the process of filling in the know ledge gap regarding the biology of DC adhesion. Critically, this work suggests plausibility fo r the rational design of biomaterials optimized for DC culture. Currently, DC-based immunotherapy approaches for diseases such as cancer [92] and autoimmune diseases like typeI diabetes [93-95] rely on ex vivo culture and expansion of patient-derived DCs onto tissue culture-treated polystyrene, without regard for the optimization of cellsubstrate interactions. In fact, the adhesive substrate provided for DCs in this ex vivo approach is typically tissue culture-treated polystyrene presenting serum proteins adsorbed from the culture media. We therefore chose to examine serum-coated tissue culture-treated polystyrene (SER substrate) as a relevant benchmark. We found that the SER substrate elicited DC cytokine production levels that

PAGE 55

55 were low in IL-12p40, high in IL-10 and produced non-optimal T-cell responses, compared to the other substrat es. These findings therefore indicate that serum-coated tissue culture-treated polystyr ene may not be the best choice for DC culture for immunotherapies requiring a Th1 type response, where robust T-ce ll proliferation and production of IFNare desired (e.g., immunotherapy for HIV). Broadly, these findings stress the need to tailor adhesive culture surfac es for a given therapeut ic application in order to optimize ex vivo culture and expansion for DC-based immunotherapies. Additionally, this work suggests the potential for DC adhesion based signals as a general mechanism that could prove to pl ay a role in various phenomena such as tissue-dependent immune responses [96-98], autoimmunity and graft-versus-host disease. For example, our findings that DCs cultured on FG and LN substrates resist LPS stimulation of the anti-inflammatory cytokine IL-10 and that LPS-stimulated DCs cultured on VN substrates produced elevated levels of IL-12 are particularly interesting in this context and warrant further invest igation into the role of DC adhesion in influencing tissue-dependent immune responses to danger signals.

PAGE 56

56 CHAPTER 3 ADHESIVE SUBSTRATE-MODULATION OF ADAPTIVE IMMUNE RESPONSES IN NOD MICE Introduction Immunotherapies and biomater ial implantations involvin g dendritic cells (DCs) is an attractive field of research that requi res effective combinations of synthetic biomaterial and biological components. Dendritic cells (DCs) ar e specialized antigen presenting cells that modulate both innat e and adaptive immune responses. In this study, we cultured non-obese diabetic mice derived DCs on different adhesive substrates and probed modulatio n in surface expression of stimulatory molecule, MHCII and co-stimulatory molecules CD80 and CD86 and, cytokine production of IL-12p40 and IL-10. Furthermore, Tcell cytokine (IL-4 and IFN) production and proliferation modulated by DC cultured on diffe rent adhesive substrates was quantified. Particularly, we found that DCs cultured on vitronectin adhesive substrate induced highest IL-12p40 production whereas collagen i nduced highest IL-10 producti on from DCs. Furthermore, it was observed that DCs cultured on vitronectin induced highest population of IL-4 producing T-cells and DCs cultured on fibr onectin substrate induced highest expression of IFNin T-cells. This work will help in advancement of the fi eld of adhesion based modulation of immune system and influence ra tional design of biomaterials and ex vivo DC-based immunotherapies for type 1 diabetes. Generation of Murine Bone Marrow-derived DCs Immature bone marrow-deriv ed DCs were generated from 7-week-old female NOD mice in accordance with protocol approv ed by the University of Florida (protocol number E751) using a modified 10-day protocol. Same procedure was followed for isolating and culturing DCs as given in Chapter 2.

PAGE 57

57 Isolation of T-Cells Isolation of T-cells was performed using t he protocols illustrat ed in Chapter 2. Protein Coating and DC Culture Extracellular matrix proteins were coated onto 12-well tissue culture-treated polystyrene plates using the same protocol as Chapter 2. Protocols for Studying DC-Functions Adhesion and proliferation studies of DCs were studied using the protocols mentioned in Chapter 2. Dendritic Cell maturation was quantified by measuring surface marker levels and cytokine production of IL-12p40 and IL-10, mentioned in detail in Chapter 2. Mixed Lymphocyte Reaction Immature NOD murine bone marrow-derived DCs (40,000 cells/well) were cultured on protein-coated (adsorbed overnight, 20 g/ml coating concentration) U-bottom tissue culture-treated 96 well plates for 24 h. CD4+ T-cells purified from spleen of BALB/cbyj mice in the ratio of 1:6 were then added to the wells and co-cultured with adherent DCs for an additional 96 h. BrdU (Beckton Dickins on) (final concentration 10 mM) was added to the co-culture along with pr otein transport inhibitor (0.7 ml for every 1 ml of media) (Beckton Dickinson) 5 h before labeling T-cells with CD4 fluorescently tagged antibodies. At the end of 96 h, T-cells were immunofluorescently stained for BrdU (BD Pharmingen) to quantify proliferation rates. A dditionally, cells were immunofluorescently stained for cell-surface markers CD4 (B D Pharmingen) and intracellular cytokine staining of IFN(BD Pharmingen) and IL-4 (BD Pharmingen). Flow cytometry was utilized for data acquisition of 20,000 cells. Analysis was performed on CD4+ gated cells and was further classified for the presence of IFN-g, IL-4 and BrdU. T-cells were stimulated with phorbol 12-myristate 13-acet ate/Ionomycin (10 ng/ml for 2 days) to

PAGE 58

58 verify T-cell proliferation pot ential (data not shown), whil e a mixed lymphocyte reaction conducted with iDCs without pre-culturing on adhesive substrates comprised the negative control. Statistical Analysis Statistical analyses were performed using general linear nested model ANOVA, linear regression analysis and/ or Pearsons correlation, as appropriate, using Systat (Version 12, Systat Software, Inc., San Jose CA). Pair-wise comparisons were made, with p-values of less than or equal to 0.05 considered to be significant. Results Endotoxin Quantification, DC Morphology an d Adhesion Quantification of endotoxin on adhesive substrates re vealed endotoxin levels below the detection limit of <0. 06 endotoxin units/mL. As an init ial indication of substrate influence on DC function, DC morphologies were investigated. Dendritic cells are a unique cell type, which modulate their surf ace morphologies to increase the surface area for antigen presentation to lymphocytes. These alterations in morphology have been linked to qualitative information in modulation of DC function. Bone marrow derived immature DCs isolated from NOD mice were cultured on tissue culture treated polystyrene pre-coated with the following adhesive proteins BSA, COL, FG, FN, LN, SER and VN and DC morphology was analyzed ( Figure 3-1). Few DCs cultured on BSA, FG and VN adhesive substrates demonst rate dendritic morphology and formation of large clusters was not observed. Cont rarily, DCs cultured on FN, SER, LN, COL and in the presence of LPS expressed both dendr ites and clustering. Furthermore, many DCs cultured on SER ( inset, Figure 3-1 ) demonstrated a veil ed morphology while a

PAGE 59

59 small number of DCs demonstr ated dendritic processes. BS A COL FG FN LN SER V N +LPS Figure 3-1. Adhesive substrates differentially modulate NOD-DC morphology. A precoated tissue culture treated plate was ut ilized to culture NOD-DCs for 24 hrs and phase contrast micrographs we re acquired (scale bar =100 m). Inset micrographs are shown to clearly demonstrate morphology of DCs on the given substrate (scale bar = 50 m).

PAGE 60

60 Clustering of vascular DCs was observ ed in atherosclerosis prone area, an indication of maturation of splenic DCs isolated from BB-DP rats and a method utilized by DCs for transferring ant igens. On the other hand pr esentation of dendrites and a veiled morphology might be indicators of activation on DCs isolated from human peripheral blood mononuclear cells. Based on morphologic al characteristics, DCs cultured on COL, FN and LN adhesive substrate can be grouped together as having both homotypic clusters and dendritic processes and hence these clusters might be categorized as containing activated and matured cells. Furthermore, DCs cultured on BSA and FG substrates manifest only dendritic processes. Dendritic cells cultured onto VN uniquely falls in the group of having both dendritic processes and veil morphology without demonstrating a homot ypic clustering. Interesti ngly, DCs cultured on SER substrate demonstrate all the three morphologies ( Figure 3-2). It is apparent that adhesive substrates have a differential e ffect on morphologies of DCs cultured on different adhesive substrates. In order to ascertain the qualitat ive comparison of maturation states derived from morphological differences, further studies were carried out to quantify differential maturation of DCs upon culture on different adhesive substrates.

PAGE 61

61 Dendrites Clusters V eil +LPS BS A COL FG FN LN SER VN Figure 3-2. Non-obese diabetic mouse derived DCs can be divided into different groups according to the typical activated mo rphological manifestations of veil, dendrites and clusters is demonstrat ed as a Venn diagram upon culture on different adhesive protein substrates : bovine serum albumin (BSA), collagen (COL), fibrinogen (FG), fibronectin (F N), laminin (LN), serum (SER) and vitronectin (VN) while DCs cultured in the presence of lipopolysaccharide (LPS) was included as a control. Dendritic cells isolated from bone marrow of NOD-mice show differential adhesion to different adhesive substrates ( Figure 3-3 BSA, COL, FG, FN and LN; ** +LPS, COL, FG, FN, LN, SER, VN and No Pre-co at; +LPS, BSA, FG, FN, SER, VN and No Pre-coat; +LPS, BSA, COL, SER, VN and No Pre-coat; +LPS, BSA, COL, SER, VN and No Pre-coat ). Dendritic cells were cultur ed for 24 hr on 96-well plate precoated with different proteins: BSA, COL, FG, FN, LN, SER, and VN. Dendritic cells cultured in the presence of LPS were also in cluded in the analysis. Interestingly, initial adhesion of NOD-DCs to COL and LN was li mited, while the la rgest number of DCs adhered to BSA. Dendritic cells cultured on SER and VN substrates and iDCs and in the presence of LPS demonstrated similar number of adherent cells. Overall significance was determined by ANOVA, p-value < 0.0003. T he trend in significant pairs between the

PAGE 62

62 number of DCs adhered onto different adhesiv e substrates is summarized by: BSA > +LPS = SER = VN = iDCs > FG = LN & LN = COL & FN = SER. +L P S BSA C O L FG FN L N S E R VN No P reco a t Number of Adherent Cells 0 100 200 300 400 500 600 700 ** Figure 3-3. Initial-adhesion of DCs cultured on tissue culture treated 96-well plates precoated with following proteins: bovine serum albumin (BSA), collagen (COL), fibrinogen (FG), fibronectin (FN), lami nin (LN), serum (SER) and vitronectin (VN) has a differential profile. Immatu re DCs (No Pre-coat) and culture with lipopolysaccharide (LPS, 1 g/mL) were included as controls. Data represents mean and standard error of at least 12 data points (replicat es) consolidated from two mice. The significant pair symbols are described in the text. Dendritic Cell Phenotype Morphological and adhesion differences between different adhesive substrates cultured NOD-DCs motivated us to quantify their activation upon culture on the adhesive substrates. Immature DCs were cultured on tissue culture treated 12 well polystyrene plates pre-coated with prot eins: BSA, COL, FG, FN, LN, SER and VN. Immature DCs and DCs cultured in the pres ence of LPS were included as negative and positive control respectively. Upon 1 day of culture DCs were lifted, stained and quantified for the expression of MHC-II, a stimulatory molecule, CD86 and CD80 co-

PAGE 63

63 stimulatory molecules. The DCs spread on the substrates as well as those loosely adherent were collected for phenotypic analysis. Statistical analysis of the data reveal ed that adhesive substrates modulate DCexpression of both stimulatory and co-stimula tory molecules. Additionally, adhesive substrates were also able to modulat e percentage population expressing MHC-II, CD86 and CD80. The data from 3 or more s eparate NOD-mouse bone marrow harvest was pooled together, statistically anal yzed and plotted as bar graphs ( Figure 3-4, Figure 35, Figure 3-6 ). Representative density plots are included for reference ( Figure 3-7). Dendritic cells cultured on FN, LN, SER, COL adhesive substrates and in the presence of LPS, showed highest increase in percent age of cells expressi ng MHC-II as compared to iDCs. Dendritic cells cultured on BSA, FG, VN and the negative control, iDCs had similar percentage of cells expressing MHCII. Overall significance was determined by ANOVA, p-value < 0.001. The trend in si gnificant pairs between percentage of cells expressing MHC-II when cultur ed on different substrates is summarized by: +LPS = COL = FN = LN = SER > VN = i DCs & BSA = FG = VN = iDCs ( Figure 3-4A FG, VN, iDCs; ** COL, LN; $ BSA, FG, VN, iDCs; +LPS, COL, FN, LN and SER; FG, VN and iDCs; $$ BSA, FG, VN, iDCs ). Soluble LPS was added to DCs cultured on different adhesive substrates to quantify t he effect of soluble maturation signal in conjunction with adhesion signaling. The percentage of DCs expressing MHC-II was the highest for the condition SER+LPS and there were no significant differences between DCs cultured on other adhesive substrates. Overall significance was determined by ANOVA, p-value < 0.001. The trend in si gnificant pairs between percentage of cells expressing MHC-II in the presence of L PS is summarized by: SER+LPS > BSA+LPS,

PAGE 64

64 COL+LPS, FG+LPS, FN+LPS, LN+LPS = VN+LPS = iDCs ( Figure 3-4C BSA+LPS, FG+LPS, FN+LPS and VN+LPS ). Geometric mean fluorescence intensity (gMFI) of MHC-II was quantified form the data acquired via flow cytometery. Dendritic cells cultured on FN, VN and BSA had the hi ghest MHC-II expression followed by DCs cultured on FG, LN and SER adhesive substr ates and iDCs. Overall significance was determined by ANOVA, p-value < 0.001. The trend in significant pairs between DCexpression of MHC-II is su mmarized by: BSA = VN & FN > SER & BSA > FG = LN = SER = iDCs ( Figure 3-4B FG, LN, SER, iDCs; $l BSA, VN; ** SER, iDCs; BSA, FN, VN; FG, LN, SER, iDCs. ). Next, MHC-II expre ssion of DCs upon culture on adhesive substrates in the presence of L PS was quantified. Dendr itic cells cultured on BSA, FG and VN adhesive substrates h ad highest MHC-II expression whereas DCs cultured on LN showed the least surfac e expression. Overall significance was determined by ANOVA, p-value < 0.001. The trend in significant pairs DC-expression of MHC-II when cultured in the presence of LPS is summarized by: BSA+LPS = VN+LPS = FG+LPS > iDCs & VN+LPS > LN+LPS & VN+LPS > SER+LPS > iDCs (Figure 3-4D iDCs; ** VN+LPS; VN+LPS and i DCs; LN+LPS, SER+LPS, iDCs; $ BSA+LPS, FG+LPS, SER+LPS, VN+LPS ). Interestingly, DCs cultured on VN, BSA and FG substrates induced the lowest per centage population of DC expressing MHC-II molecule. Next, the surface expression and percentage population expressing CD86 via flow cytomtery was quantified. The adhes ive substrates modulate the surface expression as well as the percentage of popul ation expressing CD86 a co-stimulatory molecule. Dendritic cells cultured in the presence of LPS induced highest percentage of DCs expressing CD86. Dendritic cells cult ured on FG, LN, SER and VN substrates

PAGE 65

65 along with the negative control iDCs stimulat e the lowest percentage of cells expressing CD86. Dendritic cells cultured on BSA, COL and FN adhesive substrates induced similar percentage of cells expressing CD86 ( Figure 3-5A all other conditions; +LPS, FG, iDCs; ** +LPS, FG, VN and iDCs; +LPS, BSA, COL, FN; +LPS; $ +LPS, COL ). Overall significance was determi ned by ANOVA, p-value < 0.001. The trend in significant pairs between percentage of cells expressing CD86 when cultured on different substrates is summarized by: +LPS > COL = FN > FG = VN = iDCs & BSA > FG = iDCs & LN = SER = VN = FG = iDCs The effect of soluble LPS as soluble maturation stimulation in the presence of adhesive substrates was quantified. It was observed that in the presence of LPS adhesiv e substrates modulate percentage of cells expressing CD86. Dendritic cells cultur ed on COL+LPS and LN+LPS induced highest percentage of cells expressing CD86. Dendritic cells cultured on FN+LPS and SER+LPS induced similar

PAGE 66

66 BSA+LPS C OL + LPS FG+LPS F N + LP S LN+ LPS S E R + LPS V N + L PS i D Cs gMFI MHCII 0 200 400 600 800 1000 1200 ** $ B SA+LP S C OL+LPS FG+L P S FN+L P S LN+LPS SER +L PS V N +LPS iDCs % MHCII DCs 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 (C) +L PS BSA CO L FG FN LN SER V N i DCs % MHC-II DCs 0 1 2 3 4 5 6 (D) + LP S BSA C OL FG FN L N S E R V N iDCs gMFI MHC-II 0 100 200 300 400 (B) ** $ $ (A) ** $ $$ Figure 3-4. Adhesive substrates differentially activate non-obese diabetic (NOD) bone marrow derived-dendritic cells (DCs) as evidenced through surface presentation of major histocompatibility (MHC-II). Dendr itic cells were isolated from at least 2 separate mice and each mouse handled as an independent experiment repeat. It is important to note that al l the DCs adhered or floating were included in the analysis. A) Dendritic cells positive for surface expression of MHC-II. B) Dendritic cells surface expression of MHC-II. C) Dendritic cells positive for surface expression of MHC-II when cultured in the presence of LPS. D) Dendritic cells surface expression of MHC-II when cultured in the presence of LPS. The significant pair symbols are described in the text. but moderate percentage of cells expressi ng CD86. Dendritic cells cultured on FG+LPS, VN+LPS, BSA+LPS and the negative control iDCs had similar and lower percentage of cells expressing CD86 ( Figure 3-5C COL+LPS, FN+LPS, LN+LPS, SER+LPS; BSA+LPS; FG+LPS, FN+LPS, SER+LPS, VN+LPS, iDCs; COL+LPS, LN+LPS, SER+LPS; BSA+L PS, COL+LPS, iDCs; BSA+LPS, FG+LPS, SER+LPS; ** BSA+LPS, COL+L PS, FG+LPS, LN+LPS, VN+LPS, iDCs; $

PAGE 67

67 COL+LPS, FN+LPS ,LN+LPS, SER+LPS ). Overall significance was determined by ANOVA, p-value < 1 x 10-6. The trend in significant pairs between percentage of cells expressing CD86 when cultured on different substrates in the presence of LPS is summarized by: COL+LPS = LN+LPS = SER+LPS > FG+LPS = BSA+LPS = VN+LPS = iDCs & COL+LPS > FN+LPS > iDCs. Surface expression of CD86 cultured on adhesive substrate was quantified and statistical analys is revealed an overall ANOVA, p-value < 0.001 ( Figure 3-5B COL, FG FN, LN, SER, VN, iDCs; ** COL, SER, VN, iDCs; +LPS, BSA; +LPS ). The trend in significant pairs between DC-expression of CD86 when cultured on different substrates is su mmarized by: +LPS > COL = FG = FN = LN = SER = VN = iDCs & BSA > COL = SER = VN = iDCs. Surface expression of CD86 was quantified upon DC culture on diffe rent adhesive substrate in the presence of LPS and no significant difference ( Figure 3-5D) between DC CD86 expressions was observed. Adhesive substrates have been shown to modulate the percentage of DCs expressing CD80 co-stimulatory molecule. Interestingly the negative control (iDCs) and positive control of DCs cultured in the presence of LPS had similar levels of percentage cells positive for CD80, which was similar to DCs cultured on adhesive substrate VN and FN. Dendritic cells cultured on LN, FG and BSA demonstrated the lowest percentages of DCs positive for CD80 ( Figure 3-6A BSA, COL, FG, LN; +LPS, FN, SER, VN, iDCs; ** +LPS, FN, LN, SER, VN, iDCs; BSA, COL, FG, LN; +LPS, COL, FN, SER, VN, iDCs; BSA, COL, FG, LN, VN; $ BSA, COL, FG, LN, SER ). Overall significance was determined by ANOVA, p-val ue < 0.001. The trend in significant pairs between percentage of cells expressing CD80 w hen cultured on different substrates is summarized by: iDCs = +LPS = VN = FN & +LPS > COL = FG = BSA & COL > LN.

PAGE 68

68 Surface expression of CD80 was quantifi ed when DCs were cultured on different adhesive substrates. Dendritic cells when cultured on SER substrate induced highest expression of CD80, which was similar to DCs cultured on FG (p = 0.052, are significant if 90% confidence level is considered) and LN (p = 0.085, are significant if 90% confidence level is considered). Overall significance was determined by ANOVA, pvalue < 3.4 x 10-7. The trend in significant pairs between DC-expressing CD80 when + L PS BS A CO L F G FN LN SER VN iDCs gMFI CD86 0 20 40 60 80 100 120 + LPS BSA C O L FG F N L N S ER VN i D Cs % CD86 DCs 0 2 4 6 8 10 12 B S A +L P S C OL+L P S FG+LPS FN+LPS L N+L P S S E R + L P S VN+LPS i D Cs gMFI CD86 0 20 40 60 80 100 120 BSA+LP S COL + L P S FG+LP S FN+LPS L N +L P S SER + LPS V N+ L PS i DCs % CD86 DCs 0 2 4 6 8 10 ** ** **(A) (B) (C) (D) $ $ Figure 3-5. Adhesive substrates differentially activate non-obese diabetic (NOD) bone marrow derived-dendritic cells (DCs) as evidenced through surface presentation of co-stimulatory molecule, CD86. Dendritic cells were isolated from at least 2 separate mice and each mouse handled as an independent experiment repeat. It is important to note that al l the DCs adhered or floating were included in the analysis. A) Dendritic cells positive for surface expression of CD86. B) Dendritic cells surface expression of CD86. C) Dendritic cells positive for surface ex pression of CD86 when cultured in the presence of LPS. D) Dendritic cells surfac e expression of CD86 when

PAGE 69

69 cultured in the presence of LPS. The significant pair symbols are described in the text. cultured on different substrates is summarized by: SER > +LPS = COL = FN = BSA = VN = iDCs ( Figure 3-6B +LPS, BS A, COL, FN, VN and iDCs ). +L P S BS A COL FG FN LN S ER VN iDCs % CD80 DCs 0 20 40 60 80 100 + LP S BSA C O L FG FN LN SE R VN i D C s gMFI CD80 0 1 2 3 4 5 ** (A) (B) $ Figure 3-6. Adhesive substrates differentially activate non-obese diabetic (NOD) bone marrow derived-dendritic cells (DCs) as evidenced through surface presentation of co-stimulatory molecule, CD80. Dendritic cells were isolated from at least 2 separate mice and each mouse handled as an independent experiment repeat. It is important to note that al l the DCs adhered or floating were included in the analysis. A) Percentage of DC population expressing CD80. B) Dendritic cell expression of CD80. The significant pair symbols are described in the text.

PAGE 70

70 10 10 10 10 10 Without LPS With LPS 100101102103104 100101102103104 3.86% 95.83 % 0.26% 0.05% FN 10 0 10 1 10 2 10 3 10 4 10 0 10 1 10 2 10 3 10 4 1.09 % 95.29 % 2.05 % 1.57 % LN 100101102103104 100101102103104 2.69 % 96.72 % 0.22 % 0.37 % LN 10 0 10 1 10 2 10 3 10 4 10 0 10 1 10 2 10 3 10 4 1.51 % 93.81 % 2.46 % 2.22 % COL 100101102103104 100101102103104 1.14 % 98.26 % 0.00 % 0.60% COL 0 1 2 3 4 10 0 10 1 10 2 10 3 10 4 4.84 % 93.78 % 1.09 % 0.30 % BSA 100101102103104 100101102103104 0.62% 99.10 % 0.07% 0.21 % BSA 10 010 1 10 2 10 3 10 4 10 010 110 210 310 4 0.96 % 97.11 % 0.96 % 0.96 % FG 100101102103104 100101102103104 3.01% 96.58 % 0.29% 0.12 % FG 10 0 10 1 10 210 3 10 4 10 0 10 1 10 2 10 3 10 4 1.16 % 96.14 % 1.67 % 1.03 % SER 100101102103104 100101102103104 6.49 % 89.85 % 3.07 % 0.59% SER Without LPS 10 10 10 10 10 0 1 2 3 4 10 0 10 1 10 2 10 3 10 4 2.64% 93.41 % 3.66% 0.29 % FN MHC II CD80 10010110 2 10 3 10 4 100101102103104 45.06 % 52.51 % 2.25 % 0.17 % FN 10010110 2 10 310 4 100101102103104 65.05 % 30.83 % 3.64 % 0.49 % LN 10010110 2 10 3 10 4 100101102103104 60.42 % 35.19 % 3.29 % 1.10 % COL 10010110 2 10 3 10 4 100101102103104 23.95 % 74.65 % 1.00 % 0.40 % BSA 10010110 2 10 3 10 4 100101102103104 25.40 % 73.52 % 0.81 % 0.27 % FG 10010110 2 10 3 10 4 100101102103104 30.85 % 65.17 % 3.08 % 0.90 % SER MHC II CD86

PAGE 71

71 CD86 100101102103104 10010110210310 4 3.49% 96.34% 0.17% 0.00% 100101102103104 10010110210310 4 1.20% 97.60% 0.70% 0.50% 100101102103104 100101102103104 3.74% 95.29% 0.75% 0.21% Without LPS With LPS VN MHC-II MHC-II MHC-II 0256512768102 4 100101102103104 CD11c only89.9% MHC II Purit y CD86 CD86 VN Without LPS 10010110210 3 10 4 100101102103104 15.56 % 83.57 % 0.29 % 0.58 % VN MHC-II CD80 10 0 101102103104 10 0 10 1 10 2 10 3 10 4 14.09 % 84.95 % 0.65 % 0.32 %CD80 MHC-II Figure 3-7. Representative phen otypic density plots of data summarized in Figure 3-5, 3-6 for murine NOD-derived DCs cultured on protein-coated tissue culturetreated polystyrene substrate for 24 h. Shown also is iDC purity and maturation markers. 10 0 10 1 10 2103104 10 0 10 1 10 2 10 3 10 4 0.16 % 99.52 % 0.04 % 0.28 % LN 10 0 10 1 10 2 10 3 10 4 10 0 10 1 10 2 10 3 10 4 0.20% 99.08 % 0.05% 0.66 % FN 100101102103104 100101102103104 0.25 % 99.06 % 0.05 % 0.63% COL 10 0 10 1 10 2 10 3 10 4 10 0 10 1 10 2 10 3 10 4 0.22% 99.50 % 0.07% 0.21 % BSA 10 0 10 1 10 2103104 10 0 10 1 10 2 10 3 10 4 0.21 % 99.14 % 0.07 % 0.58 % FG 100101102103104 100101102103104 0.34 % 99.26 % 0.08 % 0.33 % SER 10 0 10 1 10 210 310 4 10 0 10 1 10 2 10 3 10 4 0.14 % 99.48 % 0.04 % 0.34 % VN 100101102103104 100101102103104 0.22 % 98.79 % 0.07 % 0.92% iDCsMHC-II MHC-II MHC-II MHC-II MHC-II MHC-II MHC-II MHC-II CD86 CD86 CD86 CD86 CD86 CD86 CD86 CD86 Figure 3-8. Soluble proteins do not activa te DCs, quantified via percentage of cells expressing stimulatory (M HC-II) and co-stimulatory (CD86) molecules. Density plots of MHC-II vs. CD 86 are generated through FCS Express version 3 software and the data was obt ained using flow cytometry. The percentage of DCs positive for these mo lecules is reported in the insets.

PAGE 72

72 Dendritic Cell Cytokine Secretion Dendritic cells were cultured on adhesive substrates for 24 hours and the cytokine IL-12p40 a pro-inflammatory cytokine which may induce a Th1 type response, secreted by DCs was quantified using sandwi ch-ELISA. Dendritic cells spread on the substrates as well as those loosely adherent were collected for cytokine production analysis. Dendritic cells cultured in the presence of LPS induced highest amount of IL12p40 cytokine, while DCs cultured on all the other adhesive substrates produced either low or moderate IL-12p40. Ov erall significance was deter mined by ANOVA, p-value < 1 x 10-6. The trend in significant pairs between DC production of IL-12p40 when cultured on different substrates is summarized by: +LPS > VN > BSA > LN = SER > COL > FN > FG = iDCs ( Figure 3-9A BSA, COL, FG, FN, LN, SER, VN, iDCs; ** +LPS, COL, FG, FN, LN, SER, VN, iDCs; +LPS, BSA, FG, FN, LN, SER, VN, iDCs; +LPS, BSA, COL, FN, LN, SER, VN; $ +LPS, BSA, COL, FG, LN, SER, VN, iDCs; +LPS, BSA, COL, FG, FN, VN, iDCs,; $$ +LPS, BSA, COL, FG, FN, LN, SER, iDCs ). Dendritic cells were cultured on differ ent adhesive substrates in the presence of LPS to understand the effect of adhesive substrates on the IL-12p40 secretion by these cells. Adhesive substrates differentiall y modulated IL-12p40 cytokine production of DCs when cultured on adhesive substrates with a soluble maturation stimulus of LPS, however, DCs largely were resistant to induction of IL-12p40 when cultured on adhesive substrates without sol uble maturation stimuli ( Figure 3-9B COL+LPS, FG+LPS, FN+LPS, SER+LPS, VN+LPS, iDCs; BS A+LPS, FN+LPS, SER+LPS, iDCs; $ BSA+LPS, FN+LPS, LN+LPS, SER+LPS, i DCs; ** BSA+LPS, COL+LPS, FG+LPS, LN+LPS, SER+LPS, VN+LPS, iDCs; FG+LPS, FN+LPS, SER+LPS, iDCs; BSA+LPS, COL+LPS, FG+LPS, FN+LPS, LN+LPS, VN+LPS, iDCs; BSA+LPS,

PAGE 73

73 FN+LPS, SER+LPS, iDCs; $$ BSA+LPS, COL+LPS, FG+LPS, FN+LPS, LN+LPS, SER+LPS, VN+LPS ). Overall significance was deter mined by ANOVA, p-value < 1 x 1010. The trend in significant pairs between DC production of IL-12p40 in the presence of LPS when cultured on different substrates is summarized by: SER+LPS > FN+LPS > BSA+LPS > COL+LPS = FG+LPS = LN+LPS = VN+LPS > iDCs. Additionally, supernatant isolated from the culture of DCs on different adhesive substrates were quantified using sandwich-ELISA for production of IL-10 an antiinflammatory cytokine which is tr aditionally considered to induce a Th2 type response. The DCs cultured in the presence of LPS induced highest amount of IL-10 cytokine production followed by DCs cultured on COL ( Figure 3-9C BSA, COL, FG, FN, LN, SER, VN, iDCs; +LPS, COL, FG, VN, i DCs; ** +LPS, BSA, FG, FN, LN, SER, VN, iDCs; +LPS, BSA, COL, FN, LN, SER; +LPS, COL, FG, VN, iDCs; +LPS, COL, FG, VN and iDCs ).

PAGE 74

74 + LP S BSA COL FG F N LN SER VN iDCs Concentration IL-10 (pg/ml) 0 50 100 150 200 250 + LPS BS A COL F G FN LN S ER V N iDCs Concentration IL-12p40 (pg/mL) 0 200 400 600 800 1000 1200 ** $ $$ BS A +LP S COL+LPS FG+LPS F N +L P S LN + LPS SER+LP S VN+LP S iDCs Concentration IL-12p40 (pg/mL) 0 200 400 600 800 1000 1200 **(B) $ (C) ** BSA + LP S COL+LPS FG+L P S FN + LPS LN+L P S SE R +LPS VN+L P S iDC s Concentration IL-10 (pg/mL) 0 50 100 150 200 (D) (A) $$ Figure 3-9. Adhesive substrate modulates non-obese diabetic (NOD) mouse dendritic cells (DCs) cytokine production. Dendritic cells were isolated from at least 3 separate mice and each mouse handled as an independent experiment repeat. It is important to note that all the DCs adhered or floating were included in the analysis. A) Dendritic cells production of IL-12 upon culture on adhesive substrates. B) Dendritic cells production of IL-12 upon culture on adhesive substrates in the presence of LPS. C) Dendritic cells production of IL-10 when cultured on adhesive substrates. D) Dendritic cells production of IL-10 when cultured on adhesive substrat es in the presence of LPS. The significant pair symbols are de scribed in the text. Overall significance was determi ned by ANOVA, p-value < 1 x 10-3. The trend in significant pairs between DC production of IL10 when cultured on different substrates is summarized by: +LPS > COL > FG = VN = iDCs > FN = LN > BSA & FG = SER. Cytokine production by DCs when cultured on adhesive substrates in the presence of LPS was quantified and r eported. The condition SER+LPS induced highest IL-10 cytokine production whereas there was no significant difference between cytokine

PAGE 75

75 productions induced by other adhesive substrates (Figure 3-9D all the other conditions ). Overall significance was deter mined by ANOVA, p-value < 1 x 10-3. The trend in significant pairs between DC producti on of IL-10 in the presence of LPS when cultured on different substrates is su mmarized by: SER+LPS > BSA+LPS = COL+LPS = FG+LPS = FN+LPS = LN+LPS = VN+LPS = iDCs. Mixed Lymphocyte Reaction Mixed lymphocyte reaction of NO D isolated DCs and BALB/cByJ CD4+ T-cells was carried out to evaluate adhesive substr ates modulation of DC-mediated CD4+ T-cell function. The CD4+ T-cell proliferation and T-helper ce ll type response was quantified by flow cytometery of cells obtained fr om the co-culture of DCs and CD4+ T-cells. The CD4+ T-cells were stained with intracellular cytokine IFNand IL-4 which are the indicators of Th1 and Th2 type of response respectively. B S A C OL FG FN LN S ER VN i DCs Proliferative T-cells (%) 0 10 20 30 40 50 ** $ ** $ Figure 3-10. In a 96-hour mixed-lymphocyte reaction (1:6, DC to T-cell ratio) DCs modulate T-cell proliferation different ially. Immature DCs were cultured on these modified surfaces for 24 h bef ore adding T-cells isolated from the spleen. Each experiment was ind ependently repeated at least 3 times (NOD/LtJ for DCs and BALB/cbyj for CD4+ T-cells). At least 20,000 CD4+ Tcells were analyzed for each run. The significant pair symbols are described in the text.

PAGE 76

76 The T-cells that were positive for CD4 surface marker were analyzed for the presence of stained BrdU for the quantificati on of T-cell proliferation. The T-cell proliferation was quantified at the end of 96 hours. The prol iferation of T-cells was highest for DCs cultured on SER, COL and FG. BSA CO L F G F N LN SER VN N o P r e coa t % IFNproducing T-cells 0.0 0.5 1.0 1.5 2.0 BSA C O L FG FN LN S E R VN N o Pr ecoa t % IL-4 producing T-cells 0.0 0.5 1.0 1.5 2.0 2.5 3.0 BSA CO L F G F N LN SER VN N o Pr ec oat gMFI IFNproducing T-cells 0 2 4 6 8 10 12 14 (A) (B) (C) B SA C OL FG F N LN SER VN N o Pre-coa t gMFI IL-4 producing T-cells 0 1 2 3 4 5 6 (D) Figure 3-11. In a 96-hour mixed-lymphocyte reaction (1:6, DC to T-cell ratio) DCs modulate CD4+ T-cell cytokine production and ex pression of these cytokines. Immature DCs were cultured on thes e modified surfaces for 24 h before adding T-cells isolated from the spl een. Each experiment was independently repeated at least 3 times (NOD/LtJ for DCs and BALB/cbyj for CD4+ T-cells). At least 20,000 CD4+ T-cells were analyzed for each run. A) Percentage T cells expressing IFN. B) T cells expressing IFN. C) Percentage T cells expressing IL-4. D) T cells expressing IL-4. T he significant pair symbols are described in the text. The proliferation induced by DCs cultured on FN, LN and BSA was the minimum for all the substrates. Overall significanc e was determined by ANOVA, p-value < 1 x 10-

PAGE 77

77 3. The trend in significant pairs between CD4+ T-cell proliferation is summarized by: SER > VN = iDCs = FN = LN = BSA & SE R = COL = FG & COL = FG > LN = iDCs ( Figure 3-10 COL, FG, SER; BSA, LN, iDCs; ** SER; $ COL, FG, SER; BSA, FN, LN, VN, iDCs ). Dendritic cells cultured on different substrate did not induce differential percentage of cells producing intra-cellular cytokine IFN(Figure 3-11A). Overall significance was determined by AN OVA, p-value < 0.1. The data suggests (when 90% confidence level is considered) that iDCs were higher in inducing a population of T-cells positive for the IFNcytokine. The geometric mean intensity (gMFI) of IFNexpressed by T-cells co-cultured with DCs on the adhesive substrates were quantified and reported as a bar graph. It was observed that T-cells co-cultured with DCs on FN adhesive substrates induced highest expression of IFN(Figure 3-11B FG, SER ). Overall significance was determi ned by ANOVA, p-value < 0.03. There were no other significant differences between gMFI induced by adhesive substrates. Combining the results of gMFI and percentage T-cells expressing IFNwe can conclude that FN substrate might induce a Th1 type response and all the other substrates might induce equivalent levels of Th1 response. Similarly, the quantification of percentage of cells expressing IL-4 intracellular cytokine re vealed that the co-culture on VN adhesive substrate induced highest per centage of T-cells expressing IL-4, while all the other substrates induc ed equivalent levels, which wa s less by 2.5 times the level induced by VN. Overall significance wa s determined by ANOVA, p-value < 0.003 ( Figure 3-11C all the other conditions ). Geometric mean fluorescence intensity of T-cells expressing IL-4 was also quantified. There was no significant difference found between IL-4 expressions of T-cells induced by different substrates. Overall significance

PAGE 78

78 was determined by ANOVA, p-value = 0.25 (Figure 3-11D). Combining the results obtained from gMFI and percentage of T-cells ex pressing IL-4 we can conclude that VN substrate might induce a Th2 type response. Representative plots for percentage of CD4+ T-cells expressing IL-4, IFNand BrdU incorporation (f or proliferation) are presented ( Figure 3-12 ).

PAGE 79

79 Figure 3-12. Representative phenotypic densit y plots of data summarized in Figure 310, 3-11 for murine BALB/cbyj CD4+ T-cells co-cultured with NOD-DCs on protein-coated tissue culture-treated polystyrene substrate for 96 hours.

PAGE 80

80 FN FG LN COL VN BSA SER No Pre-coat Figure 3-13. In a 96-hour mixedlymphocyte reaction (1:6, DC to T-cell ratio) B6-DCs do not modulate T-cell prolif eration differentially w hen cultured with soluble protein. Impact of the Study Type 1 diabetes (T1D) is an autoimmune disorder pathologically described by destruction of the patients -cell islets by T-cells via inducing apoptosis through the release of granzyme B/perforin or th rough the release of cytokines TNF-/IFN. Dendritic cells are said to play a centra l role in priming lymphocyte-based autoimmune response [99]. Dendritic cells can identify the modified microenvironment of the islets and migrate to the lymph node to prime nave T-cells. During the migration towards lymph nodes if DCs get matured or activat ed they can induce an auto-reactive cytotoxic T-cell response otherwise a regulator-T-cel l response is generated [100]. The migration of DCs involves transition through the panc reas-tissue and muscle capillary basement membrane to the secondary lymphoid organ s. Several studies have shown that extracellular matrix (ECM) proteins, su ch as laminin (LN), collagen (COL) and fibronectin (FN) have abnormally high levels in the pancreas-tissue and muscle capillary basement membrane in prediabetic and diabetic pathology. The possibility of

PAGE 81

81 involvement of ECM proteins in T1D-pathology led us to investigate the effect of ECM proteins modulation of DC-maturation and s ubsequent T-cell responses. Additionally, DC-maturation was evaluated in the presence of LPS to evaluate effects of adhesive substrates in the presence of a soluble maturation signal. The differential morphological features clearly an oversimplification of the functional differences, nonetheless point to the fact that only NOD-DCs cultured on BSA, FG and VN substrates, all of which are found in blood can be characterized as DCs that form clusters. Clustering of vascular DCs has been observed in atherosclerosis prone area [101], an indication of maturation of splenic DCs isolated from BB-DP rats [101-103] and a method ut ilized by DCs for transferring antigens [104]. On the other hand presentation of dendrites and a veiled morphology might be indicators of activation on DCs isolated from human peripheral blood mononuclear cells [105]. Interestingly, although am ong these proteins VN is found in ECM, its expression has not been shown to be up-regulated in diabetes pathology. Upon comparison of the effect of these three adhesive substrates on modulation of DC-functions we observed that a) BSA substrate induces high expr ession of MHC-II and CD86 whereas VN and FG substrates induced lower expressi on of MHC-II and CD86 and b) BSA adhesive substrates induces low but similar levels of percentage of DCs expressing CD86 and MHC-II as VN and FG adhesive substrates ( Figure 3-4, Figure 3-5 ). The IL-12p40 cytokine production profile of DCs cultured on BSA and VN adhesive substrates are moderately high, whereas that of DCs cultured on FG substrat e is as low as iDCs. The IL-10 cytokine production of DCs on VN and FG substrates were similar whereas DCs cultured on BSA was even lower than the negativ e control of iDCs. This data suggests

PAGE 82

82 that DCs cultured on BSA substrates whic h had high expression of MHC-II and CD86 and high IL-12p40 with low IL-10 production and no cluster formation might induce Th1 type T-cells response, however, th is was not apparent from the IFNproduction from Tcells. Similarly, analyzing DCs cultured on FG, which is expected to have a moderate Th2 type response did not show higher per centages of T-cells expressing IL-4. However, DCs cultured on VN, which ha d moderate levels of IL-12p40 and IL-10 productions and low expression of MHC-II and CD86, with low expression of CD80 along with low percentage of DCs expressi ng MHC-II and CD86, high CD80 expression and not demonstrating the formation of clus ters induced highest percentage of IL-4 producing T-cells (~ 2.5 times higher than all the other substr ates) and low percentage of IFNproducing T-cells ( Figure 3-5, Figure 3-6, Figure 3-8, Figure 3-10 ). These data might suggest that the defective NO D-DCs are naturally inclined toward suppressing the population of IL-4 producing T-cells, unless provided with extra-cellular signaling. Dendritic cells are known to modulate the ECM and might cause the adsorbed proteins to become soluble. In order to validate that the NODDCs were not influenced by the presence of soluble proteins we cultured NOD-DCs with low concentration (10 g/mL) of different proteins. The NOD-DCs cultured on adhesive substrates had similar levels of activation (percentage of cells expressing MHC-II and CD86) as compared to iDCs, (Figure 3-7B). These data suggests that adhesive cues are required for differential activation of NOD-DCs and the degradation protein products do not play a major role in modulating NO D-DC-function. Furthermore, these data also suggests that the xenogenic immune response that might be attributed to different species-derived

PAGE 83

83 proteins used in different experiments here did not play a major role in activating NODDCs. Additionally, MLR-experiment with so luble proteins was carried out with DCs obtained from the strain of CD57BL6/j mouse (known to shown normal responses) and T-cells isolated from BALB/cByJ mouse. In this case B6-DCs induced similar levels of T-cell proliferation as iDCs on all t he substrates as induced by iDCs ( Figure 3-12 ). This data suggests that the soluble proteins do not have an appreciable effect on B6-DC maturation or T-cell proliferation. Pearsons correlation was employed to furt her investigate, t he effect of NOD-DCcytokine on the percentages of T-ce lls producing either IL-4 or IFNand their expression. We observed that DC-produced pro-inflammatory cytokine IL-12p40 had a low and moderate but negative correlation to percentage of T-cells producing IFN(Pearsons Correlation = -0. 32) and expression of IFN(Pearsons Correlation = -0.28) respectively. Traditionally, IFNproduction follows similar trend to IL-12p40 however, our results do not suggest the same and hence, further stresses the defective nature of DCs (Figure 3-14A, Figure 3-14C ). Furthermore, NOD-DC produced IL-10 had a low correlation with percentage of T-cells produci ng IL-4 (Pearsons Correlation = 0.22), but a high correlation with expressi on of IL-4 (Pearsons Correl ation = 0.68). This suggests that adhesive substrate induced differential DCproduction of IL-10 is a good indicator of the extent of generated Th2 type response ( Figure 3-14B, Figure 3-14D). Furthermore, the percentage of DCs expressing stimulator y molecules MHC-II and co-stimulatory molecules CD86 and CD80 had a low and n egative correlation with CD4+ T-cell proliferation, with a Pears ons correlation of -0.24, -0 .14 and -0.18 respectively ( Figure 3-14E). These data might suggest that stimulatory and co-stimulatory molecules do not

PAGE 84

84 play a major role in inducing T-cell proliferation when cultured on different adhesive substrates. Upon correlating T-cell prolifer ation and DC-expression of stimulatory and co-stimulatory molecules, it was observed that expression of MHC-II had a low and negative Pearsons Correlation of -0.28 with Tcell proliferation; expression of CD86 had a moderate but negative Pearsons Correlation of -0.44 with T-ce ll proliferation and expression of CD80 had a high and positive correlation with T-cell proliferation ( Figure 3-14F ). A high correlation between T-cell prol iferation and co-stimulatory molecules CD80, along with low correlation with MHC-II suggests that expression of CD80 might be the governing factor for T-cell proliferation via NOD-DCs cultured on the adhesive

PAGE 85

85 substrates. (B) IL-4+ T-cells (%) Pearsons Correlation = 0.22 IFN+ T-cells (%) (A) IL-12p40 Pearsons Correlation = -0.32 IL-12p40 IL-10 gMFI IL-4+ T-cells Pearsons Correlation = 0.68 IL-10 gMFI IFN+ T-cells Pearsons Correlation = -0.28(C) (D)CD4+ T-cell Proliferation % MHC-II DCs % CD86 DCs % CD80 DCs Pearsons Correlation = -0.24 Pearsons Correlation = -0.14 Pearsons Correlation = -0.18 gMFI MHC-II gMFI CD86 gMFI CD80 Pearsons Correlation = -0.28 Pearsons Correlation = -0.44 Pearsons Correlation = 0.70 CD4+ T-cell Proliferation (F) (E) Figure 3-14. Dendritic cell surface stim ulatory and co-stimulatory molecules and cytokine produced may modulate T-cell response, in a mixed lymphocyte reaction, determined using Pearsons co rrelation values (0.1 to 0.4 low correlation, 0.4 to 0.7 moderate correla tion and 0.7 to 1.0 high correlation). This analysis led us to compare cumula tive effect of DC-cytokine production, traditionally known to direct Th-cell type response and DC-expression of CD80 on the

PAGE 86

86 percentage T-cells expressi ng either IL-4 or IFNcytokines ( Figure 3-15). gMFI CD80 1.5 2.0 2.5 3.0 3.5 4.0 Concentration IL-10 (pg/ml) 0 20 40 60 80 100 120 140 160 gMFI CD80 1 52 02 53 03 54 0 Concentration IL-12p40 (pg/ml) 0 50 100 150 200 250 300 350 400 BS A COL FG FN LN SER VN BSA COL FN LN SER VN iDCs (A) (B) % IL-4 producing T-cells % IFNproducing T-cells iDCs Figure 3-15. A 3-factor bubble graph between expression of CD80 by DCs (gMFI-CD80) on the x-axis and DC-cytokine producti on (IL-10 and IL-12p40) on the y-axis is plotted with the diameter of the bubble representi ng the percentage of CD4+ T-cells producing IL-4 or IFN. These plots have been constructed to understand the effect of cumulative effe cts of expression of co-stimulatory molecule and cytokine produced by DCs on T-helper type responses. A) Dependence of percent of T-cells producing IL-4 on DC IL-10 production and CD80 expression. B) Dependence of percent of T-cells producing IFNon DC IL-12p40 production and CD80 expression. A bubble graph was plotted with CD80 expr ession on X-axis with either IL-12p40 or IL-10 on the y-axis. The size of the bubble represents the amount of percentage cells positive for IFNand IL-4 respectively. Comparing the two bubble plots ( Figure 3-15A, Figure 3-15B) can help understand effects of differ ent adhesive substrates on T-helper type response in a diabetic pathology. Interestingly, the no pre-coat condition of iDCs, although have low IL-12p40 production and low CD80 expression had high IFNand moderate IL-4 positive T-cells, thus suggesti ng that iDCs might not be best suited for ex vivo immunotherapy approaches. Furthermore DCs cultured on SER induced highest CD80-expression, with moderate IL-12p40 and IL-10 production and moderate to low IL-4 and IFNresponses, thus suggesting that this protein coating did not preferentially mediate the T-cell response toward toler ance (Th2-type) or inflammation (Th1-type).

PAGE 87

87 Currently, several immunotherapies involve ex vivo culture of DCs [106-108]. Such cultures involve culture and expansion of DC-population on ti ssue-culture treated polystyrene plate which has adsorbed proteins from serum as substrate. From our studies, it is clear that culturing DCs on SER, although, produces a low Th1 type response produces a low Th2 type response as well. Another, interesting surface adsorbed protein was VN, which induced high % IL-4 producing T-cells but low % IFNproducing T-cells ( Figure 3-15A, Figure 3-15B). Evidently, it is important to understand and characterize the substrate on which NOD-DCs are being cultured and choose the substrate best suited for the designed immunotherapy.

PAGE 88

88 CHAPTER 4 INTEGRIN-MEDIATED PEPTIDE ADHESI ON BASED CONTROLLED ACTIVATION OF DENDRITIC CELLS Introduction Manipulating the bodys immune respons e to boost or repair natural healing mechanisms is a powerful approach that is be ing explored using dendritic cell (DC) immunotherapies. We have engineered a ve rsatile process to produce density gradients of surface-bound species for high throughput analyses of cell responses to well-characterized receptor-ligand interacti ons. For instance, the Arg-Gly-Asp (RGD) tripeptide sequence, which is found in a variet y of extracellular ma trix proteins and is recognized by a number of integrins, c an be used to promote integrin-mediated adhesion in a concentration depe ndent manner in many cell types. In DCs, integrins are often used as markers to characterize th e tissue specificity and maturation state of DC subsets. However, the functional ro le of DC integrins in mediating immune responses remains largely unexplored. Our goal is to engineer biomimetic cell adhesive substrates to investigate the functional interaction of int egrin receptor binding and other pro-inflammatory / anti-i nflammatory signals that direct DC maturation. Chip Manufacture The chips with RGD-peptide gradient using click chemistry surface functionalization were generated by our colla borators at NIST. Briefly, Universal Substrates were derivitized with azoterminated GRGDS peptide via the highly selective 1,3 Huisgen dipolar cycloaddition reacti on, forming a covalent triazole linkage. This reaction proceeds under aqueous conditions with a high degree of dependability, complete specificity and using reac tants that are bio-compatible ( Figure 4-1).

PAGE 89

89 R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H (A) (B) (C) (D) R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D R G D C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H C O O H (A) (B) (C) (D) Figure 4-1. Gradient SAM substrates can be converted into biomolecule functionalized gradient surfaces with def ined concentration, spatial orientation and complete chemical specificity by coupling an al kyne-terminated linker through the UVoxidation generated carboxyl groups. The steps A to D show different process done to crosslink peptide on the substrate. Dendritic Cell Isolation Dendritic cells were isolated from bone marr ow of 7 week old C57Bl6/j mice using a 10 day protocol. Briefly, bone marrow was is olated from femur and tibia of the mouse. Red blood cells were removed lysed by AC K lysing buffer (Whittaker) and the isolated precursor cells were incubated with DC-media consisting of 20 ng/mL of GM-CSF (R&D Systems), DMEM/F12 (1:1) wi th L-glutamine (Cellgro, Herndon, VA) and 10% fetal bovine serum (FBS) (Hyclone), 1% sodium pyruvate (Lonza, Walkersville, MD) and 1%

PAGE 90

90 non-essential amino acid (Lonza, Walkersville, MD) for 2 days in a flask. The floating cells were collected from the flask at the end of two days and re-seeded with fresh DCmedia in a 6-well low-attachment plate (C orning Inc., NY) for 6 days. After 6 days of culture the cells were re-suspended in fresh media and seeded onto tissue-culture treated 6 well plates (Corning, Inc, NY) for 2 days. After 10 day of culture DCs were lifted using 5mM solution of Na2EDTA (Fisher Scientific) in phosphate buffer saline (PBS) (Hyclone). Dendritic cells thus isolat ed were then tested for immaturity (MHC-II+ < 6% and CD86+ < 6%), purity (using flow cytometery: CD11c+ > 90%) and viability (Trypan blue staining). Dendritic Cell Culture on Chip The chips were washed with PBS and bl ocked with 1% bovine serum albumin (BSA) (Fisher Bioreagents) for 1 h. The chips were then washed with PBS before seeding the DCs. The number of cells on co ntrol chips and the sample chip were maintained the same. Dendritic cells were cultured for 24 h on the chips. The DCs were treated with monensin (BD Biosciences) for the last 5 h before staini ng for intra-cellular cytokines IL-12p40 and IL-10. Immunostaining The chips were stained for either MHC-II and CD86 or IL-10 and IL-12p40. MHC-II and CD86 The control and the sample chips cultured with DCs were washed with PBS before blocking with 4% paraformaldehyde (USB Corpor ation) for 10 min. The chips were then blocked with 10% FBS in PBS for 1 h. An tibodies against major histocompatibility complex (MHC-II) and CD-86 cro ss-linked with phycoerythrin (PE) and fluorescein isothiocyanate (FITC) (Santa Cruz Biotechn ology Inc.) respectively were diluted 1:100

PAGE 91

91 in 1% FBS and utilized to stain the DCs on the same chip. The chips were then washed with PBS and mounted using 4',6-diamidino-2-phenylindole (DAPI) (Santa Cruz Biotechnology Inc.) containing mounting gel. IL-10 and IL-12p40 The DC-cultured chips were washed with PBS and stained with IL-10 and IL12p40 antibodies (Santa Cruz Biotechnology Inc.). The chips were incubated with cytoperm and cytofix buffers (BD Pharmingen) for 30 min. Then the chips were incubated in 10% goat serum for 1 hour for blocking, upon which the cells were stained with primary antibodies IL-10 and IL-12p40. The chips were washed with wash buffer (BD Pharmingen) and stained with respective secondary antibodies for IL-10 and IL12p40. V Staining The DCs were stained for integrin usi ng crosslinking-extraction process. The DCcultured chips were washed 3X with PBS to re move all the media. The chips were then incubated with 1mM solution of sulfo-BSOCO ES (Bioworld) made in PBS for 15 min to crosslink the cells to the chips. The chips were then washed with 1M tris-buffer (Fisher Scientific) twice. The cells were then extracted from off the chip by incubating the chip in 0.1% SDS solution (Fisher Scientific) m ade in PBS for 10 min. The chips were then blocked with 0.1% BSA solution for 1 hr. The chips were stained with primary V antibody (Santa Cruz Biotechn ology Inc.) for 1 hr, followed by biotinylated secondary antibody (Santa Cruz Biotechn ology Inc.) for 1 hr. The ch ips were then incubated in streptavidin conjugated to alkaline phos phatase for 15 min. The chips were then incubated in ELF97 substrate (Invitrogen) prepared using the manufactures protocol for

PAGE 92

92 8 min. The chips were washed in the wa sh buffer provided by the manufacturer and mounted on a microscope glass slide for image analysis. Imaging and Analysis The chips were imaged from 12.55 mm fr om the edge with minimum concentration of -RGD peptide or COOH terminated groups to the maximum concentration. Chips were imaged at 12-bit using FITC, DAPI and Rhodamine (RHOD) filt ers using Axiovert 200M Carl Zeiss inverted fluorescent mi croscope. The images were analyzed using Axiovision 4.0 software. A mask was generated using DAPI channel, by selecting the region with intensity from the edge of the nucleus to 4 m away from the nucleus. This mask was used for images obtained using bot h FITC and Rhodamine filters and the sum fluorescence intensity was calculated for individual DC. Similarly, the mask was utilized to quantify mean background intensity ( Figure 4-2). The following formula was used to quantify the corrected fluo rescence intensity per DC. Corrected Intensity / DC = (Sum fl uorescence Intensity Mean Background intensity) / {Number of DC} {Mean Background intensity} Corrected intensity of individual DC was averaged over every 500 m (breadth) x 100 m (length) area and plotted against the incr easing length which corresponds to the increasing concentration of the RGD gradi ent. A total of 150,000 450,000 DCs were analyzed per chip. The entire length of the chip was analyzed, omitting 12.5 mm from each of the length-ends and 3 mm from breadth-ends to compensate for the edgeeffects and mounting-artifacts.

PAGE 93

93 Increasing RGD-gradient Image Analysis 40 mm Increasing RGD-gradient DC Nucleus 4 m wide circumferential area chosen for analysis (A) FITC DAPI Overlay (C) (B) Figure 4-2. Dendritic cells cultured and immunofluorescence stained on the RGDgradient can be quantified fo r the expression of surface molecules such as MHC-IIstimulatory molecule; CD86-co -stimulatory molecule and intracellular anti-inflammatory cytokine IL10; pro-inflammatory cytokine IL-12p40. A) The schematic of the chip utilized for making the RGD-gradient is shown. B) A representative micrograph of a chip obtained using fluorescent microscope is shown with increasing gr adient of RGD from left to right and DCs stained for surface expression of CD86 with FITC (shown in green) and nuclei with DAPI (shown in blue). C) Image analysis was performed on the micrographs of stained DCs for nuclei and surface markers or intracellular cytokine.

PAGE 94

94 Statistical Analysis The least mean square values and standar d error were obtained using two-way ANOVA via General Linear M odel program, SYSTAT 12, with independent variables as the number of runs and the RGD-peptide su rface density. Additionally, Pearsons Correlation between the fluorescence intensity and the length of the chip corresponding to surface density of RGDpeptide was quantified. Results The DCs were cultured on the chips with RGD-gradient, control chips without RGD and chips in the presence of LPS. Upon 24 h of culture, cells were immunofluorescently stained with either the cell surface molecule s or intra-cellular cytokines. The numbers of cells on the chips were quantified via nuclear staining assuming one nucleus per cell. It was observed that the num ber of cells over 500 m (breadth) x 100 m (length) block of the chip / gradient remained the same with a representative coefficient of determination of R2 = 0.04 for the RGD-gradient chip and R2 = 0.2 for the control chip ( Figure 4-3). Dendritic cells were cultured on the RG D-gradient chip and CD86 cell surface expression was quantified by acquiring im ages using a fluorescent microscope and image analysis. The background corrected fluor escence intensity was plotted against increasing surface density of the RGD-peptide. It was observed that there is a general trend of increasing cell surface expression of CD86 with increasing surface density of RGD peptide. Two-way ANOVA revealed a coefficient of determination, R2 = 0.94 and a p-value < 1.8E-12. Furthermore, DCs cu ltured on the highest analyzed RGD surface density have ~5 times more CD86 cell surfac e expression as compared to the lowest analyzed RGD surface density ( Figure 4-4A).

PAGE 95

95 Length (mm) 07 1421283542 Normalized Cell Number 0 1 2 3 4 Length (mm)(B) RGD-surface density (pmol/cm2) 15.3 36.3 57.3 78.3 99.3 120.3 141.3 0 7 1421283542 Normalized Cell Number 0 1 2 3 4 (A) Linear Fit RGD-gradient y = (0.004.0007) x + (1.320.0155) R = 0.3 Linear Fit Control y = (-0.0050.0003) x + (1.000.006) R = 0.7 Figure 4-3. Dendritic cells cultured on RG D-gradient chips demonstrate that the number of DCs adhering does not have a correlati on with the RGD-gradient or control gradient. A) Number of DCs on RGD gradient B) Number of DCs on control gradient. Dendritic cells were cultured on the cont rol chip and CD86 cell surface expression was quantified. The background corrected fluo rescence intensity was plotted against

PAGE 96

96 increasing surface density of the RGD-peptide that corres ponds to the same distance from the 0mm in both RGD-gr adient chip and control chip. It was observed that the CD86 expression remains constant throughout t he length of the control chip. Dendritic cells cultured on the RGD-gradient chip, at the lowest RGD surface density have 10-20 times more CD86 expressi on and at the highest RGD surface density 50-100 times more CD86 expression of the DCs cultur ed on the control chip. Two-way ANOVA revealed a coefficient of determination, R2 = 0.65 and a p-value < 0.0003. Dendritic cells were cultured in the presence of LPS and CD 86 cell surface expression was quantified. The background corrected fluorescence intensit y was plotted on the same plot as the RGD-gradient chip and control-chip plot as a positive control. It was observed that DCs cultured at the highest concentration of the RGD surface density had CD86 expression comparable to the DC expression of CD86 when cultured in the presence of LPS. The standard error of the CD86 expression of DC s is shown as dashed line where as the solid line represents the least square mean value (Figure 4-4A).

PAGE 97

97 Length (mm) 0 7 1421283542 CD86 (RFI / DC) 0 200 400 600 800 LPS control Linear Fit RGD-gradient: y = (0.780.03) x + (1032) R = 0.8 Control: y = (0.050.002) x + (4.90.2) R = 0.8 Control RGD-gradient RGD-surface density (pmol/cm2) 15 36 57 78 99 120 141 Pearsons Correlation= 0.8 Pearsons Correlation= 0.8(B) (C) RGD Gradient Control RGD-Surface Densit y CD86Increasing -COOH CD86 Figure 4-4. Dendritic cells cultured on RGD-gradi ent chips demonstrat e that surface expression of CD86 co-stimulatory mo lecule is proportional to the RGDgradient. A) CD86 expression by DCs cultured on RGD gradient and control gradient. B) Pearsonss Correlation coefficient for DCs cultured on RGD gradient. C) Pearsonss Correlation coefficient for DCs cultured on control gradient. Dendritic cells were cultured on the RG D-gradient chip and MHC-II cell surface expression was quantified by acquiring im ages using a fluorescent microscope and image analysis. The background corrected fluor escence intensity was plotted against (A)

PAGE 98

98 increasing surface density of the RGD-peptide. It was observed that there is a general trend of increasing cell surface expression of MHC-II with increasing surface density of RGD peptide. Two-way ANOVA revealed a coefficient of determination, R2 = 0.97 and a p-value < 2.5E-12. Furthermore, DCs cu ltured on the highest analyzed RGD surface density have ~6 times more MHC-II cell surf ace expression as compared to the lowest analyzed RGD surface density ( Figure 4-5A). Dendritic cells we re cultured on the control chip and MHC-II cell surface ex pression was quantified. The background corrected fluorescence intensity was plotte d against increasing surface density of the RGD-peptide that corresponds to the same distance from the 0mm in both RGDgradient chip and control chip. It was obser ved that the MHC-II expression remains constant throughout the length of the contro l chip. Dendritic cells cultured on the RGDgradient chip, at the lowest RGD surf ace density have 6-25 times more MHC-II expression and at the highest RGD surface density 40-150 times more MHC-II expression of the DCs cult ured on the control chip. Two-way ANOVA revealed a coefficient of determination, R2 = 0.62 and a p-value < 0.02. Dendritic cells were cultured in the presence of LPS and MHC-II ce ll surface expression was quantified. The background corrected fluorescence intensity was plotted on the same plot as the RGDgradient chip and control-chip plot as a positive control. It was observed that DCs cultured at the highest concentration of t he RGD surface density had MHC-II expression higher than DC expression of MHC-II when cultured in the presence of LPS. The standard error of the MHC-II expression of DCs is shown as dashed line where as the solid line represents the least square mean value ( Figure 4-5A).

PAGE 99

99 Length (mm) 0 7 1421283542 MHC-II (RFI / DC) 0 200 400 600 800 Control RGD-gradient LPS control Linear Fit RGD-gradient: y = (0.84.03) x + (1002.6) R = 0.8 Control: y = (0.040.002) x + (50.2) R = 0.8 RGD-surface density (pmol/cm2) 15.3 36.3 57.3 78.3 99.3 120.3 141.3 Pearsons Correlation= 0.8 Pearsons Correlation= 0.8(B) (C) R G DG r ad i e n t Co n t r o l Figure 4-5. Dendritic cells cultured on RGD-gradi ent chips demonstrat e that surface expression of MHC-II st imulatory molecule is proportional to the RGDgradient. A) MHC-II expression by DCs cultured on RGD gradient and control gradient. B) Pearsonss Correlation coefficient for DCs cultured on RGD gradient. C) Pearsonss Correlation coefficient for DCs cultured on control gradient. Dendritic cells were cultured on the RG D-gradient chip and IL-10 intracellular cytokine production was quantified by acquiri ng images using a fluorescent microscope and image analysis. The background corrected fluorescence intensity was plotted against increasing surface density of the RGDpeptide. It was observed that there is a (A)

PAGE 100

100 general trend of constant IL-10 cytokine production with variable surface density of RGD peptide. Two-way ANOVA revealed a coefficient of determination, R2 = 0.01 and a pvalue > 0.99 ( Figure 4-6A). Dendritic cells were cultured on the control chip and intracellular cytokine IL-10 production was quantified. 07 1421 28 3542 IL-10 (RFI / DC) 0 1000 2000 3000 4000 5000 LPS control Control RGD-gradient Linear Fit RGD-gradient: y = (6.50.16) x + (52314) R = 0.9 Control: y = (-0.010.0009) x + (4.10.07) R = 0.6 Length (mm) RGD-surface density (pmol/cm2) 15.3 36.3 57.3 78.3 99.3 120.3 141.3 Pearsons Correlation=0.9 Pearsons Correlation=-0.6 (B) (C) RGD Gradient Control RGD-Surface Densit y IL-10Increasing -COOH IL-10 Figure 4-6. Dendritic cells cultured on the RGD-gr adient chips and the control chips for 24 h, were stained for the intr acellular cytokine IL-10 using immunofluorescence staining and im age analysis was performed to quantify the fluorescence intensity at varying RGD surface density present on the chip. A) IL-10 expression by DCs cultured on RGD gradient and control gradient. (A)

PAGE 101

101 B) Pearsonss Correlation coefficient for DCs cultured on RGD gradient. C) Pearsonss Correlation coefficient for DCs cultured on control gradient. The background corrected fluorescence intensity was plotted against increasing surface density of the RGD-pept ide that corresponds to t he same distance from the 0mm in both RGD-gradient chip and contro l chip. It was observed that the IL-10 expression remains constant throughout the l ength of the control chip. Two-way ANOVA revealed a coefficient of determination, R2 = 0.55 and a p-value > 0.99. Dendritic cells were cultured in the presence of LPS and IL -10 cell surface expr ession was quantified. The background corrected fluorescence intensit y was plotted on the same plot as the RGD-gradient chip and control-chip plot as a positive control. It was observed that DCs cultured at the highest concentration of t he RGD surface density had IL-10 expression lower by 7-9 than DC expression of IL-10 wh en cultured in the presence of LPS. The standard error of the IL-10 expression of DCs is shown as dashed line where as the solid line represents the least square mean value ( Figure 4-6A). Dendritic cells were cultured on the RG D-gradient chip and IL-12p40 intracellular cytokine production was quantified by acquiri ng images using a fluorescent microscope and image analysis. The background corrected fluorescence intensity was plotted against increasing surface density of the RGDpeptide. It was observed that there is a general trend of increasing of IL-12p40 produc tion with increasing surface density of RGD peptide. Two-way ANOVA revealed a coefficient of determination, R2 = 0.76 and a p-value < 7.25E-12. Furthermore, DCs cu ltured on the highest analyzed RGD surface density have ~3 times more IL-12p40 cytokin e production as compared to the lowest analyzed RGD surface density ( Figure 4-7A). Dendritic cells we re cultured on the

PAGE 102

102 control chip and IL-12p40 cytokine production was quantified. The background corrected fluorescence intensity was plotte d against increasing surface density of the RGD-peptide that corresponds to the same distance from the 0mm in both RGDgradient chip and control chip. It was obs erved that the IL-12p40 cytokine production remains constant throughout the length of t he control chip. Dendritic cells cultured on the RGD-gradient chip, at the lowest RGD surface density have similar IL-12p40 expression and at the highest RGD surface density ~6 times more IL-12p40 cytokine production of the DCs cultur ed on the control chip. Two-way ANOVA revealed a coefficient of determination, R2 = 0.55 and a p-value < 0.0001. Length (mm) 0 714 212835 42 IL-12p40 (RFI / DC) 0 1000 2000 3000 4000 5000 6000 Control RGD-gradient LPS control Linear Fit RGD-gradient: y = (16.4) x + (15035) R = 0.9 Control: y = (0.020.001) x + (1.60.1) R = 0.6 RGD-surface density (pmol/cm2) 15.3 36.3 57.3 78.3 99.3 120.3 141.3 (A)

PAGE 103

103 Pearsons Correlation=0.6 Pearsons Correlation=0.1 (B) (C) RGD Gradient Control RGD-Surface Densit y IL-12p40Increasing -COOH IL-12p40 Figure 4-7. Dendritic cells cultured on RGD-gradient chips demonstrate t hat intracellular pro-inflammatory cytokine IL-12p40 is proportional to the RGD-gradient. A) IL-12p40 expression by DCs cultured on RGD gradient and control gradient. B) Pearsonss Correlation coefficient for DCs cultured on RGD gradient. C) Pearsonss Correlation coefficient for DCs cultured on control gradient. Dendritic cells were cultured in the presence of LPS and IL-12p40 cell surface expression was quantified. The background corre cted fluorescence intensity was plotted on the same plot as the RGD-gradient chip and control-chip plot as a positive control. It was observed that DCs cultured at the highest concentration of the RGD surface density had IL-12p40 expression 3-5 times lo wer than DC production of IL-12p40 when cultured in the presence of LPS. The standard error of the IL-12p40 production by DCs is shown as dashed line where as the solid line represents the least square mean value ( Figure 4-7A). Dendritic cells were cultured on the RGDgradient chip and surface expression of V integrin was quantified by acquiring imag es using a fluorescent microscope and image analysis. The background corrected fluor escence intensity was plotted against increasing surface density of the RGD-peptide. It was observed that the V integrin expression remains low from 20-50 pmol/cm2 of RGD-surface density and then the expression increases linearly with the in crease in RGD-surface density. Two-way ANOVA revealed a coeffici ent of determination, R2 = 0.78 and a p-value < 3.6E-12.

PAGE 104

104 Furthermore, DCs cultured on the highest analyzed RGD surface density have ~30 times more V integrin expression as compared to the lowest analyzed RGD surface density ( Figure 4-8A). Dendritic cells were cultured on the control chip and surface expression of V integrin was quantified by acquiring images using a fluorescent microscope and image analysis. The backgro und corrected fluorescence intensity was plotted against increasing surface density of the RGD-peptide that corresponds to the same distance from the 0mm in both RG D-gradient chip and control chip. It was observed that the V integrin expression remains low throughout the RGD-surface density. Two-way ANOVA revealed a coefficient of determination, R2 = 0.01 and a pvalue > 0.7 ( Figure 4-8A). Length (mm) 07 14212835 42 V (RFI) 0 500 1000 1500 2000 2500 3000 Linear Fit RGD-gradient: y = (10.2.2) x + (-3303) R = 0.9 Control: y = (-0.010.001) x + (6.60.8) R = 0.07 RGD-surface density (pmol/cm2) 15.3 36.3 57.3 78.3 99.3 120.3 141.3 Control RGD-gradient Threshold Value (A)

PAGE 105

105 Pearsons Correlation=0.9 Pearsons Correlation=-0.1 (B) (C) RGD Gradient Control RGD-Surface Densit y V Increasing -COOH V Figure 4-8. Dendritic cells demonstra te increased integrin V expression with increase in RGD-peptide surface density. A) RGD bound V integrin expression by DCs cultured on RGD gradient and control gradient. Arrow shows the threshold value upon which the RGD bound V expression is different from the control. B) Pearsonss Correlation coefficient for DCs cultured on RGD gradient. C) Pearsonss Correlation coefficient for DCs cultured on control gradient. Impact of the Study Overall, this study demonstrates that the maturation of DCs based on surface density of a peptide can be predicted and controlled for in vitro experiments, thus, enabling us to develop systems where effects of certain immunological active components on matured DCs can be studi ed in a highly controlled manner. Furthermore, the results obtained through these studies can be directly correlated with the traditional immunological technique of fl ow cytometry, since, the basic principles behind analyzing the cells were maintained the same. This study lays groundwork to study the effects of other adhesion dependent DC-maturation and effects of immunologically relevant drugs and other immune system modulati ng components. Furthermore, we have demonstrated that the DC maturation can be modulated by highly controlled adhesion cues. We have developed a system where the extent of DC maturation can be controlled and quantified by image analysis on per cell basis. We

PAGE 106

106 have shown that the DC surface expression of MHC-II and CD86 is directly proportional to the surface density of RGD-peptide. Interestingly, it was observ ed that the increasing surface density of RGD-peptide resulted in increasing pro-inflammatory cytokine IL12p40 production and the anti-inflammatory cyt okine IL-10. This suggests that the extent of adhesion cues is an important fa ctor in the maturation pathway of DCs and should be further studied. This work will pr ovide incentives and tools in improving the applications of ex vivo culture of DCs and implanted biom aterials from immunological perspective.

PAGE 107

107 CHAPTER 5 A HIGH-THROUGHPUT MICROPARTICL E MICROARRAY PLATFORM FOR DENDRITIC CELL-TARGETING VACCINES Introduction Immunogenomic approaches combined with advances in adjuvant immunology are guiding progress toward rational desi gn of vaccines. Furthermore, drug delivery platforms (e.g., synthetic particles) are demonstrating promise for increasing vaccine efficacy. Currently there are scores of known antigenic epit opes and adjuvants, and numerous synthetic delivery systems accessible for formulation of vaccines for various applications. However, the lack of an efficient means to test immune cell responses to the abundant combinations available represents a significant blockade on the development of new vaccines. In order to over come this barrier, we report fabrication of a new class of microarray consisting of ant igen/adjuvant-loadable poly (d,l lactide-coglycolide) microparticles (PLGA MPs), identified as a promising carrier for immunotherapeutics, which are co-localized with dendr itic cells (DCs), key regulators of the immune system and prime targets for vaccines. The intention is to utilize this highthroughput platform to optim ize particle-based vaccines designed to target DCs in vivo for immune system-related disorders, su ch as autoimmune diseases, cancer and infection. Preparation of PLGA Microparticles A 50:50 polymer composition of poly(d,l lactide-co-glycolid e) (PLGA) with inherent viscosity 0.55-0.75 dL/g in hexafluoroisopropanol, HFIP (Lactel, AL, USA) was used to generate microparticles. Poly-vinyl alcoho l (PVA) (MW ~ 100,000 g/mol) was purchased from Fisher Science (Rochester, NY, U SA) and was used as an emulsion stabilizer. Phosphate buffered saline (PBS) solution (Hyclone, UT, USA) was used as the aqueous

PAGE 108

108 phase to form the emulsions while methylene chloride (Fisher Scientific, NJ, USA) was used as an organic solvent to dissolve PLGA polymer. Microparticles were formed using a standard water-oil-water solvent evaporation technique. Briefly, the PLGA polymer was di ssolved in methylene chloride at 20% concentration. Rhodamine, a red-fluorescent dye (RHOD) (Sigma-Aldrich), fluorescein isothiocyanate, green fluorescent dye (FITC) (Sigma Aldrich) or 9-anthracenecarboxylic acid, blue fluorescent dye (ACA) (Sigma-Aldrich) solution (100 L of 5 mg/mL in PBS) was emulsified with 1000 L of 20% PLGA solution at 26,500 rpm for 60 seconds using a tissue-miser homogenizer (Fisher Scientific, NJ, USA) to form a primary emulsion. The primary emulsion was added to 2 mL of 9% PVA solution in PBS and the homogenizing was continued at 19,500 rpm for 60 seconds to form the secondary emulsion. Then, the secondary emulsion was added to 7 mL of 9% PVA solution. The particles thus formed were agitated using a m agnetic stirrer (Fisher Scientific, NJ, USA) for 3 hours to evaporate residual methylene chloride. The remaining solution was centrifuged at 10,000 x g fo r 10 minutes to collect MPs which were subsequently washed three times with PBS. The PBS was as pirated from the centrifuged MPs, which were then flash-frozen in liquid nitrogen and kept under vacuum in dry ice overnight. The MPs were stored at -200C until used. Bovine serum albumin (BSA) (Fisher Bioreagents) was used as a representative protein to quantify encapsulation efficiency. Bovine serum albumin was encapsulated in MPs via solvent evaporation technique by incorporating 1mg BSA in the aqueous pha se of the primary emulsion.

PAGE 109

109 Characterization of Microparticles Particle Size Measurements Particle size was characterized by dynamic light scattering technique using a Nanotrac (Microtrac Inc.) particle size analyzer. A total of 35 mg of particles were resuspended in 50 mL of de-ionized (DI) wate r via sonication in a sonicating bath for 2 minutes (Branson 2510, Paragon Electronics, FL).. The laser probe was dipped in the particle suspension and the laser scatte ring data was collected for 3 minutes. Scanning Electron Microscopy Microparticle morphology was characterized by scanning electron microscope (FEG-SEM JEOL JSM 6335F, Major Analytical Instrumentation Center, University of Florida). A particle suspension in DI water was used to print particles and dried for 16 hours at room temperature. Dried particles were then coated with 5-10 nm thickness of gold and imaged at magnifications r anging from 30x to 60,000x. Efficiency of Protein Encapsulation A known weight of MPs loaded with BSA was dissolved in methylene chloride. An equal amount of PBS was added to the solu tion and sonicated for 5 minutes. The emulsion thus obtained was t hen centrifuged at 10,000 x g for 10 minutes to separate the two phases. Phosphate buffered saline, c ontaining the water-soluble BSA, was then very carefully pipetted out and saved. E qual amount of PBS was replaced in the solution and centrifuged. This process was repeated 4 times and the PBS pooled for analysis. The concentration of BSA protein in the solution was then quantified by measuring the absorbance at a wavelength of 280 nm by spectr ophotometer (Nanodrop Technologies Inc., DE, USA) and compar ing absorbance to a standard curve made from known concentration of BSA in PBS.

PAGE 110

110 Degradation of Microparticles The blue fluorescent dye, ACA incorporated MPs were printed into a flat bottom tissue culture treated 96-well plate (Corning Inc ., NJ, USA), in an array format of 1 spot per well. Phosphate buffered saline, 1N HCl and 1M NaOH were used to make solutions of pH 4.0, 5.0, and 7.4 and 13.1 usi ng glass pH electrode (Fisher Scientific). Microparticle arrays were incubated with these three solutions and fluorescent micrographs were taken at different times to identify the loss of fluorescence. The experimental conditions were set to minimize any photo-bleaching effect, by keeping the samples mounted on the microscope in a dark room. Intensity of th e signal (here, ACAloaded MPs) was corrected from the backg round fluorescence intensity by using the formula for corrected intensity: (Signal Background) (Background) The background-corrected fluorescence intensit y measured at each time-point was divided by the background-corrected fluorescence int ensity measured at time t = 0, and the multiplied by 100 to be expressed as a perce ntage. The ratio of fluorescence intensity was plotted against time and the data was fitt ed to the curve using Dynamic Fit Wizard SigmaPlot. Particle Array Fabrication Glass coverslips (22 x 22 mm2, Fisher scientific) were cleaned in an oxygen plasma etcher (Terra Universal, CA, USA) for 6 minutes. An arra y of (3-Aminopropyl) trimethoxysilane (NH2-terminated silane) (Sigma-Ald rich) as obtained from the manufacturer was diluted in DI water. The silane solution was printed on clean coverslips using a Calligrapher Miniarrayer (BioRad) contact printer, with 400 m diameter solid metal pin. The printed covers lips were then coated with 150 of titanium (Ti-99.995% pure) followed by 150 of gold (Au-99.999% pure) (Williams Advanced

PAGE 111

111 Materials, IL, USA). Gold coated coverslips were then sonicated in 70% ethanol in DI water for 15 minutes to remove gold coat ing from over the printed islands exposing NH2-terminated silane arrayed spots, while l eaving the gold coating intact around the islands for further processing ( Figure 5-1A ). The coverslip was then washed with DI water without letting it dry, to ensure the lifted gold was re moved. The coverslips were dried by blowing nitrogen gas on the coverslip. The coverslips were then incubated with 0.01 M, methyl-terminated alkanethiol (CH3(CH2)11SH, Sigma-Aldrich) for 1 hour followed by washing with 200 proof ethyl alcohol (Fisher Scientific). Substrates were then incubated in 10% pluronic F-127 (BASF Corporation, USA) in DI water, for 4 hours to render the surface around the islands cell -resistant. The coverslip was washed with DI water and re-suspended MPs of PLGA in PBS were over-printed on the exposed islands ( Figure 5-1B). Microparticles were printed using a pin of 100, 200 or 400 m diameter. The MPs on the islands were allowed to dry comple tely by incubating them at room temperature under vacuum for 30 min. The coverslip was then rinsed with DI water. Micrographs were obtained using MosaiX module of Axiovision software (CarlZeiss). The fluorescently-labeled MPs we re counted using the Automatic Object Measurement Program, Axiovision software. To design a printed array of a dilution of number of single fluorescent dye encapsul ated MPs, we re-suspended the desired MPs in 1 mL of DI-water via combination of vortex mixing and sonication. The MPconcentration was verified by taking micrographs of fluorescent-MPs on a hemocytometer and utilizing Automatic obj ect Measurement Program, Axiovision software. A serial 1:2 dilution of MPs in 20 L DI-water was generat ed, with a starting concentration of 3.26 x 106/mL.

PAGE 112

112 Figure 5-1. Microparticle/dendritic cell (M P/DC) array fabricat ion incorporates miniarraying solid pin contact pr inting equipment, silane and alkanethiol surface chemistry, and physisorption of MPs to provide MP/DC co-localization on isolated spots. A.) Surface-engineering MP array. B.) Cross-section of a Dendritic Cells Microparticles encapsulated drug-formulations (C) MP/DC Array (A) MP Array Fabrication 6. Particles overprinted onto NH2 silanespots 1 Print arrays of NH2 silane spots O2 plasma cleaned glass Miniarrayer Printing Pin Exposed spot of NH2-terminated silane Non-fouling PEG background PEG-based Pluronic F-127 adsorbed onto CH3alkanethiol monolayer provides nonadhesive background Au layer provides off-spot background substrate for CH3 alkanethiol monolayer PLGA Microparticles Physisorbed onto NH2-silane (B) Surface Chemistry 2. Deposit Ti/Au 3. Sonicate to lift metal deposited over NH2 silane spots 4. Assemble CH3 alkanethiol monolayer on Au 5. Adsorb Pluronic F-127 block Gol Titaniu Gla

PAGE 113

113 single spot in a MP-array illustrating physisorbed MPs on NH2-terminated spots with a polyethylene glycolbased non-adhesive background surface chemistry (not to scale). C.) Dendritic cells are seeded on MP arrays, selectively adherent to NH2-terminated spots providi ng co-localized DCs/MPs arrays. These suspensions were added into separate wells in a 384-well plate which was used as the source plate to over-spot MPs onto the arrayed NH2-terminated substrate. Fluorescence micrographs of the entire array were obtained and the number of MPs on each island counted using Axiovision software. In order to generate orthogonal ly-overlapping dilution arrays of MPs encapsulated with different fluorescent dyes, ACAloaded or RHOD-loaded MPs were re-suspended in DI-water separately and their individual MP-concentrations determined. Sixteen different combinatorial mixtures of RHOD-loaded and ACA-loaded MPs were generated in a 384-well source-plate using 3 different di lutions of each type of MP (final volume 40 L, final concentrations RHOD: 4.5 x 106/mL, 8.8 x 106/mL and 19.7 x 106/mL; ACA: 4.0 x 106/mL, 7.9 x 106/mL and 15.8 x 106/mL). These mixed MP solutions were then over-printed onto NH2-terminated islands. In order to minimize cross-contamination of samples, the pin was washed 3 times in a solution of 0.05 % Tr iton X-100 surfactant followed by washing in water and then drying under vacuum before printing each MPspot. Fluorescence micrographs were obt ained using a 10x objective for both Rhodamine and DAPI filters and AxioVision MosaiX software was utilized to align and compile individual micrographs. The num ber of RHOD-loaded and ACA-loaded MPs on each spot was quantified by using the Axio Vision Automatic Measurement Program. A MATLAB routine was compiled to quantitativ ely display the resulting mean and standard deviation of 4 replicate arrays fo r each spot in the 4 x 4 arrays.

PAGE 114

114 Dendritic Cell Culture and Staining Dendritic cells were isolated from bone marr ow of 7 week old C57Bl6/j mice using a 10 day protocol. Briefly, bone marrow was is olated from femur and tibia of the mouse. Red blood cells were removed lysed by AC K lysing buffer (Whittaker) and the isolated precursor cells were incubated with DC-media consisting of 20 ng/mL of GM-CSF (R&D Systems), DMEM/F12 (1:1) wi th L-glutamine (Cellgro, Herndon, VA) and 10% fetal bovine serum (Bio-Whittaker), 1% sodium pyruvate (Lonza, Walkersville, MD) and 1% non-essential amino acid (Lonza, Walkersville, MD) for 2 days in a flask. The floating cells were collected from the flask at the end of two days and re-seeded with fresh DCmedia in a 6-well low-attachment plate (C orning Inc., NY) for 6 days. After 6 days of culture the cells were re-suspended in fresh media and seeded onto tissue-culture treated 6 well plates (Corning, Inc, NY) for 2 days. After 10 day of culture DCs were lifted using 5mM solution of Na2EDTA (Fisher Scientific) in PBS. Dendritic cells thus isolated were then tested for immaturity (MHC-II+ < 6% and CD86+ < 6%), purity (using flow cytometery: CD11c+ > 90%) and viability (Trypan blue staining). A dried MP-arrayed coverslip was rinsed with PBS and re-hydrated for 30 minutes with PBS-mix containing 0.05 g/L of m agnesium chloride and 0.05 g/L of calcium chloride made by mixing equal parts of PBS with magnesium and calcium and PBS without magnesium and calcium (PBS50-50), before seeding immature DCs. Dendritic cells were seeded in a serum-free media in PBS50-50 for 15 minutes at 37 0C. Dendritic cells were then rinsed with PBS having calcium and magnesium to wash-off nonadherent cells. Fresh DC-m edia was added onto the covers lip and adherent DCs were cultured for 24 hours. Dendritic cells were fixed with, 3.7% paraformaldehyde (USB Corporation, USA) for 10 minutes and then washing with PBS to remove excess fixing

PAGE 115

115 agent. The nuclei of DCs were then stained wi th 10 mg/mL Hoechst (Invitrogen, USA) diluted to 1:10,000 times as per the manufacturers protocol. Fluorescence and phasecontrast micrographs of adherent cells on the islands were obtained using a 10x objective and compiled using MosaiX module of Axiovision software used for stitching the acquired images. A schematic of cells cultured on the arrays is described ( Figure 51C). Statistical Analysis Statistical analyses were performed using either a one-way ANOVA or a two-way ANOVA, using Systat (Version 12, Systat Software, Inc., San Jose, CA). Pair-wise comparisons were made using Tukeys HonestlySignificant-Difference, with p-values of less than or equal to 0.05 considered to be signi cant. Results Array Fabrication and Micr oparticle Characterization The aim of constructing particle arrays wa s to develop an enabling platform to test and optimize MPs with multi-parameter combinatorial formulations for immunotherapies in a high-throughput format. Using a standard solid-pin miniarray contact printer, we fabricated arrays of NH2-terminated spots with a polyethylene glycol-based nonadhesive background ( Figure 5-1 ). Miniarraying NH2-terminated silane onto clean glass allows the silane molecules to covalently bond to the glass while the drying step leaves a film spot of excess non-ligated silane. These spotted films were measured to be ~13 m thick and ~400 m in diameter on the surface of t he glass, and prevent formation of a continuous metal coating layer during t he Ti/Au deposition step. We took advantage of this surface non-uniformity and removed the Ti/Au layer deposited onto silane spots by sonication in ethanol, leaving behind the covalently-bound NH2-terminated silane islands

PAGE 116

116 surrounded by Au background. This background Au surface was then made hydrophobic by the assembly of a monolayer of CH3-terminated alkanethiol, in preparation for the adsorption of Pluronic F-127. Pluronic F127 is a tri-block copolymer whose central hydrophobic block facilitates adsorption to hydrophobic surfaces, while flanking blocks of polyethylene glycol provide resistance to cell adhesion. Note that altering the NH2-terminated silane spot area may be de sirable for specific applications and can be accomplished as required by using different pin diameters. While unmodified PLGA MPs were used in the present study, this approach is generalizable to investigate other particles eit her of synthetic or natural origin, including viruses. Standard water-oil-water solvent evaporation technique was used to generate PLGA MPs and MP properties of particle si ze, surface morphology and degradation rate were investigated. Microparticle size distri bution was characterized using dynamic light scattering analysis, and the average parti cle diameter was found to be 1.08 m, with diameters ranging from 600 nm to 2 m ( Figure 5-2). 0 2 4 6 8 10 12 14 16 18 100 1000 10000Size (nanometers) % Volume 0.76 m 1.97 m Figure 5-2. A typical size distribution curve of PLGA-microparticles quantified via dynamic light scattering analysis (based on volume estimation) demonstrates an average size of particles of 1.08 m. A bimodal poly-dispersity in the particle size is observed; with t he smaller population-set has an average

PAGE 117

117 diameter of 0.76 m, whereas the larger si ze population-set has 1.97 m average diameter. A bimodal poly-dispersity in particle size was observed, with the smaller-sized population having an average size of 0.76 m, and the larger-sized population had an average diameter of 1.97 m. While particle size is ex pected to affect cellular interactions, for the purpose of array validation, MPs were used as made without a filtering step to further restrict particle size. Next, we investigated pH-dependent parti cle degradation. This is of interest because acidic degradation reflects a majo r mechanism by which particles can be physiologically degraded. Phagocytic cells ta ke up particles into phagosomes which fuse with lysosomes, forming phagolysosome s where an acidic ~5 is maintained to promote degradation. Particle degradation was monitored under different pH conditions by quantifying the fluorescence of printed ACA-encapsulated MPs over a period of 12 hours. The mean and standard error of at least 5 data points are plotted ( Figure 5-3 ). The acquired data was fitted to a threeparameter exponential dec ay equation of the form: y = y0 + A*e-B*t, with coefficient of determination of R2 > 0.94 for all conditions over the time-frame examined. T he variable t represents the time at which fluorescence intensities were quantified. The variabl e y represents the background-corrected fluorescence intensity measured at time t, divided by the background-corrected fluorescence intensity measured at time t = 0, multiplied by 100 to be expressed as a percentage. The parameter y0 represents the saturation va lue, while the parameter A represents the difference betw een initial and saturation va lues, and the parameter B represents the decay constant. From 0 h, all conditions ex amined demonstrated similar burst-release degradation. After 2 h, pH-dependent degr adation was apparent.

PAGE 118

118 Specifically, at 12 h, MPs incubated at pH 7.4 (representing averag e physiological pH) demonstrated a degradation of 10%, whereas MPs incubated at pH 4.0 and pH 5.0 demonstrated a degradation of ~20%. Additi onally, at pH 13.1 PLGA-MPs degraded completely within 5 minutes (data not shown). The degradation properties of the PLGAMPs revealed that degradation in the acid ic pH 5.0 (phagolysosomal pH ~5) is accelerated as compared to physiological ex tracellular pH 7.4, suggesting that DCs can degrade the MPs in the phagolysosomes and analyze MP-encapsulated immunomodulatory molecules effectively. Time (hr) 0 2 46 8 101214 [ RFI(t) / RFI(t=0) ] 100 75 80 85 90 95 100 pH 7.4 pH 5.0 pH 4.0 Figure 5-3. Surface-adsorbed PLGA-microparticles degrade in a pH-dependent fashion and are quantified in situ by image analysis. Fabricated MPs were over-printed onto NH2-terminated spots via solid-pin miniarray contact printing. Microparticle spot sizes were varied in order to present optimal co-localization with cells while pr oviding for alignment error that can occur during over-printing. Microparticles were over-p rinted with different pin diameter sizes of

PAGE 119

119 100, 200 and 400 m to provide different MP spot sizes and examined by scanning electron microscope (SEM), ( Figure 5-4A) 20 X 20 X 180 X 130 X 180 X 10000 X 20 X Pin Diameter = 100 m Pin Diameter = 200 m Pin Diameter = 400 m (A) (B) Figure 5-4. Scanning electron micrographs demonstrat e constructed arrays of surfaceadsorbed PLGA MPs (microparticles) on NH2-terminated silane spots (visible as circular regions devoid of metal deposition). A.) Overspotting pin diameter is optimized for aligned delivery of MPs. B.) Representative micrograph of MPs printed on the adhesive-islands indicating MP smooth surface morphology and spherical shape (scale bar = 5 m).

PAGE 120

120 We determined that the 200 m diameter pin provided the largest MP spot diameter that simultaneously prevented ove r-lapping of MPs onto the surrounding gold resulting from pin-alignment error. Generating arrays of up to 12 x 12, a maximum alignment displacement of ~20 m was determined, while over-printing across a distance of 12 mm. Over-spot alignment error was observed to be random, without systematic misalignment, which allows construction of large high-fidelity arrays. Scanning electron microscope micrographs of MP arrays revealed that MPs were spherical in shape and had a smooth su rface morphology, suggesting the array fabrication process did not adversely affect the particle morphology ( Figure 5-4B). Furthermore, MPs were well-distributed throughout the printed spots and the Au deposited onto spots was completely remov ed, leaving no trace of gold flakes. Lastly, in order to validate the PLGA MPs as a carrier vehicle, the loading of watersoluble protein, bovine serum albumin (BSA) was quantified. Bovine serum albumin was incorporated in the water phase of the primary emulsion of the solvent extraction/evaporation proce ss. Bovine serum albumin was recovered by solvent extraction from dry PLGA-MPs and quantif ied by absorbance spectrophotometry. Given the initial BSA loading am ount of 1 mg, an encapsulat ion efficiency of 62% was obtained, providing 0.2 mg of BSA per mg of PLGA polymer. These loading amounts are in good agreement with literature. Particle Array Validation Next, we were interested in determining the sensitivity of the del ivery of MPs to the MP suspension concentration in the source plate. The number of fluorescent dyeloaded MPs arrayed onto NH2-terminated spots was quantified by fluorescence

PAGE 121

121 microscopic image analysis, and the relati onship between the numbers of printed particles to the particle concentration in t he source-plate was determined. To achieve this, different concentrations of RHOD-encap sulated MP suspensions were prepared by 1:2 serial dilutions using an MP starting concentration of 3.3 x 106/mL, and printed onto NH2-terminated silane spots. The numbers of MP s printed onto islands from the three highest source plate concentrations were distin ct from each other, as well as distinct from all other conditions. A maximum st andard deviation of 6 MPs delivered was determined, with the highest MP source pl ate concentration demonstrating the lowest standard deviation of 1 MP. While the deliv ery error in the system did not allow printing MPs in distinct quantities using a source plate concentration below 0.41 x 106/mL, the number of printed MPs on the islands revealed a linear relationship (y = 21.3x 0.1, R2 = 0.99) to the concentration of the MP suspension in the source plate over the entire dilution range ( Figure 5-5 ** with all other conditions; with 3.3 x 106/mL, 1.6 x 106/mL & 0.8 x 106/mL ). Notably, these data demonstrate the ability to reproducibly deliver a lower limit of 7 2 MPs per spot.

PAGE 122

122 1:2 1:4 1:8 1:16 1:1 1:32 1:64 1:2 1:4 1:8 1:16 1:1 1:32 1:64 (B) 0.0 0.5 1.01.52.0 2.5 3.03.5 0 20 40 60 80 Source plate particle concentration (x 10 6 /mL) (A) **y = 21.3x 0.1 R2 = 0.99 ** **# particles per spo t Figure 5-5. Delivery of microparticles (MPs) by solid pin microarray printing is controlled by source plate particle suspension concentration to deliver a lower limit of 16 2 surface-adsorbed MPs per spot. A.) Representative fluorescence micrograph of serial 1:2 dilutions of rhodamine dye-loaded PLGA MPs printed in quadruplet in a 4 x 8 array format is shown (for 4 separate preparations; scale bar = 200 m.). B.) Surface-adsorbed MP numbers are quantified by image analysis, average delivered MP values with standard deviations are plotted as a function of source plate concentration, and linear fit parameters are shown. The significant pair symbols are described in the text. In order to demonstrate the ut ility of this system to quant itatively investigate multiparameter MP combinations, we constructed a dosing array consisting of orthogonallyoverlapping dilutions of RHOD-loaded and ACA-loaded MPs. The source plate contained 16 different MP mixtures comprisi ng all possible combinations of 3 serial dilutions of RHOD-loaded MPs and 3 serial dilutions of ACA-loaded MPs. These combinatorial mixtures were then printed in 4 x 4 arrays, in the following configuration, where 0 represents the null dos e (no MPs) and increasing in tegers represent increasing

PAGE 123

123 3,32,31,30,3 3,22,21,20,2 3,12,11,10,1 3,02,01,00,0MP concentration (1 minimu m, 3 maximum), and (x,y) pairs represent (ACA, RHOD) MP loading conditions. { } Fluorescence micrographs were co mpiled of the printed arrays ( Figure 5-6). The number of RHOD-loaded and ACA-loaded MPs on each spot was quantified, and a MATLAB routine was compiled to quantitativ ely display the resulting mean and standard deviation on a per spot basis from 4 replicate arrays ( Figure 5-7A, Figure 5-7B ). Additionally, the number of RHOD-loaded and ACA-loaded MPs printed on spots was plotted against the concentration of MPs in the source plate (Figure 5-7C ). The data was then fit by linear regression and the resu ltant slopes for each MP formulation were compared. Potential fo rmulation-specific differences in these slopes would reflect differences in MP delivery. For example, it is possible that differ ential loading of organic phase-soluble molecules (i.e., fluorescent dyes) could modulate MP surface properties, thereby affecting delivery. Th is is an important considerat ion, because our goal is to

PAGE 124

124 provide precise, reproducible MP delivery, independent of formulation. (0,0) (1,0) (2,0) (3,0) (0,1) (1,1) (2,1) (3,1) (0,2) (1,2) (2,2) (3,2) (0,3) (1,3) (2,3) (3,3) Increasing number of ACA encapsulated particles Increasin g number of RHOD encapsulated particles Figure 5-6. Fluorescence micrograph of diffe rent microparticle (MP) formulations quantitatively printed using solid pin c ontact printing in different MP-dose combinations with minimal cross-contamination. Linear regression analysis of the raw data from 4 replicate arrays revealed average slopes of 18 0.9 and 16 0.9, for RHOD-loaded and ACA-loaded MPs, respectively. Statistical analysis via Student's t-test determined t hat these slope-data sets are statistically indistinct from each other (p-value > 0.14). This indicates that quantitative delivery of these organic phase-loaded MP formulations was achieved, independent of formulat ion. For PLGA MPs, it is impor tant to note that unlike organic

PAGE 125

125 phase-soluble molecules, water-soluble mole cules are dispersed in submicron-sized water droplets throughout the MP polymer ma trix, and would therefore be expected to alter MP surface properties to a lesser extent than oil-soluble molecules. These findings suggest broad applicability of this approach towa rd the investigation of a wide range of MP loading conditions. # ACA particles 1 2 3 4 1 2 3 4 0 50 100 150 200 250 300 1 0 0 1 2 2 3 3 300 250 200 150 100 50 0 0 1 1 2 2 3 3 0 1 1 2 2 3 3 # RHOD p articles 1 2 3 4 1 2 3 4 0 100 200 300 400 500 ( B ) 0 100 200 300 400 500 1 0 0 1 2 2 3 3 (A) 05 10152025 0 100 200 300 400 500 # particles per spot ACA: y = 16x 14.6 ; R2 = 0.94 RHOD: y = 18x 8.5 ; R2 = 0.99 (C) Particle concentration in source plate (x 10 6 / mL) Figure 5-7. Particles encapsulating rhodamine (RHO D, red) and 9 anthracenecarboxylic acid (ACA, bl ue) are printed and surface-adsorbed microparticle (MP) number s are quantified by fluor escence image analysis on a spot-by-spot basis. Data from 4 separate array preparations were pooled and average and standard deviations for eac h arrayed spot are presented in 3-D bar plots. Values on xand y-axes represent the RHOD or ACA source plate MP concentration while z-axis values represent the number of surfaceadsorbed MPs of either ACA (A. ) or RHOD (B. ). C.) Average delivered MP values with standard deviations are plo tted as a function of source plate concentration, and linear fi t parameters are shown. In order to reduce crosscontamination between samples, pin-washing steps incorporating detergent/de-ionized water/drying were included before each printing step. While cross-contamination was not completely eliminated, minimal cross-contamination

PAGE 126

126 of printed MPs was observed (9 5 incorre ctly-delivered MPs per spot, as judged by delivery to no MP spots, or < 3% of the highest delivered). Overall, these data demonstrate the ability to construct arrays of multi-parameter combinations of MP formulations with high precision and to quantitatively analyze delivery in situ by image analysis. Dendritic Cell Array Fabrication Due to our interest in DC-targeting MP -vaccines, we next fabricated co-localized MP/DC arrays by seeding DCs onto prepared MP arrays. Our intention was to construct arrays providing isolated DC populations, co-l ocalized with physisorbed MPs of various formulations, while preventing mixing of cell populations by constraining migration via a non-adherent polyethylene glycol-based background. Toward this effort, different nonadhesive backgrounds and cell-seeding conditions were investigated. Leukocytes are well-known for their ability to adhere to numerous non-adhesive substrates which are capable of blocking adhesion of other cell types. For example, albumin-coated substrates have recently been shown to support DC adhesion, with an equivalent number of adherent DCs compared to adhesive proteins fibronectin, laminin, vitronectin, fibrinogen and collagen. In this work, we found that polyethylene glycol (PEG) based Pluronic F-127 adsorbed onto a CH3-terminated surface pr ovided an improved and prolonged non-fouling effect compared to a self-assembled monolayer of PEGalkanethiol (HSC11EG3OH) on Au. Additionally, Pluronic F-127 had the added advantage of maintaining its nonfouling properties after drying and re-hydration, as opposed to PEG-alkanethiol which did not remain non-fouling after drying and rehydration. Furthermore, to ensure MP-stability it is imperative that printed MPs are not exposed to non-aqueous solvents such as ethanol (which is required for PEG-

PAGE 127

127 alkanethiol monolayer assembly). Hence, Pluronic F-127 was determined to be clearly better-suited for this application, and was therefore utiliz ed as the non-adhesive background. Initial efforts to constrain DC seeding to adhesive islands focused on the use of serum-free seeding conditions. However, even after multiple washing steps, a moderate number of DCs was found to remain adherent to off-spot areas when DCs were seeded in PBS. We therefore investigated a reduction in the concentration of Ca++ and Mg++ ions in the cell seeding buffer. These diva lent ions are a functional requirement for a number of cell surface receptor s, including integrins. We found that reducing the Ca++ and Mg++ ion concentration by half (final conc entration of 0.025 g/L for each ion) enabled efficient removal of cells from offspot areas (which largely remained rounded), providing a favorable differential in DC adhesion between adherent and non-adherent areas. This differential adhesion was visual ized by nuclear staining of fixed DCs cultured on MP-arrays for 24 h ( Figure 5-8A) A high density of adherent DCs (nuclei shown in blue) was preferent ially located on isolated islands, co-localized with RHODloaded MPs (shown in red), while DCs were absent on off-spot areas. Furthermore, phase-contrast micrographs of live cells confirmed that DCs pr eferentially adhered and spread onto the cell-adhesive spots whil e adhering minimally onto the background ( Figure 5-8B). Because soluble polyethylene glycol is known to be able to drive cellular fusion (albeit in high concentrations), cultures were inspected for multinucleated cells. Notably, no cell fusion was observed. Taken together, these data demonstrate the feasibility of constructing co-localized MP/DC arrays that can be utilized for the analysis of MP-mediated modulati on of DC response.

PAGE 128

128 (A) (B) Figure 5-8. Dendritic cell (DC) adhesion is re stricted to adhesive islands, and DCs are co-localized with printed microparticl es (MPs) on isolated islands against a non-fouling PEG-based background. Inte r-island DC migration is absent for up to at least 24 h. A.) Dendritic cells were seeded onto arrays of printed rhodamine-encapsulated MPs. Fluoresc ence micrograph overlay of nuclear staining, shown in blue, and MPs, shown in red, is shown (scale bar = 500 m). B.) Representative phase-contrast mi crograph of DCs cultured 24 hr on a 12 x 12 array, demonstrating that DCs are restricted to adhesive islands and are adherent and moderately spread on adhesive islands. Impact of the Study We report a new category of microarrays co-localized MP/cell arrays. This method takes advantage of standard miniarraying equipment to array spots of adhesive islands against a non-adhesive background to pr ovide co-localization of adherent cells and an adsorbed depot of MPs of desired formulations. This represents a versatile platform that can be used to probe the effe cts of different MP loading/modification formulations on any adherent cell type. Our overall goal was to develop a new platform enabling the screening of MP-based vaccine formulations to optimize DC response toward promoting desired immune responses. T he DC, identified as the most efficient antigen presenting cell, plays a central role in immune regulation, and is therefore an ideal vaccine delivery target. The co-localiz ation of MPs with is olated populations of DCs against a non-adhesive backg round in this microarray platform will permit the high-

PAGE 129

129 throughput analysis of DC responses to syn thetic immuno-modulating MPs targeting DCs. In this work, we have demonstrated th e quantitative, reproducible printing of surface-adsorbed MPs of different formulations with minimal cross-contamination. This control over the number of printed MPs and seeded DCs allows for the modulation of MP/DC ratio. Furthermore, these results demonstrate the ability to array isolated populations of DCs on adherent islands without inter-spot migration. The miniaturization achieved by this method increases throughput and reduces the required number of cells, particles and expensive reagents by many -fold. For example, in the demonstrated configuration, a >30-fold th roughput increase is achieved compared to 96-well plate, obviating need for expensive microfluidic systems. Critically, this miniaturization enables high-throughput investigation of rare cell populations, facilitating advances in personalized medicine, and in particular, personalized vaccines directed toward diseases such as cancer, autoimmune diseases and infections by in vivo targeting of DCs.

PAGE 130

130 CHAPTER 6 DENDRITIC CELL ARRAYS FOR BIOLOGICAL EVALUATION OF VACCINE PARTICLES Introduction A unique core technology we have developed, involves construction of microarrays of microparticl es co-localized with DCs on arrayed adhesive islands. Large numbers of these islands (>1,000) can be arra yed onto a single glass slide, with each island presenting a unique vaccine particle formulation, and providing a unique test sample. Dendritic cell function (e.g., activation, cytokine production, expression of putative tolerogenic markers), in response to co-localization and uptake of vaccine particles and in the presence of locally-delivered pro-tolerogenic biological factors is probed in situ, on arrayed chips, thr ough immunostaining and image analysis. The working hypothesis is that rational vaccine particle design, taken to a high-throughput level will reveal unknown cross-talk among immunomodulatory agents and indicate vaccine particle formulations optimized for their ability to promote tolerogenic DC markers. Materials and Methods A 50:50 polymer composition of poly(d,l lactide-co-glycolid e) (PLGA) with different factors encapsulated in the particles were generated using the technique described in previous chapters. Dendritic cells were isolated from C57BL6/j mice as described in previous chapters. Furthermore, co-localizati on of microparticles and dendritic cells on a chip was obtained as mentioned in previous chapters. Different factors that were encapsulated in the particles included lippopl ysaccharide (an adjuvant), tissue growth factor-beta(TGF-beta), polyinos inic-polycytidylic acid sodium salt (Poly I:C an immunostimulant), hemoglobin (t olerogenic factor), vitaminD3 (tolerogenic factor) and

PAGE 131

131 rapamycin (tolerogenic factor). Furthermore, particles were conjugated with different peptides using bioconjugate techniques. Brie fly, 1.3 mg of EDC (1-ethyl-3-(3dimethylaminopropyl) carbodiim ide) and 0.7 mg of NHS ( N -Hydroxysuccinimide) were dissolved in 1 mL of PBS containing 1 mg of PLGA particles for 15 minutes on a shaker rotating at 300 rpm. This made the carboxyl terminated polymer c hains on the particles chemically active and ready to be crosslinked to the primary amine terminal of the peptides. 0.2 mM concentration of the peptides, CS-2 (fibronectin derived), RGD (present in different proteins), P2 (binds to CD11b cell surface molecule), PD-2 (binds to CD11c cell surface molecule) were crossl inked to the surface of the particles separately. The surface modified particles were printed in different combinations with the particles encapsulated with different fact ors in different concentrations on the chip. Dendritic cells were cultured on these chips and cell function such as up-regulation of cell surface molecules and intracellular cytokine production was probed in-situ using immunostaining as described in Chapter 4. Image acquisition and analysis was performed using fluorescent microscopy and Axiovision 4.0.3 software respectively. Statistical Analysis Statistical analyses were performed using one-way ANOVA usi ng Systat (Version 12, Systat Software, Inc., San Jose, CA). Pair-wise comparisons were made using Tukeys Honestly-Significant-Difference, wi th p-values of less than or equal to 0.05 considered to be signi cant. Furthermore, graphs were plotted and regression analysis was performed using Sigmaplot (Version 10.0, Systat Software Inc. San Jose, CA). Results We show that DCs are restricted to NH2-terminated adhesive islands, in colocalization with surface-ads orbed LPS-loaded MPs. Micropar ticle to DC ratio is readily

PAGE 132

132 controlled and DCs are able to phagocytose surface-adsorbed MPs. Figure 6-1. Dendritic cells can uptake MPs encapsulated with CyPher5E dye and fluoresce in the Cy5 filter thus, suggesting that the particles have been phagocytosed. Here green-cytosol, blue-par ticles, pink-phagocytosed MPs. Phagocytosis of MPs is quantifiable us ing MPs with encapsulated pH-sensitive dye, CyPHer5E (GE Life Sciences), which in creases fluorescence over 2-fold at pH 5 (in phagolysosome) compared to pH 7.4 (extracellular) ( Figure 6-1 ). (A) # per Spot 0 50 150 MPs DCs MP/DC Arra y blue DC nucleus red microparticles DCs can lift and phagocytose surface-adsorbed microparticles red actin; green MPs Figure 6-2. Dendritic cells selectively adhere to NH2-terminated spots (presenting adsorbed MPs) and do not adhere to the PEG background. Quantitation is obtained through image analysis (e.g., 150 MPs : 50 DCs per spot). Physisorbed particles are able to lifted from the substrate (seen in SEM image) and are phagocytosable (seen in DIC image and pH-sensitive dye). Phagocytosis is quantifiable using MP loading of pH-s ensitive dye, CyPHer5E dye. MP to cell ratio can be optimized so all MPs are taken up in 16 h.

PAGE 133

133 Studies are ongoing to optimize MP:DC ra tio for efficient uptake of surfaceadsorbed MPs in the test-period of 24 hr. Critically, this data demonstrates the ability to quantify DC responses in situ, on arrayed chips, amenable to high-throughput analysis ( Figure 6-2 ). RFI RFI LPS-loading dilution encapsulated in PLGA MP 1:1 1:2 1:4 1:8 LPS-loading dilution encapsulated in PLGA MP 1:1 1:2 1:4 1:8 Blank MP NO MP Blank MP NO MP Increasing Dose of Activating Factor, LPS Time (min) RFI / cell 010203040 50 60 70 50 100 150 200 250 300 Quantifying MP Phagocytosis by pH-sensitive dye, CyPHer5E DC Activation blue nuclei red MHC-II green CD86 Randomized DC / MP Arrays Pooled IL-10 data: Students t-test on IL-10 data sets from randomized arrays yields pvalue > 0.47, indicating data sets are equivalent. This suggests limited cross-talk across arrayed islands, validating sample independence. MP loading of LPS dilution 1:1 1:21:4 1:8 IL-10 per cell (RFI) 0 .02 .04 .06 .08 .10 .12 .14 Figure 6-3. As proof-of-princi ple, DCs were arrayed, co-l ocalized with MPs loaded with an increasing dose of activating factor (LPS), with equal number of particles and cells per spot. Dendritic cells were incubated 4 hrs and then immunostained for markers of activation, MHCII & CD86. Quantification reveals over a 4-fold increase compared to unl oaded MPs for the highest LPS dose. Furthermore, after 24 h, cytokine producti on of IL-10 & IL-12 was quantified in situ through golgi-stop treatment, followed by immuno-staining for intracellular cytokines. Comparison of pooled IL-10 data with 3 different randomized DC arrays revealed the lack of positional dependence, suggesting limited cross-talk across islands. The symbols r epresents that the condition is significantly di fferent from similar symbols.

PAGE 134

134 Specifically, using LPS-loaded MPs as a model MP formulation, we quantified DC production of stimulatory molecules (MHC II), co-stimulato ry molecules (CD86, CD80) and cytokines (IL-10, IL-12) on DC arrays. Furthermore, we have investigated the potential for inter-island interactions due to di ffusion of molecules across arrayed spots. Four array configurations (1 ordered, 3 rand omized) of LPS-loaded MPs were fabricated and DC IL-10 production data sets were compar ed by Students t-test. The resultant pvalue >0.47 indicates data sets are equiva lent, suggesting limited cross-talk across islands and validating sample independence. Pooled data for DC IL-10 production is plotted. Finally, we have begun using MP/DC arrays to investigate formulations with potential to produce tolerogenic DCs (Figure 6-3). Poly I:C RGD 4N1K P2 PD2 TGF1 V IP PLGA-blank No MPs RFI p er Cell *0.0 4.0e+5 8.0e+5 1.2e+6 IL-10 ProductionDC/MP Array with Pro-Tolerogenic Factors Surface-Tethered Factor 1 10 11 2 2,1 3 3 1 3,2 4 4 ,1 4, 2 4, 3 5 5,1 5, 2 5,3 5, 4 6 6 1 6,2 6 3 6, 4 7 7,1 7 ,2 7 3 7 4 8 8 1 8 2 8 3 8,4 9 RFI/ DC 0.00 0.2 0.4 0.6 0.8 1.0 1.2 1.4 *IL-10** ** *** 1 10 11 2 2,1 3 3 1 3,2 4 4 ,1 4, 2 4, 3 5 5,1 5, 2 5,3 5, 4 6 6 1 6,2 6 3 6, 4 7 7,1 7 ,2 7 3 7 4 8 8 1 8 2 8 3 8,4 9 RFI/ DC 0.00 0.2 0.4 0.6 0.8 1.0 1.2 1.4 *IL-10** ** ***11 PEG 10 Blank MPs 9 No MPs 8 RGD 7 PD2 6 P2 5 CS 1 4 Poly I:C 3 IL 4 2 VIP 1 4N1K # Condition 11 PEG 10 Blank MPs 9 No MPs 8 RGD 7 PD2 6 P2 5 CS 1 4 Poly I:C 3 IL 4 2 VIP 1 4N1K # Condition 1 10 11 2 2,1 3 3,1 3,2 4 4,1 4,2 4,3 5 5,1 5,2 5,3 5,4 6 6,1 6,2 6,3 6,4 7 7,1 7,2 7,3 7,4 8 8,1 8,2 8,3 8,4 9 1 10 11 2 2,1 3 3,1 3,2 4 4,1 4,2 4,3 5 5,1 5,2 5,3 5,4 6 6,1 6,2 6,3 6,4 7 7,1 7,2 7,3 7,4 8 8,1 8,2 8,3 8,4 9 Figure 6-4. MPs (10:1 ratio MPs : DCs) with factors surface-tethered were incubated 24 h on MP/DC arrays and analyzed for production of IL-10, an antiinflammatory cytokine. Notably, MPs with tethered TGF1 and RGD peptide induced elevated DC production of IL-10 compared to other factors (peptides: 4N1K, P2, P-D2, VIP), and a 2-fold increase over untreated MPs (PLGAblank). Two-component mixtures of (1:1 ) MP formulations (surface-tethered: P2, PD2, 4N1K, CS-1, RGD; encapsulated: VIP, IL-4, poly I:C; surfaceadsorbed: PEG-Pluronic) were investigat ed on MP/DC arrays for the ability to induce IL-10 cytokine. Interestingly, we found surface tether ing of factors TGF-beta and RGD peptide promote DC production of anti-inflammatory cytokine IL-10, but when two-component

PAGE 135

135 mixtures of MP formulations were tested, it was the combinat ion of tethered 4N1K peptide and encapsulated IL-4 MPs that provided synergistic IL-10 production. Furthermore, it was shown that combined delivery of particles encapsulated with VD3 and TGF-beta to DCs had a synergistic effe ct, where as when these factors were provided to DCs, individually, did not show a higher response ( Figure 6-4). Impact of the Study We report a new category of microarrays co-localized MP/cell arrays. This method takes advantage of standard miniarraying equipment to array spots of adhesive islands against a non-adhesive background to pr ovide co-localization of adherent cells and an adsorbed depot of MPs of desired formulations. This represents a versatile platform that can be used to probe the effe cts of different MP loading/modification formulations on any adherent cell type. Our overall goal was to develop a new platform enabling the screening of MP-based vaccine formulations to optimize DC response toward promoting desired immune responses. T he DC, identified as the most efficient antigen presenting cell, plays a central role in immune regulation, and is therefore an ideal vaccine delivery target. The co-localiz ation of MPs with is olated populations of DCs against a non-adhesive backg round in this novel platform will permit the highthroughput analysis of DC responses to syn thetic immuno-modulating MPs targeting DCs. In this work, we have demonstrated th e quantitative, reproducible printing of surface-adsorbed MPs of different formulations with minimal cross-contamination. Critically, this control over the number of printed MPs and seeded DCs allows for the modulation of MP/DC ratio. Fu rthermore, these results demons trate the ability to array isolated populations of DCs on adherent islands without in ter-island migration. The miniaturization achieved by this method increases throughput and reduces the required

PAGE 136

136 number of cells, particles and expensive rea gents by many-fold. For example, in the demonstrated configuration, a >30-fold th roughput increase is achieved compared to 96-well plate, obviating need for expensive microfluidic systems. Critically, this miniaturization enables high-throughput investigat ion of rare cell populations, facilitating advances in personalized medicine, and in particular, personalized vaccines directed toward diseases such as cancer, autoimmune diseases and infections by in vivo targeting of DCs.

PAGE 137

137 CHAPTER 7 HIGHTHROUGHPUT PARTICLE-BAS ED VACCINE GENERATION Introduction With the advent of immunogenomic approac hes combined with advances in molecular immunology, progress is now being made toward the rational design of vaccines. Several techniques are being devel oped to speed the process of identifying target antigens. Furthermore the understanding of pattern recognition receptors has advanced adjuvant technology. Although, abundant publicati ons on these topics have been produced, there has been no concomitant increase in t he number of new effective vaccines developed, despite years of enor mous effort. Recently, there has been an effort to develop particle based immune cell-targeting vaccines that can generate the necessary immune responses. The particle based vaccines have several advantages over traditionally administered vaccines, su ch as, providing stability for roomtemperature storage and shipping, providing ease of administration (e.g., needle-free, single-dose), and improved ease of manufacture Furthermore, particle based vaccines have been shown to be effective in targeti ng dendritic cells, an ant igen presenting cell, in vivo and generate a T-cell based immune respons e. However, the number of possible combinations of the adjuvants, antigens and other immunomodulatory drugs that can be encapsulated in particle vaccines warrants a need to develop a high-throughput particle production method to produce large numbers of multi-parameter combinations of microparticle modifications and evaluate the poten cy of such particles. In order to meet this need we have developed a technique to synthesize particle based vaccines in a high-throughput manner. We leverage the use of standard miniarraying equipment with accurate over-spotting capabilities for both the high-throughput loading and fabrication

PAGE 138

138 of large numbers of combinatoriallyloaded particles in non-expensive, standard polystyrene plates using compatible organic solvent to generate poly (d,l lactide-coglycolide) (PLGA) based particles. Materials Utilized A 50:50 polymer composition of poly(d,l lactide-co-glycolid e) (PLGA) with inherent viscosity 0.55-0.75 dL/g in hexafluoroisopropanol, HFIP (Lactel, AL, USA) was used to generate particles. Poly-vinyl alcohol (PVA) (MW ~ 100,000 g/mol, Fisher Science, Rochester, NY, USA) was utilized as an em ulsion stabilizer. Phosphate buffered saline (PBS) solution (Hyclone, UT, USA) wa s used as the aqueous phase to form the emulsions while propylene carbonate (PC) (F isher Scientific, NJ, USA) was used as an organic solvent to dissolve PLGA polymer. Microparticles were formed using emulsification-diffusion technique. Fluorescent dyes, 7-di ethylamino-4-methylcoumarin (Coumarin: ex = 375nm and em = 445nm), 1,1',3,3,3',3'-H examethylindodicarbocyanine iodide (Cyanine: ex = 648nm and em = 670nm) and rhodamine 6G (RHOD: ex = 528nm and em = 550nm) were encapsulated in the particles for quantification purposes. Polystyrene based 384-well plates were us ed as generation chambers for the particles and a Calligrapher Miniarrayer (BioRad) contact printer was used to transfer solutions in these particle generation chambers. Parallel Particle Production A 384-well plate was used as a source plat e having 6 different dilutions of the 3 fluorescent dyes dissolved in PC. These diluti ons of the dyes were then printed in a 384 well plate using the contact printer to form a 6x6x6 matrix with all possible combinations of the 3 available dyes (C oumarin, RHOD and Cyanine), which resulted in

PAGE 139

139 216 different conditions. These dyes were c hosen so as to have minimum overlap of the absorption of one dye with the emission spectrums of another, in order to reduce the quenching effect. Upon printing of the dyes, 10 L PLGA dissolved in PC at a concentration of 3% was added to the well s. The plate was then sonicated in a sonicating water bath for 3 minutes. A 20 L 5% PVA solution in PBS was added to the wells and the plate was sonica ted for 5 more minutes. M P G e n e r a t i o n C h a m b e r sSource Plate M P G e n e r a t i o n C h a m b e r s Sonication Particles Washing with DI H2O to Remove PVA and Propylene Carbonate Figure 7-1. Particles were generated using a water-oil-water based emulsification diffusion method in a multi-well polystyrene plate with multiple fluorescent dyes as representative drug formulations. The plate was then incubated for 16 hours under a pressure of 0.2mTorr at 370C to evaporate PC and water. De-ionized water, 80 L, was added to each well and the plate was sonicated for 60 sec. The pl ate was then centrifuged at 1500 Gs for 50 minutes. The supernatant was then removed and the particles re-suspended by sonicating in 50 L/well of de-ionized water. This washing process of centrifuging,

PAGE 140

140 aspirating and re-suspending was repeated 5 times to remove PVA from the particles in the well. The plate was then frozen by pl acing over dry ice and dried under vacuum overnight. All the particle generation steps were performed in dark to reduce photobleaching of the dyes ( Figure 7-1). For analyzing, the particles were re-suspended in de-ionized water solution using a sonicating bath and printed on a plasma cleaned glass microscope slides (Fisher Scientific, NJ, USA). Particle Analysis The dye printed plate was analyzed using a plate reader with appropriate filters to measure fluorescence of the printed fluorescent dyes. Dyes encapsulated particles were generated and printed on a glass slide usi ng contact pin miniarrayer. This glass slide was then scanned using Typhoon 9410 (GE Healthcare) fluorescent imager with appropriate filters for the fl uorescent dyes. Furthermore, micrographs of the printed particles were obtained using Carl Zei ss 400M fluorescent microscope using the appropriate filters for each dye. Detailed im age analysis of the particle spots was carried out using Axiovision software. Furthermore, the par ticles were printed onto microarrays using contact pin printing and then the parti cles were analyzed using scanning electron micrographs. Statistical Analysis Statistical analyses were performed using one-way ANOVA usi ng Systat (Version 12, Systat Software, Inc., San Jose, CA). Pair-wise comparisons were made using Tukeys Honestly-Significant-Difference, wi th p-values of less than or equal to 0.05 considered to be signi cant. Furthermore, graphs were plotted and regression analysis was performed using Sigmaplot (Version 10.0, Systat Software Inc. San Jose, CA).

PAGE 141

141 Results Particles were generated via paralle l particle production technique through emulsification process. The dyes, Coumarin Rhodamine and Cyanine were printed in a 384 polystyrene plate from a source plate havi ng 6 dilutions of the dyes. The printing was performed using standard miniarraying equipment and 6x6x6 = 216 different combinations of the dyes were generated. T he relative fluorescence intensity of the dyes was measured using a plate reader. It was observed that the dyes coumarin, rhodamine and cyanine followed a linear curve with the coefficient of determination of R2 Coumarin = 0.99; R2 Rhodamine = 0.99 and R2 Cyanine = 0.94, when fitted to a linear curve ( Figure 7-2 ). 0 10 20 30 40 50 60 Coumarin R2 = 0.99 Rhodamine R2 = 0.99 Cyanine R2 = 0.94 No Dye1:16 1:8 1:4 1:2 1:1Normalized Relative Fluorescence Intensity 0 10 20 30 40 50 60 Coumarin R2 = 0.99 Rhodamine R2 = 0.99 Cyanine R2 = 0.94 No Dye1:16 1:8 1:4 1:2 1:1Normalized Relative Fluorescence Intensity Figure 7-2. Combination of fluorescent dyes at different dilutions can be printed in the same well accurately. The coefficient of determination was obtained by simple regression analysis with the decreasing concentration of dye in the source plate: R2 Coumarin = 0.99; R2 Rhodamine = 0.99; R2 Cyanine = 0.94. A solution of PLGA in PC was added to t he wells containing the dyes in different concentrations and micro-batches of particles were generated with PVA as the

PAGE 142

142 emulsifier. Relative fluorescence intensity of the three dyes for each micro-batch was obtained using the plate reader and plo tted against the individual condition (Figure 7-3). Figure 7-3. Combination of fluorescent dyes at different dilutions can be encapsulated in PLGA particles with uniform distribut ion of dyes within the individual population of particles. It was observed that the relative fluorescenc e intensity of particles in a micro-batch follow similar trend as the printed dyes. The particles generated using parallel particle production, were pooled together and size and shape was determined using scanning electron microscope and dynamic 1 10 100 1000 10000 100000 020406080100ConditionsRelative Fluorescence Intensity 1 10 100 1000 10000 100000 050100150200250 ConditionsRelative Fluorescence Intensit y

PAGE 143

143 light scattering technique respectively. It was observed that the particles had a smooth morphology and the average size of the particles based on volume estimation was 1.04m (Figure 7-4). Size (nm) 1e+01e+11e+21e+31e+41e+5 % Volume 0 5 10 15 20 25 Figure 7-4. Particles were produced via paralle l particle production method and the particles were analyzed using SEM and dynamic light scattering. A typical size distribution curve of PLGA-particles quantified via dynamic light scattering analysis (based on volume estimation) demonstrates an average size of particles of 1.04 m. The particles generated using this process were then printed on glass microscope slide and were imaged using typhoon 9410 and fluorescent microscope. An overlay image was plotted (Figure 7-5). 3-D surface plots were plotted using ImageJ software for each of the fluorescent micr ograph and the overlaid micrograph. This

PAGE 144

144 method is useful for quick scanning of the amount of fluorescent dye encapsulated in the particles. Figure 7-5. The fluorescence intensities of t he particles generated via PPP method can be visualized by quick scan imaging. The particles encapsulating 3 different fluorescent dyes in 6 dilutions, thus forming 216 different combinations, were printed onto a glass slide and scanned using typhoon 9410 and fluorescent microscope. This method is useful for image quick scan and visualization of the amount encapsulated factors. Mean fluorescent intensity (M FI) of individual particles were obtained from few of the micro-batches. These MFIs were then pl otted as histograms to compare the dye

PAGE 145

145 distribution within a micro-bat ch. It was observed that the dye was distributed uniformly among the particles of one micro-batch. A chip was generated consisting of cell adhesive islands surrounded by nonfouling regions. Particles generated using PPP method were printed onto the cell adhesive regions. Dendritic cells were cultured onto the chip, which co-localized with the printed particles (Figure 7-7). Impact of the Study Currently, standard methodology for the generation of polymeric PLGA particles consists of single batch processing using a double-emulsion/solvent-evaporation method. This process can take 4-5 hours and even skilled hands may be limited to producing less than a dozen MP formulations in one day. As a result of our ability to assess many particle formulations at once through our high-throughput MP array approach, we have identified a unique and critical need to quickly generate large numbers (hundreds to thousands) of different MP formulations. To meet this need, we have generated a high-throughput par allel particle production technology, utilizing solidpin miniarraying equipment for the robotic loading of component solutions into wells of a 384-well plate which then serve as particle-generation chambers.

PAGE 146

146 CHAPTER 8 CONCLUSION AND FUTURE OUTLOOK Designing biomaterials that can acti vely modulate the immune system is an emerging field that will help in understanding basic working of immune cell functions at a molecular level and further the field of immunomodulatory biomater ials. Biomaterials can help understand molecular and cellular processes that occur when immune cells such as antigen presenting cells interact with foreign bodies. Furthermore, design of biomaterials has to be influenced by the im munobiology of the disease. Generation of biomaterials influenced vaccines is a step forward in this direction. Modulation of immune system using biomat erials has shown promising results in the field of cancer research. However, use of biomaterials to actively modulate the immune system for treatment of autoimm une diseases has not been extensively studied. Biomaterials in combination with biological components can be recognized by the immune system as foreign and hence get targeted by immune cells such as T-cells and B-cells. This hypothesis has been utilized in clinical trials to generate dendritic cell (DC), a specialized antigen presenting ce ll targeted in vivo and ex vivo immunotherapies for autoimmune diseases. Such immunotherapies can be made more efficient with the use of biomaterials. For inst ance, biomaterials in the form of polymeric particles can be utilized to provide sustai ned release of therapeut ic factors in vivo targeted to DCs. Additionally, implanted biomaterials, co mbination products involving biological components and synthetic material s, and ex vivo strategies can be made more effective using biomaterials based immunomodulation. Hence, looking forward, an extensive study was performed to charac terize the adhesion based modulation of adaptive immune system for improv ing the present ex vivo strategies for generating DC

PAGE 147

147 based vaccines targeting autoimmune diso rder of type 1 diabetes. Furthermore, highthroughput microarraying techniques we re utilized to generate and characterize particle based vaccines targeted to DCs. Currently, the ex vivo immunotherapies involve culturing exogenously generated DCs onto tissue culture treated polystyrene plates adsorbed with serum proteins. However, this strategy has not been optimiz ed to provide optimal surface for DC adhesion. We have shown in this body of work that adhesion based stimuli to DCs can modulate their cell function and these DCs can subsequently modulate the adaptive immune response as quantified by T-cell re sponses. Furthermore, we show that NODmice derived DCs were differentially modul ated by adhesion stimuli than the normal wild type B6 derived DCs. Furthermore, we showed that increasing integrin expression which corresponds to increasing adhesive molecules provided follows increase in expression of maturation signa ls of DCs. This work suggests that the mechanism of adhesion that involves integrin binding to adhesive molecules such as peptides, indeed, is a governing factor that mediates DC maturation. A pept ide having RGD amino acid sequence was utilized in this study, which is present in several proteins. Looking ahead, surface modified biomaterials can be ut ilized to generate other known peptides / functional molecules of interest that modul ate DC functions. For example, DC T-cell synapse can be studied in great detail and critical questions such as extent of clustering of integrins required on DCs to generate such synapses can be answered. In parallel studies were performed to design particle based vaccines that can effectively target DCs and modulate their functions to generate the needed immunotherapy. We developed highthroughput techniques to generate multicomponent

PAGE 148

148 PLGA based particle vaccines. Furthermore we developed highthr oughput techniques using lab-on-a-chip format to characterize the particle vaccines cultured in the presence of DCs. Additionally, techniques were developed to screen these DCs for most effective particle vaccine formulations. In future, further, formulations will be tested on the chips and the particle vaccines selected will be tested in a mouse model for a desired immunotherapy. This work will further t he field of biomaterials based vaccines.

PAGE 149

149 LIST OF REFERENCES [1] Adorini, L. and Penna, G., Dendritic cell tolerogenicity: a key mechanism in immunomodulation by vitamin D re ceptor agonists, Hum Immunol 70 (5) (2009), pp. 345-352. [2] Lu, L. and Thomson, A. W., Manipulat ion of dendritic cells for tolerance induction in transplantation and autoimmune disease, Transplantation 73 (1 Suppl) (2002), pp. S19-S22. [3] Miretti, M. M. and Beck, S., I mmunogenomics: molecular hide and seek, Hum Genomics 2 (4) (2006), pp. 244-251. [4] Steinman, R. M. and Nussenzweig, M. C., Dendritic cells: features and functions, Immunol Rev 53 (1980), pp. 127-147. [5] Steinman, R. M., Dendritic cells in vivo: a key target for a new vaccine science, Immunity 29 (3) (2008), pp. 319-324. [6] van, Duin D., Medzhitov, R., and Shaw, A. C., Triggering TLR signaling in vaccination, Trends Immunol 27 (1) (2006), pp. 49-55. [7] http://clinicaltrials.gov/ (2009). [8] Giannoukakis, N., Phillips, B., and Trucco M., Toward a cure for type 1 diabetes mellitus: diabetes-suppressive de ndritic cells and beyond, Pediatr Diabetes 9 (3 Pt 2) (2008), pp. 4-13. [9] Phillips, B., Nylander, K., Harnaha, J., Machen, J., Lakomy, R. Styche, A., Gillis, K., Brown, L., Lafreniere, D., Gallo, M., Knox, J., Hogela nd, K., Trucco, M., and Giannoukakis, N., A microsphere-based vaccine prevents and reverses new-onset autoimmune diabetes, Diabetes 57 (6) (2008), pp. 1544-1555. [10] Phillips, B. E., Giannoukakis, N., and Trucco, M., Dendritic cell mediated therapy for immunoregulation of type 1 diabetes mellitus, Pediatr Endocrinol Rev 5 (4) (2008), pp. 873-879. [11] Matzinger, P., An innate sense of danger, Ann N Y Acad Sci 961 (2002), pp. 341342. [12] Medzhitov, R. and Janeway, C. A., Jr., Innate immune recognition and control of adaptive immune responses, Semin Immunol 10 (5) (1998), pp. 351-353. [13] Fernandez, N. C., Lozie r, A., Flament, C., Ricciardi-Castagnoli, P., Bellet, D., Suter, M., Perricaudet, M., Tursz, T. Maraskovsky, E., and Zitvogel, L., Dendritic cells directly trigger NK cell functions: cross-talk relevant in innate anti-tumor immune responses in vivo, Nat Med 5 (4) (1999), pp. 405-411.

PAGE 150

150 [14] Lambrecht, B. N., Salomon, B., Klatzmann, D., and Pauwels, R. A., Dendritic cells are required for the development of chronic eosin ophilic airway inflammation in response to inhaled antigen in s ensitized mice, J Immunol 160 (8) (1998), pp. 4090-4097. [15] Shi, Y., Evans, J. E., and Rock, K. L., Molecular identification of a danger signal that alerts the immune system to dying cells, Nature 425 (6957) (2003), pp. 516-521. [16] Nakajima, H. and Takatsu, K., Role of cytokines in allergic airway inflammation, Int Arch Allergy Immunol 142 (4) (2007), pp. 265-273. [17] Yates, S. F., Paterson, A. M., No lan, K. F., Cobbold, S. P., Saunders, N. J., Waldmann, H., and Fairchild, P. J ., Induction of regulatory T cells and dominant tolerance by dendritic cell s incapable of full activation, J Immunol 179 (2) (2007), pp. 967-976. [18] Kijima, M., Yamaguchi T., Ishifune, C., Maekawa, Y., Koyanagi, A., Yagita, H., Chiba, S., Kishihara, K., Shimada, M., and Yasutomo, K., Dendritic cellmediated NK cell activation is co ntrolled by Jagged2-Notch interaction 1, Proc Natl Acad Sci U S A 105 (19) (2008), pp. 7010-7015. [19] Ranjit, S. and Daz hu, L., Potential role of dendritic cells for progression of atherosclerotic lesions, Postgrad Med J 82 (971) (2006), pp. 573-575. [20] Hammad, H. and Lambrecht, B. N., Dendritic cells and epithelial cells: linking innate and adaptive immunity in asthma, Nat Rev Immunol 8 (3) (2008), pp. 193-204. [21] Hynes, R. O., Integrins: bidirect ional, allosteric signaling machines, Cell 110 (6) (2002), pp. 673-687. [22] Anderson, J. M., Rodriguez, A., and Chang, D. T., Foreign body reaction to biomaterials, Semin Immunol 20 (2) (2008), pp. 86-100. [23] Hunter, S. K., Kao, J. M., Wang, Y., Benda, J. A. and Rodgers, V. G., Promotion of neovascularization around hollow fiber bioartificial organs using biologically active substances, ASAIO J 45 (1) (1999), pp. 37-40. [24] Jenney, C. R. and Anderson, J. M ., Adsorbed serum proteins responsible for surface dependent human macrophage beh avior, J Biomed Mater Res 49 (4) (2000), pp. 435-447. [25] Keselowsky, B. G., Br idges, A. W., Burns, K. L., Tate, C. C., Babensee, J. E., LaPlaca, M. C., and Garcia, A. J., Role of plasma fibronectin in the foreign body response to biomaterials, Biomaterials 28 (25) (2007), pp. 36263631.

PAGE 151

151 [26] Shen, M. and Horbett, T. A., The effects of surfac e chemistry and adsorbed proteins on monocyte/macrophage adhes ion to chemically modified polystyrene surfaces, J Biomed Mater Res 57 (3) (2001), pp. 336-345. [27] Tang, L. and Eaton, J. W., Fibrin(ogen) mediates acute inflammatory responses to biomaterials, J Exp M ed 178 (6) (1993), pp. 2147-2156. [28] Acharya, A. P., Dolgova, N. V., Cl are-Salzler, M. J., and Keselowsky, B. G., Adhesive substrate-modulation of adaptive immune responses, Biomaterials 29 (36) (2008), pp. 4736-4750. [29] Lo, J. and Clar e-Salzler, M. J., Dendritic cell subsets and type I diabetes: focus upon DC-based therapy, Autoimmun Rev 5 (6) (2006), pp. 419-423. [30] Nencioni, A. and Brossart, P., Cell ular immunotherapy with dendritic cells in cancer: current status, Stem Cells 22 (4) (2004), pp. 501-513. [31] Ludewig, B., Oderma tt, B., Landmann, S., Hengartner, H., and Zinkernagel, R. M., Dendritic cells induce autoimmune diabetes and maintain disease via de novo formation of local lymphoid tissue, J Exp Med 188 (8) (1998), pp. 1493-1501. [32] Morel, P. A., Vasquez, A. C., and Fe ili-Hariri, M., Immunobiology of DC in NOD mice, J Leukoc Biol 66 (2) (1999), pp. 276-280. [33] Turley, S. J., Dendritic cells: in citing and inhibiting autoimmunity, Curr Opin Immunol 14 (6) (2002), pp. 765-770. [34] Peng, R., Bathjat, K., Li, Y., and Cl are-Salzler, M. J., Defective maturation of myeloid dendritic cell (DC) in NOD mice is controlled by IDD10/17/18, Ann N Y Acad Sci 1005 (2003), pp. 184-186. [35] Siperstein, M. D., U nger, R. H., and Madison, L. L ., Studies of muscle capillary basement membranes in normal s ubjects, diabetic, and prediabetic patients, J Clin Invest 47 (9) (1968), pp. 1973-1999. [36] Kaur, H., Chen, S., Xin, X., Chiu, J., Khan, Z. A., and Chakrabarti, S., Diabetesinduced extracellular matrix protein expression is mediated by transcription coactivator p300, Diabetes 55 (11) (2006), pp. 3104-3111. [37] Thomas, R. and Lipsky, P. E., Dendrit ic cells: origin and differentiation, Stem Cells 14 (2) (1996), pp. 196-206. [38] Morel, P. A. and Feili-Hariri, M., How do dendritic cells prevent autoimmunity?, Trends Immunol 22 (10) (2001), pp. 546-547. [39] Hynes, R. O., Structural biology. Changing partners, Science 300 (5620) (2003), pp. 755-756.

PAGE 152

152 [40] Takagi, J., Structural basis for ligand recognition by integrins, Curr Opin Cell Biol 19 (5) (2007), pp. 557-564. [41] Wang, Q., Klyubin, I., Wright, S., Griswold-Prenner, I., Rowan, M. J., and Anwyl, R., Alpha v integrins mediate beta-amy loid induced inhibition of long-term potentiation, Neurobiol Aging 29 (10) (2008), pp. 1485-1493. [42] Tanabe, J., Fujita, H., Iwamatsu, A., Mohri, H., and Ohkubo, T., Fibronectin inhibits platelet aggregation indep endently of RGD sequence, J Biol Chem 268 (36) (1993), pp. 27143-27147. [43] Rezania, A. and Healy, K. E., The effect of peptide surface density on mineralization of a matrix deposited by osteogenic cells, J Biomed Mater Res 52 (4) (2000), pp. 595-600. [44] VandeVondele, S., Vo ros, J., and Hubbell, J. A., RGD-grafted poly-L-lysine-graft(polyethylene glycol) copolymers block non-specific protein adsorption while promoting cell a dhesion, Biotechnol Bioeng 82 (7) (2003), pp. 784790. [45] Benoit, J. P., Fais ant, N., Venier-Julienne, M. C. and Menei, P., Development of microspheres for neurological disor ders: from basics to clinical applications, J Control Release 65 (1-2) (2000), pp. 285-296. [46] Bromberg, L., Intelligent hydrogels for the oral delivery of chemotherapeutics, Expert Opin Drug Deliv 2 (6) (2005), pp. 1003-1013. [47] dlakha-Hutcheon, G., Bally, M. B., Shew, C. R., and Madden, T. D., Controlled destabilization of a liposomal drug delivery system enhances mitoxantrone antitumor activity, Nat Biotechnol 17 (8) (1999), pp. 775-779. [48] Gao, Y., Gao, G., He, Y., Liu, T., and Qi, R., Re cent advances of dendrimers in delivery of genes and drugs Mini Rev Med Chem 8 (9) (2008), pp. 889900. [49] Reddy, S. T., Rehor, A., Schmoekel, H. G., Hubbell, J. A., and Swartz, M. A., In vivo targeting of dendritic cells in lymph nodes with poly( propylene sulfide) nanoparticles, J Control Release 112 (1) (2006), pp. 26-34. [50] Richards Grayson, A. C., Choi, I. S., Tyler, B. M. Wang, P. P., Brem, H., Cima, M. J., and Langer, R., Multi-pulse drug delivery from a resorbable polymeric microchip device, Nat Mater 2 (11) (2003), pp. 767-772. [51] Shi, L., Caulfield, M. J., Chern, R. T., Wilson, R. A., Sanyal, G., and Volkin, D. B., Pharmaceutical and immunological eval uation of a single-shot hepatitis B vaccine formulated with PLGA microspheres, J Pharm Sci 91 (4) (2002), pp. 1019-1035.

PAGE 153

153 [52] Tamber, H., Johansen, P., Merkle, H. P., and Gander, B., Formulation aspects of biodegradable polymeric microspheres for antigen delivery, Adv Drug Deliv Rev 57 (3) (2005), pp. 357-376. [53] Cohen, S., Yoshioka, T., Lucarelli M., Hwang, L. H., and Langer, R., Controlled delivery systems for proteins based on poly(lactic/glycolic acid) microspheres, Pharm Res 8 (6) (1991), pp. 713-720. [54] Elamanchili, P., Diwan, M., Cao, M., and Samuel, J., Characterization of poly(D,L-lactic-co-glycolic acid) based nanoparticulate system for enhanced delivery of antigens to dendritic cells, Vaccine 22 (19) (2004), pp. 2406-2412. [55] Waeckerle-Men, Y. and Groettrup, M., PLGA mi crospheres for improved antigen delivery to dendritic cells as cell ular vaccines, Adv Drug Deliv Rev 57 (3) (2005), pp. 475-482. [56] Waeckerle-Men, Y. Allmen, E. U., Gander, B., Scandella, E., Schlosser, E., Schmidtke, G., Merkle, H. P., and Groet trup, M., Encapsulation of proteins and peptides into biodegradable pol y(D,L-lactide-co-glycolide) microspheres prolongs and enhances antigen presentation by human dendritic cells, Vaccine 24 (11) (2006), pp. 1847-1857. [57] Bonifaz, L., Bonnyay, D., Mahnke, K., Rivera, M., Nussenzweig, M. C., and Steinman, R. M., Efficient targeting of protein antigen to the dendritic cell receptor DEC-205 in the steady state leads to antigen presentation on major histocompatibility complex cl ass I products and peripheral CD8+ T cell tolerance, J Exp Med 196 (12) (2002), pp. 1627-1638. [58] Faraasen, S., Voros, J., Csucs, G. Textor, M., Merkle, H. P., and Walter, E., Ligand-specific targeting of micr ospheres to phagocytes by surface modification with poly(L-lysine)-grafted poly(ethylene glycol) conjugate, Pharm Res 20 (2) (2003), pp. 237-246. [59] Kempf, M., Mandal, B., Jilek, S., Thiele L., Voros, J., Textor M., Merkle, H. P., and Walter, E., Improved stimulation of human dendritic cells by receptor engagement with surface-modified microparticles, J Drug Target 11 (1) (2003), pp. 11-18. [60] Elamanchili, P., Diwan, M., Cao, M., and Samuel, J., Characterization of poly(D,L-lactic-co-glycolic acid) based nanoparticulate system for enhanced delivery of antigens to dendritic cells, Vaccine 22 (19) (2004), pp. 2406-2412. [61] Reddy, S. T., Rehor, A., Schmoekel, H. G., Hubbell, J. A., and Swartz, M. A., In vivo targeting of dendritic cells in lymph nodes with poly( propylene sulfide) nanoparticles, J Control Release 112 (1) (2006), pp. 26-34.

PAGE 154

154 [62] Waeckerle-Men, Y. and Groettrup, M., PLGA mi crospheres for improved antigen delivery to dendritic cells as cell ular vaccines, Adv Drug Deliv Rev 57 (3) (2005), pp. 475-482. [63] Yoshida, M., Mata, J. and Babensee, J. E., Effect of poly(lactic-co-glycolic acid) contact on maturation of murine b one marrow-derived dendritic cells, J Biomed Mater Res A 80 (1) (2007), pp. 7-12. [64] Babensee, J. E., Inte raction of dendritic cells with biomaterials, Semin Immunol 20 (2) (2008), pp. 101-108. [65] Schlosser, E., Muelle r, M., Fischer, S., Basta, S., Busch, D. H., Gander, B., and Groettrup, M., TLR ligands and anti gen need to be coencapsulated into the same biodegradable microsphere fo r the generation of potent cytotoxic T lymphocyte responses, Vaccine 26 (13) (2008), pp. 1626-1637. [66] Elamanchili, P., Lutsiak, C. M. Hamdy, S., Diwan, M., and Samuel, J., "Pathogen-mimicking" nanoparticles for vaccine delivery to dendritic cells, J Immunother 30 (4) (2007), pp. 378-395. [67] Kasturi, S. P., Sachaphibulkij, K., and Roy, K ., Covalent conjugation of polyethyleneimine on biodegr adable microparticles fo r delivery of plasmid DNA vaccines, Biomaterials 26 (32) (2005), pp. 6375-6385. [68] Jilek, S., Zurkaulen, H., Pavlovic, J., Merkle, H. P., and Wa lter, E., Transfection of a mouse dendritic cell line by pl asmid DNA-loaded PLGA microparticles in vitro, Eur J Pharm Biopharm 58 (3) (2004), pp. 491-499. [69] Schena, M., Shalon, D., Davis, R. W., and Brown, P. O., Quantitative monitoring of gene expression patterns with a complementary DNA microarray, Science 270 (5235) (1995), pp. 467-470. [70] El-Ali, J., Sorger, P. K., and Jensen, K. F., Cells on chips, Nature 442 (7101) (2006), pp. 403-411. [71] Flaim, C. J., Chien, S., and Bhatia, S. N., An extr acellular matrix microarray for probing cellular differentiation, Nat Methods 2 (2) (2005), pp. 119-125. [72] Soen, Y., Mori, A., Palmer, T. D., and Brown, P. O., Explor ing the regulation of human neural precursor cell differ entiation using arrays of signaling microenvironments, Mol Syst Biol 2 (2006), pp. 37. [73] Schena, M., Shalon, D., Davis, R. W., and Brown, P. O., Quantitative monitoring of gene expression patterns with a complementary DNA microarray, Science 270 (5235) (1995), pp. 467-470. [74] Soen, Y., Chen, D. S. Kraft, D. L., Davis, M. M. and Brown, P. O., Detection and characterization of cellular i mmune responses using peptide-MHC microarrays, PLoS Biol 1 (3) (2003), pp. E65.

PAGE 155

155 [75] Flaim, C. J., Chien, S., and Bhatia, S. N., An extr acellular matrix microarray for probing cellular differentiation, Nat Methods 2 (2) (2005), pp. 119-125. [76] Keselowsky, B. G. Collard, D. M., and Garcia A. J., Surface chemistry modulates focal adhesion compositi on and signaling through changes in integrin binding, Biomaterials 25 (28) (2004), pp. 5947-5954. [77] Kempf, M., Mandal, B., Jilek, S., Thiele L., Voros, J., Textor M., Merkle, H. P., and Walter, E., Improved stimulation of human dendritic cells by receptor engagement with surface-modified microparticles, J Drug Target 11 (1) (2003), pp. 11-18. [78] Xia, C. Q., Peng, R ., Beato, F., and Clare-Salzler, M. J., Dexamethasone induces IL-10-producing monocyte-derived dendritic cells with durable immaturity, Scand J Immunol 62 (1) (2005), pp. 45-54. [79] Xiao, B. G., Huang, Y. M., and Link, H., Tolerogenic dendritic cells: the ins and outs of outcome, J Immunother 29 (5) (2006), pp. 465-471. [80] Johansson, U. and Londei, M., Ligation of CD47 during monocyte differentiation into dendritic cells results in reduced capacity for interleukin-12 production, Scand J Immunol 59 (1) (2004), pp. 50-57. [81] Johansson, U., Higginbottom, K., and Londei, M., CD47 ligation induces a rapid caspase-independent apoptosis-like cell death in human monocytes and dendritic cells, Scand J Immunol 59 (1) (2004), pp. 40-49. [82] Pasquali, L., Giannoukak is, N., and Trucco, M., Induc tion of immune tolerance to facilitate beta cell regeneration in type 1 diabetes, Adv Drug Deliv Rev 60 (2) (2008), pp. 106-113. [83] Yuri V.Bobryshev, Dendritic cell s in atherosclerosis, (2001), pp. 547-557. [84] Lord, R. S. and Bobr yshev, Y. V., Clustering of dendritic cells in athero-prone areas of the aorta, Atherosclerosis 146 (1) (1999), pp. 197-198. [85] Delemarre, F. G., Hoogeveen, P. G. De Haan-Meulman, M. Simons, P. J., and Drexhage, H. A., Homotypic cluster fo rmation of dendritic cells, a close correlate of their state of maturati on. Defects in the biobreeding diabetesprone rat, J Leukoc Biol 69 (3) (2001), pp. 373-380. [86] Verdijk, P., van V eelen, P. A., de Ru, A. H., H ensbergen, P. J., Mizuno, K., Koerten, H. K., Koning, F., Tens en, C. P., and Mommaas, A. M., Morphological changes during dendritic cell maturation correlate with cofilin activation and translocation to the cell membrane, Eur J Immunol 34 (1) (2004), pp. 156-164. [87] Xiao, B. G., Huang, Y. M., and Link, H., Tolerogenic dendritic cells: the ins and outs of outcome, J Immunother 29 (5) (2006), pp. 465-471.

PAGE 156

156 BIOGRAPHICAL SKETCH Abhinav Acharya was born in 1983 in Jhans i, India. Abhinav earned his bachelors degree from National Institute of Technology, Tiruchirappalli in June 2005. He came to the United States of America to pursue the degree of PhD. He started his graduate studies under the guidance of Dr. Benjam in Keselowsky, and earned his Ph.D. in materials science and engineering in May 2010. Abhinav had an extensive interaction with faculty members from different depar tments and had a multi-disciplinary research experience. He wishes to re main in academia and is looking forward to his time as a postdoctoral researcher.