<%BANNER%>

Synthesis and Characterization of DNA Nano-Meso-Microspheres as Drug Delivery Carriers for Intratumoral Chemotherapy

Permanent Link: http://ufdc.ufl.edu/UFE0021197/00001

Material Information

Title: Synthesis and Characterization of DNA Nano-Meso-Microspheres as Drug Delivery Carriers for Intratumoral Chemotherapy
Physical Description: 1 online resource (278 p.)
Language: english
Creator: Enriquez, Iris V
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: 5, bovine, dna, gadolinium, intratumoral, local, methotrexate, mitoxantrone
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: Conventional cancer chemotherapy results in systemic toxicity which severely limits effectiveness and often adversely affects patient quality of life. There is a need to find new drugs and delivery methods for less toxic therapy. Previous studies concerning DNA complexing with chemotherapy drugs suggest unique opportunities for DNA as a mesosphere drug carrier. The overall objective of this research was devoted to the synthesis and evaluation of novel DNA-drug nano-mesospheres designed for localized chemotherapy via intratumoral injection. My research presents DNA nano-meso-microspheres (DNA-MS) that were prepared using a modified steric stabilization method originally developed in this lab for the preparation of albumin MS. DNA-MS were prepared with glutaraldehyde covalent crosslinking (genipin crosslinking was attempted) through the DNA base pairs. In addition, novel crosslinking of DNA-MS was demonstrated using chromium, gadolinium, or iron cations through the DNA phosphate groups. Covalent and ionic crosslinked DNA-MS syntheses yielded smooth and spherical particle morphologies with multimodal size distributions. Optimized DNA-MS syntheses produced particles with narrow and normal size distributions in the 50 nano meters to 5 mu meters diameter size range. In aqueous dispersions approximately 200% swelling was observed with dispersion stability for more than 48 hours. Typical process conditions included a 1550rpm initial mixing speed and particle filtration through 20 mu meter filters to facilitate preparation. DNA-MS were in situ loaded during synthesis for the first time with mitoxantrone, 5-fluorouracil, and methotrexate. DNA-MS drug incorporation was 12%(w/w) for mitoxantrone, 9%(w/w) for methotrexate, and 5%(w/w) for 5-fluorouracil. In vitro drug release into phosphate buffered saline was observed for over 35 days by minimum sink release testing. The effect of gadolinium crosslink concentration on mitoxantrone release was evaluated at molar equivalences in the range of 20% to 120%. The most highly crosslinked DNA-MS exhibited the longest sustained release. The drug efficacy of mitoxantrone loaded DNA-MS was evaluated in vitro using a murine Lewis lung carcinoma cell line and a significant cytotoxic response was found at mitoxantrone doses as low as 1ppm. Drug release properties, DNA biodegradability, and observed cancer cell cytotoxicity of drug loaded DNA-MS suggest that they are appropriate for intratumoral chemotherapy evaluation aimed at improved and less toxic cancer therapy.
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 Iris V Enriquez.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Goldberg, Eugene P.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021197:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021197/00001

Material Information

Title: Synthesis and Characterization of DNA Nano-Meso-Microspheres as Drug Delivery Carriers for Intratumoral Chemotherapy
Physical Description: 1 online resource (278 p.)
Language: english
Creator: Enriquez, Iris V
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: 5, bovine, dna, gadolinium, intratumoral, local, methotrexate, mitoxantrone
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: Conventional cancer chemotherapy results in systemic toxicity which severely limits effectiveness and often adversely affects patient quality of life. There is a need to find new drugs and delivery methods for less toxic therapy. Previous studies concerning DNA complexing with chemotherapy drugs suggest unique opportunities for DNA as a mesosphere drug carrier. The overall objective of this research was devoted to the synthesis and evaluation of novel DNA-drug nano-mesospheres designed for localized chemotherapy via intratumoral injection. My research presents DNA nano-meso-microspheres (DNA-MS) that were prepared using a modified steric stabilization method originally developed in this lab for the preparation of albumin MS. DNA-MS were prepared with glutaraldehyde covalent crosslinking (genipin crosslinking was attempted) through the DNA base pairs. In addition, novel crosslinking of DNA-MS was demonstrated using chromium, gadolinium, or iron cations through the DNA phosphate groups. Covalent and ionic crosslinked DNA-MS syntheses yielded smooth and spherical particle morphologies with multimodal size distributions. Optimized DNA-MS syntheses produced particles with narrow and normal size distributions in the 50 nano meters to 5 mu meters diameter size range. In aqueous dispersions approximately 200% swelling was observed with dispersion stability for more than 48 hours. Typical process conditions included a 1550rpm initial mixing speed and particle filtration through 20 mu meter filters to facilitate preparation. DNA-MS were in situ loaded during synthesis for the first time with mitoxantrone, 5-fluorouracil, and methotrexate. DNA-MS drug incorporation was 12%(w/w) for mitoxantrone, 9%(w/w) for methotrexate, and 5%(w/w) for 5-fluorouracil. In vitro drug release into phosphate buffered saline was observed for over 35 days by minimum sink release testing. The effect of gadolinium crosslink concentration on mitoxantrone release was evaluated at molar equivalences in the range of 20% to 120%. The most highly crosslinked DNA-MS exhibited the longest sustained release. The drug efficacy of mitoxantrone loaded DNA-MS was evaluated in vitro using a murine Lewis lung carcinoma cell line and a significant cytotoxic response was found at mitoxantrone doses as low as 1ppm. Drug release properties, DNA biodegradability, and observed cancer cell cytotoxicity of drug loaded DNA-MS suggest that they are appropriate for intratumoral chemotherapy evaluation aimed at improved and less toxic cancer therapy.
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 Iris V Enriquez.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Goldberg, Eugene P.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021197:00001


This item has the following downloads:


Full Text

PAGE 1

1 SYNTHESIS AND CHARACTERIZATION OF DNA NANO-MESO-MICROSPHERES AS DRUG DELIVERY CARRIERS FOR INTRATUMORAL CHEMOTHERAPY By IRIS VANESSA ENRIQUEZ SCHUMACHER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Iris Vanessa Enriquez Schumacher

PAGE 3

3 To my sister Laura Susana and my parents Ir is and Pedro Enriquez, for being a source of constant strength and unconditi onal love and support through all of lifes endeavors. To my grandparents Cecilio and Susana Ca rtagena and Pablo and Carmen Enriquez, for instilling in our families the value and importance of a great education. To my husband Jim, for walking by my side through this journey.

PAGE 4

4 ACKNOWLEDGMENTS I must acknowledge the many advisors, family members, colleagues, and friends whose support made it possible for me to complete my doct oral studies and researc h. I express my most sincere gratitude to my advisor and committee ch airman, Dr. Eugene P. Goldberg for challenging and guiding me through this research and for always being present as a mentor in science and in life. His support, guidance, car ing, and patience allowed me to co mplete this research and to grow as a scientist. I also thank the other members of my committee, Dr. Anthony Brennan, Dr. Chris Batich, Dr. Wolfgang Stre it, Dr. Hassan El-Shall, and Dr. Ronald Baney, who have each individually guided, encourage d, and supported me through this graduate school journey. I extend a special thanks to Gill Brubaker for his friendship and support and for always sharing his knowledge on particle characteriza tion techniques and analysis. In addition, special thanks are extended to the faculty and staff at the Particle Engineering Research Center, especially Gary Scheiffele and Sophie Leone for providing trai ning on equipment and providing access to their facilities. I also thank Drs. Olajompo Moloye, Taili Thula, and John Azeke from the Batich research group for offering their assistance and knowledge in cell culture and DNA purity analysis. Additionally, I thank the Major Analy tical Instrumentation Center, especially Wayne Acree for always preparing my samples to exact specification and for providing access to their instrumentation. I acknowledge my appreciation of Paul Mar tin for his friendship, lively discussions, assistance with SEM, and for patiently sharing hi s knowledge of cell culture. I also gratefully acknowledge Jennifer Wrighton whom has not only provided considerable administrative assistance throughout my graduate studies, but whom has also provided a solid foundation of support through her caring and friendship. A very special thanks is ex tended to Dr. Richard

PAGE 5

5 Doc Connell for his caring mentorship, friendshi p, support, discussions of materials science, and lessons of life that will be carried with me always. I would like to acknowledge seve ral key individuals who laid the initial building blocks and opened the door for this research. First, I extend a special acknowledgement to Dr. Brian Cuevas, who provided a tremendous amount of me ntorship and encouragement when I first started graduate school de spite the fact that he was in the middle of completing his doctoral research and writing his disserta tion. Second, I extend a special thanks to Dr. Joshua Stopek who also provided encouragement and support du ring my first two years in the Goldberg research group and offered a multitude of knowle dge on DNA, biomaterials, and the scientific thought process. Third, I wish to acknowledge Dr Amanda York who took the time to teach me microsphere synthesis and char acterization techniques. F ourth, I sincerely thank the undergraduate research a ssistants, Karly Jacobsen, Jeanney Lew, and Nathan Hicks, who taught me, as much as I taught them, and whose help and hard work facili tated in the completion of this research. Lastly, I acknowledge the other me mbers of the Goldberg research group including Shema Freeman, Dr. Amin Elachchabi, Samesha Barnes, Dr. Lynn Peck, Hungyen Lee, Drs. Margaret and Daniel Urbaniak, and Dr. Adam Reboul whose either support, unconditional friendship, or enthusiasm have made the completion of this work possible. I also express a very special thanks to the following friends Drs. Brian and Robin Hatcher, Dr. Brett Almond, Erica Kennedy, Dr. Adam Feinbe rg, Ayelet Feinberg, Dr. Clay Bohn, Christa Partridge, Dr. Thomas Estes, Nancy Estes, Dr Leslie Wilson, Dr. Cliff Wilson, Dr. Michelle Carman, Dr. Marie Kane, Kevin Kane, Dr. Je ff Wrighton, Jason Ely, Julie Reboul, Kenneth Chung, Chris Long, Victoria Salazar, Jim Selig a, Anika Odukale Edwards, Nelly Volland, Thierry Dubroca, David Jackson, Julian Sheats, Chelsea Magin, and the family of committee

PAGE 6

6 members including Lehua Goldberg, Kathy Bre nnan, and Abby Brennan, whose love, support, encouragement, and unconditional friendship have made this process that much more worth while. I am indebted to my family for their unc onditional love, encouragement, and support which have truly made it possible for me to reach my goals and obtain my dreams. I thank my younger sister and guardian angel, Laura Susana Enriquez, who has always been by my side when I have needed it most. I especially thank my parents, Iris and Pe dro Enriquez, for setting an amazing example for my sister and me a nd whose love and encouragement helped me persevere through each challenge and obstacle placed in my path. I could not have done this without their love and support. I extend a special thanks to my grandparents Susana and Cecilio Cartagena and Carmen and Pablo Enriquez for their life lessons and unconditional love and support. I also thank the rest of my family members whose love and encouragement were also close by. In addition, I thank my in-laws, Su san, Fred, and Jennele Schumacher, Grandpa Bob and Grandma Carole Bevilacqua, Grandma Schum acher, and the rest of the family for their generous love and support. I express my most sincere appreciation to my husband, James Schumacher, whose unending love, patience, encouragement, a nd support made this journey possible.

PAGE 7

7 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ........12 LIST OF FIGURES................................................................................................................ .......14 LIST OF ABBREVIATIONS........................................................................................................23 ABSTRACT....................................................................................................................... ............27 CHAPTER 1 INTRODUCTION................................................................................................................. .29 Research Rationale............................................................................................................. ....29 Specific Aims.................................................................................................................. ........33 Aim 1: Synthesis and Characterizatio n of DNA Nano-Meso-Microspheres..................33 Aim 2: Optimization of DNA Na no-Mesosphere Synthesis...........................................34 Aim 3: In Vitro Evaluations of Mitoxantrone In Situ Loaded DNA NanoMesospheres.................................................................................................................35 Aim 4: In Vitro Evaluations of Methotre xate or 5-Fluorouracil In Situ Loaded DNA and BSA Nano-Mesospheres.......................................................................................36 2 BACKGROUND................................................................................................................... .38 Introduction................................................................................................................... ..........38 Cancer......................................................................................................................... ............39 Types of Cancer...............................................................................................................3 9 Tumor Development........................................................................................................40 Conventional Treatments.................................................................................................42 Surgery and radiation...............................................................................................42 Chemotherapy..........................................................................................................43 Intratumoral Chemotherapy....................................................................................................44 Rationale for Intratumoral Chemotherapy.......................................................................44 Controlled Release Microspheres....................................................................................46 Drug Release Kinetic Models..........................................................................................47 Release profiles........................................................................................................47 Release kinetics........................................................................................................48 Chemotherapy Drugs............................................................................................................. .49 Mitoxantrone................................................................................................................... 50 Methotrexate................................................................................................................... .52 5-Fluorouracil................................................................................................................. .53 Deoxyribonucleic Acid.......................................................................................................... .55 DNA Mesosphere Synthesis...................................................................................................58

PAGE 8

8 Steric Stabilization........................................................................................................... 58 DNA Crosslinking...........................................................................................................59 Ionic crosslinking.....................................................................................................59 Covalent crosslinking...............................................................................................59 Research Goals................................................................................................................. ......61 3 DNA NANO-MESO-MICROSPHERE SYNTHESIS...........................................................63 Introduction................................................................................................................... ..........63 Materials and Methods.......................................................................................................... .64 Materials...................................................................................................................... ....64 Synthesis and characterization.................................................................................64 Cell culture...............................................................................................................65 Synthesis equipment.................................................................................................65 Methods........................................................................................................................ ...65 Solution preparation.................................................................................................65 Crosslinking reaction study......................................................................................68 Pilot microsphere synthesis study............................................................................69 General microsphere synthesis.................................................................................69 Crosslinking determination......................................................................................71 Microsphere characterization...................................................................................73 Evaluation of fibroblast growth................................................................................76 Results........................................................................................................................ .............79 Pilot Microsphere Synthesis Study..................................................................................79 Synopsis...................................................................................................................79 Scanning electron microscopy.................................................................................79 Stabilizing Agent Study...................................................................................................80 Synopsis...................................................................................................................80 Scanning electron microscopy.................................................................................81 Crosslinking Reaction Study...........................................................................................82 General Microsphere Synthesis.......................................................................................83 Synopsis...................................................................................................................83 Particle analysis........................................................................................................83 Surface charge and dispersability.............................................................................89 Microscopy...............................................................................................................91 Evaluation of fibroblast growth................................................................................97 Discussion..................................................................................................................... ........103 Pilot Microsphere Synthesis Study................................................................................103 Stabilizing Agent Study.................................................................................................103 Crosslinking Reaction Study.........................................................................................104 General Microsphere Synthesis Studies........................................................................106 Particle analysis......................................................................................................106 Surface charge and dispersability...........................................................................110 Microscopy.............................................................................................................111 Evaluation of fibroblast growth..............................................................................113 Conclusions.................................................................................................................... .......117 Particle Analysis............................................................................................................11 8

PAGE 9

9 In Vitro Human Dermal Fibroblast Growth..................................................................119 4 OPTIMIZATION OF DNA NANO -MESOSPHERE SYNTHESIS...................................121 Introduction................................................................................................................... ........121 Materials and Methods.........................................................................................................1 22 Materials...................................................................................................................... ..122 Methods........................................................................................................................ .122 Solution preparation...............................................................................................122 DNA nano-mesosphere synthesis...........................................................................124 Mesosphere characterization..................................................................................126 Results & Discussion........................................................................................................... .129 Filtration Study..............................................................................................................1 29 Synopsis.................................................................................................................129 Particle analysis......................................................................................................129 Microscopy.............................................................................................................139 Mixer Speed and Crosslink Concentration Study.........................................................142 Synopsis.................................................................................................................142 Particle analysis......................................................................................................142 Microscopy.............................................................................................................152 Conclusions.................................................................................................................... .......157 Filtration Study..............................................................................................................1 57 Mixer Speed and Crosslink Concentration Study.........................................................157 5 IN VITRO EVALUATIONS OF MITOXANT RONE LOADED DNA NANOMESOSPHERES..................................................................................................................15 9 Introduction................................................................................................................... ........159 Materials and Methods.........................................................................................................1 60 Materials...................................................................................................................... ..160 Synthesis and characterization...............................................................................160 Cell culture.............................................................................................................161 Synthesis equipment...............................................................................................161 Methods........................................................................................................................ .161 Solution preparation...............................................................................................161 Synthesis procedure................................................................................................164 Particle characterization.........................................................................................165 In vitro DNA-MXN-MS characterization procedures............................................168 Assessment of DNA-MXN-MS cytotoxicity.........................................................170 Results and Discussion......................................................................................................... 173 Particle Analysis............................................................................................................17 3 Percent yield...........................................................................................................173 Dry particle size.....................................................................................................176 Hydrated particle size.............................................................................................179 Surface charge analysis..........................................................................................182 Scanning electron microscopy...............................................................................184 Energy dispersive x-ray spectroscopy....................................................................185

PAGE 10

10 In Vitro DNA-MXN-MS Characterization....................................................................187 MXN loading efficiency.........................................................................................187 In vitro MXN release..............................................................................................189 Assessment of DNA-MXN-MS cytotoxicity.........................................................192 Conclusions.................................................................................................................... .......195 Particle Analysis............................................................................................................19 5 In Vitro MXN Loading and Release..............................................................................196 In Vitro Cytotoxicity Analysis.......................................................................................196 6 IN VITRO EVALUATIONS OF DRUG LOAD ED DNA AND BSA NANO-MESOMICROSPHERES................................................................................................................198 Introduction................................................................................................................... ........198 Materials and Methods.........................................................................................................1 99 Materials...................................................................................................................... ..199 Methods........................................................................................................................ .200 Solution preparation...............................................................................................200 Synthesis procedure................................................................................................202 Particle characterization.........................................................................................204 In vitro drug evaluation procedures.......................................................................205 Results and Discussion......................................................................................................... 207 MTX and 5-FU In Situ Loaded DNA-MS and BSA-MS..............................................207 Particle analysis......................................................................................................207 In vitro MTX and 5-FU loading efficiency............................................................219 In vitro MTX and 5-FU release..............................................................................222 DNA-MXN-MS Studies................................................................................................225 Particle analysis......................................................................................................225 In vitro MXN loading efficiency............................................................................232 In vitro MXN release..............................................................................................234 Conclusions.................................................................................................................... .......237 MTX and 5-FU In Situ Loaded DNA-MS and BSA-MS..............................................237 Particle analysis......................................................................................................237 In vitro MTX and 5-FU loading efficiency and release.........................................238 Overall conclusions................................................................................................239 DNA-MXN-MS Studies................................................................................................240 Particle analysis......................................................................................................240 In vitro MXN loading efficiency and release.........................................................241 Overall conclusions................................................................................................242 7 CONCLUSIONS.................................................................................................................. 243 Overview....................................................................................................................... ........243 DNA Nano-Meso-Microspheres Synthesis..........................................................................244 Drug Loading and Release....................................................................................................245 In vitro Cell Growth and Cytotoxicity..................................................................................246 8 FUTURE STUDIES.............................................................................................................24 8

PAGE 11

11 APPENDIX A EDS CHARACTERISTIC X-RAY DE SIGNATIONS AND ENERGIES FOR DNA NANO-MESO-MICROSPHERES.......................................................................................251 B FOLIC ACID MODIFICATION OF MI TOXANTRONE LOADED DNA-MS AND BSA-MS PRELIMINARY RESULTS.................................................................................252 Introduction................................................................................................................... ........252 Materials and Methods.........................................................................................................2 52 Materials...................................................................................................................... ..252 Methods........................................................................................................................ .253 Results........................................................................................................................ ...........253 Preliminary Conclusions.......................................................................................................2 56 C FRUIT AND VEGETABLE D NA EXTRACTION PROTOCOLS....................................257 Overview....................................................................................................................... ........257 Materials...................................................................................................................... .........257 Protocols...................................................................................................................... .........258 Homogenization Medium..............................................................................................258 Enzyme Solution...........................................................................................................258 Banana DNA Extraction................................................................................................258 Onion DNA Extraction..................................................................................................260 Tomato DNA Extraction...............................................................................................261 LIST OF REFERENCES............................................................................................................. 263 BIOGRAPHICAL SKETCH.......................................................................................................277

PAGE 12

12 LIST OF TABLES Table page 2-1 Classes and subclasses of anticancer agents......................................................................50 2-2 Methotrexate dosage ranges fo r various administration routes.........................................53 2-3 5-fluorouracil solubil ity in various media..........................................................................54 2-4 5-fluorouracil dosage ranges for various administration routes........................................55 3-1 The yields, dry mean particle diameter s, and crosslink con centrations for DNA-MS prepared with ionic and cova lent crosslinking agents.......................................................83 3-2 The dry mean particle diameters and size ranges for DNA-MS prepared with ionic and covalent crosslinking agents.......................................................................................84 3-3 Mean and median hydrated particle size values for ionically crosslinked DNA-MS........87 3-4 Percent change in size values for ioni cally and covalently crosslinked DNA-MS............89 3-5 Zeta potential values and dispersab ility times for DNA-MS crosslinked with gadolinium, chromium, and glutaraldehyde......................................................................91 4-1 The yields and percent d ecrease in yield values for DNA-MS synthesized with ionic and covalent crosslinking agents.....................................................................................129 4-2 The dry mean particle diameter values for the 20m-filtered and the non-filtered DNA-MS synthesized with ionic and covalent crosslinking agents................................130 4-3 The dry mean particle diameter values for the 20m-filtered and the non-filtered DNA-MS synthesized with ionic and covalent crosslinking agents................................133 4-4 The dry and hydrated mean particle diamet ers, percent change in size values, and crosslink concentrations for DNA-MS............................................................................139 4-5 Percent yield values for DNA-MS prepar ed at varying mixer speeds and gadolinium crosslink concentrations...................................................................................................143 4-6 Mean dry particle size and size range va lues for DNA-MS prepared at varying mixer speeds and gadolinium crosslink concentrations.............................................................144 4-7 Dry and hydrated mean particle size and percent swelling values for DNA-MS prepared at the 1550rpm mixer speed condition..............................................................150 4-8 Zeta potential values fo r DNA-MS prepared with 20%MEQ, 50%MEQ, and 120%MEQ crosslink concentrations...................................................................................................152

PAGE 13

13 5-1 Yield and theoretical yield values for DNA-MXN-MS prepared at the 20%MEQ, 50%MEQ, and 120%MEQ crosslink conditions.................................................................174 5-2 Dry mean particle diameter a nd size range values for DNA-MXN-MS.........................177 5-3 Dry mean particle diameter valu es for DNA-MXN-MS and blank DNA-MS................177 5-4 Mean dry and hydrated particle diameter s with percent change in size due to swelling....................................................................................................................... .....180 5-5 Zeta potential values for DNA-MXN-MS and blank DNA-MS with their respective change in surface charge..................................................................................................182 5-6 MXN loading and loading efficiencies for DNA-MXN-MS, BSA-MS, and GEL-MS..188 6-1 Percent yield and theoretical yield valu es for drug loaded DNA-MS and BSA-MS.......208 6-2 Dry mean particle diameter and size range values for MTX in situ loaded BSA-MS and DNA-MS...................................................................................................................21 0 6-3 Dry mean particle diameter values for 5-FU in situ loaded BSA-MS and DNA-MS.....213 6-4 Loading efficiency values for MTX and 5-FU in situ loaded BSA-MS and DNA-MS..219 6-5 Percent yield and theoretical yield values for DNA-MXN-MS.......................................226 6-6 Dry mean particle diameter values for crosslinked and non-crosslinked DNA-MXNMS and DNA-MS............................................................................................................227 6-7 Percent loading and loading e fficiency values for DNA-MXN-MS...............................233 A-1 Characteristic x-ray designations and en ergies for elements analyzed via energy dispersive x-ray spectroscopy..........................................................................................251 B-1 Dry mean particle diameters for FA and MXN loaded DNA-MS and BSA-MS............254

PAGE 14

14 LIST OF FIGURES Figure page 2-1 The 4 phases of the life cycle of a cell...............................................................................41 2-2 Physiological capabilities of ge netically altered cancerous cells......................................42 2-3 Therapeutic dose chart illustrating a toxi c dose (red line), moderate dose (blue line), and a controlled release curve (dashed line)......................................................................45 2-4 Diffusion (blue line) and erosion (o range line) drug release profiles................................47 2-5 Drug release due to diffusion or erosion............................................................................48 2-6 First order profiles for (A) drug release and (B) release kinetics......................................48 2-7 Drug release profiles for A) zero order release kinetics and B) Higuchi release kinetics....................................................................................................................... ........49 2-8 Chemical structure for mitoxantrone hydrochloride..........................................................51 2-9 Chemical structure of methotrexate...................................................................................52 2-10 Chemical structure of 5-fluorouracil..................................................................................54 2-11 A drawing of DNA.......................................................................................................... ...56 2-12 Cartoon of the molecular structure of a DNA subunit.......................................................57 2-13 A brief schematic of the DNA-MS synthesis process.......................................................59 2-14 Chemical structur e of glutaraldehyde................................................................................60 2-15 Chemical structures of the four DNA base groups............................................................60 2-16 Chemical structure of genipin............................................................................................6 1 3-1 A cartoon of the DNA generi c repeat unit used for a ll crosslink calculations..................71 3-2 Schematic drawing of possible DNA crossl inking sites for tr ivalent cations, M = Gd3+, Cr3+, or Fe3+..............................................................................................................72 3-3 Schematic drawing of a possible mechan ism for glutaraldehyde to covalently crosslink DNA. (Note: R1 and R2 represent the remainder of DNA molecule.)...............73 3-4 The chemical conversion of MTS to formazan..................................................................77 3-5 SEM micrograph of chromium cro sslinked DNA-MS (Magnification:1,000x)................80

PAGE 15

15 3-6 SEM micrographs of DNA-MS prepared with CAB concentrations of A) 3% (w/v), B) 5% (w/v), C) 10% (w/v), D) 25% (w/v) (Magnifications: 2,000x), and E) 25% (w/v) (Magnification: 13 0x; Scale bar: 200m)................................................................81 3-7 Particle size distributions of ionically crosslinked DNA-MS under dry conditions..........85 3-8 Particle size distributions for covalently crosslinke d DNA-MS under dry conditions......86 3-9 Particle size distribut ions of ionically crossl inked DNA-MS under hydrated conditions..................................................................................................................... ......87 3-10 Particle size distributi on of the glutaraldehyde cro sslinked DNA-MS under hydrated conditions..................................................................................................................... ......88 3-11 Zeta potential chart for DNA-MS crossli nked with gadolinium (Gd), chromium (Cr), and glutaraldehyde (GTA).................................................................................................90 3-12 Optical microscopy images of DNA-MS ioni cally crosslinked with A) chromium, B) gadolinium, and C) iron trivalent cations (Magnification: 400x)......................................92 3-13 Optical microscopy images of DNA-MS covalently crosslinked with A) glutaraldehyde, and B) geni pin (Magnification: 400x). Note: Red circles highlight un-crosslinked DNA strands..............................................................................................93 3-14 SEM images of DNA-MS prepared with A) chromium and B) gadolinium trivalent cationic crosslinking agents (Magnification: 2000x).........................................................93 3-15 SEM image of a DNA-MS aggregate cr osslinked with triv alent iron cations (Magnification: 4,000x).....................................................................................................94 3-16 SEM images of ionically crosslinked DNA-MS with smooth surf ace topographies in the nano-mesosphere size range: A) ch romium (Magnification: 9,500x) and B) gadolinium crosslinked DNA-MS (Magnification: 8,500x)..............................................94 3-17 SEM images of ionically crosslinked DNA-MS: A) aggregated chromium and B) discrete gadolinium crosslinked DNA-MS (Magnification: 550x)...................................95 3-18 SEM images of DNA-MS prepared with A) genipin (Magnification: 4,000x) and B) glutaraldehyde (Magnification: 2,00 0x) covalent crosslink agents...................................95 3-19 EDS spectra collected on DNA-MS crosslinked with gadolinium....................................96 3-20 EDS spectra collected on DNA-MS crosslinked with chromium......................................97 3-21 EDS spectra collected on DNA-MS crosslinked with iron................................................97 3-22 Fibroblast proliferation prof iles for DNA treatment conditions........................................98

PAGE 16

16 3-23 Optical microscopy images of crystal violet stained normal human dermal BJ fibroblast cells exposed to A) media w ith cells, B) 100g DNA, and C) 500g DNA treatment conditions (Magnification: 50x)........................................................................99 3-24 Fibroblast proliferation values for cr osslinked DNA-MS and DNA treatment groups at the 100g condition.......................................................................................................99 3-25 Optical microscopy images of crystal violet stained normal human dermal BJ fibroblast cells exposed to DNA-MS prepared with A) ga dolinium, B) chromium, C) iron, and D) glutaraldehyde crosslinking ag ents at the 100g condition and E) media with cells control group condi tion (Magnification: 50x).................................................101 3-26 Fibroblast proliferation va lues for the crosslinked DNAMS treatment groups at the 25g condition................................................................................................................. 102 3-27 Optical microscopy images of crystal violet stained normal human dermal BJ fibroblast cells exposed to DNA-MS prepar ed with A) gadolinium, B) chromium, and C) glutaraldehyde crosslinking agents at the 25g condition and D) media with cells control group condition (Magnification: 50x).........................................................102 3-28 Schematic drawing of possible interac tions between chromium and iron trivalent cations and phosphate oxygens and base pair guanine and adenine N7 nitrogen atoms. (Note: M = Cr3+ or Fe3+.).....................................................................................107 3-29 A drawing of the 4 phase s of the cell life cycle...............................................................113 3-30 A drawing of the 4 phases of the cell life cycle, including the G0 quiescence phase......114 4-1 A particle size distribution comparison of non-filtered and 20m-filtered chromium crosslinked DNA-MS u nder dry conditions....................................................................131 4-2 A particle size distribution comparison of non-filtered and 20m-filtered gadolinium crosslinked DNA-MS u nder dry conditions....................................................................132 4-3 A particle size distribution comp arison of non-filtered and 20m-filtered glutaraldehyde crosslinked DNA-MS under dry conditions............................................132 4-4 An illustrative depiction of the five mechanisms responsible for interparticle bonding: A.) solid bridges, B.) liquid br idges, C.) van der Waals forces, D.) electrostatic forces, a nd E.) interlocking bonds...............................................................134 4-5 Hydrated particle size distribution comparisons of the 20m-filtered ionically crosslinked DNA-MS.......................................................................................................135 4-6 A particle size distribution comparison of the 20m-filtered, chromium crosslinked DNA-MS under dry and hydrated conditions..................................................................136

PAGE 17

17 4-7 A particle size distribution comparison of 20m-filtered, gadolinium crosslinked DNA-MS under dry and hydrated conditions..................................................................136 4-8 A particle size distribution comparison of 20m-filtered, glutaraldehyde crosslinked DNA-MS under dry and hydrated conditions..................................................................137 4-9 A particle size distribution comparison of 20m-filtered and non-filtered chromium crosslinked DNA-MS unde r hydrated conditions............................................................138 4-10 A particle size distribution comparison of 20m-filtered and non-filtered gadolinium crosslinked DNA-MS unde r hydrated conditions............................................................138 4-11 A particle size distribution comp arison of 20m-filtered and non-filtered glutaraldehyde crosslinked DNAMS under hydrated conditions...................................139 4-12 Optical images of the 20m-filtered DNAMS crosslinked with A) gadolinium, B) chromium, and C) glutaralde hyde (Magnification: 200x)...............................................140 4-13 Scanning electron microscopy images of the 20m-filtered DNA-MS crosslinked with A) chromium (Magnification: 4, 500x), B) gadolinium (Magnification: 2,000x), and C) glutaraldehyde (Magnification: 1,000x)...............................................................141 4-14 Graphical representation of yield values generated by each DNA-MS synthesis condition...................................................................................................................... ....143 4-15 A dry particle size distribution comparis on of DNA-MS synthesized at varying mixer speeds at the 20%MEQ crosslink concentration condition...............................................145 4-16 A dry particle size distribution comp arison of DNA-MS synt hesized at varying crosslink densities at the 950rpm mixer speed condition................................................146 4-17 A dry particle size distribution comp arison of DNA-MS synt hesized at varying crosslink densities at the 1250rpm mixer speed condition..............................................146 4-18 A dry particle size distribution comp arison of DNA-MS synt hesized at varying crosslink densities at the 1550rpm mixer speed condition..............................................147 4-19 Dry and hydrated particle size distribu tions for DNA-MS prepared at 1550rpm and crosslinked with gadolinium to 20%MEQ........................................................................148 4-20 Dry and hydrated particle size distribu tions for DNA-MS prepared at 1550rpm and crosslinked with gadolinium to 50%MEQ........................................................................149 4-21 Dry and hydrated particle size distribu tions for DNA-MS prepared at 1550rpm and crosslinked with gadolinium to 120%MEQ......................................................................149 4-22 Zeta potential values fo r DNA-MS prepared with 20%MEQ, 50%MEQ, and 120%MEQ crosslink concentrations...................................................................................................151

PAGE 18

18 4-23 SEM images of DNA-MS crosslinked with 20%MEQ gadolinium at the A) 950rpm, B) 1250rpm, and C) 1550rpm mixer speed conditions (Magnifications: 2,000x)...........152 4-24 SEM images of DNA-MS crosslinked with 50%MEQ gadolinium at the A) 950rpm, B) 1250rpm, and C) 1550rpm mixer speed conditions (Magnifications: 1,500x)...........153 4-25 SEM images of DNA-MS crosslinked with 120%MEQ gadolinium at the A) 950rpm, B) 1250rpm, and C) 1550rpm mixer speed conditions (Magnifications: 1,000x, 2,000x, and 1,500x, respectively)....................................................................................154 4-26 EDS spectra collected on DNA-MS pr epared at the 1550rpm mixer speed and 20%MEQ crosslink concentration condition.....................................................................155 4-27 EDS spectra collected on DNA-MS pr epared at the 1550rpm mixer speed and 50%MEQ crosslink concentration condition.....................................................................156 4-28 EDS spectra collected on the DNA-MS prepared at the 1550rpm mixer speed and 120%MEQ crosslink concentration condition...................................................................156 5-1 The first step in the reaction betw een lactate from the LDH enzyme and NAD+ in the MTS cytotoxicity assay used for these studies.......... ......................................................171 5-2 The second and final step in the MTS assay used for these studies where NADH reacts with the tetrazolium salt to produce NAD+ and a red formazan dye.....................171 5-3 Yield values for DNA-MXNMS prepared at the 20%MEQ, 50%MEQ, and 120%MEQ crosslink conditions.........................................................................................................17 4 5-4 Theoretical yield values include the weight of the in situ loaded MXN for DNAMXN-MS prepared at the 20%MEQ, 50%MEQ, and 120%MEQ crosslink conditions......175 5-5 Graphical comparison of blank DNA-MS yield and DNA-MXN-MS theoretical yield values......................................................................................................................... ......175 5-6 A dry particle size distribution co mparison of DNA-MXN-MS synthesized at varying crosslink concentrations......................................................................................176 5-7 A comparison of DNA-MXN-MS and bla nk DNA-MS dry particle size distributions at the 20%MEQ crosslink condition..................................................................................178 5-8 A comparison of DNA-MXN-MS and bla nk DNA-MS dry particle size distributions at the 50%MEQ crosslink condition..................................................................................178 5-9 A comparison of DNA-MXN-MS and bla nk DNA-MS dry particle size distributions at the 120%MEQ crosslink density condition...................................................................179 5-10 Hydrated particle size distributions for DNA-MXN-MS prepared at the 50%MEQ and 120%MEQ crosslink density conditions...........................................................................180

PAGE 19

19 5-11 A dry and hydrated particle size dist ribution comparison of DNA-MXN-MS at the 50%MEQ crosslink condition...........................................................................................181 5-12 A dry and hydrated particle size dist ribution comparison of DNA-MXN-MS at the 120%MEQ crosslink condition..........................................................................................181 5-13 A zeta potential comparison of DNA-MXNMS and blank DNA-MS prepared at the 20%MEQ crosslink condition...........................................................................................182 5-14 A zeta potential comparison of MXN load ed and blank DNA-MS prepared at the 50%MEQ crosslink condition...........................................................................................183 5-15 A zeta potential comparison of MXN load ed and blank DNA-MS prepared at the 120%MEQ density condition............................................................................................183 5-16 Zeta potential values for DNA-MXN-MS pr epared at varying crosslink conditions......184 5-17 Scanning electron micrographs of DNA-MXN-MS prepared at the A) 20%MEQ (Magnification: 2,000x), B) 50%MEQ (Magnification: 3,000x), and C) 120%MEQ crosslink conditions (M agnification: 2,000x)..................................................................185 5-18 EDS spectra of DNA-MXNMS prepared at the 20%MEQ crosslink condition..............186 5-19 EDS spectra of DNA-MXNMS prepared at the 50%MEQ crosslink condition..............186 5-20 EDS spectra of DNA-MXNMS prepared at the 120%MEQ crosslink condition............187 5-21 Percent MXN loading comparisons of DNA-MXN-MS, BSA-MS, and GEL-MS.........188 5-22 MXN loading efficiency comparisons for DNA-MXN-MS, BSA-MS, and GEL-MS...189 5-23 First 24 hour MXN release profile for DNA-MXN-MS at varying crosslink concentrations................................................................................................................. .190 5-24 MXN release profiles for DNA-MXNMS at Hour 1, Day 1, and Day 75......................190 5-25 Total MXN release profile for vary ing crosslink DNA-MXN-MS conditions................191 5-26 Schematic representation of crosslin k concentration on drug release, A) high crosslink and B) low cr osslink conditions.......................................................................192 5-27 In vitro cytotoxicity profiles of DNA-MXNMS and DNA-MS on mLLC cells............193 5-28 Cytotoxicity profiles fo r free MXN dose conditions.......................................................194 6-1 Percent yield values for 15% (w/w) MTX in situ loaded BSA-MS and DNA-MS.........209 6-2 Percent yield values for 30% (w/w) 5-FU in situ loaded BSA-MS and DNA-MS.........209

PAGE 20

20 6-3 Dry particle size dist ribution for 15% (w/w) MTX in situ loaded BSA-MS prepared at 1250rpm with 8% (w/w) GTA.....................................................................................210 6-4 Dry particle size distri butions of 15% (w/w) MTX in situ loaded BSA-MS prepared at 1250rpm and 1550rpm.................................................................................................211 6-5 Dry particle size di stribution of 15% (w/w) in situ loaded DNA-MS.............................212 6-6 Dry particle size di stribution for BSA-MS prepared at 1550rpm and in situ loaded with 15% (w/w) or 30% (w/w) MTX..............................................................................212 6-7 Dry particle size distri bution for 30% (w/w) 5-FU in situ loaded BSA-MS...................213 6-8 Dry particle size distri bution for 30% (w/w) 5-FU in situ loaded DNA-MS..................214 6-9 SEM micrograph of 15% (w/w) MTX in situ loaded BSA-MS prepared at the A) 1250rpm mixer speed and B) 1550rpm mi xer speed (Magnifications: 3,000x)..............215 6-10 SEM micrograph of 15% (w/w) MTX in situ loaded BSA-MS prepared at the 1550rpm mixer speed with 2mL of GTA (Magnification: 11,000x)...............................215 6-11 SEM micrographs of the 30% (w/w) MTX in situ loaded BSA-MS at magnifications of A) 3,000x and B) 10,000x...........................................................................................216 6-12 SEM micrographs of 15% (w/w) MTX in situ loaded DNA-MS at magnifications of A) 2,000x and B) 6,000x..................................................................................................217 6-13 SEM micrographs of 30% 5-FU in situ loaded A) BSA-MS and B) DNA-MS (Magnifications: 3,000x)..................................................................................................217 6-14 SEM micrograph of 30% (w/w) 5-FU lo aded BSA-MS (Magnification: 20,000x)........218 6-15 SEM micrograph of 30% 5-FU in situ loaded DNA-MS (Magnification: 10,000x).......218 6-16 Loading efficiency comparison chart for 15% (w/w) MTX in situ loaded BSA-MS prepared at different mixer speeds...................................................................................220 6-17 Loading efficiency comparison chart for BSA-MS loaded with different MTX concentrations prepared at the 1550rpm mixer speed......................................................220 6-18 Loading efficiency comparison chart of the 15% (w/w) MTX in situ loaded DNAMS and BSA-MS prepared at the 1550rpm mixer speed................................................221 6-19 Loading efficiency compar ison chart of the 30% 5-FU in situ loaded DNA-MS and BSA-MS......................................................................................................................... ..221 6-20 MTX release profiles for in situ loaded DNA-MS and BSA-MS for the first 24 hours..222

PAGE 21

21 6-21 MTX release profiles for in situ loaded DNA-MS and BSA-MS for duration of study.......................................................................................................................... .......222 6-22 Higuchi square root time kinetics MTX in vitro release for the 30% (w/w) MTX in situ loaded BSA-MS condition (Hours 1 through 8).......................................................223 6-23 5-FU release profiles for in situ loaded DNA-MS and BSA-MS....................................224 6-24 Higuchi square root time kinetics 5-FU in vitro release for the 23% (w/w) 5-FU in situ loaded DNA-MS condition (Hours 1 through 8)......................................................224 6-25 Higuchi square root time kinetics 5-FU in vitro release for the 27% (w/w) 5-FU in situ loaded BSA-MS condition (H our 1 through Day 35)...............................................225 6-26 Theoretical yield values for DNAMXN-MS prepared at varying MXN concentrations................................................................................................................. .226 6-27 Percent yield values for MXN in situ loaded and non-loaded DNA-MS........................227 6-28 Dry particle size distribution for DNA-MXN-MS prepared at varying MXN concentrations................................................................................................................. .228 6-29 Dry particle size distribution compar ison of DNA-MXN-MS prepared with and without gadolinium crosslinking......................................................................................229 6-30 Dry particle size distribution comp arison of non-crosslinked DNA-MXN-MS and blank DNA-MS................................................................................................................229 6-31 SEM micrographs of DNA-M XN-MS prepared at the A) 10% (w/w), B) 15% (w/w), and C) 25% (w/w) MXN concentra tions (Magnifications: 2,000x)................................230 6-32 SEM micrographs of DNA-MXN-MS prepar ed with A) gadolinium crosslinking and B) no gadolinium crosslinki ng (Magnifications: 2,000x)................................................231 6-33 SEM micrographs of DNA-MS prepared w ith no gadolinium crosslinking, A) blank DNA-MS and B) MXN in situ loaded DNA-MS (Magnifications: 3,000x)...................231 6-34 Loading efficiency comparison chart for DNA-MXN-MS prepared at varying MXN concentrations................................................................................................................. .232 6-35 Loading efficiency comparison chart for DNA-MXN-MS prepared at with and without gadolinium crosslinking......................................................................................233 6-36 The first 24 hour MXN release profiles for DNA-MXN-MS prepared at varying MXN concentrations........................................................................................................235 6-37 MXN release profiles for DNA-MXN-MS prep ared at varying MXN concentrations...235

PAGE 22

22 6-38 The first 24 hour MXN release profil es for DNA-MXN-MS prepared with and without gadolinium crosslinking......................................................................................236 6-39 MXN release profiles for DNA-MXN-MS prepared with and without gadolinium crosslinking................................................................................................................... ...236 B-1 Particle size distributi on comparison of FA and MXN loaded DNA-MS conditions.....253 B-2 Particle size distributi on comparison of FA and MXN loaded BSA-MS conditions......254 B-3 SEM micrographs of DNA-MS with A) 0.5% FA, B) 1% FA, and C) no FA (Magnifications: 3,000x)..................................................................................................255 B-4 SEM micrographs of DNA-MXN-MS with A) 0.5% FA, B) 1% FA, and C) no FA (Magnifications: 2,000x)..................................................................................................255 B-5 SEM micrographs of BSA-MS with A) 0.5% FA, B) 1% FA, and C) no FA (Magnifications: 3,000x)..................................................................................................255 B-6 SEM micrographs of BSA-MXN-MS with A) 0.5% FA, B) 1% FA, and C) no FA (Magnifications: 3,000x)..................................................................................................256

PAGE 23

23 LIST OF ABBREVIATIONS %MEQ Percent molar equivalence % (w/v) Percent weight per volume % (w/w) Percent weight per weight %Yield Percent yield g Microgram (1 x 10-6g) L Microliter (1 x 10-6L) m Micrometer (1 x 10-6m) DNA Density of aqueous DNA solution used in MS preparation 5-FU 5-Fluorouracil ANOVA Analysis of variance BSA Bovine serum albumin BSA-MS BSA nano-meso-microsphere(s) BSS Balanced salt solution CAB Cellulose acetate butyrate dissolved in 1,2-dichloroethane Cr Chromium trivalent cation CDNA Concentration of the aqueous DNA solution used in MS preparation d Particle diameter DD Dry DNA-MS diameter DH Hydrated DNA-MS diameter DMEM Dulbeccos Modified Eagles Media with L-glutamine DNA Deoxyribonucleic acid DNA-MS DNA nano-meso-microsphere(s) DNA-MXN-MS Mitoxantrone in situ loaded DNA nano-meso-microsphere(s)

PAGE 24

24 EDS Energy dispersive x-ray spectroscopy EDTA Ethylenediamine tetraacetic acid EMEM Eagles Minimum Essential Modified Media FBS Fetal bovine serum Fe Iron trivalent cation FSEM Field emission scanning electron microscope G0 Quiescence or non-prolifer ating phase of the cell cycle G1 Intermitotic phase of the cell cycle G2 Premitotic phase of the cell cycle Gd Gadolinium trivalent cation GEL Gelatin GEL-MS Gelatin nano-meso-microsphere(s) Genipin GEN GTA Glutaraldehyde hr Hour(s) KeV Kiloelectron volt (1 x 103eV) LDH Lactate dehydrogenase M Molar or Mitotic phase in the cell cycle MEM Minimum essential media mg Milligram (1 x 10-3g) min Minute(s) mL Milliliter (1 x 10-3L) mLLC Murine Lewis lung carcinoma

PAGE 25

25 mm Millimeter mM Millimolar (1 x 10-3M) MS Nano-meso-microsphere(s) MTS Colorimetric cell growth or viability assay MTX Methotrexate mV Millivolts MXN Mitoxantrone NADH Nicotinamide adenine dinucleotide linked dehydrogenase NAD+ Nicotinamide adenine dinucleotide NEAA Non-essential amino acid nm Nanometer (1 x 10-9m) PBS Phosphate buffered saline rpm Revolutions per minute S DNA synthesis phase of the cell cycle SEM Scanning electron microscopy TCA Trichloroacetic acid UV-Vis Ultraviolet visible spectroscopy VDNA Volume of aqueous DNA solution used in MS preparation WF Final weight of DNA-MS WTSOLID True solid weight of BSA or DNA WTDNA True solid weight of DNA WX Solid weight of crosslinki ng agent added during MS synthesis WY Weight of the in situ loaded mitoxantrone

PAGE 26

26 XgDrug Weight of methot rexate or 5-fluorouracil XgMXN Weight of mitoxantrone

PAGE 27

27 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SYNTHESIS AND CHARACTERIZATION OF DNA NANO-MESO-MICROSPHERES AS DRUG DELIVERY CARRIERS FOR INTRATUMORAL CHEMOTHERAPY By Iris Vanessa Enriquez Schumacher August 2007 Chair: Eugene P. Goldberg Major: Materials Science and Engineering Conventional cancer chemotherapy results in systemic toxicity which severely limits effectiveness and often adversely a ffects patient quality of life. There is a need to find new drugs and delivery methods for less toxic therapy. Previous studies concerning DNA complexing with chemotherapy drugs suggest unique opportunities for DNA as a mesosphere drug carrier. The overall objective of this research was devoted to the synthesis and evaluation of novel DNA-drug nano-mesospheres designed for localized chemotherapy via intr atumoral injection. My rese arch presents DNA nano-mesomicrospheres (DNA-MS) that were prepared us ing a modified steric stabilization method originally developed in this lab for the preparation of album in MS. DNA-MS were prepared with glutaraldehyde covalent crosslinking (genipin crosslinking was attempted) through the DNA base pairs. In addition, novel crossli nking of DNA-MS was demonstrated using chromium, gadolinium, or iron cations through the DNA phosphate groups. Covalent and ionic crosslinked DNA-MS syntheses yielded smooth and spherical particle morphologies with multimodal size distributions.

PAGE 28

28 Optimized DNA-MS syntheses produced particles with narrow and normal size distributions in the 50nm to 5m diameter size range. In aqueous dispersions approximately 200% swelling was observed with dispersion stability for more than 48 hours. Typical process conditions included a 1550rpm initial mixing speed and particle filtration through 20m filters to facilitate preparation. DNA-MS were in situ loaded during synthesis for the first time with mitoxantrone, 5fluorouracil, and methotrexate. DNA-MS drug in corporation was 12%(w/w) for mitoxantrone, 9%(w/w) for methotrexate, and 5%(w/w) for 5-fluorouracil. In vitro drug release into phosphate buffered saline was observed for over 35 days by minimum sink release testing. The effect of gadolinium crosslink concentrati on on mitoxantrone release was ev aluated at molar equivalences in the range of 20% to 120%. The most hi ghly crosslinked DNA-MS exhibited the longest sustained release. The drug efficacy of mitoxantrone loaded DNA-MS was evaluated in vitro using a murine Lewis lung carcinoma cell line and a significant cytotoxic response was found at mitoxantrone doses as low as 1ppm. Drug release properties, DNA biodegradability, and observed cancer cell cytotoxicity of drug loaded DNA-MS suggest th at they are appropriate for intratumoral chemotherapy evaluation aimed at im proved and less toxic cancer therapy.

PAGE 29

29 CHAPTER 1 INTRODUCTION Research Rationale Cancer is the second most common cause of death in the United States, following heart disease, and accounts for every 1 in 4 deaths nationwide according to the 2005 and 2006 cancer statistic reports published by th e American Cancer Society.1, 2 Over one-half of these reported deaths can be attributed to cancers of the brea st, prostate, lung and bronchus, and colon and rectum for both men and women.1 Leukemia and cancer of the br ain and central nervous system account for one-half of a ll children’s fatalities.2 Primary modalities of treatment for these and other forms of cancer incl ude surgery, radiation, and systemic chemotherapy.3 Conventional systemic chemotherapy is often used as the standard of care for most cancer cases, and, in many circumstances, is given in combination with surger y or radiation therapy in order to decrease the size of the tumor prior to ex cision or enhance the effects of radiation on cancerous cells.3 Conventional systemic chemotherapy functions by attacking rapidly di viding cancer cells while traveling through the bloodstream. When used as a sole form of treatment, conventional systemic chemotherapy often provides limited eff ectiveness and compromises the quality of life for its patients.4 Patients undergoing systemic chemot herapy not only experience a variety of acute side effects such as nausea, mouth sores, alopecia (hair loss), irrita tion of the veins, and induced menopause, but may also experience debilitating long term effects such as multiple organ failure and even death.3, 5 Adverse side effects, such as those just previously mentioned, arise because chemotherapeutic agents are incap able of differentiating between cancerous cells and healthy rapidly dividing cells such as hair follicle, gastroin testinal, and bone marrow cells. The inability for chemotherapy agents to diffe rentiate between healt hy cells and cancerous cells are not solely to blame for their lack of therapeutic efficacy. When delivered

PAGE 30

30 intravenously, chemotherapeutic agents are very effective against blood borne cancers such as leukemia and lymphoma, however, they tend to be less effective against solid tumor cancers such as that of the breast, lung and bronchus, pr ostate, colon and rectum, and head and neck.6, 7 The therapeutic deficiency of systemic chemotherapy against solid tumor cancers can be attributed to the inability for chemotherapy agents to pene trate the tumor cells due to their intricate vasculature and interstitium.6, 7 It is estimated that there will be 1,399,760 new cases of cancer diagnosed this year resulting in a vast demand for conventiona l treatment modalities such as systemic chemotherapy.2 This large demand for systemic chemotherapy produces an urgent need for researchers to develop new delivery methods to localize chemotherapy to thereby reduce its systemic toxicity and increas e its therapeutic efficacy. The idea of localizing chemotherapy has been around since the onset of its use in the 1940s.8 In the 1950s, scientists such as C.T. Kl opp and others experimented with localizing chemotherapy agents such as nitrogen mustard by intra-arterial infusion.8, 9 They found that the intra-arterial infusion of these agents led to increased efficacy, however, the dose required for this type of administration did not eradicate prob lems with systemic toxicity. Over the next 40 years, these findings led to the development of dr ug delivery carriers in the form of microspheres synthesized from proteins, polysaccharides, and synthetic polymers.10 The most successful of the aforementioned microsphere formulations were those synthesized with bovine serum albumin or human serum albumin.10 The success of albumin protein as a microsphere drug carrier can be attributed to its high st ability, biodegradability, and ease of processing.10 Past research has shown the ability of albumin to load various dr ugs such as doxorubicin, mitomycin C, cisplatin, 5-fluorouracil, and mitoxantrone.4, 10, 11 Recently it has been shown that the intratumoral

PAGE 31

31 delivery of mitoxantrone loaded albumin microspheres localized the activity of mitoxantrone and increased survival time in a muri ne mammary adenocarcinoma model.4, 12 This research also indicated that the maximum tolerable dose of m itoxantrone increased 8-fold when delivered intratumorally rather than intravenously.12 The use of albumin as a chemotherapy drug carri er has been successful with a variety of anti-tumor agents, however, it has one major disadvantage in that it has a low drug-carrying capacity.10 The maximum drug payload for mitoxantrone loaded albumin microspheres has been limited to 15% (w/w).4, 11 Studies have been conducted to increase the payload of cationic drugs, such as doxorubicin, into albumin microspheres by incorporating anionic constituents into the albumin carrier matrix.10 These studies have shown that th e incorporation of anionic materials help to increase the loading cap acity of the albumin microspheres by 4 times the original value, however, the downfall was that the rele ase of drug via ionic binding was rapid.10 Recent research has shown similar results ex cept that drug release was prolonged.13 Studies conducted recently illustrated that the incorporation of anionic poly(glutamic) acid into gelatin microspheres did not increase the payload of cat ionic mitoxantrone, however, it did extend its release over 100 days.13 Many know DNA for its role as a carrier of ge netic information; however, very few have given thought to the possible bi omaterial functions of this su pramolecular biopolymer. For the past 30 years scientists have focused their effo rts on the delivery of viral DNA, otherwise known as plasmid DNA, for its function as a gene expr essor and for its novel use as a therapeutic for cancer and genetic disorders.14-16 With most attention focused on plasmid DNA, scientists have overlooked the attractive properties that chromosomal DNA has to offer. It has only been within

PAGE 32

32 the past decade that scientists have shown inte rest in the biomaterial functions of chromosomal DNA. In the late 1990s, the use of DNA as a natu ral occurring functional biopolymer surfaced. Scientists began to report th e successful coating of non-so luble DNA-alginic acid films on Millipore filters for the successful absorption and removal of toxic interc alating pollutants such as benzopyrene.17 Insoluble DNA films cast on polysulf one dialysis membranes were also researched and scientists illu strated that the DNA films in creased the hydrophilicity and hemocompatibility of the hydrophobic polysulfone surfaces.18 Scientists have even shown that DNA-lipid films can be used as drug deliv ery vehicles and antimicrobial surfaces by incorporating agents such as ethidium bromide and silver ions into the film.19, 20 Most recently, scientists have shown that DNA may be used as a structural biopolymer for the development of biodegradable scaffolds for tissue engineering.21 In the early 1980s, Tr ouet et al complexed DNA to adriamycin and daunorubicin chemothera peutic agents and injected the DNA-drug complex intravenously into rabbit and mice models.22, 23 These studies found that the DNA-drug complexes reduced the cardiotoxicity a nd improved the half-lives of the drugs.22, 23 The novelty of the use of DNA as a functional biopolymer and the need for developing new methods for delivering chemotherapy is what inspired the research presented in this dissertation. The research pr esented here describes the development and use of DNA nanomeso-microspheres as drug carriers. Shortly af ter these studies began, a collaboration with Dr. Amanda York looked at incorporating 1.5% (w/w) DNA into albumin microspheres.24 The results from these studies indi cated that DNA blended albumin mi crospheres not only loaded the same amount of drug as the albumin microsphe res, but also extended release over 1000 hours.13, 24 The optimistic payload and release character istics of the DNA blended albumin microspheres,

PAGE 33

33 which had not yet been reported in the literature, further solidifi ed the concept of using DNA as a sole drug carrying material. Specific Aims The research presented in this dissertation was devoted to the development of a novel drug delivery carrier for localized or re gional chemotherapy using DNA. Aim 1: Synthesis and Characteriza tion of DNA Nano-Meso-Microspheres Nano-meso-microspheres from DNA derive d from herring testes (DNA-MS) were synthesized to a target dry mean diameter range of 50nm to 20m, wher e at least 60% of all particles prepared were within the mesosphere size range of 1m to 10m and < 5% of all particles were greater than 10m in size. Pa rticles less than 1m in diameter were also acceptable. Hydrated particle diameters were to be less than 25m. DNA-MS were sought to obtain aqueous dispersion stabilit y of over 24 hours and elicit minima l toxic effect on fibroblast cells in culture. DNA-MS were prepared using a modified steric stabilization process developed in this lab and were either covalently crosslinked with ge nipin or glutaraldehyde or ionically crosslinked with chromium (III) potassium sulfate, gadolin ium (III) chloride, or iron (III) nitrate. A crosslinking reaction study was conducted, prior to any synthesis studies, to determine the time needed to establish DNA crosslinking with the covalent or ionic crosslinking agents. Crosslinking time was estimated by measuring the time required for each agent to precipitate DNA from solution. The crosslinking reaction study was conducted at room temperature with a maximum crosslink time of two hours to mimic conditions observed during DNA-MS synthesis. Preliminary studies were also c onducted to determine the optimal concentration of the stabilizing agent used during DNA-MS synthesis.

PAGE 34

34 The morphology and topography of the DNA-MS were examined by optical microscopy and scanning electron microscopy. The presen ce of trivalent cations in the DNA-MS was assessed by energy dispersive x-ray spectrosc opy. Crosslinking was confirmed through DNAMS dispersability testing in 0.05M phosphate buffered saline at a pH of 7.4. Dry and PBS swollen particle sizes were quant itatively characterized using an LS Coulter 13 320 particle size analyzer. The DNA-MS surface charge was meas ured by zeta potential an alysis in 0.01M PBS at a pH of 7.4. Normal human dermal BJ fibrobl ast cells were used in culture to evaluate the effect of DNA and crosslinked DNA-MS on cell growth. Concentrations of 100g and 500g for DNA, and 25g and 100g for crosslinked DNA-MS were tested. A colorimetric MTS assay was used to measure fibroblast growth at the 0, 24, and 72 hour time points. Aim 2: Optimization of DNA Nano-Mesosphere Synthesis The process parameters for DNA-MS synthe sis were optimized to produce controlled particle size distributions with mean dry diameters in the mesosphere size range of 1m to 10m and mean hydrated diameters of less than 25m. DNA-MS were prepared to where at least 60% of all particles fell within the mesosphere size range of 1m to 10m and < 5% of all particles were greater than 10m in size. Particles less th an 1m in diameter were also acceptable. DNAMS were sought to obtain hydrated particle diam eters of less than 25m with aqueous dispersion stabilities of over 24 hours. DNA-MS synthesis parameters were optimized by incorporating a filtra tion step at the end of the MS synthesis process with a filter pore size of 20m to reduce or eliminate the presence of aggregates from the yield. The DNA-MS synt hesis parameters were further optimized by analyzing the effects of mixer speed and cross link concentration on par ticle diameter, swelling, morphology, and size distribution. Mixer speed s of 950rpm, 1250rpm, and 1550rpm with ionic crosslink concentrations of 20%, 50%, and 120% molar equivalence (MEQ) were assessed for

PAGE 35

35 these studies. Synthesized DNAMS were characterized using the characterization techniques employed in Aim 1. Aim 3: In Vitro Evaluations of Mitoxantrone In Situ Loaded DNA Nano-Mesospheres Mitoxantrone in situ loaded DNA-MS (DNA-MXN-MS) were prepared using optimized synthesis conditions from Aim 2 to obtain particles with controlled size di stributions where at least 60% of all particles prepar ed were within the mesosphere size range of 1m to 10m and < 5% of all particles were greater than 10m in size. Particles less than 1m in diameter were also acceptable and hydrated particle diameters were to be less than 25m. In addition, MXN in situ loaded DNA-MS were sought to obtain loadings of 12% (w/w) MXN and release MXN for 24 hours or more in phosphate buffered saline unde r minimum sink conditions. DNA-MXN-MS at MXN concentrations of 1g/mL, 10g/mL, and 25g/ mL were also sought to elicit a cytotoxic response to that of free drug at the same MXN concentrations on murine Lewis lung carcinoma (mLLC) cells in culture. Gadolinium cation crosslinked DNA-MS were in situ loaded with mitoxantrone using 20%MEQ, 50%MEQ, and 120%MEQ crosslink concentrations. The particle diameter, size distribution, morphology, drug loading, and perc ent drug release of the DNA-MXN-MS were evaluated with respect to th e crosslink concentration. The particle diameter, size distribution, morphology, and pres ence of gadolinium trivalent cations were assessed using characterization techniques from Aim 2. Drug loading was determined by incubating the DNA-MXN-MS under st irred conditions in an enzymatic digestion buffer at 37C for 48 hours. DNA-MXN-MS were pr epared to obtain mitoxa ntrone loadings of 12% (w/w). The released dr ug concentrations were then analyzed using UV-visible spectroscopy against a MXN sta ndard curve. The percent dr ug release was measured under minimum sink conditions (i.e. lo w volume, 1.25mL) to simulate the tumor environment. Drug

PAGE 36

36 release data was obtained by incubating the DNA-MXN-MS in phosphate buffered saline under constant agitation at 37C for a mi nimum of fourteen days in triplic ate. At specific time points, aliquots were taken and measured using UV-visi ble spectroscopy against a MXN standard curve to determine drug concentration. The cytotoxi city of DNA-MXN-MS crosslinked with 120% MEQ gadolinium was evaluated using a murine Le wis lung carcinoma cell line at concentrations of 1g/mL, 10g/mL, and 25g/mL. These concentra tions were used to determine the effects of MXN dose on the tumor cell line. The cytotoxic ity of free drug and blank DNA-MS were also evaluated. An MTS assay was used to measure the cellular viability of the mLLC cells at days 0 through 4. Aim 4: In Vitro Evaluations of Methotrexate or 5-Fluorouracil In Situ Loaded DNA and BSA Nano-Mesospheres Deoxyribonucleic acid (DNA) a nd bovine serum albumin (BSA) nano-meso-microspheres (MS) were in situ loaded with methotrexate (MTX) or 5-fluorouracil (5-FU) using optimized DNA-MS synthesis conditions from Aim 2 to further analyze th e drug loading capabilities of DNA and BSA. MTX and 5-FU in situ loaded DNA-MS and BSA-MS were synthesized to produce particles with controlled size distributions where at leas t 60% of all particles prepared were within the mesosphere size range of 1m to 10m and < 5% of all particles were greater than 10m in size. Particles less than 1m in diameter were also acceptable and hydrated particle diameters were to be less than 25m. In addition, DNA-MS and BSA-MS were sought to obtain drug loadings of 5% (w/w) MTX or 5-FU and releas e drug for more than 24 hours in phosphate buffered saline under minimu m sink conditions. MTX or 5-FU in situ loaded DNAMS and BSA-MS were compared with respec t to particle diameter, size distribution, morphology, topography, drug loading, and percent drug release.

PAGE 37

37 DNA-MXN-MS were also prepared with a 120%MEQ gadolinium crosslink concentration and were in situ loaded with 10% (w/w), 15% (w/w), and 25% (w/w) MXN to determine the maximum drug loading ability of DNA. The pa rticle diameter, size distribution, morphology, topography, drug loading, and percent drug releas e of the DNA-MXN-MS were evaluated with respect to MXN concentration. In addition, DNA-MXN-MS were prepared with no gadolinium crosslinking to determine if MXN serves as a crosslinking agent for DNA-MS. DNA-MXN-MS were prepared to obtain in situ MXN loadings of 10% (w/w). The DNA-MXN-MS were characterized as mentioned above and comp ared with respect to crosslinking. The particle diameters and size distributi ons and morphologies a nd topographies were obtained using characterization techniques from Aim 2. Drug loading and release were determined by using the analysis techniques from Aim 3. MTX, 5-FU, and MXN standard curve were constructed to determine drug concentra tion and release from the drug loaded DNA-MS and BSA-MS.

PAGE 38

38 CHAPTER 2 BACKGROUND Introduction There are over 100 types of cancers known to man which may arise from a variety of organs within the body.25, 26 The American Cancer Society estim ates that in the USA, there will be 1,444,920 new cases of cancer and over 1 million cases of basal and squamous cell skin cancers diagnosed this year (2007), whic h will result in approximately 560,000 deaths.27 These estimates are consistent with data collected in 2005 and 2006 by the National Cancer Advisory Board.28 Cancer continues to be a national and worldwide problem and since 1999 has surpassed heart disease as the leading cause of death for people of age 85 and under.1 In addition, cancer is the second leading cause of death fo r children between ages 1 and 14.1 The global incidence of cancer is expected to increase from 10.3 million to 14.7 million by the year 2020.29 The increase in cancer incidence places a financial burden on nati onal and international economies. In 2006, the National Institutes of Hea lth estimated that the overall cancer medical, indirect morbidity, and indirect mo rtality costs were over $206 billion.27 The incurred cost of cancer treatments such as radiation and ch emotherapy was approximately $72.1 billion in 2004 and is expected to increase in the forthcoming years.30 Primary treatment modalities such as radiation and chemotherapy not only place a financia l burden on cancer patients, but also impact their quality of life due to the sy stemic toxicity and limited effec tiveness of the treatments. The increasing rates of new cancer cases, their attribut ed costs, and impact on human life generate a significant demand to develop inexpensiv e drugs and new treatment modalities. Conventional chemotherapy is delivered orally or intravenously which results in systemic toxicity. Traditional alkylating and antimeta bolite cancer drugs function by attacking rapidly dividing cells such as those seen in tumor cells however, many normal tissue cells also divide

PAGE 39

39 rapidly and thus normal tissue toxicity occurs.5 Normal tissue toxicity results from poor tumor cell specificity which may be remedied through localized delivery techniques such as those obtained with intratumoral delivery.4, 5 The potential benefits of in tratumoral therapy (IT) are a localized super dose delivery to the tumor, greatly reduced system ic toxicity, and the opportunity to prolong local drug effectiveness. Recent research has shown that the intratumoral delivery of biodegradable carriers loaded with mitoxantrone greatly im proved survival and decreased toxicity in a murine breast cancer model.12 These findings suggest that IT may be a useful alternative to conventional chemotherapy. Previous studies concerning the intrav enous delivery of DNA complexed with anthracycline chemotherapy drugs have show n that DNA reduces the cardiotoxicity and increases the half-life of the chemotherapy drugs in vivo .22, 23, 31-33 The therapeutic value of the DNA-drug complexes was also found to be comparable to that of free drug.22, 23, 31-33 These studies suggested that cross linked but biodegradable DNA me sosphere compositions, which have not been previously reported, might have interesting and useful IT drug delivery properties. Cancer Types of Cancer The most common cancers found in humans arise from mutations in epithelial cells leading to the formation of solid tumor carcinomas.34 Solid tumor carcinomas account for 80% of all cancer-related deaths and incl ude cancers of the mouth, st omach, esophagus, skin, mammary glands, pancreas, lung, liver, ovary, gallbladder, urinary bladder, and the small and large intestines.34 There are two major classifications of carcinomas which include squamous cell carcinomas and adenocarcinomas.34 Squamous cell carcinomas arise from mutations in epithelial cells that serve as a protective lining for underlying cells such as the skin and esophagaus.34 Adenocarinomas arise from secretory cel ls that are contained within epithelial

PAGE 40

40 cells that protect the cavities or channels that they line such as mammary glands, lung, uterus, and cervix.34 Nonepithelial tumors account for the remaini ng 20% of all malignant cancers and include sarcomas, lymphomas, leukemias, neuroectoder mal tumors (i.e. gliomas, neuroblastomas, schwannomas, etc.), melanomas, and small-cell lung carcinomas.34 Sarcomas arise due mutations in connective tissue cells such as fibr oblasts (secrete collag en), osteoblasts (form bone), and myocytes (form muscle).34 Furthermore, sarcomas may al so arise from alterations in adipocyte cells whose prim ary functions are to st ore fat from cytoplasm.34 Lymphomas and leukemias develop from blood-forming cells whic h include the red and white blood cells and are usually classified as systemic cancers.34 Neuroectodermal tumors account for 2.5% of all cancer deaths (~14,000 deaths/year) and arise from cells derived from the central and peripheral nervous systems.28, 34, 35 Melanomas derive from pigmented cells that line the skin and retina whereas small-cell lung carcinomas or iginate from endodermal ce lls that line the lung. Tumor Development Cancer is a progressive dis ease displaying benign, maligna nt, and metastatic stages.25, 34-37 Initially, cancer was thought to be a disease de fined by uncontrolled cell proliferation, however, the identification of many malignant tumors which are slow growing in humans and diseases that demonstrate very rapid cell proliferation with no evidence of neoplastic transformation, have proven otherwise.36 Current models have shown that cance r develops due to genetic alterations that result in a cell’s inability to respond to intraand ex tracellular signals that control proliferation, differentiation, and death. 35-38 Genetic alterations may ar ise due to inherited gene mutations, ionizing or UV radiati on, exposure to physical or chem ical carcinogens, free radical attacks, or errors during DNA synthesis.35, 37

PAGE 41

41 Two to six types of genetic alte rations arise with respect to each individual type of cancer with each alteration having a specific capability.25, 35 Normal cell proliferation is attained through the successful progression through the cell life cycle.38-42 The normal life cycle of a cell involves four phases including the Gap1 or “G1” phase, the DNA Synthesi s or “S” phase, the Gap2 or “G2” phase, and the Mitosis or “M” phase, Figure 2-1.41-43 Normal cells will grow (G1), replicate and synthesize their DNA (S), continue to grow with two sets of DNA with matching chromosomes (G2), and finally divide to produ ce two cells with matching DNA (M).38-43 S G 1 S G 1 S G 2 S G 2 M G 2 M G 2 M G 1 M G 1 G1 S G2 M S G 1 S G 1 S G 2 S G 2 M G 2 M G 2 M G 1 M G 1 G1 S G2 M Figure 2-1. The 4 phases of the life cycle of a cell.41-43 Genetically altered cells also progress through the cel l life cycles; however they attain new capabilities with each alteration that lead to th e development of malignant growth, Figure 2-2.25, 44 These six new capabilities include self-suffici ency in growth signals (SS), insensitivity to anti-growth signals (IAS), tissue invasion and meta stasis (TI&M), limitless replication potential (LRP), sustained angiogenesis (S A), and apoptosis evasion (AE).25, 45 These six new capabilities allow cancerous cells to self-stimulate into an active proliferative state (SS) a nd desensitize themselves to anti-growth signals (IAS).25 Genetically altered cancer cells thus divide a nd grow quicker than normal cells.25, 44 The formation of large

PAGE 42

42 cancerous cell colonies allow them to invade su rrounding tissues and detach to become systemic (TI&M).25, 37 Their acquisition of limitless replica tion potential and their ability to evade apoptosis represent the truly deva stating effects of this disease and introduces the urgent demand for new chemotherapy drug and delivery methods.25, 37 SS IAS TI&M LRP SA AE CANCER SS IAS TI&M LRP SA AE CANCER Figure 2-2. Physiological capabilities of genetically altere d cancerous cells. Conventional Treatments Conventional cancer treatments include surgery, radiation, and chemotherapy and improvements in cancer survival often resu lt in using a combination of the three.3, 26, 45, 46 The choice of treatment regimen varies with each patient and is dependent on th e type and location of the cancer and the physical condition of the patient.3 Radiation therapy and surgery are localized forms of treatment whereas chemotherapy is a systemic treatment.3, 46 Surgery and radiation Surgery is the oldest form of cancer treatment known dating back to Egyptian times when benign lipomas and polyps were surgically removed with a knife or red-hot iron.47 Instances of excisions and cauterizations of tu mors and ulcers have also been documented dating back to the Medieval time period between 500 and 1500 A.D..47 This form of localized treatment continued

PAGE 43

43 on through the Renaissance and Reformation periods to current day.47 Surgery is a very common form of cancer treatment and is often used to confirm the diagnosis of cancer, determine the cell type, and to reduce the size of the tumor through excision.3, 26 Advances in surgical technologies have lead to less invasive form s of surgery which spare as much normal tissue as possible. In some cases, radiation or chemotherapy will be used before surgery to decrease the size of the tumor and after surgery to eliminat e any further traces of the cancer.3 Radiation therapy is the sec ond oldest of the three conve ntional cancer treatments and involves the administration of x-rays or gamma rays directly to the tumor.3 The localization of radiation therapy destroys cancer cells and mi nimizes toxic effects on healthy adjacent cells.3 Radiation therapy functions by damaging the DNA of the cancer cell t hus hindering it from functioning normally.3 The first treatment of cancer usi ng radiation was in 1896 following the discovery of the radioactiv ity of uranium by Becquerel.47 The treatment of uterine carcinoma with radium soon followed in 1905 after radium’s initial discovery by Marie and Pierre Curie in 1898.47 Chemotherapy The first use of “anti-cancer” drugs was f ound during the Medieval time period when arsenic pastes were used to locally treat tumors.47 Arsenic pastes continued to be used until 1865 when a potassium arsenite solution was give n to a patient to treat chronic leukemia.45, 47 At the turn of the twentieth century, Paul Ehrlich, “t he father of chemotherapy”, made tremendous advances in the field of chem otherapy and tested many new systemic chemical agents based on arsenic compounds on rodent models to pr edict their effectiv eness in humans.47, 48 Later during the early 1940s, the first succe ssful clinical studies using systemic chemotherapy were reported.45, 47 Patients at Yale University with Hodgki n’s disease were treat ed with nitrogen mustard, a derivative of mustard gas, and were found to respond positively to the systemic

PAGE 44

44 treatments (i.e. tumor size decreased).47 However, patients relapsed after a short period of time and no true successes were found with syst emic chemotherapy agents until 1956 when methotrexate cured over 50% of patients with choriocarcinoma.47 Chemotherapy is most commonly administered intravenously, however it may also be delivered orally in pill form or through other me thods such as muscular, subcutaneous, or intraarterial injections.3, 46 The route of chemotherapy admini stration depends strongly on the drug’s in vivo pharmacological properties such as half-life and metabolim.46 The ultimate clinical objective of systemic chemotherapy is to kill cancer cells in the body wh ile leaving all other healthy and normal functioning cells intact.46 This is difficult to accomplish since chemotherapy drugs function by attacking rapidl y dividing cells while circulati ng throughout the bloodstream. Since many normal functioning cells also divide rapidly such as hair cells, bone marrow cells, and cells that line the mouth and gastrointestinal tr act, toxic side effects re sult such as hair loss, nausea, and even death.3, 45, 46, 49 Systemic chemotherapy is mo st effective against systemic cancers such as leukemia and lymphomas and is less effective against solid tumor cancers such as the most prominent breast, lung, colon, and rectum cancers because these cancers have a lower number of rapidly dividing cells.45 It is therefore important to find new delivery methods that localize the cancer drugs in order to increase their effectiveness and decrease systemic side effects. Intratumoral Chemotherapy Rationale for Intratumoral Chemotherapy It has long been noted the th erapeutic effectiveness of ma ny chemotherapy drugs is often limited by their inability to reach the location of solid tumor cells and permeate the tumor cells’ membranes.6, 50 Intratumoral chemotherapy, through dr ug loaded microsphere delivery, offers one possible solution to this reoccurring problem. The main goal of intratumoral chemotherapy

PAGE 45

45 is to increase the exposure of the chemotherapy drugs to the cancer cells by means safer than those achieved through systemic intravenous or oral delivery.12, 51 Conventional systemic chemotherapy is constantly balancing the admini stration of a toxic dose with an effective therapeutic dose, generally desc ribed as the therapeutic ratio which for most chemotherapy drugs has a value of one.4, 46, 52, 53 A therapeutic ratio value of one implicates that damage to normal tissues is dose dependent which in turn limits the frequency of chemotherapy regimens.26, 46 Thus, most chemotherapy treatments are given at concentrations below the therapeutic ratio in order to not comprise healthy ra pidly dividing cells and are given in cycles long enough to allow cells in the bone marrow to recover.46 Advances in cancer care have le d to longer patient survivals, however, it has also exposed long-term chemotherapy side effects such as in fertility and carcinogenicity which have been predominant in young patients who have been cu red of leukemia, testicular cancer, and Hodgkin’s disease.46 Therefore, great attent ion has been placed on the development of localized drug delivery methods, such as intratumoral chemotherapy, that inco rporate zero-ordered controlled release devices, such as microspheres, that can maintain a desired concentration of drug at the site of the tumor without reaching toxic or minimum effective levels, Figure 2-3.52, 54 Time Drug ConcentrationToxic Level Minimum Effective Level Time Drug ConcentrationToxic Level Minimum Effective Level Figure 2-3. Therapeutic dose chart illustrating a toxic dose (red line), moderate dose (blue line), and a controlled releas e curve (dashed line).52, 54

PAGE 46

46 Intratumoral chemotherapy using controlled release devices increase cancer patient quality of life by enhancing the safety and efficacy of chemotherapy drugs and decreasing chemotherapy frequency.4, 55 Controlled Release Microspheres Controlled release devices in the form of microspheres have been prepared using biodegradable materials such as albumin, gelatin, casei n, and deoxyribonucleic acid.11, 24, 56-59 Biodegradable materials are capable of being loaded with cytotoxic drugs and can be prepared in microsphere form. The spherical morphology of th ese devices make them ideal for intratumoral injections since locally delivered drug loaded microspheres are trapped in the micro-circulation of tumors and release cytotoxic agents slowly with minimal systemic exposure.60 Decrease in systemic toxicities have been proven through pha rmacokinetic studies which have shown that cytotoxic agent serum levels after microsphere injections are much lower than those obtained when free drug is administered intravenously and also locally.5, 7, 60-65 Locally delivered drug loaded microspheres also increase drug efficacy not only by increasing drug half-lives, but by also prolonging drug exposure to newly replicating cancer cells.61, 66 The drive to use microspheres as drug deliver y vehicles may be explained by the enhanced permeability and retention (EPR) effect that was introduced in 1986.67, 68 The EPR effect states that drug loaded microspheres improve the half-life of chemotherapy agents because microspheres do not exhibit the same lymphatic cl earance rates as free drug and are capable of being trapped within the tumor vasculature.67, 68 The question that then arises is: "What is the optimal particle size for microspheres used in intratumoral chemotherapy?". The general consensus appears to be that microspheres sma ller than 25m in diamet er are preferred since they have a high surface area-tovolume ratio, are able to be en trapped in tumor tissue, and are also small enough to pass through dilated tumor vessels.69 Microspheres in the 20m to 40m

PAGE 47

47 size range are also preferable since they are also able to be entrapped in tumor tissue and can be administered intratumorally without complications.9 For studies presented in this dissertation, particles were prepared in the injectable mesosphere size range (i.e. 1m < d < 10m). Drug Release Kinetic Models Release profiles Drug elution from drug delivery devices ma y occur either through drug diffusion or through the erosion of th e drug carrier matrix.67, 70 The drug release profiles for each system will be different and it is possible for a drug carrier to release drug through a combination of diffusion and erosion.67 The profile for diffusion based release will display an initial burst followed by a continual slow release over time as shown in Figure 2-4. The profile for erosion based release will initiall y release slowly and then increase exponentially over time, Figure 2-4. TimeDrug Release TimeDrug Release Figure 2-4. Diffusion (blue line) and eros ion (orange line) drug release profiles.67, 70 The initial burst release illustra ted in the diffusion release prof ile is attributed to the quick diffusion of the drug molecules that are entr apped on the carrier surface and hydrophilic outer layers, Figure 2-5.67, 70 The quick release exhibited towards the end of the erosion release profile

PAGE 48

48 is attributed to the release of the drug molecule s that are entrapped within the carrier matrix due to the dissolution or erosion of the matrix, Figure 2-5. Drug Carrier Diffusion Erosion Drug Carrier Diffusion Erosion Figure 2-5. Drug release due to diffusion or erosion.67, 70 Release kinetics Drug release from biodegradable controlled release matrix devices (i.e. microspheres) generally follow first order release meaning th at the drug release rate is proportional to the remaining drug concentration.71 First order release kinetics ar e typically observed with diffusion based release where the drug release ra te decreases over time, Figure 2-6 (A).71 The first order release kinetics model is descri bed by plotting the log of the amount of drug remaining in the carrier versus the amount of time released and is distinguished as a li near curve, Figure 2-6 (B).72-74 TimePercent Drug ReleasedA TimeLog (Percent Drug Still in Carrier) Br2> 0.9 TimePercent Drug ReleasedA TimePercent Drug Released TimePercent Drug ReleasedA TimeLog (Percent Drug Still in Carrier) B TimeLog (Percent Drug Still in Carrier) TimeLog (Percent Drug Still in Carrier) Br2> 0.9 Figure 2-6. First order profiles for (A) drug release and (B) release kinetics.

PAGE 49

49 Ideal drug delivery release is obtained when the drug release curve is linear and is generally described using zero order and Higuchi drug release kinetics models.67 Zero order release is obtained when the active drugs are re leased over time at a constant rate and is distinguished as a linear curve on a percent drug released versus tim e plot, Figure 2-7 (A).67, 71 TimePercent Drug Released r2> 0.9A Time1/2Percent Drug Released r2> 0.9B TimePercent Drug Released r2> 0.9A TimePercent Drug Released r2> 0.9 TimePercent Drug Released TimePercent Drug Released TimePercent Drug Released r2> 0.9A Time1/2Percent Drug Released r2> 0.9B Time1/2Percent Drug Released r2> 0.9 Time1/2Percent Drug Released r2> 0.9B Figure 2-7. Drug release profile s for A) zero order release ki netics and B) Higuchi release kinetics. The Higuchi release model was created in th e early 1960s by Takeru Higuchi and was one of the first models ever used to describe the release of therapeutic drugs from a matrix carrier based on Fickian diffusion principles.75-77 The Higuchi release model is one of the models most commonly used to describe drug release for c ontrolled drug delivery devices and is obtained when the active drugs are released at a constant rate over the square root of time, Figure 2-7 (B).75-77 Chemotherapy Drugs Chemotherapeutic drugs functi on by interfering with one or more processes involved with cell replication.46, 49 This interference causes either cance r cell apoptosis (cytotoxic drugs) or cessation of cancer cell gr owth (cytostatic drugs).46, 49 There are five major standard classes of anticancer agents which include the antimeta bolites, covalent DNA bi nding drugs, noncovalent

PAGE 50

50 DNA binding drugs, chromatin function inhibitors and drugs affecting endocrine function.78 The standard classes of anticancer agents with their respective subc lasses are given in Table 2-1. The majority of chemotherapy drugs fall under the antimetabolite and noncovalent DNA binding drug classes and are known for their effectiveness against systemic and solid tumor cells with a high ratio of dividing cells.26 Mitoxantrone, methotrexate, an d 5-fluorouracil which are drugs in the antimetabolite and noncovalent DNA binding classe s, were chosen for the research presented in this dissertation due to th eir broad treatment spectrum and their ease of availability. Table 2-1. Classes and subclasses of anticancer agents.78 Antimetabolites Covalent DNA binding drugs Noncovalent DNA binding drugs Chromatin function inhibitors Endocrine function drugs Folate antagonists Nitrogen mustards Anthracyclines Topoismerase inhibitors Glucocorticoids Pyrimidine antagonists Aziridines Mitoxantrone Microtubule inhibitors Estrogens Purine antagonists Alkane sulfonates Dactinomycin Antiestrogens Sugar-modified analogs Nitrosoureas Bleomycin Progesterins Ribonucleotide reductase inhibitors Platinum compounds Plicamycin Androgens Monoalkylating agents Antiandrogens Aromatase inhibitors Mitoxantrone Principle attention throughout th e research presented in this dissertation was given to the use of mitoxantrone due to its broad treatment spectrum, blue chromaphore for analysis, and excellent in vivo murine adenocarcinoma results th at were obtained in this lab.11, 12, 56 Mitoxantrone is a synthetic anthracenedione wh ich is a member of the anthracycline class of antitumor antibiotics.78-81 It has a planar tetracyclic arom atic ring structure that can easily intercalate into DNA.78, 79 Mitoxantrone has two symmetrical aminoalkyl side arms with no

PAGE 51

51 sugar moiety which is common in mo st anthracycline agents, Figure 2-8.78-81 Its trade name is NovantroneTM and is also known as dihydroxya nthracenedione dihydrochloride.81 Mitoxantrone has a molecular weight of 517.4 g/mol and is ava ilable as a blue powder that is hygroscopic and soluble in water and ethanol.81 In a clinical setting, mitoxantr one has been shown to produce less cardiac toxicity and a diminished potential to att ack healthy rapidly dividing cells as compared to other anthracyclines.79 O O OH OH NH NH N OH HCl HCl H N OH H Figure 2-8. Chemical structure for mitoxantrone hydrochloride. Mitoxantrone produces antitumor activity th rough two main interactions with DNA.79, 81 The first interaction is considered a combinati on of high-affinity inter calation of the planar mitoxantrone structure into opposing DNA strands and lower-affinity electrostatic binding.78, 79, 81 Intercalation occurs most preferentially at the G-C ba se pairs, which disrupts DNA synthesis.78, 79, 81 This intercalation is thought to be further stabil ized through electrostatic interactions between the positively charged nitr ogens on the two alkyl side chains and the negatively charged ribose phosphates on the DNA.79, 81 The second interaction involves impairment of cell division through the inhibitio n of topoisomerase II activity which is an enzyme most prominently found in the G2 phase of the cell cycle.78, 79, 81

PAGE 52

52 Mitoxantrone is administered by intravenous injection and is primarily used to treat leukemias, lymphomas, and advanc ed or recurrent breast cancer.78, 81 The therapeutic dosage for mitoxantrone ranges from 10 to 14 mg/m2 daily for 1 to 3 days.80 Mitoxantrone is known to cause dose-limiting toxic side effects even thoug h it has lower cardiac toxicity as compared to other commonly used anthracycline drugs.79, 81 The most significant side effects are nausea, vomiting, and stomatitis, however, myelosuppression, alopecia, and phlebitis may also occur.79-81 Methotrexate Methotrexate was discovered in the late 1940s after leukemic child ren went into cancer remission after being treated w ith folic acid antagnonists.78, 80, 82 Methotrexate is categorized as a folate antagonist in the antimetabolite cancer dr ug class. It is a weak acid with a chemical structure that consists of 95% N-[4-[[2,4-diamino-6-pteridi nyl)methyl]methylamino]benzoyl]-Lglutamic acid, Figure 2-9.81 Methotrexate has a molecular wei ght of 452.5 g/mol and is available as sodium methotrexate which is readily soluble in water.81 N N N N N N N CH3 N O OH HO O O H H H H H H Figure 2-9. Chemical stru cture of methotrexate. Methotrexate produces antitumor activity by bi nding to dihydrofolate reductase (DHFR) which is a key enzyme in DNA synthesis.78, 80-82 When methotrexate is bound to DHFR, DHFR

PAGE 53

53 activity is inhibited and dihydrof olate is not reduced to tetr ahdyrofolic acid which ceases de novo purine synthesis.78, 80, 81 For this reason, methotrexate is cons idered an S phase cell cycle specific drug.80 Methotrexate is most frequently administer ed intravenously; however, it may also be administered through intratumoral or intramuscular injections.80 The dosage range for each administration method is given in Table 2-2. Methotrexate is a fi rst-line drug for the treatments of breast cancer, osteogenic sarcoma, gestationa l choriocarcinoma, and leukemia and under high doses, can cross the blood brain barr ier to treat can cers of the brain.80, 82 Methotrexate has been clinically shown to be as effective as and less toxic than its folic acid antagonist counterpart aminopterin; however, it still causes toxic side effects.78 The two major sites of methotrexate toxicity are located in the bone marrow a nd in the endothelium of the oropharynx and gastrointestinal tract.78 This leads to a va riety of toxic side effects which include myelosuppression, mucositis, renal tubular obstruction, hepatoxicity, pneumonitis, hypersensitivity, and neurotoxicity.82 Table 2-2. Methotrexate dosage ranges for various administration routes. Administration method Dosage range (mg/m2) Intravenous – low dose 10 to 50 Intravenous – medium dose 100 to 500 Intravenous – high dose > 500 Intratumoral 10 to 15 Intramuscular 25 5-Fluorouracil 5-fluorouracil is a fluorinated pyrimidine uracil within with a fluorine atom substituted for a hydrogen atom at the carbon number 5 posi tion on the pyrimidine ring, Figure 2-10.81, 83 5fluorouracil falls within the antimetabolite cance r drug class and has a molecular weight of 5fluorouracil is 130.08 g/mol.80, 81, 83 Its full chemical name is 5-fluoro-2,4(1H,3H)-

PAGE 54

54 pyrimidinedione and its regi stered name is Adrucil.80, 81 5-Fluorouracil is se nsitive to light and easily precipitates out of solution at low temperatures or if let standing at room temperature for prolonged periods of time.80, 81 It is partially soluble in wa ter or ethanol, Table 2-3, and is commercially available as an alkalin e buffered solution with a pH of 9.81 N N O F O H H Figure 2-10. Chemical st ructure of 5-fluorouracil.81, 83 Table 2-3. 5-fluorouracil so lubility in various media.81 Media Solubility (mg/mL) Water 12.2 Ethanol 5.5 Chloroform 0.1 5-Fluorouracil produces antitumor activity by be having as a “false” pyrimidine ultimately inhibiting the formation of thymidine synthe tase, an enzyme necessary for DNA synthesis.78, 80, 81 The antitumor activity of 5-fluorouracil is governed by three main mechanisms which include 1) the inhibition of thymidine synthetase by the generation of fluorodeoxyuridine monophosphate (FdUMP), 2) the incorporation of fluorouridine triphosphate (FUTP) into RNA, and 3) the incorporation of FUTP into DNA.81, 83 During its primary mechanism of action, FdUMP binds tightly to thymidine synthetase and pr events the formation of thymidylate.83 Thymidylate is one of the four essential deoxyr ibonucleotides required for DNA s ynthesis and without it causes toxicity in actively dividing cell.81, 83 Therefore, 5-fluorouracil is considered as an S phase cell cycle specific drug.78, 81, 83

PAGE 55

55 5-fluorouracil is primarily administered in travenously; however it may also be administered by hepatic or by head and neck infusions.78, 80, 81 The dosage ranges varies for each different administration method and are given in Table 2-4.80, 81 Table 2-4. 5-fluorouracil dosage ranges for various administration routes.80, 81 Administration method Schedule Dosage range Intravenous Once a week 12 to 15 mg/kg Intravenous Every day for 5 days every 4 weeks 12 mg/kg Intravenous Every week for 5 weeks 500 mg/m2 Hepatic infusion Eight hours for 5 to 21 consecutive days 22 mg/kg in 100mL D5W Head and neck infusion Continuous for 4 to 5 days 1,000 mg/m2 5-fluorouracil was discovered in 1958 and since then has been used to treat a variety of solid tumor cancers including commonly occurri ng cancers such as breast, head and neck, colorectal, gastric, and pancreatic cancers.78, 81 5-fluorouracil is also used to treat renal cell carcinoma, squamous cell carcinoma of the esophagus prostate cancer, bladder cancer, and basal cell carcinoma; however, it is not recommended for invasive skin cancers.78, 81 The most doselimiting toxic side effects of 5-fluorouracil ar e stomatitis, nausea, vomiting, diarrhea, and anorexia since 5-fluorouracil fre quently attacks rapidly dividing cells within the gastrointestinal tract.80, 81, 83 In addition, 5-fluorouracil causes such toxic side effects as myelosuppression, dermatologic toxicity, ocular t oxicity, neurotoxicity, bone marro w suppression, cardiac toxicity, and biliary sclerosis.78, 81, 83 Deoxyribonucleic Acid Many know DNA for its role as a carrier of ge netic information; however, very few have given thought to the possible bi omaterial functions of this su pramolecular biopolymer. For the past 30 years scientists have focused their effo rts on the delivery of viral DNA, otherwise known as plasmid DNA, for its function as a gene expressor.15, 16 Plasmid DNA, unlike chromosomal or genomic DNA, are small circular DNA molecules th at are regarded as genetic mobile elements

PAGE 56

56 because they have the ability to independently replicate themselves and have their genetic information transferred between cells.84 The delivery of plasmid DNA has been sought after for such applications as biosensing for the detection of chemical c ontaminants, gene expression for its use as a novel therapeutic for cancer and ge netic disorders, and ti ssue engineering as bone, skin, and nerve cell regenerative stimuli.15,14,85 With all of this attention focused on plasmid DNA, scientists have overlooked the attractive prope rties that chromosomal DNA has to offer. It has only been within the past decade that scie ntists have shown intere st in the biomaterial functions of chromosomal DNA. Chromosomal DNA is a macromolecular polynuc leotide with a rigid right-handed double helical molecular conformation, Figure 2-11.86 The molecular structure of chromosomal DNA was discovered in 1953 by James Watson and Fran cis Crick and consists of two phosphodiester deoxyribose backbone chains that connect toge ther through adenine-t hymine and cytosineguanine base pairs, Figure 2-12.87 This naturally occurring bi opolymer has a unique base pair configuration for each individual living creation, from humans to plants, bacteria, etc. In addition, the molecular weights of DNA range from the hundreds of th ousands to easily into the hundreds of millions which is also depe ndent on the original source of the DNA.86 Figure 2-11. A drawing of DNA.

PAGE 57

57 N N N N N O H O H H H CH2 H P OH O O N O O N O H O H H H CH2 H P O N N O N O H O H H H H2C H P O N N N O N N O H H H H CH2 H O P OH O O Adenine Thymine Guanine CytosineH H H HO HO H H H H H HO O O HO Figure 2-12. Cartoon of the molecular structure of a DNA subunit (Note: Bonds are not drawn to scale). In the late 1970s and early 1980s, scientists began to explore the use of chromosomal DNA for biomedical purposes.22, 23, 32, 33, 88 During this time, chromosomal DNA was used as a conjugated carrier for chemotherapy drugs such as doxorubicin and was found that DNA helped to reduce the cardiotoxicity of such drugs when injected intravenously.22, 23, 32, 33, 88 Into the late 1990s, scientists further explor ed the biomaterial properties of DNA and used DNA-lipid films as drug delivery vehicles for ethidium bromide.19, 20 DNA-alginate films were also shown to have been cast onto Millipore filt ers and remove contaminants such as benzopyrene and acridine orange from aqueous solutions.17 DNA films have also been coated onto nonwoven cellulose fabrics and loaded with silver ions to gi ve the fabric surface an tibacterial properties.20 In

PAGE 58

58 addition, in 2003 DNA films were shown to increase the blood compatibility of hemodialysis membrane polysulfone fiber surfaces.18 DNA films increased the hydr ophilicity and decreased the protein absorption of the polysulfone fibers.18 In 2003, DNA was used to produce regenerative tissue scaffolds and in 2006, DNA film s were cast onto titanium medical devices.21, 89 The combination of the biomaterial properties of DNA presented in this section suggest that DNA may be a desirable explorat ory material for mesosphere synthesis for intratumoral chemotherapy applications. DNA Mesosphere Synthesis Steric Stabilization The DNA microspheres (i.e. d > 10m), me sospheres (i.e. 1m > d > 10m), and nanospheres (i.e. d < 1m) (DNA-MS) presented in th is dissertation were pr epared using a steric stabilization process developed in this la b for the synthesis of albumin microspheres.11, 56-59 This process creates a sterically stab le colloidal microemulsion thro ugh the complete solvation of a stabilizing agent (i.e. cellulos e acetate butyrate in 1,2-dichloroet hane) at the surface of aqueous DNA droplets.90 In brief, the DNA-MS synthesis process invol ves adding a small volume of an aqueous DNA solution to a large volume of an organic po lymer solution and mixing these two solutions at high speed (i.e. speeds greater than or equal to 900rpm) for 20 minutes in order to create a DNA microemulsion. The DNA microemulsion is th en crosslinked using covalent or ionic crosslinking agents and the mixer speed is redu ced to 600rpm. Crosslinking reactions continue for another hour and 40 minutes at which time acetone is added to the system to compensate for any evaporation during synthesis and to initia te the DNA microsphere wa shing process, Figure 2-13. The synthesis process continues for an additional hour and then the DNA-MS are washed in acetone four times, collected through centrifugation, a nd allowed to air dry over night at room

PAGE 59

59 temperature. All DNA-MS synthesis processes are carried out at room temperature and are given in more detail in each subse quent chapter in this dissertation. DNA MicroDispersion Low Speed, 1hr 40min Synthesis Completion in Acetone Low Speed, 1 hour DNA & Surfactant High Speed, 20min Covalent or Ionic Xlinked DNA MS DNA MicroDispersion DNA MicroDispersion Low Speed, 1hr 40min Low Speed, 1hr 40min Synthesis Completion in Acetone Synthesis Completion in Acetone Low Speed, 1 hour Low Speed, 1 hour DNA & Surfactant DNA & Surfactant High Speed, 20min High Speed, 20min Covalent or Ionic Xlinked DNA MS Covalent or Ionic Xlinked DNA MS Figure 2-13. A brief schematic of the DNA-MS synthesis process. DNA Crosslinking Ionic crosslinking It has long been shown that multivalent metal cat ions have the ability to crosslink plasmid or chromosomal DNA via the phosphate groups found along the DNA backbone or via the electron-donor groups found within the base pairs.91, 92 Recent studies cited in the literature have shown that divalent cations, such as Co2+ and Zn2+, and multivalent cations, such as spermidine (tertiary valence) and spermine (quaternary valence) destabilize the double helix of DNA by disrupting the base pairing of DN A due to metal ion crosslinks.93 Therefore, the concept to use trivalent cations, such as those obtained in chromium (III) potassium sulfate, gadolinium (III) chloride, and iron (III) nitrate, as crosslinking agents fo r DNA-MS synthesis was developed in this lab and set forth in this dissertati on. The mechanisms describing possible DNA-MS crosslinking with each of the af orementioned trivalent cations is given in detail in Chapter 3. Covalent crosslinking Glutaraldehyde has been the primary crosslink ag ent used in this lab for the preparation of albumin, gelatin, poly(glutamic acid) and blended microspheres since the early 1980s.11, 13, 56-59,

PAGE 60

60 94 Glutaraldehyde is a dialdehyde Figure 2-14, that reacts easily with primary amine sites commonly found in the biomaterials menti oned above through a Schiff base reaction.11, 13 Therefore, glutaraldehyde was chosen as an e xperimental covalent cr osslinking agent for DNAMS synthesis due to the amino groups found on th e adenine, guanine, and cytosine base groups within DNA, Figure 2-15. Possible crosslink mechanisms between glutaraldehyde and DNA are given in detail in Chapter 3. O O Figure 2-14. Chemical stru cture of glutaraldehyde. N N N N N Adenine N N N N O N Guanine N N N O Cytosine N N O O Thymine H H H H H H H H H H H H Figure 2-15. Chemical structures of the four DNA base groups.

PAGE 61

61 More recently, the use of genipin as a natura lly occurring biomaterial crosslinking agent has been receiving much attenti on due to its significantly lower in vitro cytotoxicity as compared to glutaraldehyde.57, 95 Genipin is an iridoid glucosid e extract from the fruits of the Gardenia jasminoides Ellis, Figure 2-16.95, 96 It also has the ab ility to react with primary amines and was thus chosen as an experimental covalent cr osslinking agent for DNA-MS synthesis. Possible crosslink mechanisms between genipin and DNA are given in detail in Chapter 3. O H3CO O OH OH Figure 2-16. Chemical structure of genipin. Research Goals The increase in cancer inci dence rates and financial burdens on worldwide economies places an urgent need on new cancer diagnostics drugs, and treatment modalities. The limited effectiveness of chemotherapy agents on solid tumor cancers suggest that localized treatment through intratumoral chemotherapy may provide increased drug therapeutic efficacy and may improve the quality of life for cancer patients. The possible biomaterial applications of DNA are limitless and its demonstrated ability to load drugs such as ethidium bromide and doxorubicin make it a very desirable material for mesos pheres synthesis for intratumoral chemotherapy applications. Therefore, the obj ective of the research presented in this dissertation was the synthesis and characterization of drug loaded DNA nano-meso-microspheres for intratumoral cancer therapy. Studies were ai med at producing DNA-MS with > 60% of all particle diameters

PAGE 62

62 prepared falling in the mesosphe re size range (i.e. 1m > d > 10m) < 5% of all particles greater than 10m in diameter, swollen diamet ers < 25m, and dispersion times of 24 hours for injectability and stability purpos es. Studies were also aimed at preparing DNA-MS capable of loading 12% (w/w) MXN, loading 5% (w/w) MTX or 5-FU, releasing drug for 24 hours, and producing cytotoxic effects on murine Lewis l ung carcinoma cells at concentrations as low as 1g/mL MXN.

PAGE 63

63 CHAPTER 3 DNA NANO-MESO-MICROSPHERE SYNTHESIS Introduction This chapter is devoted to the studies conduc ted on the synthesis a nd characterization of nano-meso-microspheres (MS) prepared from deoxyribonucleic acid (DNA) and crosslinked with covalent and ionic agents. To date, there is little in the li terature regardi ng the use of DNA as a biopolymer for chem otherapy applications. DNA-MS were prepared using a st eric stabilization pr ocess that was deve loped in this lab for the synthesis of albumin (BSA) and gelatin (GEL) MS.11, 57 This steric stabilization process was adapted to prepare DNA-MS that were cros slinked while in suspension covalently with either genipin or glutaraldehyde or ionically with either chromium (III) potassium sulfate, gadolinium (III) chloride, or iron (III) nitrate. The steric st abilization process involved a 20 minute microemulsion process followed by a one hour and 40 minute crosslink process based on previous results obtained with glutaraldehyde crosslinked BSA-MS and genipin crosslinked GEL-MS synthesis.11, 57 DNA-MS were prepared by incr easing the concentration of the stabilizing agent from 3% (w/v) to 5% (w/v) in order to obtain better steric repulsion between the higher molecular weight DNA-MS. DNA-MS were crosslinked via the aqueous phase unlike BSA-MS and GEL-MS which are crosslinked with glutaraldehyde or genipin via the continuous organic phase.11, 57 Crosslinking through the continuous or ganic phase was attempted with the DNA-MS, however, crosslinking via the aqueous phase yielded particles that were less aggregated and agglomerated and more spherical in morphology. The objectives for this chapter were to synthe size DNA-MS to a target dry mean diameter range of 50nm to 20m, where at least 60% of all particles prepared were within the mesosphere size range of 1m to 10m and < 5% of all particle s were greater than 10m in size. Particles

PAGE 64

64 less than 1m in diameter were also acceptable. H ydrated particle diameters were to be less than 25m. DNA-MS were sought to obtain aqueous dispersion stability of over 24 hours and elicit minimal toxic effect on fibroblast cells in culture. The morphology and topography of the DNA-MS were examined by optical microscopy and scanning electron microscopy. The presen ce of trivalent cations in the DNA-MS was assessed by energy dispersive x-ray spectrosc opy. Crosslinking was confirmed through DNAMS stability testing in 0.05M phosphate buffere d saline (PBS) at a pH of 7.4. Dry and PBS swollen particle sizes were quant itatively characterized using an LS Coulter 13 320 particle size analyzer. The DNA-MS surface charge was meas ured by zeta potential an alysis in 0.01M PBS at a pH of 7.4 Normal human dermal BJ fibrobl ast cells were used in culture to evaluate the effect of DNA and crosslinked DNA-MS on cell growth. Concentrations of 100g and 500g for DNA, and 25g and 100g for crosslinked DNA-MS were tested. A colorimetric MTS assay was used to measure fibroblast growth at the 0, 24, and 72 hour time points. Materials and Methods Materials Synthesis and characterization The following were purchased from the Si gma-Aldrich Company: DNA sodium salt derived from herring testes Type XIV, cellulose acetate butyrate, HPLC grade 1,2dichloroethane, methanol, chromium (III) pot assium sulfate dodecahydrate, gadolinium (III) chloride hexahydrate, iron (III) nitrate nonahydrat e, 25% (w/w) Grade II aqueous glutaraldehyde solution, and Grey’s Balanced Salt Solu tion. Acetone, sodium phosphate monobasic monohydrate, sodium phosphate dibasic anhydrous, and sodium chloride, each A.C.S. certified, 60mm by 15mm polystyrene radiatio n sterilized petri dishes, Fish erbrand microscope slides, and 50mL and 15mL polypropylene centrifuge tubes were purchased from Fisher Scientific

PAGE 65

65 International. Genipin was obt ained from Challenge Bioproducts Co., Ltd. Type I and Type II deionized ultrapure water was prepared usi ng a Barnstead NANOpure Ultrapure Water System and is termed ultrapure water th roughout. The resistivity of the ultrapure water was at least 16 M -cm-1 for all aqueous solutions prepared. Cell culture Corning 75cm2 cell culture flasks, Fisherbrand cl ear polystyrene 96-well plates, MEM non-essential amino acid soluti on, Cellgro heat inac tivated fetal bovine serum, and 0.22m Corning cellulose nitrate filters were purchased from Fisher Scientific International. The CellTiter 96 Non-Radioactive Cell Prolifera tion Assay was purchased from the Promega Corporation. Normal human dermal BJ fibrobl ast cells, Eagle’s Minimum Essential Modified Media (EMEM), L-glutamine, trypsin-EDTA so lution, and penicillin-streptomycin were obtained from the American Type Cultu re Collection (ATCC, Manassas, VA). Synthesis equipment Solutions prepared or washed by vortex mixing were conducted on a Genie 2 vortex. Centrifugation used a Dynac II bench top centr ifuge. General microsphere syntheses were carried out using Caframo Model BDC6015 and Li ghtnin Model LIU08 mechanical lab mixers in 300mL Labconco lyophilization flasks. Methods Solution preparation Deoxyribonucleic acid. Aqueous 5% (w/v) DNA solutions were prepared at room temperature by adding 0.5g of DNA to 5mL of ultrapure water in a 50mL polypropylene centrifuge tube. The solution was mixed on a vortex mixer for 30 seconds. Three milliliters of ultrapure water was then added to the DNA and placed on the rotary shaker for at least two hours until the DNA had completely dissolved. Once the DNA had dissolved, the volume was brought

PAGE 66

66 up to 10mL and vortexed for 30 seconds. The DNA so lution was then placed in the refrigerator over night to ensure the comple te collapse of bubbles generated during vortex and rotary mixing. The percent solid concentra tion of the DNA solution was qua ntified using a Metler LJ16 Moisture Analyzer at 130C for 60 minutes. The concentration of the DNA solution was then adjusted until the concentr ation was within 10% of the desire d concentration. Once the desired concentration was reached, the aqueous DNA soluti on was placed in the refr igerator until further use. Cellulose acetate butyrate. Solutions of cellulose acetate butyrate in 1,2-dichloroethane (CAB) were used as the water-immiscible c ontinuous phase for the emulsion stabilization process during DNA-MS synthesis. The CAB solu tions were used at a concentration of 5% (w/v) and prepared by adding 25g of cellulose ace tate butyrate to 500mL of 1,2-dichloroethane. The CAB solution was mixed at room temperat ure on a magnetic stir plate on high until the cellulose acetate butyrate had completely dissolved in the 1,2-dichloroethane. The resulting CAB solution was capped, parafilmed, and stored at room temperature. Genipin. Aqueous genipin solutions were prepar ed to a concentration of 5% (w/v) by adding 1.25g of genipin to 25mL of ultrapure water. The genipin solutions were mixed on a vortex at room temperature until th e genipin had completely dissolv ed in the ultrapure water. The genipin solutions were capped, paraf ilmed, and stored in the refrigerator. Glutaraldehyde. Aqueous glutaraldehyde solutions we re prepared to a concentration of 5% (w/v) by diluting the 25% (w /w) aqueous glutaraldehyde solu tion purchased from the Sigma Aldrich Company. The aqueous glutaralde hyde solution was diluted by adding 20mL of ultrapure water to 5mL of the 25% (w/w) solutio n. The glutaraldehyde so lutions were prepared

PAGE 67

67 in 30mL glass jars and mixed on a vortex at room temperature for 1 minute. After mixing, the aqueous glutaraldehyde solutions were paraf ilmed and stored in the refrigerator. Chromium (III) potassium sulfate. Aqueous chromium (III) pot assium sulfate solutions were prepared to a concentration of 0.1M by adding 24.97g of chromium (III) potassium sulfate dodecahydrate to 500mL of ultrapure water. The chromium (III) solutions were mixed on a magnetic stir plate over night at room temperature until all the chromium had dissolved in the water. After complete mixing, the 0.1M chromi um (III) solution was parafilmed and stored at room temperature. Gadolinium (III) chloride. Aqueous gadolinium (III) chloride solutions were prepared to a concentration of 0.1M by adding 18.59g of gado linium (III) chloride hexahydrate to 500mL of ultrapure water. The gadolinium (III) solutions we re mixed on a magnetic stir plate over night at room temperature until all the gadolinium had disso lved in the water. After complete mixing, the 0.1M gadolinium (III) solution was parafilm ed and stored at room temperature. Iron (III) nitrate. Aqueous iron (III) nitrate solutions were prepared to a concentration of 0.1M by adding 2.02g of iron (III) nitrate nonahydrate to 50mL of ultrapure water. The iron (III) solutions were mixed in a 50mL polypropylene cent rifuge tube on a rotary shaker over night at room temperature until all the ir on had dissolved in the water. After complete mixing, the 0.1M iron (III) solution was parafilmed a nd stored at room temperature. Phosphate buffered saline. Four liters of a 0.05M phosphate buffered saline (PBS) solution with a pH of 7.4 was prepared in the lab for measuring the sw elling properties of the DNA-MS. The PBS solution was prepared by mi xing 2.9L of a 0.05M sodium phosphate dibasic solution with 1L of a 0.05M sodium phosphate monobasic solution. During mixing, the pH of

PAGE 68

68 the PBS solution was measured and sodium phos phate monobasic solution was added until the target pH of 7.4 was reached. A PBS solution with a concentration of 0.01M at a pH of 7.4 was used for the zeta potential measurements and was prepared by d iluting a 0.1M PBS solution and adjusting the pH back to 7.4. The 0.1M PBS solution was prepar ed by mixing 2.9L of a 0.1M sodium phosphate dibasic solution with 1L of a sodium phosphate monobasic solution. The two solutions were mixed and the pH of the resulting solution wa s brought to 7.4 by adding monobasic solution. The prepared PBS solutions were stored at room temperature until needed. Cell culture media. Cell culture media was prepared by adding 50mL of fetal bovine serum (FBS), 10mL of non-esse ntial amino acid solution, 10mL of L-glutamine, and 10mL of penicillin-streptomycin to 500mL of Eagle’s Minimum Essentia l Modified Media (EMEM) in order to obtain a 10%FBS, 2%non-essential am ino acid (NEAA), 2% L-glutamine, and 2% penicillin-streptomycin treated media and will be referred to as treated media from this point forward. The treated media solutions were mixed manually and placed in the refrigerator until further use. DNA solutions used in the cell cu lture study were prepared as mentioned above; however; they were sterilized with a 0.22m filter immediately prio r to adding treatment. Crosslinking reaction study The time to crosslink DNA with covalent or ionic crosslinking agents was estimated by measuring the time needed for each agent to pr ecipitate DNA from solution. For these studies, genipin and glutaraldehyde covale nt crosslinking agents were tested by adding 2mL of the 5% (w/v) crosslinking agent solutio ns to 3mL of the 5% (w/v) aqueous DNA solution. For ionic crosslinking, 2mL of the 0.1M chromium (III), gadolinium (III), or ir on (III) solutions were added to 3mL of the 5% (w/v) aqueous DNA solution. The amount of time required for the individual agent to crosslink the DNA, as meas ured by the precipitation of the DNA, was then

PAGE 69

69 recorded. These studies were conducted at room temperature w ith a two hour reaction time in order to mimic conditions observed during MS synthesis. Pilot microsphere synthesis study DNA-MS were initially prepared using ch romium (III) potassium sulfate dodecahydrate to determine if the ch romium (III) cations (Cr3+) could ionically crosslink the DNA molecule. For this study, DNA-MS were synthesized us ing a 1% (w/v) aqueous DNA solution, a 0.1M Cr3+ solution, and a 3% (w/v) CAB solution. DNA-MS were prepared by adding 2mL of the 1% (w/v) aqueous DNA solution to 16mL of the 3% (w/v) CAB solution in a 50mL polypropylene centrifuge tube. The centrifuge tube was capped and then mixed on the vortex at level 8 for 2 minutes. After 2 minutes, 1mL of the 0.1M Cr3+ solution was added and mixed on the vortex at level 8 for an additional 5 minutes. After 5 mi nutes, the DNA-MS suspension was separated into three separate 50mL polypropylene centrifuge tu bes and acetone was added up to 35mL. The DNA-MS were then rinsed in the acetone for 30 s econds using the vortex at level 8 and collected by centrifugation at 2000rpm for 10 minutes. Afte r centrifugation, the acet one was decanted and the DNA-MS were combined into two centrifuge t ubes. The acetone rinse was repeated and then the DNA-MS were combined into one centrifuge t ube and the acetone rinse was repeated. After the final rinse, the acetone was decanted and th e DNA-MS were allowed to air dry at room temperature by securing a Kimwipe over the mouth of the open centrifuge tube with a rubber band. General microsphere synthesis Based on results obtained from the pilot st udy and previous research conducted on BSAMS and GEL-MS synthesis, DNA-MS were prepar ed using an emulsion st abilization technique that sterically stabilizes the DNA molecule into spherical conformations and crosslinks them while in suspension. This emulsion stabilization process involved dispersing 3mL of a 5% (w/v)

PAGE 70

70 aqueous DNA solution (i.e. the aqueous phase) into 47mL of a 5% (w/v) CAB solution (i.e. the continuous phase) in a 300mL Labconco lyoph ilization flask. After dispersion, a DNA microemulsion was created by vigorously mixing th e two solutions at 1250rpm for 20 minutes at room temperature, using a paddle mixer with a two inch, two blade propeller. The DNA microemulsion was then covalently or ionically crosslinked while in suspension by reducing the speed of the paddle mixer to 600rpm and adding 2m L of a crosslinking agent. In the case where genipin was used as a crosslinking agent, 9mL was added instead of 2mL because the amount of genipin needed to establish the same amount of crosslinking as glutar aldehyde in albumin was found to be 4.5 times more.57 Therefore, the amount of genipi n used for these studies was also increased by a factor of 4.5. The DNA microe mulsion then underwent crosslinking for 1 hour and 40 minutes at which time 50mL of acetone was added and any further reactions were allowed to reach completion for another hour. After synthesis was complete, the DNA-MS underwent four rinses in acetone to remove any re sidual organic phase or crosslinking agent. The DNA-MS were rinsed by separating the resu ltant DNA-MS suspension into four separate 50mL polypropylene centrifuge tubes. To thes e tubes, acetone was added up to 35mL and the tubes were capped and vortexed on high for 30 seconds. After vortexing, the DNA-MS were collected by centrifuging the tubes at 2600rpm for 10 minutes. After centrifugation, the acetone was decanted and fresh acetone was added again up to 35mL. The acetone rinse was repeated once more as mentioned above and then twice more by consolidating the contents of 4 tubes to 2 tubes and then 2 tubes to one t ube. After the final acetone ri nse, the centrifuge tube was uncapped, the acetone was decanted, and a Kimwip e was secured over the mouth of the tube using a rubber band. The DNA-MS were then allo wed to dry overnight at room temperature.

PAGE 71

71 Crosslinking determination DNA-MS were crosslinked by either ioni c bonding through the phos phate groups on the DNA backbone or by covalent bonding presumably through the amino groups within the base pairs of the DNA molecule. Cro sslinking efficiency was estimated on the basis of percent molar equivalence (%MEQ) of crosslink agent to moles of po ssible reactive sites found within one DNA repeat unit. For these crosslinking estimates, a generic repeat unit structure, independent of DNA source, was established and considered to in clude phosphate, deoxyribose, adenine, thymine, cytosine, and guanine groups as shown in Figur e 3-1. The number of repeat units within a DNA molecule was estimated from the molecular we ight of the DNA molecule and the molecular weight of the repeat unit. A molecular we ight of 1.3million Daltons was used for these calculations as reported from the Sigma-Aldrich Company. N N N N N O H O H H H CH2 H P OH O O N O O N O H O H H H CH2 H P O N N O N O H O H H H H2C H P O N N N O N N O H H H H CH2 H O P OH O O Adenine Thymine Guanine CytosineH H H HO HO H H H H H HO O O HO Figure 3-1. A cartoon of the DNA ge neric repeat unit used for all crosslink calculations (Note: Bonds are not drawn to scale).

PAGE 72

72 Using the generic repeat unit model, the number of amino groups and phosphate groups per repeat unit were calculated to obtain the numbe r of sites available fo r possible covalent or ionic crosslinking. The number of amino groups and phosphate groups per repeat unit was multiplied by the total number of DNA repeat un its in the 1.3 million Daltons chain to obtain the total number of possible reac tive sites; which were 3,231 amino groups and 4,308 phosphate groups. The extent of crosslinking was estimated from these available sites. Ionic crosslinking was based on the assumption that the trivalent cations can bind to 3 possible phosphate groups as shown in Figure 3-2. O H O H H H CH2 H P O O H O H H H CH2 H P O O P O O OThymine Cytosine OO OO O H O H H H CH2 H P O O H H H H CH2 H O P O O P O O OAdenine Guanine OO OO M M Figure 3-2. Schematic drawing of possible DNA crosslinking s ites for trivalen t cations, M = Gd3+, Cr3+, or Fe3+. (Note: Bonds are not drawn to scale.)

PAGE 73

73 For covalent crosslinking conditions, it was assumed that one glutaraldehyde molecule could establish a crosslinking unlike genipin for which it wa s assumed that two genipin molecules would be needed for crosslinking.11, 96 Glutaraldehyde cros slinking was assumed to occur through a Schiff base reaction with the am ino groups found within the adenine, cytosine, and guanine nucleobases of DNA. A possible mechanism for DNA covalently crosslinking with glutaraldehyde is shown in Figure 3-3. Future studies should be conducted to determine the true mechanism for DNA to crosslink with glutaral dehyde. A possible schematic drawing for DNA covalently crosslinking with ge nipin was not included since geni pin was found to not crosslink DNA efficiently (i.e. DNA-MS did not turn blue and immediately dissolved in water.). + O O Glutaraldehyde N N N N N N N O N Adenine Cytosine R1 R2 H H H H 2H2O + N N N N N N N O N Adenine Cytosine R1 R2 Schiffbase + O O Glutaraldehyde N N N N N N N O N Adenine Cytosine R1 R2 H H H H 2H2O + N N N N N N N O N Adenine Cytosine R1 R2 Schiffbase Figure 3-3. Schematic drawing of a possible m echanism for glutaraldehyde to covalently crosslink DNA. (Note: R1 and R2 represent the remainder of DNA molecule.) Microsphere characterization Yield analysis. The yield of each DNA-MS condition was calculated and expressed as a percent yield value. The percent yield was cal culated by dividing the final weight of the DNAMS by the amount of weight used to synthesize the MS. Equation 3-1 i llustrates the expression for calculating the percent yield where WF is the final weight of the DNA-MS, VDNA, DNA, and

PAGE 74

74 CDNA are the volume, density, and concentrati on of the aqueous DNA solution used, and WX is the weight of the crosslinking agent added during synthesis 100 ) W ) V ((C W ( %YieldX DNA DNA DNA F % (3-1) Dry particle size analysis. The mean dry diameters and pa rticle size distributions of the DNA-MS were obtained using a Coulter LS 13 320 particle size analyzer Approximately 2mg of the dried DNA-MS were suspended in 2mL of methanol. The suspension was then sonicated for 30 seconds in order to break up any aggregat es and tested. The Coulter LS 13 320 particle size analyzer was set to run at a pump speed of 73% using a protein/ DNA particle diffraction model. Standards were tested in methanol before the first run of the first batch to ensure that the instrument was performing adequate ly. Each condition was tested th ree times in which each test consisted of two runs. This method of testing produced six independent particle diameters and size distributions. Data collected from these experiments were statistically analyzed using SigmaStat 3.0 software. Hydrated particle size analysis. The mean swollen diameters and particle size distributions of the DNA-MS were obtained usin g a Coulter LS 13 320 particle size analyzer. Approximately 2mg of the DNA-MS were suspe nded in 2mL of 0.05M PBS with a pH of 7.4. The suspension was then sonicated for 30 seconds in an attempt to break up any MS aggregates. The DNA-MS were then allowed to swell in the PBS for an additional two minutes and thirty seconds. After swelling the DNA-MS were tested in the Coulter LS 13 320 particle size analyzer using a pump speed of 73% and a protein/DNA partic le diffraction model. Standards were tested in PBS before the first run to ensure that th e instrument was performing adequately. Each condition was tested three times in which each test consisted of two runs. This method of testing produced six independent particle diameters a nd size distributions. Af ter obtaining the mean

PAGE 75

75 swollen particle diameters, the data was used to calculate the percent change in size following Equation 3-2, where DD is the dry diameter and DH is the swollen or hydrated diameter. A negative percent change in size depicted a decrea se in particle size (i.e degradation) whereas a positive percent change in size depicted an increase in particle size (i.e. swelling). Data collected from these experiments were statistically analyzed using SigmaStat 3.0 software. 100 ) D D D ( Size in Change %D D H (3-2) Surface charge analysis. DNA-MS surface charge was characterized to determine dispersability and obtained using a Brookhaven Zeta Plus zeta potential anal yzer with ZetaPALS software. Approximately 2mg of dry DNA-MS were suspended in 1.5mL of 0.01M PBS with a pH of 7.4. Each DNA-MS condition was sample d three times in which each sample underwent ten runs. This method of testing produced thir ty independent zeta potent ial values. The data collected from zeta potential analys is was statistically analyzed using SigmaStat 3.0 software. Dispersability. The dispersability of the DNA-MS was measured by suspending approximately 13mg of dried DNA-MS in 20mL of cold Grey’s Balanced Salt Solution (BSS) in a 50mL polypropylene centrifuge. The centrif uge tube was capped and the DNA-MS were mixed on a vortex at level 8 for thirty seconds. After mixing, the centrifuge tube was placed in a stand and the time for DNA-MS to fall out of dispersion was measured using a stop watch. Each condition was sampled three times resulting in th ree separate dispersability measurements. Optical microscopy. The morphology of the DNA-MS was observed through optical microscopy. Approximately 1mg of the DNA-MS were spread over a Fisherbrand microscope slide and observed under a Zeiss Ax ioplan2 optical microscope. Im ages were taken with a Carl Zeiss AxioCam HR camera under 40x magnification and saved usi ng AxioVision 3.1 software.

PAGE 76

76 Scanning electron microscopy. The morphology and surface topography of the DNAMS were observed using scanning electron micr oscopy (SEM). Approximately 1mg of dry DNA-MS was mounted onto aluminum SEM stubs us ing double sided tape. The DNA-MS were then coated with gold-palladium for 2 minutes using a Technix Hummer V sputter coater. Images were taken either on a JEOL 6400 SE M using an accelerati ng voltage of 5KeV, condenser lens setting of 10, obj ective lens setting of 117, and a working distance of 15mm, or on a JEOL 6335F Field Emission SEM at an ac celerating voltage of 5KeV and a working distance of 15mm. Energy dispersive x-Ray spectroscopy. The presence of trivalent cations in the DNAMS after washing and drying was observed using en ergy dispersive x-ray spectroscopy (EDS). DNA-MS were mounted onto a piece of silicon wafe r. The silicon wafer was then secured to aluminum SEM stubs using carbon double sided ta pe. The DNA-MS were then coated with carbon for 2 minutes using a Technix Hummer V sputter coater. EDS spectra on the DNA-MS were collected using a JEOL 6400 SEM at an accelerating voltage of 15KeV and working distance of 15mm. A dead time of 20% to 40% was allowed for each condition tested. Evaluation of fibroblast growth Synopsis. Normal human dermal BJ fibroblast cells were used in culture to determine the effects DNA and crosslinked DNA-MS on ce llular growth. DNA was evaluated at concentrations of 100g and 500g, and DNA-MS were tested at concentrations of and 25g and 100g. The effects of DNA and DNA-MS on fi broblast growth was determined using a colorimetric MTS assay which measures the numb er of metabolically active cells in culture.35 The MTS tetrazolium salt, which is also known as Owen’s reagent, is bioreduced to a colored formazan product through the functioning electron transport systems of viable cells.35, 97, 98 This conversion is assumed to take place through th e interaction of the tetrazolium salt with

PAGE 77

77 nicotinamide adenine dinucleotide (NADH) lin ked dehydrogenases, which are produced during cellular respiration.35, 99 This interaction results in the oxidized form of nicotinamide adenine dinucleotide (NAD+) and a colored formazan product which absorbs at the 490nm wavelength, Figure 3-4.35 The fibroblast cells were evaluated at the 1, 24, and 72 hour time points. An increase in cell growth was defined by an increase in absorbance at 490nm for any given treatment group. OCH2COOH N N N N SO3 N S CH3 CH3 MTSTetrazoliumCompound (Owen'sReagent) O P HO O P O HO O H H O OH OH H N N O O OH OH H N N N N N NADH +H H H H O +N A D+ OCH2COOH NN N N H N S CH3 CH3 SO3 FormazanO P HO O P O HO O H H O OH OH H N N O O OH OH H N N N N N +H H H H O + OCH2COOH N N N N SO3 N S CH3 CH3 MTSTetrazoliumCompound (Owen'sReagent) O P HO O P O HO O H H O OH OH H N N O O OH OH H N N N N N NADH +H H H H O +N A D+ OCH2COOH NN N N H N S CH3 CH3 SO3 FormazanO P HO O P O HO O H H O OH OH H N N O O OH OH H N N N N N +H H H H O + Figure 3-4. The chemical conversion of MTS to formazan.35, 97, 99 Culture. Fibroblast cells were cultured in 75cm2 cell culture flasks with 20mL of treated media. The cells were incubated at 37C in humid ified air with 5% carbon dioxide for at least 24 hours. After 24 hours, the cells were monitored daily until the cells reached 95% confluency. The cells were then trypsinized and counted in order to determine the cel l density. If the cell density was below a concentration of 1 x 104 cells/mL, the cells were split and re-cultured until the cell density was between 1 x 104 cells/mL and 1 x 105 cells/mL. Once the desired cell density was obtained, the cell susp ension was used to seed the co rresponding 96-well plates used for the study. Cell seeding. A cell density of 4.4 x 104 cells/mL was obtained for these studies and 100L of the acquired cell suspension were added to specified wells in six separate 96-well

PAGE 78

78 plates. Eight well plates were used for this study representing the two c ontrol plates used to measure the absorbance of treatment groups in th e absence of cells and the six treatment plates representative of the 1, 24, and 72 hour time points. The six seeded well plat es used to evaluate the cytotoxicity of the treatment groups were incubated at 37 C in humidified air with 5% carbon dioxide in order to reach ade quate attachment for at least 24 hours before adding the first treatment. After 24 hours the media was removed fr om the wells and the treatments were added. The cell culture procedures for this prolifer ation study were conducted with the help and guidance of Paul Martin. Treatment groups. The treatment groups for this study evaluated the effects of DNA and crosslinked DNA-MS on fibroblast growth. To evaluate DNA as a sole biomaterial without the influence of possible cytotoxic crosslinking agents, 0.50g of DNA derived from herring testes was dissolved in 10mL of ultrapure water. Th e aqueous DNA solution was then sterilized using a 0.22m filter and 2L of the st erile DNA solution was added to 98L of the treated media to obtain a concentration of 100g of DNA in th e wells. Similarly, 10L of the filtered DNA solution was added to 90L of the treated media to obtain a concentration of 500g. DNA-MS synthesized with covalent and ioni c crosslinking agents were assessed at concentrations of 25g and 100g to determine if the leaching of the cr osslinking agents during DNA-MS degradation caused a negative response on fibroblast growth. For these conditions, 2mg of crosslinked DNA-MS were sterilized by thoroughly rinsing th e DNA-MS with 1mL of 70% ethanol by vortexing them on high for 30 sec onds. The DNA-MS were then centrifuged at 3,000rpm for 5 minutes and collected by decanting the ethanol and allowing the sterilized DNAMS to air dry over night. Th e sterilized DNA-MS were then resuspended in 1mL of treated media and added to designated cells using approp riate volumes to reach desired treatment group

PAGE 79

79 concentrations. For the 25g tr eatment conditions, 13L of the MS/media suspension was added to 87L of the treated media. For the 100 g treatment conditions, 50L of the MS/media suspension was added to 50L of the treated media. There were three control groups used in this study which consisted of 1) treated media in the absence of cells, 2) treated media with cells, 3) treatment groups in treated media in the absence of cells. Absorbance valu es taken of the media and treat ment groups in the absence of cells were used to correct cell cult ure data obtained in the study. To access the morphology of the fibroblast cells after treatment, the cells were fixed with 10% formalin and stained with crystal violet im mediately following the proliferation assay. All control, DNA, and DNA-MS conditions were evalua ted in replicates of six. All cellular proliferation data collected was statistically an alyzed using SigmaStat 3.0 Software. Results Pilot Microsphere Synthesis Study Synopsis A pilot study was conducted to determin e if DNA-MS could be prepared through chromium trivalent cation crosslinking. Synthesis proced ures were carried out in a 50mL polypropylene centrifuge tube at room temperatur e. Prepared chromium crosslinked DNA-MS were analyzed via scan ning electron microscopy. Scanning electron microscopy The pilot study was successful in producing chromium triv alent cation crosslinked DNAMS. Evidence of successful chromium bonding was visually denoted by the blue color of the resultant DNA-MS. It was visually observed th at during the synthesis process, the chromium crosslinked DNA-MS aggregated during wa shing and furthermore upon drying. SEM micrographs confirmed aggregation and discrete particles were not produced. However, the

PAGE 80

80 SEM micrographs also illust rated that the pilot study pr oduced DNA-MS with spherical morphologies as shown in Figure 3-5. Figure 3-5. SEM micrograph of chromium crosslinked DNA-MS (Magnification:1,000x). Stabilizing Agent Study Synopsis A study was designed to evaluate the effects of th e stabilizing agent c oncentration (CAB) on synthesized DNA-MS particle discreteness due to particle aggregation encountered in the pilot study. The concentrations of the CAB soluti ons were varied for this study and assessed at 3% (w/v), 5% (w/v), 10% (w/v), and 25% (w/v). In order to incr ease the yield of the synthesized DNA-MS, the concentration of the aqueous DNA so lution was increased from 1% (w/v) to 3% (w/v). Prepared DNA-MS were characte rized using scanning electron microscopy.

PAGE 81

81 Scanning electron microscopy SEM micrographs displayed varied results for the CAB concentrations used for this study. The 5% (w/v) CAB condition produced DNA-MS that were disperse during washing and dried into nice discreet powders. DNA-MS prep ared with the 3% (w/v ), 10% (w/v), and 25% (w/v) CAB concentrations produ ced particles that aggregated during acetone rinsing which resulted in agglomerated clum ps upon drying. SEM visually c onfirmed these observations as shown in Figure 3-6. Figure 3-6. SEM micrographs of DNA-MS prepared with CAB concen trations of A) 3% (w/v), B) 5% (w/v), C) 10% (w/v), D) 25% (w/v) (Magnifications: 2,000x), and E) 25% (w/v) (Magnification: 13 0x; Scale bar: 200m). The 5% (w/v) CAB condition produced the most discrete DNA-MS particles as compared with the other three conditions examined. Th e 25% (w/v) CAB conditi on produced the highest level agglomeration, followed by the 10% (w/v) and 3% (w/v) CAB conditions. SEM micrographs also illustrated a trend in decr easing particle diameter size as the CAB

PAGE 82

82 concentration increased with the 25% (w/v) CAB condition producing DNA-MS with the smallest diameters. Crosslinking Reaction Study The time to crosslink DNA with covalent or ionic crosslinking agents used for general microsphere synthesis was investigated. Cross linking time was estimated by measuring the time needed for each agent to pr ecipitate DNA from solution. Th is study found that the ionic crosslinking agents displayed the quickest reaction with DNA, with the glutaraldehyde and genipin following after. The chromium and gadolinium crosslinking agents reacted instantaneously with the DNA and appeared to produce films that were crosslinked homogenously. The iron reacted with the DNA rapidly; however, the reaction was not homogeneous and produced clumps of iron cros slinked DNA instead of a uniform DNA film. It took approximately two hours for the glutaral dehyde to react with the DNA and produce a crosslinked film. The film app eared to be crosslinked homoge nously. The genipin crosslinking agent did not appear to react with the DNA sin ce no noticeable change in color was observed (i.e. The DNA solution did not turn from white to blue which is indicativ e of genipin reacting with openly functional primary amine sites.95). It was speculated that the genipin solution was no longer reactive since there was no interaction between the DNA and genipin. A side study was therefore conduc ted to determine if the genipin solution was still reactive. To test this, 9mL of the 5% (w/v) genipin solution was added to 3mL of 5% (w/v) lysine, 3mL of irra diated 5% (w/v) DNA, and 3mL of 5% (w/v) DNA separately. The genipin and lysine and genipin and DNA solu tions were mixed on the vortex mixer on level 8 for 30 seconds and then placed on a rotary mixer. The genipin reacted completely with the lysine solution after two hour s. The color of the lysine solution changed from white to a dark purplish blue. On the other hand, it took approximately 72 hours for the

PAGE 83

83 genipin to react with both DNA solutions and the reaction never reached completion. It was assumed that the genipin would react quicker with the irradiated DNA since it would be degraded and thus have functional amino sites more readily accessible than the non-degraded DNA, but this was not the case. Both DNA solutions changed from a white color to a reddish brown to finally a dark bluish yellow indicati ng a small reaction between the genipin and the amino groups within the DNA molecule. General Microsphere Synthesis Synopsis Full scale studies were designed to evaluate and characterize DNA-MS synthesized with covalent and ionic crosslinking agents based on the successes encountered with the pilot and stabilizing agent studies. The concentra tion of the beginning aqueous DNA solution was increased from 3% (w/v) to 5% (w/v) to increa se the resulting DNA-MS yield for these studies. Particle analysis Percent yield. DNA-MS synthesis app eared to be successful with each crosslink agent condition producing yields over 50%. The ga dolinium and genipin crosslinked DNA-MS conditions produced discrete part icle yields; however, the iron, chromium, and glutaraldehyde conditions produced aggregates upon washing and drying. The gadolinium crosslinked DNAMS performed the best producing a yield of 82%. The percent yield values for each condition are tabulated in Table 3-1. Table 3-1. The yields, dry mean particle diameters, and cros slink concentrations for DNA-MS prepared with ionic and c ovalent crosslinking agents. Crosslinking agent Percent yield (%) Crosslink concentration (%MEQ) Chromium 77 120 Gadolinium 82 120 Iron 54 120 Glutaraldehyde 61 540 Genipin 63 1070

PAGE 84

84 Dry particle size analysis. The dry particle sizes of the DNA-MS were obtained in methanol. DNA-MS prepared with covalent an d ionic crosslinking agents produced particles with mean dry diameters of less than 20m. St atistical analysis using a one way analysis of variance test (ANOVA) found there to be no sign ificant differences among the mean particle diameter values for all conditions tested. Ho wever, the gadolinium and genipin crosslinked DNA-MS conditions appear to have performed th e best, producing the highest percentage of particles in the mesosphere size range and the lo west percentage of particles with diameters greater than 10m. The gadolin ium condition produced mean dry particle diameters of 2.6m 2.8m and the genipin condition produced mean dr y particle diameters of 3.1m 2.9m. The glutaraldehyde crosslink condition performed similarly well with a mean dry particle diameter of 6.4m 9.7m; however produced over 5% of partic les with diameters greater than 10m. Both the chromium and iron crosslink agent conditions produced the largest mean particle diameters as compared to all other conditions tested with values of 10.3m 13.9m and 14.1m 16.7m, respectively. The mean dry particle diameters and the size ranges for each of the synthesized DNA-MS conditions are shown in Table 3-2. Table 3-2. The dry mean particle diameters and size ranges for DNA-MS prepared with ionic and covalent crosslinking agents. Crosslinking agent Mean dry particle diameter (m) DNA-MS in 1m to 10m range (%) DNA-MS larger than 10m (%) Chromium 10.3 13.9 32 44 Gadolinium 2.6 2.8 81 2 Iron 14.1 16.7 31 50 Glutaraldehyde 6.4 9.7 64 16 Genipin 3.1 2.9 73 3 The gadolinium condition produced the most controlled and narrow particle size distribution of the three ionic crosslinking agent conditions test ed. The chromium crosslink agent condition displayed an irregular and mu ltimodal distribution over a very broad range of

PAGE 85

85 particle diameter sizes. The ir on condition exhibited a bimodal di stribution with a larger volume percent of DNAMS over a range of larger particle sizes. Each of the three ionic agent crosslink conditions produced particles within the nano-me sosphere range of 67nm to 2m, with the gadolinium crosslink condition prod ucing the highest volume percent of the three. The particle size distributions for ionically cross linked DNA-MS is shown in Figure 3-7. 0 1 2 3 4 5 0.010.1110100100010000Particle Size (m)Volume % Fe Gd Cr Figure 3-7. Particle si ze distributions of ionically cro sslinked DNA-MS under dry conditions. The covalently crosslinked DNA-MS performe d similarly well with dry mean diameters less than 10m and mostly nor malized particle size distri butions. DNA-MS covalently crosslinked with glutaraldehyde displayed a narrow particle size distribution; however, aggregates were visually noted above 10m. DNA-MS prepared with genipin also exhibited a normalized and narrow particle size distributi on; however, resulting DNA-MS were white in color instead of blue indicating a lack of r eaction between the genipin and DNA. Similar to DNA-MS synthesized with ionic cr osslinking agents, the covalent crosslinking agent conditions also produced nano-mesospheres in the 60nm to 2m range. Particle size distributions for covalently crosslinked DNA-MS can be found in Figure 3-8.

PAGE 86

86 0 1 2 3 4 0.010.1110100100010000Particle Size (m)Volume % GEN GTA Figure 3-8. Particle si ze distributions for covalently cr osslinked DNA-MS under dry conditions. Hydrated particle size analysis. DNA-MS synthesized with covalent and ionic crosslinking agents also underwent particle size analysis under hydrated conditions in 0.05M PBS at a pH of 7.4. Approximately 2mg of the DNA-MS were suspended in PBS and sonicated for 30 seconds in a 15mL centrifuge tube. Upon suspension into PBS, it was noted that the DNA-MS began to stick to the walls of the centrif uge tube and the particles began to clump or aggregate together. The samples were sonica ted to disperse the DNA-MS; however, hydrated particle size analysis for both i onic and covalent conditions illus trated that sonication was not successful in breaking up aggregates. Hydrated particle diameters for the ionical ly crosslinked DNA-MS ranged from 11m to 45m and each condition, with the exception of the iron condition, displayed hydrated diameters of greater than 25m. However, it was suspec ted that these values mostly represented the aggregates that formed during hydration instead of the true values of hydrated DNA-MS. Upon closer observation, it appeared th at the median particle diameter values better represented the

PAGE 87

87 true hydrated size of the DNA-MS and were theref ore tabulated and compared to mean hydrated particle size values, Table 3-3. Table 3-3. Mean and median hydrated particle size values for ionically crosslinked DNA-MS. Crosslinking Agent Mean dry particle diameter (m) Mean hydrated Particle diameter (m) Median hydrated particle diameter (m) Chromium 10.3 13.9 25.8 29.6 14.9 Gadolinium 2.6 2.8 45.1 85.7 12.1 Iron 14.2 17.1 11.9 10.9 9.3 Glutaraldehyde 6.4 9.7 128 222 27.6 Genipin 3.1 2.9 0 0 0 0 The ionic crosslink agent condi tions produced mostly normali zed hydrated particle size distributions with evidence of aggregation most noted in the chromium and gadolinium conditions, Figure 3-9. The iron particle size distribution did not illustrate signs of aggregation, however, it was noted that duri ng particle size anal ysis, the size of the DNA-MS began to decrease rapidly suggesting that the iron crossl inked DNA-MS were dissolv ing or degrading in the PBS. 0 1 2 3 4 5 6 0.010.1110100100010000Particle Size ( m ) Volume % Fe Gd Cr Figure 3-9. Particle size dist ributions of ionically cro sslinked DNA-MS under hydrated conditions.

PAGE 88

88 The glutaraldehyde crosslink agent condition pr oduced a mean hydrated particle size of 128m. However, it was also suspected that this value was a better representation of the aggregates that formed upon hydration instead of the true swollen diameter size of the DNA-MS. Therefore, the median hydrated particle size diam eter was tabulated and compared to the mean hydrated value, Table 3-3. The glutaraldehyde crosslink agent condition wa s the only condition to display stability in PBS. The DNA-MS prepared with genipin im mediately dissolved and were unable to undergo hydrated particle size analysis. Glutaralde hyde crosslinked DNA-MS illustrated multimodal hydrated particle size distributions with a large percent of DNA-MS exhibiting hydrated diameter values over the 100m to 1000m size range, indicative of a high de gree of aggregation upon hydration, Figure 3-10. 0 1 2 3 4 0.010.1110100100010000Particle Size (m)Volume % GTA Figure 3-10. Particle size distribution of the glutaraldehyde crosslinked DNA-MS under hydrated conditions. Statistical analysis was conducted on all co llected mean and median hydrated data to further compare each DNA-MS condition under hydration. A one way ANOVA found no significant differences among all mean and median hydr ated particle size valu es tested. The lack

PAGE 89

89 of statistical difference among mean and median values can be attributed to the variability of sizes due to the formation of aggregates upon hydration. In order to quantify the effect of hydration on particle size, the mean and median percent change in size values were calculated for a ll DNA-MS conditions tested using the mean dry, mean hydrated, and median hydrated diameter va lues. Equation 3-2 was used for all percent change in size calculations and values for each DNA-MS condition tested ar e given in Table 3-4. Table 3-4. Percent change in size values for ionically and covalently crosslinked DNA-MS. Crosslinking Agent Mean percent change in size (%) Median percent change in size (%) Chromium (+) 151 (+) 45 Gadolinium (+) 1640 (+) 365 Iron (-) 19 (-) 35 Glutaraldehyde (+) 1900 (+) 331 Genipin (-) 100 (-) 100 The gadolinium crosslinked DNA-MS condition pr oduced the highest percent increase in size under hydrated conditions, with glutaraldeh yde and chromium following after due to the formation of aggregates upon hydration. Th e chromium crosslinked DNA-MS condition performed the best with only a 151% mean or 45% median increase in size under hydrated conditions. The iron crosslink agent condition displayed the least stability under aqueous conditions of all the ionic crosslink agent condi tions tested with a 19 % mean or 35% median decrease in size. The genipin crosslinked DNAMS condition displayed th e least stability and immediately dissolved upon hydration producing a 100% decrease in size. Surface charge and dispersability Surface charge. The surface charge was measured for DNA-MS that were stable under hydrated conditions. Therefore, only the glutar aldehyde, chromium, and gadolinium crosslinked DNA-MS were evaluated. The surface charges of the DNA-MS were measured using a zeta potential analyzer in 0.01M PBS at a pH of 7.4 and expressed in millivolts (mV). The

PAGE 90

90 gadolinium crosslinked DNA-MS condition displayed th e most negative zeta po tential values, -45.3mV, as compared to the chromium and glut araldehyde crosslinked conditions, Figure 3-11. -50 -40 -30 -20 -10 0Zeta Potential (mV) Gd Cr GTA Figure 3-11. Zeta potential chart for DNA-MS crosslinked wi th gadolinium (Gd), chromium (Cr), and glutaraldehyde (GTA). The glutaraldehyde crosslinked condition disp layed the second most negative zeta potential value of -38.0mV, followed by the ch romium crosslinked DNA-MS which displayed the least negative zeta potential, -29.6mV. Statistical analysis using a one way ANOVA test found the zeta potential values for all conditions to be signif icantly different from one another (p < 0.001). Further analysis using a Tukey Test for all pairwise multiple comparisons found the zeta potential values for chromium to be si gnificantly lower from those obtained for the gadolinium (p < 0.001) and glutaraldehyde (p < 0.001) conditions, and the zeta potential values for glutaraldehyde to be significantly lower th an those obtained for the gadolinium (p = 0.002) condition. The zeta potential values with thei r respective standard errors for each condition tested are given in Table 3-5.

PAGE 91

91 Table 3-5. Zeta potential va lues and dispersability times for DNA-MS crosslinked with gadolinium, chromium, and glutaraldehyde. Crosslinking agent Zeta potential (mV) Dispersability time (hr) Gadolinium -45.3 2.448 3 Chromium -29.6 1.912 1 Glutaraldehyde -38.0 3.01 1 Dispersability. It is important for DNA-MS to be stable and dispersi ble under aqueous conditions for their intended application of intr atumoral chemotherapy. Therefore, gadolinium, chromium, and glutaraldehyde cros slinked DNA-MS were tested in Grey’s balanced salt solution for dispersability. The gadolin ium crosslinked DNA-MS displayed the most stable dispersability with dispersion times exceeding 24 hours. The gadolinium crosslinked DNA-MS maintained their dispersion for 48 hours upon which slight ag itation could resuspend the microspheres for another 48 hours. The chromium crosslinke d DNA-MS did not perform as well as the gadolinium crosslinked MS and di splayed dispersion times that onl y lasted 12 hours. For this condition, approximated 25% of a ll dispersed MS fell out of solution within the first two minutes; and within four hours, approximately 60% of the MS were out of dispersion. The remaining DNA-MS continued to fall out of di spersion for the remaining eight hours. The glutaraldehyde crosslinked DNA-MS conditions pe rformed the worst as compared to all other conditions tested in which 60% of all MS clum ped together during vorte x mixing and fell out of dispersion within the first two minutes. The rema ining 40% fell out of dispersion within the first hour. Dispersability times for each DNA-MS condition tested are given in Table 3-5. Microscopy Optical microscopy. Dried DNA-MS were observed under optical microscopy to evaluate their discreteness, mor phology, and particle size after synt hesis. Optical images of the ionically crosslinked DNA-MS displayed par ticles with small diameters and spherical morphologies, Figure 3-12. Each ionic crosslink ag ent condition produced disc rete particles with

PAGE 92

92 the exception of the iron crosslink agent condition which app eared to exhibit a moderate amount of aggregation. Optical images of the chromium crosslink agent conditio n also displayed some particles that appeared aggregated and had slig htly irregular shapes. Optical images also confirmed data obtained through part icle size analysis illustrating that the chromium crosslinked condition produced DNA-MS with larger partic le sizes than the gadolinium condition. A B C A B A B C Figure 3-12. Optical microscopy images of DNAMS ionically crosslinked with A) chromium, B) gadolinium, and C) iron trival ent cations (Magnification: 400x). The optical images of the ionically crosslinke d DNA-MS revealed particles sizes that were much smaller than those produced with covalent crosslinking agents. The DNA-MS that were covalently crosslinked with glutaraldehyde produ ced the largest microsphere sizes as compared to all other conditions tested. Optical images of DNA-MS cro sslinked with genipin displayed images of spherical particles and strands of un-crosslinked DNA further indicating that the

PAGE 93

93 genipin did not fully react with the DNA in solution during synt hesis. DNA-MS that were covalently crosslinked produced discrete particle s with spherical morphologies and are shown in Figure 3-13. Figure 3-13. Optical microscopy images of DNA-MS covalently crosslinked with A) glutaraldehyde, and B) geni pin (Magnification: 400x). Note: Red circles highlight un-crosslinked DNA strands. Scanning electron microscopy. SEM images were taken of the DNA-MS to characterize their morphology and surface topography. SEM images illustrated that the ionically crosslinked DNA-MS displayed spherical mor phologies and smooth surface topog raphies, with the exception of the iron crosslink agent condition which pr oduced an agglomerate of aggregated DNA and iron crosslinked MS, Fi gures 3-14 to 3-16. Figure 3-14. SEM images of DNA-MS prepared w ith A) chromium and B) gadolinium trivalent cationic crosslinking agents (Magnification: 2000x).

PAGE 94

94 Figure 3-15. SEM image of a DNA-MS aggregat e crosslinked with tr ivalent iron cations (Magnification: 4,000x). Figure 3-16. SEM images of ionically cross linked DNA-MS with smooth surface topographies in the nano-mesosphere size range: A) ch romium (Magnification: 9,500x) and B) gadolinium crosslinked DNAMS (Magnification: 8,500x). SEM images taken of the DNA-MS crossli nked with chromium and gadolinium also illustrated particle sizes under 10 m which visually confirmed resu lts obtained with the particle size analyzer. These SEM images also depicted particles that fell within the nano-mesosphere particle size range (i.e. d < 10 m). Upon further observation of the chromium and gadolinium crosslink agent conditions, SEM im ages illustrated that the gado linium crosslink agent condition produced discrete particles, whereas the ch romium crosslink agent condition produced aggregated particles, Figure 3-17.

PAGE 95

95 Figure 3-17. SEM images of ionically cross linked DNA-MS: A) aggregated chromium and B) discrete gadolinium crosslinke d DNA-MS (Magnification: 550x). SEM images taken of the covalently crosslinked DNA-MS produced spherical morphologies confirming observations made with the optical microscope, Figure 3-18. Figure 3-18. SEM images of DNA-MS prepared w ith A) genipin (Magnifi cation: 4,000x) and B) glutaraldehyde (Magnification: 2,00 0x) covalent crosslink agents. The SEM images also displayed variations in particle size between the genipin and glutaraldehyde crosslink agent conditions. Th ese images illustrated that the glutaraldehyde crosslink condition produced micr ospheres with larger diameters and a broader size distribution than that of the genipin, further confirming results obtained with the partic le size analyzer. Upon closer observation of the SEM images, it was also noted that the gl utaraldehyde crosslink

PAGE 96

96 condition produced DNA-MS with rougher surface topographies than the DNA-MS synthesized with the genipin crosslink condition. Energy dispersive x-ray spectroscopy. The presence of the tr ivalent cations in the DNAMS were confirmed using EDS. EDS spectra co llected on the chromium gadolinium, and iron crosslinked DNA-MS conditions confirmed the pres ence of the trivalent cations in the dried DNA-MS indicating that the cations are not being washed out by ace tone during rinsing indicating that they are indeed chemically bonding with the DNA. The EDS spectra also depicted large phosphorous peaks indicative of the phosphate groups in the DNA molecule. Silicon peaks were also seen in the spectra due to the silicon wafer used for EDS sample preparation. The collected ED S spectra for each ionic cros slink condition can be found in Figures 3-19 to 3-21. Figure 3-19. EDS spectra collected on DNA-MS crosslinked with gadolinium.

PAGE 97

97 Figure 3-20. EDS spectra collected on DNA-MS crosslinked with chromium. Figure 3-21. EDS spectra collected on DNA-MS crosslinked with iron. Evaluation of fibroblast growth DNA. The effect of DNA on the cell growth of normal human dermal BJ fibroblast cells was assessed in culture using an MTS proliferati on assay. Results from the assay indicated that

PAGE 98

98 fibroblast proliferati on decreased at the 100g and the 500g DNA conditions between hours 1 and 24, but increased between hours 24 and 72, Figure 3-22. -0.4 -0.2 0 0.2 0.4 0.6 0.8Media/Cells500g100gAbsorbance @ 490nm Day#0 Day#1 Day#3 Figure 3-22. Fibroblast proliferation profiles fo r DNA treatment conditions. Statistical analysis of these values, however, illustrated th at at Day#0 (Hour 1), the 100g DNA condition produced significantly higher prolifer ation rates than the control (p < 0.001) and the 500g condition (p = 0.021). Sta tistical analysis also illustra ted that by Day#1 (Hour 24), the 500g condition produced significantly lower prolifer ation rates than the control (p < 0.001) and the 100g condition (p = 0.003). There were no si gnificant differences between the control and the 100g or 500g conditions by Day#3 (hour 72 ); however, the 100g condition did produce significantly higher proliferation rates than the 500g condition (p = 0.017). Further analysis using optical microscopy images displayed fibrobla st cells that were ve ry similar in morphology and confluency, Figure 3-23.

PAGE 99

99 h AB C h AB C Figure 3-23. Optical microscopy images of cr ystal violet stained normal human dermal BJ fibroblast cells exposed to A) media w ith cells, B) 100g DNA, and C) 500g DNA treatment conditions (Magnification: 50x). Microsphere treatment groups. At the 100g DNA-MS condition, the gadolinium and glutaraldehyde treatment groups displayed increa sed cellular proliferatio n from hour 1 to hour 72. The chromium, iron, and DNA treatment groups illu strated a slight decrea se in proliferation between hours 1 and 24; however, they displaye d an increase in prolif eration between hours 24 and 72. The iron crosslinked DNA-MS condition was the only condition to demonstrate proliferation values lower than the media w ith cells control treatment group, Figure 3-24. -0.2 0.0 0.2 0.4 0.6 0.8 1.0Media w/CellsGdCrFeGTA DNAAbsorbance @ 490nm Day#0 Day#1 Day#3 Figure 3-24. Fibroblast pro liferation values for crosslinked DNA-MS and DNA treatment groups at the 100g condition.

PAGE 100

100 Statistical analysis of thes e values using a one way ANOVA followed by Tukey’s test of all pairwise multiple comparisons illustrated th at at 1) Day#0 (Hour 1), only the chromium and DNA conditions produced higher proliferation rates than the media with cells control treatment group (p < 0.001). On Day#1 (Hour 24), there we re no significant differences among the DNA, crosslinked DNA-MS, and media with cells cont rol treatment groups. By Day#3 (Hour 72), the gadolinium, glutaraldehyde, a nd DNA treatment group conditions all produced significantly higher proliferation rates than the media with cells control treatment group (p < 0.001). The chromium treatment group condition also produced significantly higher pro liferation rates (p = 0.016) and the iron treatment group condition produced significantly lower prol iferation rates (p < 0.001) than the media with cells control treatment group. Further analysis using optical microscopy images displayed fibroblast cells at the gadolinium, chromium, and iron treatment gr oup conditions that were very similar in morphology and confluency to the media with cells control treatment group condition, Figure 325. The gadolinium condition appeared to have the closest resemblance to the media with cells control group. The cell density for the chromium condition was si milar to the cell density found in the control condition, however, the cells appeared to be more dehydrated or stretched than those seen in the control condition. The iron cond ition exhibited similar cell morphologies to the control group as well; however, in some wells, the cell density was slightly lower than the control. The glutaraldehyde condition exhibi ted similar cellular morphology as the control condition; however, a lower concentration of cells were present in the images which indicate that the fibroblast cells were less adherent under this condition and may have experienced a slight cytotoxic response to the glutaral dehyde in the crosslinked DNA-MS.

PAGE 101

101 AB D E C AB D E C Figure 3-25. Optical microscopy images of cr ystal violet stained normal human dermal BJ fibroblast cells exposed to DNA-MS prepared with A) ga dolinium, B) chromium, C) iron, and D) glutaraldehyde crosslinking ag ents at the 100g condition and E) media with cells control group c ondition (Magnification: 50x). At the 25g DNA-MS conditions, each crossli nk agent exhibited a slight decrease in cellular proliferation from hour 1 to hour 24, how ever, displayed an increase in proliferation from hour 24 to hour 72, Figure 3-26. Cellular pr oliferation profile for each DNA-MS crosslink agent treatment group also exhibite d higher proliferation rates than the media with cells control treatment group. Statistical analysis using a one way ANOVA followed by Tukey’s test for all pairwise multiple comparisons displayed that at Day#0 (Hour 1), the gadolinium(p < 0.001), chromium (p < 0.001), and iron (p = 0.002) tr eatment groups produced significantly higher proliferation rates than the media with cells control treatment group. On Day#1 (Hour 24), statistical analysis found no si gnificant differences between the DNA-MS treatment conditions and the media with cells cont rol treatment group. By Day#3 (Hour 72), the gadolinium DNAMS treatment group was the only co ndition to display higher prolif eration rates than the media with cells control trea tment group (p < 0.001).

PAGE 102

102 -0.1 0.1 0.3 0.5 0.7 0.9 1.1Media w/CellsGdCrFeGTA Absorbance @ 490nm Day#0 Day#1 Day#3 Figure 3-26. Fibroblast prolifer ation values for the crosslinked DNA-MS treatment groups at the 25g condition. Further analysis using optical microscopy images displayed fibroblast cells at the gadolinium, chromium, and glutaraldehyde treat ment group conditions with similar cellular morphologies as the fibroblast cells present in the media with cells control treatment group, Figure 3-27. Analyses of the iron treatment groups produced images with no cells present indicating that the cells were washed off during fixation and/or staining suggesting that the fibroblast cells were less adhe rent under this condition. B ACD B ACD Figure 3-27. Optical microscopy images of cr ystal violet stained normal human dermal BJ fibroblast cells exposed to DNA-MS prepared with A) ga dolinium, B) chromium, and C) glutaraldehyde crosslinki ng agents at the 25g conditi on and D) media with cells control group condition (Magnification: 50x).

PAGE 103

103 Discussion Pilot Microsphere Synthesis Study The pilot study that was conducted to determ ine if DNA-MS could be prepared through chromium trivalent cation crosslinking was su ccessful. Chromium crosslinked DNA-MS were prepared with spherical morphologies with diamet ers that fell within th e 1m to 10m diameter range, Figure 3-5. These results were exp ected since DNA behaves as a polyanion under aqueous conditions and therefore attracts multivalent cations, such as Cr3+, under the same conditions.92 Evidence of successful chromium bonding was visually confirmed by the blue coloration of the dried synthesized DNA-MS. It is believed that the triv alent chromium cations crosslinked the DNA via electros tatic binding with the phosphate groups on the DNA backbone. This mechanism has been cited as the princi pal mechanism for multivalent metal ion binding with DNA at neutral pHs.92 Stabilizing Agent Study Chromium crosslinked DNA-MS produced fa vorable morphological and particle size results, however, upon washing and drying, th e DNA-MS became aggregated and discrete particles were not produced. It wa s therefore believed that the st eric repulsions created with the stabilizing agent used was not enough to disper se the DNA-MS during synt hesis thus causing the formation of agglomerations due to crosslinking of adjacent particles. A stabilizing agent study was conducted to determine if higher stabilizing agent concentrations c ould better disperse the DNA-MS during synthesis and re duce particle aggregation. Stabilizing agent concentrations of 3% (w/v), 5% (w/v), 10% (w/v), and 25% (w/v) were evaluated for this study. It was observed that the particle agglomera tion increased as the stabilizing agent concentration increased with the exception of the DNA-MS prepared at the 5% (w/v) CAB condition. It was also in teresting to note that the partic le diameter size decreased as

PAGE 104

104 the CAB concentration increased which is consis tent with findings obtained with protein MS.11 A resulting decrease in particle size is a result of the emulsifying stabilizing agent viscosity increasing as its concentration is increased.10 In this case, increas ing the concentration of cellulose acetate butyrate in the 1,2-dichloroeth ane increased the viscosity of the stabilizing agent thus decreasing th e resultant size of the synthesized DNA-MS. The onset of MS aggregation was first noti ced during the acetone rinsing step in the synthesis procedure where larger degrees of aggr egation were present as the CAB concentrations increased. A possible explanation for this may be at tributed to the fact that the stabilizing agent was soluble in acetone. As th e stabilizing agent c oncentration increased, the DNA-MS particle size decreased, and DNA-MS packing increased. This increased packing caused the DNA-MS to collapse on one another upon removal of the dispersi ng agent. Therefore, lower stabilizing agent concentrations allowed for lesser DNA-MS pack ing thus creating a more stable system for removal of the stabilizing agent. Based on these findings and the data obtained, it was determined the 5% (w/v) CAB was the most optimal stabilizing agen t concentration for DNAMS synthesis. Crosslinking Reaction Study The study to determine the time for covalent or ionic agents to crosslink DNA illustrated that the ionic crosslinking agents reacted qui ckest with the DNA followed by glutaraldehyde and genipin. The instantaneous r eaction between the tr ivalent cations and the DNA was to be expected since it has been well documented in the literature that DNA is known to interact electrostatically with multivalent ca tions within the millisecond time range.100 It took approximately two hours for the glutaraldehyde to crosslink the DNA, whereas it took the genipin over 72 hours to slightly crosslink the DNA molecules. Si gns of the genipin interacting with the DNA molecules after 72 hours was consistent with similar studies found in the literature

PAGE 105

105 which state that glutaraldehyde cr osslinks significantly faster than genipin and that genipin undergoes changes in color from yellow, to browni sh-red, to blue when interacting with amino groups.96, 101 The slow interaction of these two covale nt crosslinking agents may be explained by examining the theoretical crosslinking m echanism assumed to occur between DNA and glutaraldehyde or DNA a nd genipin. It was hypothesized th at the glutaraldehyde and genipin would react with the amino gr oups in the base pairs of the DNA molecule as it has been presented in the litera ture and shown in Figure 3-3 for glutaraldehyde.95, 96 Under static aqueous conditions and neutral pHs, these amino groups are unavailable for bonding because the base pairs within the DNA molecule are held t ogether through hydrogen bonding between hydrogen groups and oxygen and nitrogen groups.102 Therefore, the reactive sites on the glutaraldehyde and genipin molecules were unable to react with the amino groups found within the base pairs due to hydrogen bonding. However, it was expected for these covalent crosslinking agents to perform better during DNA-MS synthesis. Du ring synthesis, the DNA molecules would be subjected to high shear rates which would in turn unzip the DNA double helix, exposing the amino groups within the base pairs to the covalent crosslinking agents. The chromium and gadolinium crosslinking ag ents reacted instantaneously with the DNA and produced films that appeared homogenously crosslinked. The iron crosslinking agent reacted rapidly with the DNA; however, the reaction did not a ppear to be homogeneous and produced clumps of iron crosslinked DNA instead of a uniform film. Trivalent cations are known to crosslink or collapse DNA by interacting with either the phosphate backbone, the base pairs of the DNA molecule (more specifically th e N7 nitrogen on the guanine and/or adenine nucleosides), or with both the phosphate and base pairs of DNA.92, 103, 104 It has been cited that different trivalent cation complexes can have differing effects on the confirmation of the DNA

PAGE 106

106 molecule upon interaction.92 These various changes in confirmation of the DNA molecule can lead to its aggregation and collapse thus leadi ng to the appearance of non-uniform crosslinking as observed with the iron crossl inking agent condition. Based on this ideology, it may be safe to assume that gadolinium and chromium cation comp lexes used for these studies interact more favorably with the DNA molecule used than th e iron cation complex because the formation of aggregation was less evident with these two crosslinking agents. General Microsphere Synthesis Studies Particle analysis Dry particle size. DNA-MS prepared with both cova lent and ionic crosslinking agents produced particles with mean diameters less than 20m. The gadolinium and genipin crosslink DNA-MS conditions produced the most discrete partic les with the highest pe rcentage of particles in the mesosphere size range and the lowest percen tage of particles with diameters greater than 10m. Gadolinium and genipin crosslinked DNA-MS conditions appear to have performed the best and displayed more controlled and narrowe r size distributions than the DNA-MS prepared with the chromium, iron, and glutaraldehyde co nditions, Figures 3-7 and 3-8. The chromium, iron, and glutaraldehyde crossl inked DNA-MS became visibly aggregated upon washing and drying, however, it was unclear if these or other aggregates had fo rmed during synthesis. Thus the mean particle dry diameters and size distribu tions reported for these conditions better reflect particle aggregation rather than the tr ue diameter of the individual DNA-MS. The formation of aggregates during synthesi s with the chromium and iron crosslinking agents may have arisen due to the different in teractions each cationic complex had with the DNA molecule.92 Most trivalent cations are known to collapse or aggregate DNA into various condensed structures, such as spheres or toroids.92 The type of structure that DNA will condense into is dependent on the classification of the in teracting metal ion. Metal ions fall into two

PAGE 107

107 different classifications: Class A or “hard” metal ions which prefer interacti ons with the oxygen donor atoms found on the phosphate groups of th e DNA backbone, and the Class B or “soft” metal ions which prefer binding with the nitroge n donor atoms found within the base pairs of the DNA molecule.92, 103, 105 Chromium and iron fall between cl asses A and B and are considered borderline ions with both hard and soft metal ion properties.92 Metal ions with these characteristics can form bonds with both the phosphate oxygens and N7 nitrogen atoms found within the guanine and adenine base pair groups of DNA, Figure 3-28. N N N N N O H O H H H CH2 H P O-O O N N N O N N O H H H H CH2 H O P O-O O Adenine GuanineH H O H H H N O O N O H O H H H CH2 H P O N N O N O H O H H H CH2 H P O Thymine CytosineH O H H OO O -O M M P O OP OO Figure 3-28. Schematic drawing of possible inte ractions between chromium and iron trivalent cations and phosphate oxygens and base pair guanine and adenine N7 nitrogen atoms. (Note: M = Cr3+ or Fe3+.)

PAGE 108

108 The ability of the metal ions to bind at both oxygen and nitrogen sites can cause a destabilizing effect on the DNA mol ecule and lead to more aggregat ed structures, such as toroids instead of spheres.92 These aggregated structures arise more frequently when there is joint bonding between phosphate and base pair groups. Th e trivalent cationic natu re of chromium and iron allow them to interact instantaneously with the phosphate oxygen an d since the nitrogen atoms found within the base pairs interact the strongest and more frequently with d orbital transition metals such as chromium and iron, it al lows them to interact with the base pairs as well.105 The ability for chromium and iron to bind with both oxygen and nitrogen groups explains the broad particle diameter ranges and multimodal size distributions produced with these two crosslinking agents. Gadolinium on the ot her hand is a class A hard metal lanthanide cation which predominantly binds ionically to the oxygen atoms found on the phosphate groups of the DNA backbone and will neve r thermodynamically bind to th e nitrogen atoms found within the DNA base pairs.92, 103, 105 It has also been noted th at even though la nthanides (i.e. gadolinium) are known to favorably bind with ph osphate oxygen groups, th ey are also known to prevent complementary interactions between sing le DNA strands and in this case interactions between neighboring DNA-MS.92 For this reason, gadoliniu m crosslinked DNA-MS produced discrete particles with more norma lized particle size distributions. The amount of aggregate formation seen fo r the chromium and iron crosslink agent conditions during the washing and drying steps of the DNA-MS synthesis procedure may be attributed to further ionic inte ractions between un-reacted cationi c crosslinking agents and free phosphate group sites. The ionic ra dii of the trivalent cations used in this study led to either electrostatic interactions (as in the case of the chromium and iron crosslinking agents) or electrostatic repulsions (as in the case of the gadolinium crosslinking agent) during the washing

PAGE 109

109 and drying steps of the DNA-MS synt hesis process. Electrostatic interactions led to particle aggregation, whereas electrostatic repulsions led to particle discrete ness. The extent of particle aggregation or discreteness may th en be explained by the amount of electrostatic interaction or repulsion each cationic crosslinking agent presen ts which can be measured by their respective ionic radii. The ionic radius for the trivalent cations are 0.63 angstroms for chromium, 0.64 angstroms for iron, and 0.938 angstroms for gadolinium.106 The ionic radii for the chromium and iron were similar in size and smaller than th at of gadolinium. This may explain why the chromium and iron crosslinking agent conditions produced more aggregates than the gadolinium crosslink agent condition. Since the chromium and iron cations had smaller ionic radii, the smaller radii allowed the DNA-MS to be closer and interact more with one another. The size of the gadolinium trivalent cation may have been la rge enough to generate sufficient electrostatic repulsion among the DNA-MS to produce a discrete particle distribution.90 The broad size distributions and the slightly larger particle diameters obtained using the glutaraldehyde crosslinking agent may be attribut ed to the amount of time needed to obtain an interaction between glutaraldehyde and the ami no groups in DNA. It has been cited in the literature that glutaraldehyde in teracts more efficiently with D NA at higher temperatures because the hydrogen bonds holding the DNA tertiary structure together are broken.96 The same case can be made for the ability of glutaraldehyde to in teract with DNA when it is exposed to high shear rates during DNA-MS synthesis. The shear rates are large enough to disrupt hydrogen bonding between DNA base pairs thus exposing the amino groups for glutaraldehyde interaction. Since the interaction between glutaralde hyde and the amino groups within the base pairs is a lot slower than that seen between cationic crosslinking agen ts and the phosphate oxy gen groups, it is to be expected to obtain particles that are larger and less homogeneous in size than those as seen with

PAGE 110

110 the cationic crosslinking agents. The mu ltimodal size distributions obtained with the glutaraldehyde suggest that it may be an unfavorable crosslinking agent for DNA-MS. Hydrated particle size. The stability of dispersions can be best described using two major colloidal dispersion systems: 1) a lyocratic syst em which requires solvation at the surface for dispersion stability and 2) an electrocratic syst em which requires electros tatic repulsion between particles for dispersion stability.90 During synthesis, a DNA-MS emulsion is created via a lyocratic system in which cellulose acetate butyr ate dissolved in 1,2-dichloroethane completely wets each individual DNA-MS and ster ically stabilizes it in that form. During hydrated particle size analysis, the system changes from a lyocra tic system to an electrocratic system. The stability of the dispersion is then dependent on the electrostatic repulsi on between the particles instead of the steric repulsions. In an electrocratic system, it is noted in the literature that the addition of a small amount of salt to the disper sion will cause the particles in the system to coagulate or aggregate if there is insufficient electrostatic repulsions between the particles.90 DNA-MS prepared with glutaral dehyde produced the most mu ltimodal hydrated particle size distributions indicating the there were insufficient electrostatic repulsions between the particles to prevent aggregate formation. Surface charge and dispersability The zeta potential is the measur ement of the electric potentia l at the plane of shear of a particle and is used to determine the ne t interparticle forces in a dispersion.90 The lifetime of a stable dispersion can therefore by estimated by m easuring the zeta potential of the particles in dispersion. Particles with larg e zeta potential values have la rge magnitudes for interparticle repulsion and thus have longer and more stable dispersion lifetimes.90 Dispersability experiments conducted in Grey’s BSS indica ted that the gadolinium crosslinked DNA-MS displayed the most stable dispersability w ith dispersion times exceeding 48 hours. The

PAGE 111

111 chromium and glutaraldehyde crosslink agent co nditions both displayed dispersion times of 12 hours or less. The dispersion st ability of the gadolinium cross linked DNA-MS may be attributed to its zeta potential values which were much larg er than the those obtained for the chromium and glutaraldehyde crosslink agent conditions (-45.3mV for gadolinium as compared to -38.0mV for chromium and -29.6mV for glutaraldehyde). At these larger zeta potential values, the gadolinium crosslinked DNA-MS condition may have produced larg e enough interparticle repulsions to create a stable dispersion.90, 107 Microscopy Optical and scanning electron microscopy. Images acquired using optical microscopy and SEM further confirmed crosslinking reactio n observations and results obtained during dry particle size analysis. As seen in the Cross linking Reaction Study, DNA-MS that were prepared with gadolinium produced small, discrete, sp herical particles with smooth topographies suggesting homogeneous crosslinking of the gadolinium, Figures 3-14, 3-16, and 3-17. DNAMS prepared with chromium produ ced particles that were mostly spherical, however, there were a few particles that were aggregat ed and irregularly shaped as sim ilarly seen in the iron crosslink agent condition, Figures 3-15 and 3-17. The a ggregations seen in the chromium and iron conditions may be attributed to the instantaneous collapse or condensing of the DNA molecule upon interaction with these different cationic complexes since it is well stated in the literature that most trivalent cationic complexes collapse DNA into condensed toroidal or non-spherical confirmations.92 Agglomerations observed in the chro mium and iron conditions may also be a result of multi-site crosslinki ng since the chromium and iron comp lexes are capable of binding or crosslinking with the nitroge n at the N-7 coordination site in the guanine nucleobase.103, 105, 108, 109 The gadolinium chloride cationic complex, however, was only capable of binding with

PAGE 112

112 phosphate oxygen groups found along the DNA b ackbone and thus generated spherical particles.92, 103, 105 Optical and scanning electron microscopy imag es also illustrated that glutaraldehyde crosslinked DNA-MS produced larger particle sizes than the ioni cally crosslinked conditions. The larger size of the glutar aldehyde DNA-MS may be attribut ed to the time required for glutaraldehyde to bond with the amino groups in DNA. Since litera ture and experiments conducted have shown that glut araldehyde takes much longer to interact with DNA than the cationic crosslinking agents, it may be safe to assu me that the glutaraldeh yde is not interacting with the DNA base pairs until the latter part of the synthesis procedure which is conducted at 600rpm rather than 1250rpm. At these lower stir rates, the DNA particles are capable of existing at a larger size and thus are crosslinked at a larg er size. When using cationic crosslinking agents, the interaction with DNA is inst antaneous. This quicker inter action with DNA leads to smaller particle sizes since the cationi c agents are able to crosslink the DNA-MS immediately after mixing at 1250rpm. Energy dispersive x-ray spectroscopy. Elemental analysis via EDS utilizes inelastic electron scattering to cause inner shell excitations that result in emissions of characteristic x-rays of an element.110 The energies obtained for each x-ray si gnal given is specifi c to each individual element.110 EDS analysis of DNA-MS confirmed th e presence of the cationic crosslinking agents strongly suggesting that the trivalent ca tions are indeed chemically bonding with the DNA molecule, Figures 3-19 to 3-21. The emitted x-ray energies obtained from the EDS spectra were compared against x-ray wavelength energies fo r each element examined. Energy values for characteristic x-rays were confirmed for each el ement analyzed and are summarized in Table A-1 in Appendix A.

PAGE 113

113 Evaluation of fibroblast growth Overview. The effects of DNA and crosslinke d DNA-MS normal human dermal BJ fibroblast cell growth in culture was measured using a colorimetric MTS cell proliferation assay. In order for a cell to prolifer ate or reproduce, it must go th rough the life cycle of a cell.38-42 The life cycle of a cell involves four phases including the Gap1 or “G1” phase, the DNA Synthesis or “S” phase, the Gap2 or “G2” phase, and the Mitosis or “M” phase, Figure 3-29.41-43 S G 1 S G 1 S G 2 S G 2 M G 2 M G 2 M G 1 M G 1 G1 S G2 M S G 1 S G 1 S G 2 S G 2 M G 2 M G 2 M G 1 M G 1 G1 S G2 M Figure 3-29. A drawing of the 4 phases of the cell life cycle. The proliferation of a cell involves all four pha ses of the cell life cycl e. The cell will grow while in the G1 phase. Then during the S phase, the DNA fr om the parent cell is synthesized and duplicated and the cell moves on to the G2 phase where the existing ce ll now carries two sets of DNA and matching chromosomes. Finally in the M phase, the duplicated chromosomes separate and the cell divides or reproduces to create two cells with matching DNA. 38-43 A cell population will continue to undergo cellular proliferat ion through the cell cycle as long as their environment is untouched and ha s all the necessary nutrients for growth and reproduction. The addition of foreign agents or materials to the media sustaining a cell population can cause enough of an imbalance to al ter the amount of time necessary to complete

PAGE 114

114 the cell cycle. Altering the environment of a cell population can lead so me cells to enter the Gap0 or “G0” phase as shown in Figure 3-30, where the cel l enters a resting or quiescent state in which it ceases to proliferate, but retains its cap acity to re-initiate its progress through the cell cycle at a later time.41 S G 1 S G 1 S G 2 S G 2 M G 2 M G 2 M G 1 M G 1 G1 S G2 M G0 S G 1 S G 1 S G 2 S G 2 M G 2 M G 2 M G 1 M G 1 G1 S G2 M G0 Figure 3-30. A drawing of the 4 phases of the cell life cycle, including the G0 quiescence phase. This phase is often seen in the culturing of fibroblast cells which can enter the G0 phase upon alteration of their culture media.39 The environmental altera tion of a cell population can also cause a change in the G1 phase duration leading to a l onger or shorter cell cycle.41 However, if the cell decides to not re-ent er the cell cycle, the cells will round up and break their surface attachments within their cu lture environment and undergo apoptosis or cell death.39 A cell responding to an environmental change in this fashion would indicate that the new environment has become cytotoxic.

PAGE 115

115 DNA. Results obtained from the cell prolifer ation studies conducted on DNA displayed a slight decrease in pr oliferation betwee n the first two time points, hour 1 and hour 24, Figure 322. The control group consisting of the cells with media alone did not display this same trend and instead exhibited a significant increase in pr oliferation between the first two and last two time points. The decrease in pro liferation noted within the treatment groups may be attributed to the cells entering a short G0 phase or a lengthened G1 phase due to the temporary destabilization of their environment (i.e. addition of the DNA treatment). However, by the third time point, hour 72, proliferation had incr eased significantly and the tr eatment groups displayed no statistical differences in proliferation from the control group indicating that the DNA does not elicit an anti-prolif erative response or a cytotoxic response in vitro Optical images of each treatment group illustrated similar cellular dens ities and morphologies to the control group further confirming this conclusion. Microsphere treatment groups. At the 100g condition, the gadolinium and glutaraldehyde conditions contin ued to increase in prolifera tion through all three time points similar to the media with cells control gr oup. On the other hand, the chromium and iron treatment conditions demonstrated similar trends as seen above for the DNA conditions in which proliferation decreased between the first two time points, hour 1 and hour 24, and then significantly increased between the last two ti me points, hour 24 and hour 72, Figure 3-24. As mentioned before, the observed decrease in fibr oblast proliferation may be attributed to the disturbance of the culture media upon adding treatm ent leading to a change in the cell growth cycle.38-41 The same trends were seen in the 25g treatment groups for the gadolinium, chromium, and iron conditions.

PAGE 116

116 The glutaraldehyde treatment group was the only condition to exhibit the trend of continued cellular proliferation through all three time points at both 25g and 100g conditions; however, upon further observation of the glutar aldehyde condition at the 100g concentration, optical images displayed wells with lower cell dens ities than those seen in the media with cells control condition indicating that the fibroblasts cells were less adherent. These observations suggested that the glutaraldehyde elicited a cytotoxic response at the 100g concentration and not at the 25g concentration.39 The cytotoxic response of the glutaraldehyde observed at the 100g concentration was expected since glutar aldehyde has been shown to decrease normal human dermal fibroblast prolifera tion and elicit cytotoxic responses in vitro at concentrations as low as 1ppm.95 At both treatment concentr ations, the iron c ondition produced significantly lower proliferation rates than the medi a with cells control group and all other conditions indicating that the iron concentration in the DNAMS may be enough to elicit a an anti-proliferative response in vitro Optical micrographs taken of the iron trea tment conditions tend to support this conclusion illustrating that there was a slightly lower cell de nsity at the 100g condition than the media with cells control condition and that at the 25g cond ition there were no cells present in the images indicating a lack of adherence of the cells to the 96-well plate. The lack of cellular adherence at the 25g condition indicates that the iron conc entration in the DNA-MS was enough to elicit a cytotoxic response.39 A cytotoxic response in vitro was expected since it has been noted in the literature that iron (III) (Fe3+) is known to be highly cytotoxi c and cause massive cell death at concentrations greater than 10M.111 In vivo Fe3+ would be expected to be even more cytotoxic since Fe3+ is known catalyze the Haber-Weiss reacti on and create highly reactive and toxic hydroxyl radicals through Fenton Ch emistry, Equations 3-3 and 3-4. 112, 113

PAGE 117

117 Fe3+ + O2 •Fe2+ + O2 (3-3) Fe2+ + H2O2 Fe3+ + OH+ OH• (3-4) Reactive hydroxyl radicals are know to cau se oxidative damage under physiological conditions and are the main cause of several ne urodegenerative diseases such as Alzheimer’s Disease and Parkinson’s Disease.112-115 Overall the chromium and gadolinium conditions at both the 25g and 100g concentrations produced not only proliferation rate s that either exceeded or were comparable to the media with cells control condition, but also comparable cell densities and morphologies to the control. The positive result s obtained in this study were furt her supported by literature that indicates that certain complexe s of chromium (III) are known to be non-cytotoxic and that many gadolinium (III) complexes are used at low concentrations with minimal cytotoxic responses in vivo for magnetic resonance imaging of tumor phys iology of various cancers such as breast and prostate cancer.116-119 Conclusions The objectives of these studies were to s ynthesize DNA-MS in good yield with covalent (i.e. glutaraldehyde or ge nipin) or ionic (i.e. chromium, gado linium, or iron) crosslinking and to produce particles with a target dr y mean diameter range of 50nm to 20m, where at least 60% of all particles prepared were within the mesosphe re size range of 1m to 10m and < 5% of all particles were greater than 10m in size. Hydrated particle diamet ers were to be less than 25m. In addition, DNA-MS were sought to obtain aqueous dispersion stability of over 24 hours and elicit minimal toxic effect on fibroblast cells in culture. Pilot studies were conducted to determine the optimum processing conditions to achieve these goals. Pilot studies focused on determining an optimal stabilizing agent concentration for DNA-MS synthesis and on verifying the amount of time required to establish DNA crosslinking

PAGE 118

118 with the covalent and ionic crosslink agents. General DNA-MS synthesis studies were then conducted and resulting particle diameters, si ze distributions, morphol ogies, topographies, and surface charges were evaluated and compared. In addition, normal human dermal BJ fibroblast cells were used in culture to evaluate the e ffect of DNA and crosslinked DNA-MS on cell growth at concentrations of 100g and 500g for DNA, and 25g and 100g for crosslinked DNA-MS. A colorimetric MTS assay was used to measure fibroblast growth at the 0, 24, and 72 hour time points. Particle Analysis DNA-MS were successfully synthesized with io nic and covalent crosslinking agents with mean dry diameters of less than 20m. Pilot studies found that a conc entration of 5% (w/v) CAB produced the most optimal conditions for DNA-MS synthesis. Crosslinking reactions studies found that ionic crossl inking agents displayed an in stantaneous reaction with DNA, whereas it took glutaraldehyde 2 hours to react with DNA and genipin 72 hours. EDS analysis of DNA-MS confirmed the presence of the cationi c crosslinking agents st rongly suggesting that the trivalent cations are indeed chem ically bonding with the DNA molecule. The gadolinium crosslinked DNA-MS produced the smallest particles (2.6m 2.8m), with the most narrow size distri butions and the largest percentage of particles that fell within the mesosphere size range (i.e. 1m to 10m) of all the crosslinking agents tested. The gadolinium crosslinked DNA-MS also produced excellent median hydrated partic le size values of 12.1m. The multimodal dry particle size distributions obtained with ch romium, iron, and glutaraldehyde suggest that they may not be optimal cr osslinking agents for DNA-MS; however, the glutaraldehyde condition did come close to ma tching the goals of the initial particle size objectives and future studies should focus on opt imizing its processing conditions to obtain

PAGE 119

119 DNA-MS with at least 60% of all particles with diameters in the mesosphere size range and less than 5% of all particles with diameters greater than 10m. The genipin crosslinked DNA-MS produced excelle nt dry particle sizes and distributions; however, they were unstable in PBS and immediat ely dissolved. Their instability in PBS was indicative of the inefficient crosslinking between genipin and DNA. The crosslink reaction time used to prepare genipin DNA-MS was insufficient and future studies will have to be conducted with longer reaction times to s ee if genipin can be used as a crosslinking agent for DNA-MS. Dispersability experiments conducted in Gr ey’s BSS indicated that the gadolinium crosslinked DNA-MS displayed th e most stable dispersability with dispersion times exceeding 48 hours. The dispersion stability of the gadol inium crosslinked DNA-MS were attributed to their zeta potential values which were much la rger than those obtained for the chromium and glutaraldehyde crosslinked DNA-MS. In Vitro Human Dermal Fibroblast Growth Human dermal fibroblast proliferation data and optical microscopy images obtained suggest that DNA derived from he rring testes does not elicit an anti -proliferative response or a cytotoxic response in vitro Iron crosslinked DNA-MS produced significantly lower proliferation rates than the medi a with cells control group indicat ing that the iron concentration in the DNA-MS may be enough to elic it an anti-proliferative response in vitro The glutaraldehyde crosslinked DNA-MS also elicited a negative re sponse at the 100g concentration and not at the 25g concentration. DNA-MS crosslinked with chromium and gadolinium produced the best in vitro results with proliferation rates that either exceeded or were comparable to the media with cells control condition at bot h the 25g and 100g concentrations. Since the chromium condition did not produce a toxic response on the fibroblast cells in culture, future studies should focus on optimizing the synthesi s parameters for chromium crosslinked DNA-MS

PAGE 120

120 in order to obtain at least 60% of all particles with diameters in the mesosphere size range and less than 5% of all particles with diameters greater than 10m.

PAGE 121

121 CHAPTER 4 OPTIMIZATION OF DNA NANO -MESOSPHERE SYNTHESIS Introduction This chapter presents studies conducted on th e optimization of processing parameters for DNA nano-mesospheres (DNA-MS) synthesis. DNAMS synthesis parameters were optimized to produce controlled size distribu tions where at least 60% of all particles prepared fell within the mesosphere size range of 1m to 10m and < 5% of all particles were greater than 10m in size. Particles less than 1m in diameter were also acceptable and DNA-MS were sought to obtain hydrated particle diam eters of less than 25m. DNA-MS synthesis parameters were optimi zed with a filtration study to eliminate aggregates from the yield and narrow particle size distributions. A filtering step was added at the end of the synthesis process using a nylon filter w ith a pore size of 20m. The resulting particle diameters and size distributions were then evaluated and compared to non-filtered controls. DNA-MS processing parameters were further op timized by analyzing the effects of mixer speed and crosslink concentration on partic le diameter, swelling, morphology, and size distribution. Mixer speeds of 950rpm, 1250rpm, and 1550rpm with ionic crosslink concentrations of 20%, 50%, and 120% molar equivalence (MEQ) were assessed for these studies. The dry and swollen particle size s of the DNA-MS were quantitativ ely characterized using an LS Coulter 13 320 particle size analyzer. The morphology and topography of the DNA-MS were analyzed using optical microscopy and scanning el ectron microscopy. The presence of trivalent cations was assessed using energy dispersive xray spectroscopy and crosslinking was confirmed through stability stud ies in 0.05M phosphate buffere d saline at a pH of 7.4.

PAGE 122

122 Materials and Methods Materials DNA sodium salt derived from herring testes Ty pe XIV, cellulose ac etate butyrate, HPLC grade 1,2-dichloroethane, methanol, chromium (III) potassium sulfate dodecahydrate, gadolinium (III) chloride hexahydr ate, iron (III) nitrate nonahydr ate, and 25% (w/w) Grade II aqueous glutaraldehyde solution were purchased from the Sigma-Aldrich Company. Sodium phosphate monobasic monohydrate, sodium phosphate dibasic anhydrous, and sodium chloride, each A.C.S. certified, were purchased from Fisher Scientific International. Acetone, 20m and 70m Spectra/Mesh Nylon filters, and 50mL a nd 15mL polypropylene centrifuge tubes were also obtained from Fisher Scientif ic International. Type I and Type II deionized ultrapure water were prepared with a resi stivity of at least 16 M -cm-1 using the Barnstead NANOpure Ultrapure Water System in the lab and is termed ultrapure water throughout. Methods Solution preparation Deoxyribonucleic acid. Aqueous 5% (w/v) solutions of DNA were prepared at room temperature by adding 0.5g of DNA to 5mL of ultrapure water in a 50mL polypropylene centrifuge tube. The solution was mixed on a vortex for 30 seconds. Three milliliters of ultrapure water were then added to the DNA and placed on the rotary shak er for at least two hours until the DNA had completely dissolved. Once the DNA had fully dissolved, the volume was brought up to 10mL and vortexed for 30 seconds The DNA solution was then placed in the refrigerator over night to ensu re the complete collapse of bubbles generated during vortex and rotary mixing. The percent solid concentration of the DNA solution was quantified using a Metler LJ16 Moisture Analyzer at 130C for 60 minutes. The concentration of the DNA solution was then adjusted until the concentration was with in 10% of the desired concentration. Once the

PAGE 123

123 desired concentration wa s reached, the aqueous DNA solution wa s placed in the refrigerator until further use. Cellulose acetate butyrate. Solutions of cellulose acetate butyrate in 1,2-dichloroethane (CAB) were used as the water-immiscible c ontinuous phase for the emulsion stabilization process during DNA-MS synthesis. The CAB solu tions were used at a concentration of 5% (w/v) and prepared by adding 25g of cellulose ace tate butyrate to 500mL of 1,2-dichloroethane. The CAB solution was mixed at room temperat ure on a magnetic stir plate on high until the cellulose acetate butyrate had completely dissolved in the 1,2-dichloroethane. The resulting CAB solution was capped, parafilmed, and stored at room temperature. Glutaraldehyde. Aqueous glutaraldehyde solutions we re prepared to a concentration of 4% (w/v) by diluting the 25% (w /w) aqueous glutaraldehyde solu tion purchased from the SigmaAldrich Company. The aqueous glutaralde hyde solution was diluted by adding 21mL of ultrapure water to 4mL of the 25% (w/w) solutio n. The glutaraldehyde so lutions were prepared in 30mL glass jars and mixed on a vortex at room temperature for 1 minute. After mixing, the aqueous glutaraldehyde solutions were paraf ilmed and stored in the refrigerator. Chromium (III) and gadolinium (III). Aqueous solutions of chromium (III) potassium sulfate and gadolinium (III) chloride were prep ared to a concentration of 0.1M by adding 24.97g of chromium (III) potassium sulfate dodeca hydrate or 18.59g of gadolinium (III) chloride hexahydrate to 500mL of ultrapure water. Th e chromium (III) and gado linium (III) solutions were mixed on a magnetic stir plate over night at room temperature until all the chromium or gadolinium had dissolved in the water. Afte r complete mixing, the 0.1M ionic crosslinking agent solutions were parafilmed a nd stored at room temperature.

PAGE 124

124 Phosphate buffered saline. Four liters of a 0.05M phosphate buffered saline (PBS) solution with a pH of 7.4 was prepared in the lab for measuring the sw elling properties of the DNAMS. The PBS solution was prepared by mixing 2.9L of a 0.05M sodium phosphate dibasic solution to 1L of a 0.05M sodium phosphat e monobasic solution. The pH of the resulting solution was measured and the sodium phosphate monobasic solution was added until the target pH of 7.4 was reached. A PBS solution with a concentration of 0.01M at a pH of 7.4 was used for zeta potential measurements and was prepared by diluting a 0.1 M PBS solution and adjusting the pH back to 7.4. The 0.1M PBS solution was prepared by mi xing 2.9L of a 0.1M sodium phosphate dibasic solution with 1L of a sodium phosphate monobasic solution. The two solutions were mixed and the pH of the resulting solution was brought to 7.4 by adding monobasic so lution. The prepared PBS solutions were left out at room temperature until needed. DNA nano-mesosphere synthesis Filtration study procedure. DNA-MS were prepared using an emulsion stabilization technique that sterically stabilizes the DNA molecu le into spherical conformations and crosslinks them while in suspension. This emulsion stab ilization process involved dispersing 3mL of an aqueous DNA solution (i.e. the aqueous phase) into 47mL of a CAB solution (i.e. the continuous organic phase) in a 300mL Labconco lyophiliza tion flask. A DNA microemulsion was then created by vigorously mixing the two solutions at 1250rpm for 20 minutes at room temperature using a paddle mixer with a two inch, two blade propeller. The DNA microemulsion was covalently or ionically crossli nked while in suspension by reducing the speed of the paddle mixer to 600rpm and adding 2mL of the crosslinking agent. The DNA microemulsion then underwent crosslinking for 1 hour and 40 minutes at wh ich time 50mL of acet one was added and any further reactions were allowed to reach comp letion for another hour. After synthesis was

PAGE 125

125 complete, the DNA-MS underwent f our rinses in acetone to rem ove any residual organic phase or crosslinking agent. The DNA-MS were rinsed by separating the resultant DNA-MS suspension into four separate 50mL polypropylene cen trifuge tubes. To these tubes, acetone was added up to 40mL and the tubes were cappe d and vortexed on high for 30 seconds. The DNAMS were then collected by centrifuging th e tubes at 2600rpm for 10 minutes. After centrifugation, the acetone was decanted and fresh acetone was added again up to 40mL. The acetone rinse was repeated once more as mentione d above and then twice more by consolidating the contents of 4 tubes to 2 tube s and then 2 tubes to one tube. After the final acetone rinse, the DNA-MS were resuspended in 30mL of acetone and vortexed on high for 30 seconds. The resuspended DNA-MS were then fi ltered using a stainless steel vacuum filtration device with a 20m Spectra/Mesh Nylon filter. The device was then rinsed with 10mL of acetone to further filter any remaining DNA-MS under 20m. After the final filter rinse, th e centrifuge tube was capped, vortexed on high for 30 seconds, and centri fuged at 2600rpm for 5 mi nutes to collect the microspheres. The acetone was then decanted and a Kimwipe was secured over the mouth of tube using a rubber band in order to allow the DNA-MS to dry overnight at room temperature. Mixer speed and crosslink concentration study procedure. DNA-MS were prepared using the same emulsion stabilization techni que described above fo r the Filtration Study, however, mixer speeds and crosslink agent concentra tions were varied in order to determine the effects of these factors on the diameters of the resulting DNA-MS and their respective size distributions. This emulsion stabilization pr ocess involved dispersing 3mL of the aqueous DNA solution (i.e. the aqueous phase) into 47mL of the CAB soluti on (i.e. the continuous organic phase) in a 300mL Labconco lyophilization flask. A DNA microemulsion was then created by vigorously mixing the two solutions at 950rpm 1250rpm, or 1550rpm for 20 minutes at room

PAGE 126

126 temperature, using a paddle mixer with a two inch, two blade propeller. The DNA microemulsion was then ionically crosslinked with trivalent gadolinium cations while in suspension by reducing the speed of the paddle mixer to 600rpm and adding 0.3314mL, 0.8285mL, or 2mL of the 0.1M aqueous ga dolinium solution to obtain the 20%MEQ, 50%MEQ, or 120%MEQ crosslink concentration conditions. The DNA microemulsion then underwent crosslinking for 1 hour and 40 minutes at wh ich time 50mL of acet one was added and any further reactions were allowed to reach comp letion for another hour. After synthesis was complete, the DNA-MS underwent f our rinses in acetone to rem ove any residual organic phase or crosslinking agent. The DNA-MS were ri nsed using the same washing procedure as mentioned above for the Filtration Study, however a 70m Spectra/Mesh Nylon filter was used in place of the 20m filter in order to not comple tely eliminate the possible size effects of the varying mixer speeds and gadolinium crosslink con centrations. The 70m filter was also used to minimize the number of aggregates in the resulting yield. Mesosphere characterization Yield analysis. The yield of each condition synthesi zed was calculated and expressed as a percent yield value. The percent yield was cal culated by dividing the final weight of the DNAMS by the amount of weight used to synthesize the DNA-MS. The equation for the percent yield is expressed in Equation 4-1 where WF is the final weight of the DNA-MS, VDNA, DNA, and CDNA are the volume, density, and concentration of the aqueous DNA solution used respectively, and WX is the weight of the crosslinking agent added during synthesis. 100 ) ) W ) C ((V W ( Yield %X DNA DNA DNA F (4-1) Dry particle size analysis. The particle size distributi ons and diameters of the DNA-MS were obtained under dry conditions using a Coulter LS 13 320 part icle size analyzer. DNA-MS

PAGE 127

127 were sonicated for 10 to 30 seconds prior to analysis to aerate the particles and then approximately 2mg of the DNA-MS were suspende d in 2mL of methanol. The suspension was then sonicated for 30 seconds to break up any aggregates and tested. The Coulter LS 13 320 particle size analyzer was set to run at a pump speed of 73% using a protein/DNA particle diffraction model. Standards were tested in methanol before th e first run to ensure that the instrument was performing adequate ly. Each condition was tested th ree times in which each test consisted of two runs. This method of testing produced six independent and size distributions. Data collected from these experiments were sta tistically analyzed using SigmaStat 3.0 software. Hydrated particle size analysis. The mean swollen diameters and particle size distributions of the DNA-MS were obtained usin g a Coulter LS 13 320 particle size analyzer. Approximately 2mg of the DNA-MS were suspende d in 2mL of 0.05M PBS with a pH of 7.4. The suspension was then sonicated for 30 sec onds to break up any DNA-MS aggregates. The DNA-MS were then allowed to swell in the PB S for an additional two minutes and thirty seconds. After swelling, the DNA-MS were te sted in the Coulter LS 13 320 particle size analyzer using a pump speed of 73% and a prot ein/DNA particle diffracti on model. Standards were tested in PBS before the first run of the first batch to ensure that the instrument was performing adequately. Each condition was tested three times in which each test consisted of two runs. This method of testing produced six independent partic le diameters and size distributions. After obtaining the mean particle swollen diam eters, the data was used to calculate the percent change in size us ing Equation 4-2, where DD is the dry diameter and DH is the swollen or hydrated diameter. A negative perc ent change in size depicted a d ecrease in particle size (i.e. degradation) whereas a positive percent change in size depicted an increase in particle size (i.e.

PAGE 128

128 swelling). Data collected from these experiment s were statistically analyzed using SigmaStat 3.0 software. 100 ) D D D ( Size in Change %D D H (4-2) Surface charge analysis. The surface charge of DNA-MS prepared in the Mixer Speed and Crosslink Concentration Study was measured to determine the effects of crosslink concentration on surface charge The surface charge of the DNAMS was obtained using a Brookhaven ZetaPlus zeta potential analyzer with ZetaPALS software. Approximately 2mg of the DNA-MS were suspended in 1.5mL of 0.01M PB S solution with a pH of 7.4. Each condition was sampled three times in which each sample underwent ten runs. This method of testing produced thirty independent zeta potential values. The data collected from the zeta potential analyzer was statistically analy zed using SigmaStat 3.0 software. Scanning electron microscopy. The morphology and surface topography of the DNAMS were observed using scanning electron micr oscopy (SEM). Approximately 1mg of dry DNA-MS were mounted onto a small piece of sili con wafer which in turn was mounted onto an aluminum SEM stub using double sided tape. The DNA-MS were then coated with goldpalladium for 2 minutes using a T echnix Hummer V sputter coater. Images were taken either on a JEOL 6400 SEM using an accelerating voltage of 5KeV, condenser lens setting of 10, objective lens setting of 117, and a working distance of 15mm, or on a JEOL 6335F Field Emission SEM at an accelerating voltage of 5KeV and a working distance of 15mm. Energy dispersive x-ray spectroscopy. The presence of trival ent cations in the DNA-MS after washing and drying was observed using en ergy dispersive x-ray spectroscopy (EDS). DNA-MS were mounted onto a piece of silicon wafer. The silicon wafer was then secured to an aluminum SEM stub using carbon double sided tape. The DNA-MS were then coated with

PAGE 129

129 carbon for 2 minutes using a Technix Hummer V sputter coater. EDS spectra on the DNA-MS were collected using a JEOL 6400 SEM at an accelerating voltage of 15KeV and working distance of 15mm. A dead time of 20% to 40% was allowed for each condition tested. Results & Discussion Filtration Study Synopsis A filtration study was conducted to optimize DNA-MS synthesis parameters, eliminate aggregates from the yield, and narrow particle size distributions. A filtering step was added at the end of the synthesis process using a nylon filt er with a pore size of 20m. Since a particle size of less than 25m is desired for optimal intr atumoral delivery, a Spectra/Mesh Nylon filter with a 20m pore size was used for this study.69, 120 Particle analysis Percent yield. The filtration step used in this stu dy successfully tailored the particle size of the DNA-MS that were prepared with both co valent and ionic crossl inking agents by reducing the yields in most cases over 50%. The ga dolinium crosslinked DNA-MS produced the best particle yields after filtration with a yield of 67%. However, the chromium and glutaraldehyde crosslinking agent conditions continued to produce aggregat es upon washing and drying and their yields were reduced after filtration to 24% from 77% for chromium and to 23% from 61% for glutaraldehyde. The percent yield values as calculated by Equation 4-1 for each condition with their respective decreases due to filtration are tabulated in Table 4-1. Table 4-1. The yields and percen t decrease in yield values fo r DNA-MS synthesized with ionic and covalent crosslinking agents. Crosslinking agent Percent yield after filtration (%) Percent yield before filtration (%) Percent decrease in yield due to filtration (%) Chromium 24 77 69 Gadolinium 67 82 18 Glutaraldehyde 23 61 62

PAGE 130

130 Dry particle size analysis. The dry particle sizes of the DNA-MS were obtained in methanol. Filtered DNA-MS prepared with both ionic and covalent crosslinking agents produced particles with mean diameters less th an 10m in size consistent with observations presented in Chapter 3. The filtered gado linium and glutaraldehyde crosslinked DNA-MS conditions performed similarly to their non-filtere d counter parts, producing dry mean particle diameters of 3.3m 2.8m and 5.5m 6.5m, respectively with the same percentage of particles falling within the meso sphere size range (i.e. 1m to 10 m) and particles greater than 10m. The gadolinium condition s till produced the most controlled size distributions of the three. The chromium crosslink condition disp layed the most significant deviance from its nonfiltered counter part with a mean particle diameter of 4.2m 6.2m from 10.3m 13.9m and an increase in particles within the mesosphere size range from 32% to 55%. However, the chromium condition still produced the largest par ticle diameters of the three conditions tested. A one way analysis of variance (ANOVA) te st was conducted on all collected data to quantify the differences between filtered and nonfiltered particle size values and illustrated that all differences between filtered and non-filtered m ean dry particle diameters were not significant. The test also illustrated no significant partic le size differences among the crosslink agent conditions used. Table 4-2 lists each condition with their respec tive non-filtered and filtered dry mean particle diameter values and their particle diameter size ranges. Table 4-2. The dry mean particle diameter values for the 20m-filtered and the non-filtered DNA-MS synthesized with ionic a nd covalent crosslinking agents. Crosslinking agent Non-filtered mean dry particle diameter (m) 20m-filtered mean dry particle diameter (m) DNA-MS in 1m to 10m size range (%) DNA-MS larger than 10m (%) Chromium 10.3 13.9 4.2 6.0 55 16 Gadolinium 2.6 2.8 3.3 2.8 81 2 Glutaraldehyde 6.4 9.7 5.5 6.5 64 16

PAGE 131

131 The addition of the filtration step not only d ecreased the resultant mean particle diameter values for the chromium crossl ink agent condition tested, but it also normalized and narrowed the particle size distributions by reducing the amount of aggregates in the yield, Figure 4-1. The filtered gadolinium crosslinked DNA-MS condition displayed a normal and narrow particle size distribution as well, Figure 4-2, which was very similar to its non-filtered distribution seen in Chapter 3. However, the filtered glutaraldehyde crosslinked DNA-MS c ondition still displayed evidence of aggregate formation within the 50m to 70m range even though it too demonstrated a more normalized distribution, Figure 4-3. 0 1 2 3 0.010.1110100100010000Pariticle Size (m)Volume % No Filter Filtered Figure 4-1. A particle size distribution compar ison of non-filtered and 20m-filtered chromium crosslinked DNA-MS under dry conditions.

PAGE 132

132 0 1 2 3 4 0.010.1110100100010000Particle Size (m)Volume% No Filter Filtered Figure 4-2. A particle size distribution compar ison of non-filtered and 20m-filtered gadolinium crosslinked DNA-MS under dry conditions. 0 1 2 3 0.010.1110100100010000Paticle Size (m)Volume % No Filter Filtered Figure 4-3. A particle size distribution comparison of non-filtered and 20m-filtered glutaraldehyde crosslinked DNA-MS under dry conditions. Overall, the filtration step appeared to have had two benefits: 1.) the 20m filter helped to remove particles and aggregates over 20m in diamet er and 2.) the 20m filter helped to separate or disperse the DNA-MS in a process somewhat similar to sifting. The filtration step

PAGE 133

133 mechanically facilitated the break up of DN A-MS aggregates within the yield and thus subsequent particle diameters appeared reduced because the true diameters of the DNA-MS were measured and reported instead of the diameters of the aggregates. Reported sizes for the DNAMS decreased by 59% for the chromium and 14% for the glutaraldehyde conditions, and increased by 27% for the ga dolinium condition, Table 4-3. Table 4-3. The dry mean particle diameter values for the 20m-filtered and the non-filtered DNA-MS synthesized with ionic a nd covalent crosslinking agents. Crosslinking agent Non-filtered mean dry particle diameter (m) 20m-filtered mean dry particle diameter (m) Percent change in size (%) Chromium 10.3 13.9 4.2 6.0 (-) 59 Gadolinium 2.6 2.8 3.3 2.8 (+) 26 Glutaraldehyde 6.4 9.7 5.5 6.5 (-) 14 The normalization effect illustrated in the pa rticle size distributi ons for the DNA-MS may have not only been attributed to the filtration step added in the s ynthesis procedure, but also to the sonication step added prior to particle size analysis. This additional sonication step may have further broken up aggregates and aerated th e DNA-MS prior to analysis by rupturing intermolecular bonds between the particles th at formed during synthesis or drying. When particle powders are synthesized or pr epared, cohesive forces tend to join the particles together throug h an adhesive bridge.121 This adhesive brid ging leads to particle agglomerations which occur more frequently in powders with higher volume fractions of small particles as those seen when preparing DNA-MS.122, 123 There are five mechanisms which can bring about interparticle bonding which include solid bridges which arise from the melting of the particles or from the diffusion of molecules be tween particles, liquid bridges which arise from surface tension or capillary pressu re, van der Waals forces which re sult from dipole interactions on the molecular level over short distances, el ectrostatic forces which are longer ranging and

PAGE 134

134 result from the surface charge of a particle, a nd interlocking bonds which arise from mechanical interlocking between particles due to surface irregularities and protrusions, Figure 4-4.122 B A C + + + + + +D E B B A A C C + + + + + +D + + + + + + + + + + + +D E E Figure 4-4. An illustrative depi ction of the five mechanisms responsible for interparticle bonding: A.) solid bridges, B.) liquid br idges, C.) van der Waals forces, D.) electrostatic forces, a nd E.) interlocking bonds.122 The aggregates seen in the particle size dist ributions presented in this section may have been attributed to electrostatic forces between particles since th is type of interaction plays a significant role for interparticle bonding in particles 10m in diameter.124, 125 The production of aggregates may have also been attributed to liquid bridging interactions between the particles which arise during DNA-MS drying or van der Waals attractions which arise once the particles are placed in the methanol and sonicated. However, the van der Waals interactions were probably the least contributive of the three mentioned. The normalization of the particle size distributions presented in this study may have been produced during sonication both before and afte r the DNA-MS were added to the methanol analysis. The normalization effect may have arisen by the tempor ary disruption of the electrostatic forces between DNAMS, which created a more disperse system than those that were obtained in Chapter 3.125-127 It is also believed that th e van der Waals and electrostatic interparticle interactions were further disrupted with the use of the filtration step immediately after synthesis which caused a sifting or dispersion effect on the particles. Hydrated particle size analysis. The particle size of the DNA-MS synthesized with covalent or ionic crosslinking ag ents was also tested under hydrat ed conditions in 0.05M PBS at

PAGE 135

135 a pH of 7.4. Each crosslink agent condition pr oduced hydrated mean diameters of less than 25m. The ionically crosslinked DNA-MS di splayed similar normalization trends under hydrated conditions as compared to their dry dist ributions and swelled to approximately 200% of their original size as shown in Figures 4-5 to 4-7. The normalization trend seen in the ionically crosslinked DNA-MS suggests that the DNA-MS were disperse a nd dominated by electrostatic repulsions between the particles.90, 107, 122, 124 The glutaraldehyde cr osslinked DNA-MS condition did not follow the same normalization trend and continued to display multimodal particle size distributions indicating that this condition was dominate d by particle aggregation. Glutaraldehyde crosslinked DNA-MS aggregated because the particles were more influenced by the attractive van der Waals interparticle forces within the PBS more so than the ionically crosslinked DNA-MS, Figure 4-8. 90, 107, 125, 128, 129 Aside from the aggregation present, the glutaraldehyde condition also performed well exhibiting a 14 .4m mean hydrated particle diameter and only a 150% increase in size. 0 1 2 3 4 5 0.010.1110100100010000Particle Size (m)Volume % Gd Cr Figure 4-5. Hydrated particle size distribution comparisons of the 20m-filtered ionically crosslinked DNA-MS.

PAGE 136

136 0 1 2 3 4 5 0.010.1110100100010000Particle Size ( m)Volume % Dry Hydrated Figure 4-6. A particle size distribution compar ison of the 20m-filtered, chromium crosslinked DNA-MS under dry and hydrated conditions. 0 1 2 3 4 5 0.010.1110100100010000Particle Size ( m)Volume % Dry Hydrated Figure 4-7. A particle size distribution comp arison of 20m-filtered, gadolinium crosslinked DNA-MS under dry and hydrated conditions.

PAGE 137

137 0 4 8 120.010.1110100100010000 Particle Size (m)Volume % Dry Hydrated Figure 4-8. A particle size distribution compar ison of 20m-filtered, glutaraldehyde crosslinked DNA-MS under dry and hydrated conditions. Statistical analysis was conducted on all data co llected to quantify the ef fects of the filtered DNA-MS on hydrated particle size values. A t test was conducted on all hydrated data illustrated no significant differences between the filtered and non-filtered hydr ated mean particle size values. However, particle si ze distributions illustrate that th e filtration step normalized and even reduced the amount of aggregate formation during hydration shifting th e mean particle size peaks over to smaller values, Figures 4-9 to 4-11. Each crosslink agent condition with its respective hydrated mean diameter and percent ch ange in size as calculated by Equation 4-2 is given in Table 4-4.

PAGE 138

138 0 1 2 3 4 5 0.010.1110100100010000Particle Size ( m)Volume % Filtered Non-Filtered Figure 4-9. A particle size distribution compar ison of 20m-filtered a nd non-filtered chromium crosslinked DNA-MS unde r hydrated conditions. 0 1 2 3 4 5 0.010.1110100100010000Particle Size (m)Volume % Filtered Non-Filtered Figure 4-10. A particle size distribution comparison of 20m-filtered and non-filtered gadolinium crosslinked DNA-MS under hydrated conditions.

PAGE 139

139 0 4 8 12 0.010.1110100100010000Particle Size ( m ) Volume % Filtered Non-Filtered Figure 4-11. A particle size distribution comparison of 20m-filtered and non-filtered glutaraldehyde crosslinked DNA-MS under hydrated conditions. Table 4-4. The dry and hydrated m ean particle diameters, percen t change in size values, and crosslink concentrations for DNA-MS. Crosslinking agent Mean dry diameter (m) Mean hydrated diameter (m) Percent change in size (%) Crosslink concentration (%MEQ) Chromium 4.2 6.0 12.6 8.5 (+) 203 120 Gadolinium 3.3 2.8 10.3 7.0 (+) 211 120 Glutaraldehyde 5.5 6.5 14.4 7.4 (+) 125 540 Microscopy Optical microscopy. Dried synthesized DNA-MS were observed under optical microscopy in order to evaluate the discretene ss, morphology, and particle size of the MS after synthesis. DNA-MS ionically crosslinked with gadolinium a nd chromium, and covalently crosslinked with glutaraldehyde, pr oduced discrete par ticles with small diameters and spherical morphologies, Figure 4-12. Optical images illustrate d that the particle sizes were much smaller for the ionically crosslinked DNA-MS than the c ovalently crosslinked DNA-MS. This may be attributed to the amount of time it takes glutar aldehyde to crosslink wi th DNA. As mentioned from Chapter 3, glutaraldehyde reacts more slow ly with the DNA molecu le than chromium or gadolinium, and thus it would be expected that the glutaraldehyde condition with produce larger particles than the ionically crosslinked DNA-MS since the emulsion stir speed is much lower

PAGE 140

140 after crosslinking (i.e. 600rpm instead of 1250rpm ) leading to larger dr oplets in the emulsion which would lead to larger MS in the yield. A B C A B C Figure 4-12. Optical images of the 20m-filtered DNA-MS crossli nked with A) gadolinium, B) chromium, and C) glutaral dehyde (Magnification: 200x). Scanning electron microscopy. SEM images were taken of the synthesized DNA-MS in order to characterize their mor phology and surface topography. SEM images illustrated that the filtration step was successful in removing particles or aggregates with diameters greater than 20m in size for both the ionically and cova lently crosslinked DNA-MS, Figure 4-13. SEM images depicted discreet particles for conditions synthesized with the ioni c crosslinking agents. SEM images of the glutaraldehyde crosslink ag ent condition displayed 1m diameter particles

PAGE 141

141 that aggregated together to form particles la rger than 20m in diameter, Figure 4-13. Since particles larger than 20m in diameter were no t capable of passing through the 20m filter, these aggregates could have only been produced by liquid bridging or elec trostatic interactions between particles upon drying.107, 121, 122 C AB C AB Figure 4-13. Scanning electron microscopy imag es of the 20m-filtered DNA-MS crosslinked with A) chromium (Magnification: 4, 500x), B) gadolinium (Magnification: 2,000x), and C) glutaraldehyde (Magnification: 1,000x). The SEM images also displayed DNA-MS with spherical morphologies and smooth surface topographies for all conditions tested with the exception of the gadolinium crosslinked DNA-MS condition which produced some partic les with irregular morphologies and rough surface topographies. Despite several morphologi cal irregularities, DNA-MS at each crosslink

PAGE 142

142 agent condition displayed particle sizes under 10 m further confirming results obtained from particle size analysis and optic al microscopy observations. Mixer Speed and Crosslink Concentration Study Synopsis The effects of mixer speed and crosslink concentration on DNA-MS particle diameters and size distributions were analyzed to determ ine optimum processing. Mixer speeds of 950rpm, 1250rpm, and 1550rpm along with crosslink concentrations of 20%MEQ, 50%MEQ, and 120%MEQ were assessed in this study. Crosslink ag ent concentrations were obtained using 0.1M gadolinium crosslinking agent solution volumes of 0.3314mL for the 20%MEQ condition, 0.8285mL for the 50%MEQ condition, and 2mL for the 120%MEQ condition during synthesis. The term MEQ was used to denote the molar equivale nce of gadolinium groups to phosphate groups and was calculated for each crosslink con centration condition. The gadolinium crosslink agent was the only crosslinking agent used for th is study since it produced the highest yield with good particle size characteristics after filtration. The DNA-MS were synthesized, washed, and filtered with a 70m nylon filter in order to not co mpletely eliminate the possible size effects of the varying mixer speeds and gadolinium crossl ink concentrations. Resulting DNA-MS were characterized by particle size, morphology, surf ace topography, and elemental analysis. All synthesis and characterization for this study was conducted w ith the assistance of Karly Jacobsen. Particle analysis Percent yield. The yield of each condition tested was calculated and expressed as a percent yield value after complete drying of the DNA-MS was achieved using Equation 4-1. Each synthesized condition ge nerated yields over 60% with the exception of the 950rpm20%MEQ condition which produced a yiel d of 49% and the1250rpm-20%MEQ condition which

PAGE 143

143 produced a yield of 54%. The low yields obta ined with these two conditions may have been attributed to the formation of aggregates with diameters larger than 70m during synthesis. DNA-MS synthesized with a cros slink concentration of 20%MEQ and/or mixer speeds of 950rpm performed the worst with low yields of less than 60%, whereas DNA-MS prepared with the 1250rpm-50%MEQ, 1250rpm-120%MEQ, and the 1550rpm-120%MEQ conditions performed the best, with high yields of over 80%, Figure 4-14. Each DNA-MS condition with its respective yield is given in Table 4-5. 0 20 40 60 80 100950rpm20%MEQ 950rpm50%MEQ 950rpm120%MEQ 1250rpm20%MEQ 1250rpm50%MEQ 1250rpm120%MEQ 1550rpm20%MEQ 1550rpm50%MEQ 1550rpm120%MEQ Figure 4-14. Graphical representation of yi eld values generated by each DNA-MS synthesis condition. Table 4-5. Percent yield values for DNA-MS prepared at varying mixer speeds and gadolinium crosslink concentrations. Synthesis condition Percent yield (%) 950rpm-20%MEQ 49 950rpm-50% MEQ 73 950rpm-120% MEQ 65 1250rpm-20% MEQ 54 1250rpm-50% MEQ 87 1250rpm-120% MEQ 87 1550rpm-20% MEQ 70 1550rpm-50% MEQ 72 1550rpm-120% MEQ 88

PAGE 144

144 The production of these high yiel ds may have been a result of the higher mixer speeds used during synthesis which not only reduced the forma tion of aggregates, but also produced particles with smaller diameters.10, 107, 130, 131 DNA-MS with smaller particle sizes were created due to the decrease in the interfacial tension between the aqueous DNA phase and the continuous CAB phase by increasing the stir speeds during synthesis.10, 107, 130, 131 Table 4-5 above lists each synthesis condition with its respective generated yield. Dry particle size analysis. The dry particle size of ach DNA-MS condition prepared in this study was tested in methanol. Each synthe sized condition produced pa rticles with dry mean diameters of less than 20m, Table 4-6. Table 4-6. Mean dry particle size and size range values for DNAMS prepared at varying mixer speeds and gadolinium crosslink concentrations. Synthesis condition Mean dry particle size (m) DNA-MS in 1m to 10m size range (%) DNA-MS larger than 10m (%) 950rpm-20%MEQ 14.7 28.5 4619 950rpm-50%MEQ 4.9 6.9 4814 950rpm-120%MEQ 6.3 9.5 5418 1250rpm-20%MEQ 3.7 5.5 774 1250rpm-50%MEQ 6.8 14.4 5914 1250rpm-120%MEQ 5.7 12.7 91 1550rpm-20%MEQ 17.9 21.9 5731 1550rpm-50%MEQ 3.3 3.7 723 1550rpm-120%MEQ 2.2 2.1 681 Each DNA-MS condition at the 950rpm mixer sp eed produced particles with less than 60% of the diameters ranged in the mesosphere size ra nge (i.e. 1m to 10m) and much greater than 5% were larger than 10m. At the 1250rpm mixer speed, the 20%MEQ condition produced favorable results; however, the 50%MEQ condition produced many aggr egates (i.e. 14% of all particles were larger than 10m) and the 120%MEQ condition produced many particles in the nanosphere size range (i.e. 90% of all particles were smaller th an 1m). At the 1550rpm mixer

PAGE 145

145 speed, the 50%MEQ and 120%MEQ conditions produced particles th at had greater than 60% of the diameters in the mesosphere size range and le ss than 5% of all diameters greater than 10m. DNA-MS prepared at the 950rpm and 1550rpm stir speeds and 20%MEQ crosslink concentration, produced particle s with mean diameters greater than 10m and standard deviations greater than 20m indicating that these conditions produced many large aggregates during synthesis, Figure 4-15. This may be a result of non-uniform crosslinking due to the low concentration of crosslink agent available in the system. With the exception of the 20%MEQ crosslink concentration condition, no further par ticle size distribution tr ends were observed solely on crosslink concentration. These results are consis tent with previous research and current literature which has cited that unlike emulsion stir speed, cross link concentration does not have an effect on resulting microsphere sizes.11, 57, 132, 133 0 1 2 3 4 50.010.1110100100010000Particle Size ( m)Volume % 1550rpm 1250rpm 950rpm Figure 4-15. A dry particle size distribution comparison of DNA-MS synthesized at varying mixer speeds at the 20%MEQ crosslink concentration condition. Trends denoting the effects of mixer speed on particle diameter and size distribution were clearly observed. DNA-MS prepared at 950rpm pr oduced particles with broad size ranges and multimodal distributions whereas DNA-MS synt hesized at 1250rpm and 1550rpm displayed narrower and more normalized particle si ze distributions, Figures 4-16 to 4-18.

PAGE 146

146 0 1 2 3 0.010.1110100100010000Particle Size ( m ) Volume % 121%MEQ 50%MEQ 20%MEQ Figure 4-16. A dry particle size distribution comparison of DNA-MS synthesized at varying crosslink densities at the 950rpm mixer speed condition. 0 1 2 3 4 5 0.010.1110100100010000Particle Size ( m)Volume % 121%MEQ 50%MEQ 20%MEQ Figure 4-17. A dry particle size distribution comparison of DNA-MS synthesized at varying crosslink densities at the 1250rpm mixer speed condition.

PAGE 147

147 0 1 2 3 4 0.010.1110100100010000Particle Size ( m ) Volume % 121%MEQ 50%MEQ 20%MEQ Figure 4-18. A dry particle size distribution comparison of DNA-MS synthesized at varying crosslink densities at the 1550rpm mixer speed condition. This trend was to be expected since it ha s been well cited in the literature that by increasing the stir speed and in turn increasing the agitation in the system, the formation of aggregates during synthesis is prevented or reduced.10, 125, 130, 131 DNA-MS prepared at the 1550rpm stir speed at crosslink concentrations greater than 50%MEQ produced the smallest particles with the most narrow size distributions. This increase in the volume percent of smaller particles can be attributed to the higher stir speed used during synthesi s which created a finer emulsion due to the increase in sh ear force exhibited on the system.10, 57, 107, 130, 131, 134-136 In order to quantify the effect s of mixer speed and crosslink concentration on the particle sizes of the DNA-MS, a one way ANOVA test was conducted on all collected data. The test illustrated there to be no significant differences among all mean particle size values at each mixer speed and crosslink concen tration condition. However, a st rong trend was observed in that the particle size distributions for all crossli nk concentration conditions at the 1550rpm mixer speed condition produced the best particle sizes and most normalized distributions, Figure 4-18. It is to be noted, however, that a 70m filter was us ed in this study instead of a 20m filter. If a

PAGE 148

148 20m filter had been used, aggregates seen past the 20m mark would have been eliminated as seen in the results for the gadolin ium condition in th e Filter Study above. Hydrated particle size analysis. DNA-MS synthesized at the 1550rpm mixer speed condition underwent particle size analysis under hydrated conditions in 0.05M PBS at a pH of 7.4. All other conditions were not tested due to their multimodal and broad particle size distributions. The DNA-MS prepared at th e 1550rpm mixer speed condition each produced multimodal distributions upon hydration, Figures 4-19 to 4-21. These results are not consistent with observations made in the Filtration Study and may be attributed to the formation of aggregates during the DNA-MS dr ying process. However, it is assumed that the hydrated size distributions would be more normalized with fewer aggregates present if a 20m filter had been used in place of the 70m filter. 0 1 2 3 4 5 6 0.010.1110100100010000Particle Size (m)Volume % Dry Hydrated Figure 4-19. Dry and hydrated particle size dist ributions for DNA-MS prepared at 1550rpm and crosslinked with gadolinium to 20%MEQ.

PAGE 149

149 0 2 4 6 8 0.010.1110100100010000Particle Size (m)Volume % Dry Hydrated Figure 4-20. Dry and hydrated particle size dist ributions for DNA-MS prepared at 1550rpm and crosslinked with gadolinium to 50%MEQ. 0 1 2 3 4 5 6 0.010.1110100100010000Particle Size (m)Volume % Dry Hydrated Figure 4-21. Dry and hydrated particle size dist ributions for DNA-MS prepared at 1550rpm and crosslinked with gadolinium to 120%MEQ. Each condition also produced hydrated mean diameters larger than 50m; however, it was expected that these large diameter values bett er reflected the aggregates that formed during drying and then hydration more th an the actual hydrated values of the individually swollen

PAGE 150

150 particles. Upon closer analysis, it was assume d that the first peak better represented the distribution of the individually swollen DNA-MS as seen above in the Filtration Study. Therefore, the mean hydrated particle size and percent swelling values for those peaks were calculated and represented as the theorized mean pa rticle size and percent swelling values for the DNA-MS. In order to determine the effects of hydration on the particle size values of the DNA-MS at the 1550rpm mixer speed condition, a one way ANOV A was conducted on all dry, hydrated, and theorized hydrated diameter values at each cr osslink concentration. The test illustrated no significant differences among each of the crosslink concentrations tested. The theorized mean particle size and percent swelling values as ca lculated by Equation 4-2 are reported along with the initial values in Table 4-7. Table 4-7. Dry and hydrated mean particle size and percent swelling values for DNA-MS prepared at the 1550rpm mixer speed condition. Condition Mean dry particle size (m) Mean hydrated particle size (m) Percent swelling (%) Theorized mean hydrated particle size (m) Theorized percent swelling (%) 20%MEQ 17.9 21.9 58.2 47.4 225 23.8 19.3 33 50%MEQ 3.3 3.7 94.9 88.3 2780 23.8 21.3 620 120%MEQ 2.2 2.1 69.2 62.5 3050 13.6 26.8 520 To further compare the crosslink concentra tion conditions at the 1550rpm mixer speed, the percent swelling for each condition was calculated. The 20%MEQ crosslink concentration condition displayed the lowest amount of sw elling upon hydration by swelling only to 225% of its original size. Surprisingly, the higher cr osslink concentration conditions each produced swelling over 2000%; however, these larger values may rather reflect sizes of the aggregates that formed upon hydration and than the true swol len diameters of the DNA-MS. Thus a more realistic swollen value may be represented using the median values at each crosslink concentration condition. Using these values a 33% increase in size was reported for the 20%MEQ

PAGE 151

151 condition, a 620% increase in size was reported for the 50%MEQ condition, and 520% increase in size was reported for the 120%MEQ condition, Table 4-7. These valu es also illustrated that the 120%MEQ crosslink concentration condition still produced significant swelling; however, the mean swollen diameter value was still lower than 20m. The standard deviation of this value would be expected to decrease upon using a 20m f ilter instead of 70m filte r at the end of the DNA-MS synthesis process since it would facilitate the break up of aggregates that form during washing and drying. Surface charge analysis. The surface charge of the DNA-MS prepared at the 1550rpm mixer speed was measured in 0.01M PBS at a pH of 7.4 to determine the effects of crosslink concentration on yielding surface charge. Zeta potential measurements illustrated that DNA-MS values became more negative as the crossl ink concentrations increased, Figure 4-22. -70 -50 -30 -101Zeta Potential (mV) 20%MEQ 50%MEQ 120%MEQ Figure 4-22. Zeta potential valu es for DNA-MS prepared with 20%MEQ, 50%MEQ, and 120%MEQ crosslink concentrations (Note: Erro r bars represent standard error.). DNA-MS prepared at the 120%MEQ crosslink concentration produ ced the most negative zeta potential values of the three crosslink density va lues tested with a surface charge of -55.2mV,

PAGE 152

152 Table 4-8. A one way ANOVA illustrated that this condition was significantly different from both the 20%MEQ (p = 0.001) and 50%MEQ (p = 0.011) crosslink co ncentration conditions. Table 4-8. Zeta potential values for DNA-MS prepared with 20%MEQ, 50%MEQ, and 120%MEQ crosslink concentrations. Crosslink concentration condition Zeta potential (mV) Standard error (mV) 20%MEQ -55.2 3.0 50%MEQ -48.0 3.4 120%MEQ -46.3 2.4 Microscopy Scanning electron microscopy. The morphology and topography of the synthesized DNA-MS was analyzed using SEM. SEM images illustrated that the pa rticle size of the DNAMS became more uniform as the mixer speed increased from 950rpm to 1550rpm, visually confirming results obtained through particle size analysis Figures 4-23 to 4-25. AB C AB C Figure 4-23. SEM images of DNA-MS crosslinked with 20%MEQ gadolinium at the A) 950rpm, B) 1250rpm, and C) 1550rpm mixer speed conditions (Magnifications: 2,000x).

PAGE 153

153 A B C A B C Figure 4-24. SEM images of DNA-MS crosslinked with 50%MEQ gadolinium at the A) 950rpm, B) 1250rpm, and C) 1550rpm mixer speed conditions (Magnifications: 1,500x).

PAGE 154

154 AB C AB C Figure 4-25. SEM images of DNA-MS crosslinked with 120%MEQ gadolinium at the A) 950rpm, B) 1250rpm, and C) 1550rpm mixe r speed conditions (Magnifications: 1,000x, 2,000x, and 1,500x, respectively). The DNA-MS synthesized at the 1550rpm mi xer speed conditions produced spherical morphologies and smooth topographies with norma lized and mostly uniform particle size distributions; whereas the DNAMS synthesized at the 950 rpm and 1250rpm mixer speed conditions produced irregularly shaped part icles with uneven surface topographies and heterogeneous particle size distributions. The SEM images illustrated that the DNA-MS synthesized at the 1550rpm mixer speed produced the most optimal particle morphologies, topographies, and si ze distributions.

PAGE 155

155 Energy dispersive x-ray spectroscopy. Elemental analysis vi a EDS was conducted on the DNA-MS prepared at the 1550rpm mixer speed to confirm the presence of gadolinium. EDS measures the intensity of characteristic x-rays from different elements resulting from inner-shell excitations brought about by an electron beam.110 The data is plotted in intensity versus energy where larger plotted intensities are indicative of a larger number of excitations received by the EDS detector.110 EDS analysis illustrated that the DNA-MS synthesized at the 120%MEQ condition produced spectra with higher x-ray co unts of gadolinium than seen in the 50%MEQ and 20%MEQ conditions. These higher x-ray counts are to be expected due to the larger gadolinium crosslink concentration used.110 The chlorine x-ray counts also increased from the 20%MEQ condition to the 120%MEQ condition further demonstrating an increase in the crosslinking agent concentration. The chlorine peaks were visi ble in the EDS spectra since the DNA-MS where crosslinked with a 0.1M ga dolinium chloride solution, Figures 4-26 to 4-28. Figure 4-26. EDS spectra collected on DNA-MS prepared at the 1550rpm mixer speed and 20%MEQ crosslink concentration condition.

PAGE 156

156 Figure 4-27. EDS spectra collected on DNA-MS prepared at the 1550rpm mixer speed and 50%MEQ crosslink concentration condition. Figure 4-28. EDS spectra collected on the DNAMS prepared at the 1550rpm mixer speed and 120%MEQ crosslink concen tration condition.

PAGE 157

157 Conclusions Filtration Study The objectives of this study we re to minimize or eliminate the presence of aggregates in the resulting DNA-MS yields and to produce contro lled size distributions where at least 60% of all particles prepared fell with in the mesosphere size range of 1m to 10m and less than 5% of all particles were greater than 10m in size. Particles less than 1m in diameter were also acceptable and DNA-MS were sought to obtain hydrated particle di ameters of less than 25m. To achieve these goals the synthesis para meters for producing DNA-MS were altered by adding a filtration step to the end of the DNA-MS synthesis procedure using a nylon filter with a pore size of 20m. A 20m pore size was used for this study to better target the diameter range desired for optimal intratumoral chemotherapy.69, 120 The resulting particle diameters and size distributions were then evaluated and compared to non-filtered controls. Data obtained in this study illustrated that the filtration step was succe ssful in removing aggregates and particles over 20m in diameter. The filtration step also normaliz ed dry and hydrated partic le size distributions by mechanically rupturing inte rmolecular bonds between the particles that formed during synthesis and drying for both ionically and cova lently crosslinked DNA-MS Data illustrated that the gadolinium crosslinked DNA-MS produced the best conditions after filtration with normalized dry and hydrated particle size distribu tions, dry diameters less than 5m, 81% of all particles in the mesosphere size range, 2% gr eater than 10m, hydrated diameters less than 20m, and only an 18% decrease in yiel d as compared to the non-filtered yield. Mixer Speed and Crosslink Concentration Study The objective of this study was to optimi ze DNA-MS synthesis parameters to produce narrow and controlled size distributi ons where at least 60% of all particles prepared fell within the mesosphere size range of 1m to 10m and less than 5% of all particles were greater than

PAGE 158

158 10m in size. Particles less than 1m in di ameter were also accep table and DNA-MS were sought to obtain hydrated particle diameters of less than 25m. The effects of mixer speed and crosslink c oncentration on particle diameter, swelling, morphology, and size distribution were analy zed. Mixer speeds of 950rpm, 1250rpm, and 1550rpm with gadolinium cross link concentrations of 20%MEQ, 50%MEQ, and 120%MEQ were assessed for these studies. The resulting DNA-MS were characterized by yield, particle size, elemental analysis, and surface topography and morphology. Data obtained in this study re vealed that the particle size distributions for all crosslink concentration conditions tested nor malized as the stir speed increas ed. This study also revealed that the 20%MEQ crosslink concentration condition produced the lowest yields at each of the three different mixer speeds tested indicating that this crosslink concentration was responsible for producing aggregates larger than 70m in diameter. Data obtai ned in the study also illustrated that the 950rpm mixe r speed produced DNA-MS with the largest percentage of particles with diameters greater than 10m indicating that this speed produced many aggregates. Overall, the data obtained in this study illustrated that the 120%MEQ crosslink concentration condition tested at the 1550rpm mi xer speed produced the most optimal results with yields over 85%, dry particle diameters less than 5m, 68% of all particles in the mesosphere size range, 1% of all particles greater than 10m, hydrated particle diam eters less than 25m, and excellent stability in PBS.

PAGE 159

159 CHAPTER 5 IN VITRO EVALUATIONS OF MITOXANTRONE LOADED DNA NANO-MESOSPHERES Introduction Mitoxantrone (MXN) in situ loaded DNA nano-mesospheres (DNA-MXN-MS) were prepared using optimized synthesis conditions fro m Chapter 4 to obtain particles with controlled size distributions where 60% of a ll particles prepared were within the mesosphere size range of 1m to 10m and < 5% of all particles were greate r than 10m in size. Particles less than 1m in diameter were also acceptable and hydrated particle diameters we re to be less than 25m. In addition, MXN in situ loaded DNA-MS were sought to obtain loadings of 12% (w/w) MXN and release MXN for 24 hours or more in phosphate buffered saline under minimum sink conditions. DNA-MXN-MS were also s ought to elicit a cy totoxic response to that of free drug on murine Lewis lung carcinoma (mLLC) cells in culture. DNA-MXN-MS were prepared with a MXN c oncentration of 15% (w/w) and gadolinium crosslink concentrations of 20%, 50%, and 120% molar equivalence (MEQ). The particle diameter, size distribution, morphology, drug lo ading, and percent drug release of the DNAMXN-MS were evaluated with respect to the cro sslink concentration. The particle diameters and size distributions were obtained using an LS Coulter 13 320 particle size analyzer. The morphology and presence of gadolinium trivalen t cations were assessed using a scanning electron microscope with energy dispersive xray spectroscopy. Drug loading was determined by incubating the DNA-MXN-MS under stirred condi tions in an enzymatic digestion buffer at 37C for 48 hours. The released drug concentr ations were then analyzed using UV-visible spectroscopy against a MXN sta ndard curve. The percent dr ug release was measured under minimum sink conditions in order to simulate th e tumor environment. Drug release data was obtained by incubating the DNA-MXN-MS in phospha te buffered saline under constant agitation

PAGE 160

160 at 37C for a minimum of fourteen days in triplicate. At specific times, aliquots were taken and measured using UV-visible spectroscopy agains t a MXN standard curve to determine drug concentration. The cytotoxicity of DNA-MXN-MS crosslinked with 120% MEQ gadolinium was evaluated using a murine Lewis lung carcinoma (mLLC) cell line at MXN concentrations of 1g/mL, 10g/mL, and 25g/mL and was compared to the same MXN concentrations of free drug. The cytotoxicity of blank DNA-MS were al so evaluated. An MTS assay was used to measure the cellular viability of th e mLLC cells at days 0 through 4. Materials and Methods Materials Synthesis and characterization DNA sodium salt derived from herring testes Type XIV (DNA), cellulose acetate butyrate, HPLC grade 1,2-dichloroethane, methanol, gadol inium (III) chloride hexahydrate, mitoxantrone dihydrochloride, L-cysteine hydrochloride hydrate, papain from papaya latex, deoxyribonuclease I from bovine pancreas, ethylenediaminetetraace tic acid disodium salt dihydrate (EDTA), and trichloroacetic acid were purchased from th e Sigma-Aldrich Company. Sodium phosphate monobasic monohydrate, sodium p hosphate dibasic anhydrous, and sodium chloride, each A.C.S. certified, acetone, 20m and 70m Spectra/Mesh Nylon filters, 16x125mm and 13x100mm borosilicate glass test tubes, Ultrafree-CL with Du rapore Membrane Centrifugal Filter tubes (0.1m pore, 2mL), Corning ster ile 0.45m pore filter bottles, Fisherbrand Semimicro methacrylate 1.5mL disposable cu vettes, and 15mL and 50mL polypropylene centrifuge tubes were purchased from Fisher Sc ientific International. Type I and Type II deionized ultrapure water was prepared with a resistivity of at least 16 M -cm-1 using the Barnstead NANOpure Ultrapure Water System in the lab.

PAGE 161

161 Cell culture Corning 75cm2 cell culture flasks, Fisherbrand cl ear polystyrene 96-well plates, MEM non-essential amino acid solution (NEAA), Cellgro heat inactivated fetal bovine serum (FBS), and 0.22m Corning cellulose nitr ate filters were each purchased from Fisher Scientific International. The CytoTox 96 Non-Radioactiv e Cytotoxicity Assay was purchased from the Promega Corporation. The mLLC cells, Dulbecco ’s Modified Eagle’s Media with L-glutamine (DMEM), trypsin-EDTA solution, and penicillin -streptomycin were purchased from the American Type Culture Collect ion (ATCC, Manassas, VA). Synthesis equipment Solutions were washed or prepared on a Genie 2 vortex mixer. Centrifugation was conducted using a Dynac II bench top centrifuge General DNA-MXN-MS syntheses were carried out using Caframo Model BDC6015 and Li ghtnin Model LIU08 mechanical lab mixers in 300mL Labconco lyophilization flasks. Methods Solution preparation MXN in situ loaded DNA. MXN was in situ loaded into 5% (w/v) aqueous solutions of DNA at a concentration of 15% (w/w). Aque ous DNA solutions were prepared at room temperature by adding 0.5g of DNA to 5mL of ultrapure water in a 50mL polypropylene centrifuge tube. The solution was mixed on hi gh on a vortex mixer for 30 seconds. Three milliliters of ultrapure water were then added to the DNA and placed on the rotary shaker for at least two hours until the DNA had completely dissolved. The volume was then brought up to 10mL and vortexed on high for 30 seconds. The DNA solution was then placed in the refrigerator over night to ensu re the complete collapse of bubbles generated during vortex and rotary mixing. The percent solid concentration of the DNA solution was quantified using a

PAGE 162

162 Metler LJ16 Moisture Analyzer at 130C for 60 minutes. Once the concentration was determined, 3.5mL of the DNA solution was placed into a 15mL polypropylene centrifuge tube. Using the true solid weight of the DNA in soluti on obtained during moisture analysis, the weight of the MXN, XgMXN, needed to obtain a 15% (w/w) loading was calculated by multiplying the true DNA solid weight in 3.5mL of DNA solution (WTDNA) by 15 and then dividing the product by 100, Equation 5-1. DNA 100g W MXN 15g MXN XgTDNA (5-1) The desired mass of MXN was weighed and adde d to the 3.5mL of the DNA solution. The DNA-MXN solution was then vortexed on high for 30 seconds and placed on the rotary mixer for 2 hours. The solution was then placed in the refrigerator until further use. Cellulose acetate butyrate. Solutions of cellulose acetate butyrate in 1,2-dichloroethane (CAB) were used as the water-immiscible c ontinuous phase for the emulsion stabilization process during DNA-MXN-MS synthesi s. The CAB solutions were used at a concentration of 5% (w/v) and prepared by adding 25g of ce llulose acetate butyrate to 500mL of 1,2dichloroethane. The CAB solution was mixed at room temperature on a magnetic stir plate on high until the cellulose acetate butyrate had comple tely dissolved in the 1,2-dichloroethane. The resulting CAB solution was capped, parafilm ed, and stored at room temperature. Gadolinium (III) chloride. An aqueous solution of gadolinium (III) chloride was prepared to a concentration of 0.1M by addi ng 18.59g of gadolinium (III) chloride hexahydrate to 500mL of ultrapure water. The gadolinium (III) solution was mixed on a magnetic stir plate over night at room temperature until all the gadolinium had dissolved in the water. After complete mixing, the 0.1M gadolinium (III) cros slinking agent solution was parafilmed and stored at room temperature.

PAGE 163

163 Phosphate buffered saline. Four liters of a 0.05M phosphate buffered saline (PBS) solution with a pH of 7.4 was prepared in the lab for measuring the sw elling properties of the DNA-MXN-MS and for in vitro drug release analysis. The PBS solution was prepared by mixing 2.9L of a 0.05M sodium phosphate dibasi c solution to 1L of a 0.05M sodium phosphate monobasic solution. The pH of the resulting so lution was measured and the sodium phosphate monobasic solution was added until the target pH of 7.4 was reached. The PBS solution was then sterilized using a 0.45m Co rning 1L cellulose acetate filter system and left out at room temperature until needed. A PBS solution with a concentration of 0.1M at a pH of 7.4 was used for determining the MXN loading efficiency of the DNA-MXN-MS. The 0.1M PBS solution was prepared by adding 2.9L of a 0.1M sodium phosphate dibasi c solution into 1L of 0.1M sodium phosphate monobasic solution and the pH was adjusted by adding the monobasic to the solution until a pH of 7.4 was reached. The 0.1M PBS solution wa s then sterilized using a 0.45m Corning 1L cellulose acetate filter system. A 0.01M PBS solu tion at a pH of 7.4 was used for zeta potential measurements and was prepared by diluting a 0.1M PBS solution and adjusting the pH back to 7.4. The solutions were left out at room temperature until needed. Enzymatic digestion buffer. An enzymatic digestion buffe r was used to digest the DNAMXN-MS to determine their MXN loading efficiency. The enzymatic digestion buffer was prepared by adding 1,800mg of EDTA, 200mg of L-cysteine hydr ochloride hydrate, 125mg of papain from papaya latex, and deoxyribonuclease I from bovine pancreas to 250mL of the 0.1M PBS solution (pH = 7.4). The enzymatic digest ion buffer components we re stirred at room temperature until completely dissolved and used immediately after preparation.

PAGE 164

164 Cell culture media. Cell culture media was prepared by adding 50mL of FBS, 10mL of NEAA solution, and 10mL of peni cillin-streptomycin to 500mL of DMEM to obtain a 10% FBS, 2% NEAA, and 2% penicillin-str eptomycin treated media. This media mixture will be referred to as treated media throughout. The treated medi a solutions were mixed manually and placed in the refrigerator until further use. Synthesis procedure DNA-MXN-MS were prepared using a modifi ed emulsion stabilizat ion technique that was developed in this lab that sterically stabilizes the DNA molecule into spherical conformations and crosslinks th em while in suspension. This emulsion stabilization process involved dispersing 3mL of the aqueous DNA-MXN solution (i.e. the aqueous phase) into 47mL of the organic CAB solution (i.e. the conti nuous phase) in a 300mL Labconco lyophilization flask. A DNA-MXN microemulsion was then creat ed by vigorously mixing the two solutions at 1550rpm for 20 minutes at room temperature using a paddle mixe r with a two inch, two blade propeller. The DNA-MXN microemuls ion was then ionically crosslinked while in suspension by reducing the speed of the paddle mixer to 600rpm and adding 0.1M gadolinium (III) solution in volumes of 0.3314mL to obtain the 20%MEQ, 0.8285mL to obtain the 50%MEQ, or 2mL to obtain the 120%MEQ crosslink concentration conditions The DNA-MXN microemulsion then underwent crosslinking for 1 hour and 40 minutes at which time 50mL of acetone was added and any further reactions were allowed to reach co mpletion for another hour. After synthesis was complete, the DNA-MXN-MS underwent four rinses in acetone to remove any residual organic phase, crosslinking agent, or non-loaded drug. The DNA-MXN-MS were rinsed by separating the resultant DNA-MXN-MS suspension into fo ur separate 50mL polypropylene centrifuge tubes. Acetone was added to 40mL and the tubes were capped and vortexed on high for 30 seconds. The DNA-MXN-MS were then collec ted by centrifuging the tubes at 2600rpm for 10

PAGE 165

165 minutes and decanting the acetone. Fresh acetone was added again to 40mL mark and the acetone rinse was repeated once more as mentione d above and then twice more by consolidating the contents of 4 tubes to 2 tube s and then 2 tubes to one tube. After the final acetone rinse, DNA-MXN-MS were resuspended in 30mL of acetone and vortexed on high for 30 seconds. The resuspended DNA-MXN-MS were then filtere d using a stainless st eel vacuum filtration device with a 70m Spectra/Mesh Nylon filter. A 70m Spectra/Mesh Nylon filter was used in place of the 20m filter to not completely elimin ate the possible size effects of the crosslink concentrations and the MXN drug loading on th e DNA-MXN-MS. The device was then rinsed with 10mL of acetone to further filter a ny remaining DNA-MXN-MS under 70m. The centrifuge tube was then capped, vortexed on hi gh for 30 seconds, and centrifuged at 2600rpm for 5 minutes to collect the DNA-MXN-MS. The acetone was then decanted and a Kimwipe was secured over the mouth of tube using a r ubber band to allow the DNA-MXN-MS to dry overnight at room temperature. Particle characterization Yield analysis. The yield and theoretical yiel d of each condition synthesized was calculated and expressed as a percent yield valu e. The percent yield was calculated by dividing the final weight of the DNA-MXN-MS by the am ount of weight used to synthesize the DNAMXN-MS, Equation 5-2. For the theoretical yield, the weight of the MXN that was in situ loaded into the DNA solution was added to the e quation, Equation 5-3. The equations for the percent yield and percent theoretica l yield are expressed below where WF is the final weight of the DNA-MXN-MS, VDNA, DNA, and CDNA are the volume, density, and concentration of the aqueous DNA solution used respectively, WX is the weight of the crosslinking agent added during synthesis, and WY is the theoretical weight of the in situ loaded MXN.

PAGE 166

166 100 ) ) W C ((V W ( Yield %X DNA DNA DNA F (5-2) 100 ) ) W W C ((V W ( Yield l Theoretica %Y X DNA DNA DNA F % (5-3) Dry particle size analysis. Dry DNA-MXN-MS particle di ameters and size distributions were obtained using a Coulter LS 13 320 particle size analyzer. DNA-MXN-MS were sonicated for 10 to 30 seconds to aerate and separate the pa rticles prior to analysis and then approximately 2mg of the DNA-MXN-MS were suspended in 2m L of methanol. The suspension was then sonicated for 30 seconds to further break up any DNA-MXN-MS aggregates and tested. The Coulter LS 13 320 particle size analyzer was se t to run at a pump speed of 73% using a protein/DNA particle diffraction model. Standards were tested in me thanol before the first run to ensure that the instrument was performing ade quately. Each condition was sampled three times in which each sample’s particle size measurements consisted of two runs. This method of testing produced six independent particle diameters and size distributions. Data collected from these experiments were statistically anal yzed using SigmaStat 3.0 software. Hydrated particle size analysis. Swollen DNA-MXN-MS particle diameters and size distributions were obtained in 0. 05M PBS at a pH of 7.4 using a Coulter LS 13 320 particle size analyzer. DNA-MXN-MS were sonicated for 10 to 30 seconds prior to analysis to aerate and separate the particles and then approximatel y 2mg of the DNA-MXN-MS were suspended in 2mL of PBS. The suspension was then sonica ted for 30 seconds to further break up any DNAMXN-MS aggregates. The DNA-MXN-MS were th en allowed to swell in the PBS for an additional two minutes and thirty seconds. The DNA-MXN-MS were then tested in the Coulter LS 13 320 particle size analyzer using a pum p speed of 73% and a protein/DNA particle diffraction model. Standards were tested in PBS before the first r un to ensure that the instrument

PAGE 167

167 was performing adequately. Each condition was sampled three times in which each sample’s particle size measurements consisted of two runs. This method of testing produced six independent particle diameters a nd size distributions. The mean sw ollen particle diameters were then used to calculate the percent change in size by a using Equation 5-4 where DH is the mean hydrated particle diameter and DD is the mean dry particle diameter. A negative percent change in size depicted a decrease in particle size wher eas a positive percent chan ge in size depicted an increase in particle size, or swelling. Data co llected from these experiments were statistically analyzed using SigmaStat 3.0 software. 100 ) D D D ( Size in Change %D D H (5-4) Surface charge analysis. DNA-MXN-MS surface charge was measured to further characterize the in situ loading of MXN. DNA-MXN-MS surface charge was obtained in a 0.01M PBS solution at a pH of 7.4 using a Brook haven ZetaPlus zeta potential analyzer with ZetaPALS software. Approximately 2mg of the DNA-MXN-MS were suspended in 1.5mL of the PBS solution. Each condition was sampled three times in which each sample underwent ten runs. This method of testing produced thirty independent zeta potential values. The data collected from the zeta potential analyzer wa s statistically analyzed using SigmaStat 3.0 software. Scanning electron microscopy. The morphology and surface topography of the DNAMXN-MS were observed using scanning electron microscopy (SEM). Approximately 1mg of dry DNA-MXN-MS were mounted onto a small piece of silicon wafer which in turn was mounted onto an aluminum SEM stub using doubl e sided tape. The DNA-MSN-MS were then coated with gold-palladium for 2 minutes using a Technix Hummer V sputter coater. Images

PAGE 168

168 were taken on a JEOL 6335F Field Emission SEM at an accelerating vo ltage of 5KeV and a working distance of 15mm. Energy dispersive x-ray spectroscopy. The presence of trivalent cations in the DNAMXN-MS after washing and drying was observe d using energy dispersive x-ray spectroscopy (EDS). DNA-MXN-MS were mounted onto a piece of silicon wafer which was then secured to an aluminum SEM stub using carbon double sided tape. The DNA-MXN-MS were then coated with carbon for 2 minutes using a Technix Hummer V sputter coater. EDS spectra on the DNAMXN-MS were collected using a JEOL 6400 SEM at an accelerating vo ltage of 15KeV and working distance of 15mm. A dead time of 20% to 40% was allowed for each condition tested. In vitro DNA-MXN-MS characterization procedures MXN loading efficiency. The MXN loading efficiency of the DNA-MXN-MS was determined via enzymatic digestion followed by phot ometric analysis of th e recovered entrapped drug. Approximately 5mg of the DNA-MXN-MS were weighed out in to labeled 16x125mm glass test tubes and recorded. Te n milliliters of the enzymatic digestion buffer were added to the DNA-MXN-MS and the test tubes were cap ped, parafilmed, and incubated at 37 C under stirred conditions for 48 hours. Control MXN solutions we re also tested for drug degradation by adding 200l of a 1,000g/mL MXN solution into 10mL of the enzymatic digestion buffer. The control solutions were also incubated in 16x125mm glass te st tubes at 37 C under stirred conditions for 48 hours. The test tubes were then taken out of the incubator and the solu tions were allowed to cool to room temperature. While the solu tions were cooling, a 10% (w/v) solution of trichloroacetic acid (TCA) in ultrapure water was prepared. Two milliliters of each DNA-MXNMS solution were placed into labeled 13x100mm gl ass test tubes. Two milliliters of the 10% (w/v) TCA solution was then added to each samp le. The test tubes were capped and the two solutions were allowed to react for 30 minutes at room temperature. The samples were then

PAGE 169

169 centrifuged at 2,600rpm for five minutes and the supernatants from each sample were collected into 1.5mL methacrylate disposable cuvettes for UV-Vis analysis. Each sample was analyzed on a Perkin Elmer Lambda 3 spectrophotometer at a wavelength of 610nm against a MXN in 5% (w/v) TCA standard curve with concentrati ons ranging from 1g/mL to 50g/mL and a correlation coefficient of 0.99966 to determ ine the MXN concentration. The MXN concentration was then used to determine th e percent drug loading of MXN in the DNA-MXNMS using Equation 5-5.137, 138 The loading efficiency was then calculated using Equation 5-6.137139 Each condition was tested in triplicate and all collected data was statistically analyzed using SigmaStat 3.0 software. 100 ) DNAMXNMS o f Mass MS in MXN of Mass ( Loading Drug % (5-5) 100 ) Loading l Theoretica Loading al Experiment ( Efficiency Loading (5-6) In vitro MXN release. The in vitro release of MXN from DNA-MXN-MS under minimum sink conditions was tested in sterile filtered 0.05M PBS at a pH of 7.4. Approximately 2mg of each DNA-MXN-MS condition were weighed into labeled 2mL filt er centrifuge tubes (0.1m pore size) to which 1.25mL of the sterile filtered 0.05M PBS solution was added. The tubes were then capped, para filmed, and incubated at 37 C under stirred conditi ons. At specified time points, the filter centrifuge tubes were re moved from the incubator and centrifuged for 10 minutes at 3,000rpm. Aliquots were then removed and placed into 1.5mL methacrylate disposable cuvettes for UV-Vis analysis. The cu vettes were capped, labeled, parafilmed, and stored in the refrigerator until analysis. Each centrifuge filter tube was then replenished with 1.25mL of fresh 0.05M PBS, capped, parafilmed and placed back into the incubator at which time the incubation process was repeated until the next time point. The aliquots for each

PAGE 170

170 condition were collected at hours 18, days 1-7, and days 10, 15, 20, 30, 40, 50, and 75. The aliquots for each condition were analyzed on a Perkin Elmer Lambda 3 spectrophotometer at a wavelength of 609nm against a M XN in 0.05M PBS standard curve with concentrations ranging from 1g/mL to 50g/mL and a correlation coeffi cient of 0.99669. Each condition was tested in triplicate and all collected data was statistically analyzed using SigmaStat 3.0 software. Assessment of DNA-MXN-MS cytotoxicity Synopsis. The cytotoxic effect of free MXN, DNA-MXN-MS, and blank DNA-MS on the cellular viability of mLLC cells was evaluated. Free MXN a nd DNA-MXN-MS were tested at MXN doses of 1g/mL, 10g/mL, and 25g/mL and blank DNA-MS we re tested at a concentration of 100g. A colorimetric MTS as say which quantifies cell death by measuring the amount of lactate dehydrogenase (LDH) that is re leased upon cell lyses wa s used to assess the cellular viability of the mLLC.140 The lactate from LDH that is re leased from dead cells reacts with the oxidized form of nico tinamide adenine dinucleotide (NAD+) in the assay to produce pyruvate and the reduced form of nicotinamide adenine dinucleotide (NADH), Figure 5-1.99, 140 NADH then reacts with a tetrazolium salt to form NAD+ and a red formazan dye that is read photometrically at a wavele ngth of 490nm, Figure 5-2.140 The mLLC cells were evaluated at hour 1, and days 1-4. A cytotoxic response was defined by an increase in absorbance at 490nm for any given treatment group.

PAGE 171

171 H3CC H OH COOH Lactate NAD++ O O OH Pyruvate+N A DHO P HO O P O HO O H H O OH OH H N N O O OH OH H N N N N N H+ H H H H O +O P HO O P O HO O H H O OH OH H N N O O OH OH H N N N N N H H H H O + Figure 5-1. The first step in the reaction betw een lactate from the LDH enzyme and NAD+ in the MTS cytotoxicity assay used for these studies.99, 140 N A DH+TetrazoliumSalt N A D++Formazan (red)O P HO O P O HO O H H O OH OH H N N O O OH OH H N N N N N H H H H O + O P HO O P O HO O H H O OH OH H N N O O OH OH H N N N N N H H H H O Figure 5-2. The second and final step in th e MTS assay used for these studies where NADH reacts with the tetrazolium salt to produce NAD+ and a red formazan dye. Culture. The mLLC cells were cultured in 75cm2 cell culture flasks with 20mL of treated media. The cells were incubated at 37C in humid ified air with 5% carbon dioxide for at least 24 hours. After 24 hours, the cells were monitored daily until the cells reached 95% confluency. The cells were then trypsinized and counted to de termine the cell density. If the cell density was below a concentration of 1 x 104 cells/mL, the cells were split and re-cultured until the cell

PAGE 172

172 density was between 1 x 104 cells/mL and 1 x 105 cells/mL. The cell suspension was then used to seed the corresponding 96-well plates used for this study. Cell seeding. A cell density of 9.3 x 104 cells/mL was obtained and 20L of the acquired cell suspension were added to specified wells in five separate 96-well plates. The mLLC cells were seeded at a concentration of 1,860 cells/well which is consistent with cell seeding densities reported in the literature for the time frame used in this study.141, 142 Six well plates were used for this study representing the one control plat e used to measure the absorbance of treatment groups in the absence of cells and the five trea tment plates for the following time points: hours 1, 2, and 6, and days 1 and 2. The five seeded well plates used to evaluate the cytotoxicity of the treatment groups were incubated at 37C in humid ified air with 5% carbon dioxide in order to reach adequate attachment for at least 24 hours befo re adding the first treatment. The media was then removed from the wells and the treatments were added. Treatment groups. The treatment groups for this study evaluated the cellular viability of mLLC cells exposed to free MXN and DNAMXN-MS doses of 1g/mL, 10g/mL, and 25g/mL and blank DNA-MS at a concentration of 100g. Each treatment group was sterilized prior to cytotoxicity analysis. DNA-MXN-MS a nd blank DNA-MS were sterile rinsed with 70% ethanol, vortexed on high for 30 seconds, and cent rifuged at 3,000rpm for 5 minutes. They were then collected by decanting the ethanol and allowe d to air dry over night in a sterilized hood. The sterilized MS were then resuspended in tr eated media to a concentration of 2,000g/mL and added to designated cells using appropriat e volumes to reach desired treatment group concentrations. For the DNA-MXN-MS conditions 4L of the MS/media suspension were added for the1g/mL treatment group, 39L were added for the 10g/mL treatment group, and 96mL were added for the 25g/mL treatment group. For the 100g blank DNA-MS treatment

PAGE 173

173 condition, 50L of the MS/media suspension were added to specified wells. For the free MXN conditions, a stock 0.0001% MXN in treated media solution was prepared and sterilized using a 0.22m sterile syringe filter. The sterile 0.0001% MXN stock solution was then added to specified wells. One microliter was added for the 1g/mL treatment condition, 10L were added for the 10g/mL treatment condition, and 25L were added for the 25g/mL treatment condition. The treated wells were then filled with treated media to a total volume of 100L. There were three control groups used in this study which consisted of 1) treated media in the absence of cells, 2) treated media with cells, 3) treatment groups in treated media in the absence of cells. Absorbance valu es taken of the media and treat ment groups in the absence of cells were used to correct cytotoxicity data obtained in the study. A ll control and treatment conditions were evaluated in re plicates of six. All cellula r viability data collected was statistically analyzed usi ng SigmaStat 3.0 Software. Results and Discussion Particle Analysis Percent yield The percent yield and theore tical yield values for the DNA-MXN-MS prepared at the 20%MEQ, 50% MEQ, and 120%MEQ crosslink concentrations were calculated using Equations 5-2 and 5-3. Each of the three cr osslink conditions tested produce d good yield and theoretical yield values of over 60%. Three batches were prepar ed for each condition to determine the batch-tobatch consistency at each condition. The 120%MEQ crosslink concentration condition produced the smallest yield and theoretical yield standard deviations indicating that it generated the most consistent and reproducible proce ssing condition, Table 5-1. It wa s also noted that the batch-tobatch consistency increased as the crosslink concentration incr eased suggesting that DNA-MXNMS synthesis parameters become more stable and reproducible as the crosslink co ncentration is

PAGE 174

174 increased. This may be attributed to more uniform or homogenous crosslinking of the DNAMXN-MS. Table 5-1. Yield and theore tical yield values for DNA-M XN-MS prepared at the 20%MEQ, 50%MEQ, and 120%MEQ crosslink conditions. Crosslink concentration condition Yield (%) Theoretical yield (%) 20%MEQ 69 1261 11 50%MEQ 84 976 8 120%MEQ 83 375 4 The 50% MEQ and 120%MEQ crosslink conditions produced the highest yield and theoretical yield values of over 80% and 75%, re spectively; however, statistical analysis found no significant differences in yield values among all three crosslink cond itions. The yield and theoretical yields values are depicted graphically in Fi gures 5-3 and 5-4. 0 20 40 60 80 100 20%MEQ 50%MEQ 120%MEQ %Yield Figure 5-3. Yield values for DNA-MXN-MS prepared at the 20%MEQ, 50%MEQ, and 120%MEQ crosslink conditions.

PAGE 175

175 0 20 40 60 80 10020%MEQ50%MEQ120%MEQ%Theoretical Yield Figure 5-4. Theoretical yield va lues include the weight of the in situ loaded MXN for DNAMXN-MS prepared at the 20%MEQ, 50%MEQ, and 120%MEQ crosslink conditions. The yield values of blank DNA-MS were compar ed to the theoretical yield values of DNAMXN-MS for further comparisons, Figure 5-5. Gra phical representation of the yields illustrated no significant differences between blank a nd MXN-loaded DNA-MS indicating that the in situ loading of MXN did not affect the yield values. 0 20 40 60 80 10020%ME Q 50%ME Q 120%ME Q %Yield Blank DNA-MS DNA-MXN-MS Figure 5-5. Graphical comp arison of blank DNA-MS yiel d and DNA-MXN-MS theoretical yield values.

PAGE 176

176 Dry particle size The dry particle diameters and size distribut ions of the DNA-MXN-MS were obtained in methanol. DNA-MXN-MS prepared at the 120%MEQ crosslink condition produced the most narrow size distributions with mean diameters of 2.1m and st andard deviations of 2.8m, Figure 5-6. These findings are consistent with data obtained in the Mixer Study in Chapter 4 which demonstrated that the 120%MEQ crosslink condition at the 1550rpm mixer speed produced the most narrow and normalized particle size distribution. 0 1 2 3 4 0.010.1110100100010000Particle Size (m)Volume % 120%MEQ 50%MEQ 20%MEQ Figure 5-6. A dry particle size distribution comparison of DNA-MXN-MS synthesized at varying crosslink concentrations. Each crosslink condition produced mean diameter values of less than 5m with narrow and fairly normalized particle size distributions, Table 5-2. These findings are also consistent with previous data obtained in Chapte r 4 which indicated that particle size distributions normalize as stir speed is increased due to the high shear forces that reduce aggregate formation during synthesis and produce smaller diameter particles.11, 57, 107, 130, 135 A one way ANOVA illustrated no significant mean diameter size differences among each condition tested. This is also consistent with current literature which states that crosslink concentra tion or density does not

PAGE 177

177 affect yielding particle size values.11, 57, 132, 143, 144 The collected data also illustrated that the 120%MEQ crosslink concentration produced the largest percentage of particles in the mesosphere size range and the lowest percentage of par ticles with diameters greater than 10m. Table 5-2. Dry mean particle diamet er and size range va lues for DNA-MXN-MS. Crosslink concentration condition DNA-MXN-MS dry mean particle diameter (m) DNA-MXN-MS in 1m to 10m size range (%) DNA-MXN-MS larger than 10m (%) 20%MEQ 2.9 4.8 55 4 50%MEQ 4.2 6.4 55 10 120%MEQ 2.1 2.8 61 1 The dry mean diameter values for DNA-MXN-MS were compared to dry mean diameter values obtained in Chapter 4 for blank DNA-MS, to determine if MXN in situ loading affects MS particle diameter values or size distribut ions, Table 5-3. A one way ANOVA followed by a Tukey’s test for multiple comparisons illustrated a significant difference between the mean dry diameter values of DNA-MXN-MS and blank DNA-MS at the 20%MEQ crosslink condition (p = 0.023). This difference in size was mainly attr ibuted to aggregates that formed during the synthesis of blank DNA-MS at the 20%MEQ crosslink concentration c ondition. Thus the data obtained from the statistical analysis indicated that the in situ loading of MXN did not affect the particle diameter of resulting DNA-MXN-MS. This data reflects previous research and current literature which state that the in corporation of MXN does not aff ect resulting albumin, gelatin, chitosan, and gelatin-car boxymethyl cellulose blende d MS diameter values.11, 57, 143, 145 Mean dry particle diameter values for DNA-MXN-MS and blank DNA-MS for each crosslink condition are given in Table 5-3 and gr aphical comparisons are give n in Figures 5-7 to 5-9. Table 5-3. Dry mean particle diameter values for DNA-MXN-MS and blank DNA-MS. Crosslink concentration condition DNA-MXN-MS dry mean particle diameter (m) Blank DNA-MS dry mean particle diameter (m) 20%MEQ 2.9 4.8 17.9 21.9 50%MEQ 4.2 6.4 3.3 3.7 120%MEQ 2.1 2.8 2.2 2.1

PAGE 178

178 0 1 2 30.010.1110100100010000Particle Size ( m ) Volume % Blank DNA-MS DNA-MXN-MS Figure 5-7. A comparison of DNA-MXN-MS and blank DNA-MS dr y particle size distributions at the 20%MEQ crosslink condition. 0 1 2 3 4 0.010.1110100100010000Particle Size (m)Volume % Blank DNA-MS DNA-MXN-MS Figure 5-8. A comparison of DNA-MXN-MS and blank DNA-MS dr y particle size distributions at the 50%MEQ crosslink condition.

PAGE 179

179 0 1 2 3 4 0.010.1110100100010000Particle Size (m)Volume % Blank DNA-MS DNA-MXN-MS Figure 5-9. A comparison of DNA-MXN-MS and blank DNA-MS dr y particle size distributions at the 120%MEQ crosslink density condition. Hydrated particle size DNA-MXN-MS hydrated particle diameters and si ze distributions were obtained in 0.05M PBS at a pH of 7.4. DNA-MXN-MS prepared at the 50%MEQ and 120%MEQ crosslink concentration conditions displayed excellent disp ersion behavior in PBS and exhibited hydrated diameters of less than 20m. DNA-MXN-MS prepared at the 20%MEQ condition gelled together and aggregated upon exposure to PBS and thus we re unable to undergo hydrated particle size analysis. A t-test illustrated no significant diffe rences between the hydrated diameter values for the 50%MEQ and 120%MEQ crosslink conditions, suggesting th at the crosslink concentration did not influence particle swelling. This finding was supported by th e narrow hydrated particle size distributions each condition produced with most of the swollen particle diameters falling within the 5m to 20m range, Figure 5-10. These result s are contradictory to what is currently found in the literature which states that the swelling properties of part icles and films are restricted by increasing the crosslink c oncentration or density.57, 137, 146 Thus it would be expected that the DNA-MXN-MS prepared at the 120%MEQ crosslink concentrati on would exhibit smaller

PAGE 180

180 hydrated particle sizes than part icles prepared at the lower cr osslink concentrations. These findings suggest that maximum crosslinking occurs at 50%MEQ. 0 2 4 60.010.1110100100010000Particle Size (m)Volume % 120%MEQ 50%MEQ Figure 5-10. Hydrated partic le size distributions for DNA-M XN-MS prepared at the 50%MEQ and 120%MEQ crosslink density conditions. DNA-MXN-MS were further charact erized by comparing the hydrat ed size distributions to the dry size distributions, Figures 5-11 and 5-12. Both conditi ons produced distributions that peaked at about the 20m particle size. The swelling properties at each DNA-MXN-MS crosslink concentration were then measured by comparing the hydrated mean particle sizes to the dry mean particle sizes. These values were used to calculate the percent change in size using Equation 5-4. Calculated values illustrate d that the DNA-MXN-MS prepared at the 50%MEQ and 120%MEQ crosslink conditions exhibite d a 200% and 450% increase in size, Table 5-4. Table 5-4. Mean dry and hydrated particle diameter s with percent change in size due to swelling. Crosslink concentration condition Mean dry particle diameter m) Mean hydrated particle diameter (m) Percent change in size (%) 50%MEQ 4.2 6.412.8 11.1 (+) 200 120%MEQ 2.1 2.811.5 10.4 (+) 450

PAGE 181

181 0 1 2 3 4 5 0.010.1110100100010000Particle Size ( m ) Volume % Dry Hydrated Figure 5-11. A dry and hydrated particle size distribution comparison of DNA-MXN-MS at the 50%MEQ crosslink condition. 0 1 2 3 4 5 0.010.1110100100010000Particle Size (m)Volume % Dry Hydrated Figure 5-12. A dry and hydrated particle size distribution comparison of DNA-MXN-MS at the 120%MEQ crosslink condition.

PAGE 182

182 Surface charge analysis The effect of in situ MXN loading on yielding DNAMXN-MS surface charge was measured in 0.01M PBS at a pH of 7.4. Zeta potential values for DNAMXN-MS illustrated over a 40% decrease indicating the successful loadi ng of the cationic drug MXN, Table 5-5. Zeta potential values for DNA-MXN-MS ranged from -13.2mV to -28.3mV which was a decrease from the -46.3mV to -55.2mV values obtaine d for blank DNA-MS prepared with the same conditions. T-tests conducte d on the zeta potential values obtained for DNA-MXN-MS and blank DNA-MS at each crosslink co ncentration illustrate d that the incorporat ion of the cationic MXN significantly decreased the surface charge of the DNA-MS (p < 0.001), Figures 5-13 to 5-15. These findings are consiste nt to trends found in current literature which state that the charge of a particle will change upon the loading of a cationic or anionic drug.10, 147, 148 Table 5-5. Zeta potential values for DNA-M XN-MS and blank DNA-MS with their respective change in surface charge. Crosslink concentration condition Zeta potential for DNA-MXN-MS (mV) Zeta potential for blank DNA-MS (mV) Percent change in surface charge (%) 20%MEQ -13.2 3.7 -46.3 2.4 (-)71 50%MEQ -28.7 2.1 -48.0 3.4 (-) 40 120%MEQ -24.2 4.0 -55.2 3.0 (-) 56 -50 -30 -101Zeta Potential (mV) Blank DNA-MS DNA-MXN-MS Figure 5-13. A zeta potential comparison of DNA-MXN-MS and blank DNA-MS prepared at the 20%MEQ crosslink condition.

PAGE 183

183 -60 -40 -20 01Zeta Potential (mV) Blank DNA-MS DNA-MXN-MS Figure 5-14. A zeta potential comparison of M XN loaded and blank DNA-MS prepared at the 50%MEQ crosslink condition. -60 -40 -20 01Zeta Potential (mV) Blank DNA-MS DNA-MXN-MS Figure 5-15. A zeta potential comparison of MXN loaded and blank DNA-MS prepared at the 120%MEQ density condition. A one way ANOVA followed by Tukey’s test for multiple comparisons conducted on the zeta potential values obtained for the DNA-M XN-MS conditions illustrated that the zeta potential values for the 20%MEQ crosslink condition were sign ificantly lower than the 50%MEQ and 120%MEQ crosslink conditions (p < 0.001), Figur e 5-16. A lower zeta potential value

PAGE 184

184 explains why the 20%MEQ condition coagulated once it was introduced to PBS during hydrated particle size analysis. The surface charge of a pa rticle, as measured by zeta potential, determines its colloidal stability by determining its net inte rparticle forces while in an aqueous dispersion.149 A particle with a large zeta poten tial value with have a high magn itude of interparticle repulsive forces leading to a stable dispersion.107, 149 DNA-MXN-MS prepared at the 20%MEQ condition had a zeta potential value of 13.2mV which was lower than the zeta potential values obtained with the 50%MEQ and 120%MEQ crosslink conditions. Therefore the 20%MEQ crosslink condition was unable to produce suffi cient interparticle repulsive forces to create a stable dispersion. -35 -25 -15 -51Zeta Potential (mV) 20%MEQ 50%MEQ 120%MEQ Figure 5-16. Zeta potential va lues for DNA-MXN-MS prepared at varying crosslink conditions. Scanning electron microscopy SEM images confirmed results obtained with dry particle size an alysis and depicted particles that were under 5m and had very narrow size distributions, Figure 5-17. SEM images

PAGE 185

185 also illustrated that the DNA-MXN-MS produced discrete and spherical particles with smooth surface topographies. A B C A B C Figure 5-17. Scanning electron micrographs of DNA-MXN-MS prepared at the A) 20%MEQ (Magnification: 2,000x), B) 50%MEQ (Magnification: 3,000x), and C) 120%MEQ crosslink conditions (M agnification: 2,000x). Energy dispersive x-ray spectroscopy Elemental analysis by EDS was conducted to confirm the presence of the gadolinium in the DNA-MXN-MS. EDS analysis illustrated that the DNA-MXN-MS followed the same increase in gadolinium peak trends as seen in the crosslink concentr ation study. DNA-MXN-MS prepared at the 120%MEQ condition produced spectra with higher x-ray counts of gadolinium

PAGE 186

186 and chlorine than seen in the 50%MEQ and 20%MEQ crosslink conditions demonstrating an increase in the crosslink agent concentration at the higher crosslink condit ion, Figures 5-18 to 520.110 Figure 5-18. EDS spectra of DN A-MXN-MS prepared at the 20%MEQ crosslink condition. Figure 5-19. EDS spectra of DN A-MXN-MS prepared at the 50%MEQ crosslink condition.

PAGE 187

187 Figure 5-20. EDS spectra of DN A-MXN-MS prepared at the 120%MEQ crosslink condition. In Vitro DNA-MXN-MS Characterization MXN loading efficiency The MXN percent loadings and loading e fficiencies of the DNA-MXN-MS crosslink condition were measured in triplicate by enzyma tic digestion followed by photometric analysis. MXN loadings were calculated using Equation 5-5 and ranged from 10% (w/v) to 13.5% (w/v) for each of the DNA-MXN-MS crosslink condit ions tested, Table 56. A one way ANOVA conducted on all DNA-MXN-MS data illustrated that the 50%MEQ crosslink condition loaded a significantly lower amount of MXN than the 20%MEQ condition (p = 0.021), but not the 120%MEQ condition, Figure 5-21. MXN loadings fo r DNA-MXN-MS were also comparable to MXN loadings previously reported for al bumin (BSA-MS) and gelatin (GEL-MS) MS.11, 57 A one way ANOVA conducted on all collected D NA-MXN-MS data and data obtained from previous studies conducted on BSA-MS and GEL-MS showed no significant MXN loading

PAGE 188

188 differences between the 20%MEQ (p = 0.055), 50%MEQ (p = 0.076), and 120%MEQ (p = 0.666) DNA-MXN-MS conditions and the BSAMS or GEL-MS conditions. Table 5-6. MXN loading and loading effici encies for DNA-MXN-MS, BSA-MS, and GEL-MS. Crosslink concentration condition MXN lo ading (%) MXN loading efficiency (%) 20%MEQ 13.5 0.4 91.2 2.6 50%MEQ 10.4 0.3 71.1 2.3 120%MEQ 12.4 1.7 84.3 11.2 Hadba albumin MS11 12.2 0.2 81.3 1.2 Cuevas gelatin MS57 11.5 1.3 77 9 0 5 10 1520%MEQ 50%MEQ 120%MEQ HADBA BSA-MS CUEVAS GEL-MS%MXN Loadin g Figure 5-21. Percent MXN loading comparis ons of DNA-MXN-MS, BSA-MS, and GEL-MS. MXN loading efficiencies were calculated using Equation 5-6 and data collected demonstrated that the DNA-MXN-MS produced loading efficiencies of over 70% for all crosslink conditions te sted with the 20%MEQ crosslink condition produc ing the highest loading efficiency with a value of 91%, Table 5-6. This data is contradictory to certain citations in the literature which state that increasing the cross link concentrations increases drug loading and therefore the loading efficiency in microspheres.137, 144 Other sources and pr evious research in this lab have stated that crosslink concentrati on has no statistically signi ficant effect on the drug loading efficiency of albu min and gelatin microspheres.11, 57, 143 Results obtained from a one way

PAGE 189

189 ANOVA followed by a Tukey’s multiple comparisons test was supported by the latter claim and illustrated that the loading efficiency of the 50%MEQ crosslink condition wa s significantly lower than the 20%MEQ condition (p = 0.008), but not the 120%MEQ condition, indicating no true correlation between crosslink concentration a nd loading efficiency, Figure 5-22. A one way ANOVA also showed no significant MXN loadi ng efficiency differences between the DNAMXN-MS and the BSA-MS or GEL-MS. 0 20 40 60 80 10020%MEQ50%MEQ120%MEQHADBA BSA-MS CUEVAS GEL-MSMXN Loading Efficiency (%) Figure 5-22. MXN loading efficiency comp arisons for DNA-MXN-MS, BSA-MS, and GELMS. In vitro MXN release The in vitro MXN release properties of DNA-MXN-MS were measured in 0.05M PBS at a pH of 7.4. The DNA-MXN-MS were tested in triplicate using mini mum sink conditions and were incubated in 1. 250mL of PBS at 37 C to simulate the tumor environment. Each DNAMXN-MS condition tested released over 30% of the entrapped MXN within the first 24 hours producing a “burst effect” as shown in Figure 5-23. DNA-MXN-MS at each crosslink condition continued to release MXN for the next 75 hours; however, a one way ANOVA followed by a

PAGE 190

190 Tukey’s multiple comparison test illustrated that the 120%MEQ crosslink condition was the only condition to produce a significant increase in MXN release between Hour 1 and Day 1 (p = 0.002), Figure 5-24. 0 20 40 60 80 100 0510152025Time (hrs)% MXN Released 120%MEQ 50%MEQ 20%MEQ Figure 5-23. First 24 hour MXN release prof ile for DNA-MXN-MS at varying crosslink concentrations. 0 20 40 60 80 100Hour 1Da y 1Da y 75%MXN Released 20%MEQ 50%MEQ 120%MEQ Figure 5-24. MXN release profiles for DNA-MX N-MS at Hour 1, Day 1, and Day 75.

PAGE 191

191 There were no significant incr eases in MXN release between Day 1 and Day 75 for each DNA-MXN-MS crosslink condition tested sugges ting that the loaded MXN is almost fully released within the first 24 hours, Figure 5-25. Previous research conduc ted on the release of MXN from BSA-MS and GEL-MS also found MXN release to cease within 24 hours.11, 57 However, it should be noted that the DNA-MXNMS were not fully degraded by Day 75 and were still blue in color suggesti ng that the MXN had not fully rele ased. It is assumed that under in vivo conditions, the MXN release pr ofiles would differ due to na turally occurring digestive enzymes in the body which may facilitate in the release of MXN past Day 1. 0 20 40 60 80 100 01020304050607080Time ( Da y s ) % MXN Released 120%MEQ 50%MEQ 20%MEQ Figure 5-25. Total MXN release profile fo r varying crosslink DNA-MXN-MS conditions. Further statistical analysis using a on e way ANOVA followed by a Tukey’s test for multiple comparisons illustrated that the 20%MEQ crosslink condition releas ed significantly more MXN than the 50%MEQ (p = 0.030) and the 120%MEQ (p = 0.040) crosslink conditions. These findings are also consistent w ith previous research and curr ent literature which state that increasing the crosslink concentrat ion or density of a drug carrier will decrease or decelerate its drug release.10, 11, 57, 145, 146, 150-152 Drug release is diffusion dr iven and is dependent on the

PAGE 192

192 swelling properties and biodegradation of the MS carrier.67, 153 MS carriers with high swelling properties will allow the aqueous environment to penetrate the MS thus releasing the drug from the matrix quicker, Figure 5-26. Higher swel ling properties also enable the MS carrier to solubilize or biodegrade more rapidly within the aqueous media. Therefore, swelling and biodegradation are reduced by increases in cro sslink concentration, and thus it would be expected that the 20%MEQ crosslink condition would rel ease MXN faster than the 50%MEQ and 120%MEQ crosslink concen tration conditions.133, 144, 154 AB AB Figure 5-26. Schematic represen tation of crosslink concentra tion on drug release, A) high crosslink and B) low crosslink conditions. Assessment of DNA-MXN-MS cytotoxicity The cytotoxicity of DNA-MXN-MS on mLLC ce lls in culture was tested on Hour 1 and Days 1 through 4 using an MTS cytotoxicity a ssay. An increase in absorbance at 490nm indicated increased cell death. Each condition was tested in replicates of 6 at DNA-MXN-MS and free MXN doses of 1g/mL, 10g/mL, and 25 g/mL and DNA-MS concentration of 100g. Data obtained from the MTS assay was statisti cally analyzed using a one way ANOVA followed by a Tukey’s test for multiple comparisons. Cytotoxicity data obtained illustrated that the 25g/mL DNA-MXN-MS condition produced the most cytotoxic response on the mLLC cells as

PAGE 193

193 compared to the media with cells control by Day 4 (p < 0.001), Figure 5-27. Additionally, there were no significant differences in cytotoxicity responses among the 25g/mL, 10g/mL, and 1g/mL conditions at Day 4. 0 1 2 3 4 5Hr1Da y 1Da y 2Da y 3Da y 4Absorbance @ 490nm Media w/Cells 25g/mL MXNMS 10g/mL MXNMS 1g/mL MXNMS 100g DNA Figure 5-27. In vitro cytotoxicity profiles of DNA-MXNMS and DNA-MS on mLLC cells. There were no significant differe nces in cytotoxicity at H our 1 between th e control and DNA-MXN-MS dose conditions, however, by Day 1, cytotoxic responses were observed for the 10g/mL (p = 0.021) and 25g/mL (p = 0.004) DNA-MXN-MS dose conditions. The 1g/mL DNA-MXN-MS dose did not elicit a cytotoxic res ponse until Day 2 (p < 0.001) illustrating the cytotoxicity of the mLLC cells was dose depe ndent. Similar MXN dose responses on the in vitro viability of rat glioma, human myeloid leukemi a, B-chronic lymphocytic leukemia and breast cancer cells have been re ported in the literature.13, 57, 155-157 The 100g DNA-MS elicited similar cytotoxicity responses up to Day 2; however by Day 3 and 4, there were no significant differences between this condition and the media with cells control. These results are consistent with fibroblast growth data obtained in the blank DNA-MS study which illustrated that cell

PAGE 194

194 growth decreased upon addition of treatment, howev er, once the culture e nvironment stabilized, cell growth increased and produced no significant growth differen ces as compared to the media with cells control. Free MXN dose conditions exhibited sim ilar trends as the DNA-MXN-MS dose conditions, however, by Day 4, a significant cytotoxicity response was found only for the 25g/mL free MXN condition (p = 0.023) as compared to the media with cells control condition, Figure 5-28. 0 0.5 1 1.5 2 2.5 3Hr1Day1Day2Day3Day4Absorbance @ 490nm Media w/Cells 1g/mL FMXN 10g/mL FMXN 25g/mL FMXN 100g DNA Figure 5-28. Cytotoxicity prof iles for free MXN dose conditions. Further data analysis also indicated that DNA-MXN-MS c onditions elicited significantly higher cytotoxicities than the free MXN conditions at each dose tested (p < 0.05). This data suggested one of two things: 1) gadolin ium crosslinked DNA contributes to the in vitro cytotoxicity of DNA-MXN-MS or 2) DNA-MXN-MS are more readily taken up by the mLLC cells than free MXN.

PAGE 195

195 Conclusions The objectives of these studies were to prepare MXN in situ loaded DNA-MS with narrow and controlled size distributions where 60% of all particles prepared were within the mesosphere size range of 1m to 10m and < 5% of all particle s were greater than 10m in size. Particles less than 1m in diameter were also acceptable a nd hydrated particle diameters were to be less than 25m. In addition, MXN in situ loaded DNA-MS were sought to obtain loadings of 12% (w/w) MXN and release MXN for 24 hours or more in phosphate buffered saline under minimum sink conditions. DNA-MXN-MS at M XN concentrations of 1g/mL, 10g/mL, and 25g/mL were also sought to elicit a cytotoxic response to that of free drug at the same MXN concentrations on murine Lewis lung carcinoma (mLLC) cells in culture. DNA-MXN-MS were prepared with three diffe rent crosslink concentrations to also determine the effect of crosslink concentrati on on MXN loading and release. The resulting DNA-MXN-MS were characterized by yield, pa rticle size, elemental analysis, surface topography and morphology. DNA-M XN-MS also were evaluated in vitro for MXN loading, loading efficiency, release, and cytotoxicity on mLLC cells. Particle Analysis DNA-MXN-MS were prepared with good yields of over 60%, with narrow and controlled dry and hydrated size distributions. Resulting DNA-MXN-MS also produced discreet and spherical particles with smooth surface topographi es. EDS analysis confirmed the presence of gadolinium in each of the DNA-MXN-MS crosslink concentration conditi ons tested, with the 120%MEQ condition producing the largest counts of both gadolinium and chlorine. DNA-MXN-MS prepared with the 120%MEQ crosslink concentration produced the best dry and hydrated particle sizes with a dry mean di ameter of 2.1m, a hydrated mean diameter of 11.5m, 61% of all particles within the mesosphere size range, 1% of all particles greater than

PAGE 196

196 10m, and were readily dispersibl e and stable in PBS. The 20%MEQ crosslink concentration produced the least favorable DNA-MXN-MS conditi ons by coagulating immediately after being dispersed into PBS. A one way ANOVA illust rated no significant differences between the hydrated diameters for the 50%MEQ and 120%MEQ crosslink concentrations suggesting that maximum crosslinking may be achieved at the 50%MEQ concentration. In Vitro MXN Loading and Release Zeta potential analysis confirmed the loadi ng of the cationic drug MXN into each of the three DNA-MXN-MS crosslink concentration conditions tested by producing a significant increase in DNA-MS surface charge (p < 0.001). MXN loadings ranging from 10.4% to 13.5% and loading efficiencies rangi ng from 70% to 91% at each of the DNA-MXN-MS crosslink conditions tested were found to be statistically comparable to those obtained with BSA-MS or GEL-MS. MXN release was measured up to 75 days for each DNA-MXN-MS crosslink concentration condition tested; however, each cond ition produced a “drug burst” release within the first 24 hours. The 20%MEQ crosslink concentra tion condition produced the highest release of the three conditions tested; however, the 120%MEQ crosslink concentration which was the only condition to produce a significant increase in release between Day 1 and Day 75. In Vitro Cytotoxicity Analysis Data obtained from the cytotoxic ity analysis concluded that the in vitro viability of mLLC cells was dose dependent. The 10g/mL and 25g/mL DNA-MXN-MS dose conditions exhibited cytotoxicity respons es by Day 1; however, the 1g/ mL DNA-MXN-MS dose did not elicit a cytotoxic respon se until Day 2. Free MXN dose cond itions exhibited similar trends as the DNA-MXN-MS dose conditions. Cytotoxicity data also concluded that by Day 4, a significant cytotoxicity response was found only for the 25g/mL free MXN and DNA-MXNMS conditions.

PAGE 197

197 Additionally, the DNA-MXN-MS conditions elic ited significantly higher cytotoxicities than the free MXN conditions at each dose tested suggesting one of two things: 1) gadolinium crosslinked DNA contributes to the in vitro cytotoxicity of DNA-MXN-MS or 2) DNA-MXNMS are more readily taken up by the mLLC cells than free MXN.

PAGE 198

198 CHAPTER 6 IN VITRO EVALUATIONS OF DRUG LOAD ED DNA AND BSA NANO-MESOMICROSPHERES Introduction Deoxyribonucleic acid (DNA) a nd bovine serum albumin (BSA) nano-meso-microspheres (MS) were in situ loaded with methotrexate (MTX) and 5-fluorouracil (5-FU) using optimized DNA-MS synthesis conditions pres ented in Chapter 4 to furt her analyze the drug loading capabilities of DNA and BSA. MTX and 5-FU in situ loaded DNA-MS and BSA-MS were synthesized to produce particles with controlled size distributions where at least 60% of all particles prepared were within the mesosphere size range of 1m to 10m and < 5% of all particles were greater than 10m in size. Pa rticles less than 1m in diameter were also acceptable and hydrated particle di ameters were to be less than 25m. In addition, DNA-MS and BSA-MS were sought to obt ain drug loadings of 5% (w/w) MTX or 5-FU and release drug for more than 24 hours in phosphate buffered sa line under minimum sink conditions. Drug loaded BSA-MS and DNA-MS were compared with respect to particle diameter, size distribution, morphology, topography, drug loading, and percent drug release. Mitoxantrone (MXN) in situ loaded DNA-MS were also prepared to determine the maximum drug loading ability of DNA. D NA-MXN-MS were prepared with a 120%MEQ gadolinium crosslink concentra tion and were loaded with 10 % (w/w), 15% (w/w), and 25% (w/w) MXN. The particle diameter, size dist ribution, morphology, topography, drug loading, and percent drug release of the DNA-MXN-MS were evaluated with respect to MXN concentration. DNA-MXN-MS were also pr epared with no gadolinium crosslinking to determine if MXN serves as a crosslinking agent to DNA-MS. DNA-MXN-MS were prepared to obtain in situ MXN loadings of 10% (w/w). The DNA-MXN-MS were characterized as mentioned above and compared with respect to crosslinking.

PAGE 199

199 The particle diameters and size distributions were obtained using an LS Coulter 13 320 particle size analyzer. The pa rticle morphologies and topographi es were assessed using a field emission scanning electron microscope. Drug lo ading was determined by incubating the drug loaded MS under stirred conditions in an enzy matic digestion buffer at 37C for 48 hours. The released drug concentrations were then anal yzed using UV-visible sp ectroscopy against a MTX, 5-FU, or MXN standard curve. The percen t drug release was measured under minimum sink conditions in order to simulate the tumor e nvironment. Drug release data was obtained by incubating the drug loaded MS in phosphate buffe red saline under constant agitation at 37C for a minimum of fourteen days in triplicate. At specific time points, aliquots were taken and measured using UV-visible spectroscopy agains t a MTX, 5-FU, or MXN standard curve to determine drug concentration. Materials and Methods Materials DNA sodium salt derived from herring testes Type XIV, albumin from bovine serum (BSA), cellulose acetate butyrate, HPLC grad e 1,2-dichloroethane, methanol, gadolinium (III) chloride hexahydrate, 25% (w/w ) Grade II aqueous glutaralde hyde solution, mitoxantrone dihydrochloride, methotrexate hydr ate, 5-fluorouracil, L-cysteine hydrochloride hydrate, papain from papaya latex, deoxyribonuclease I from bovine pancreas, proteinase from Bacillus licheniformis, ethylenediaminetetraacetic acid disodium salt dihydrate (EDTA), and trichloroacetic acid were purchased from th e Sigma-Aldrich Company. Sodium phosphate monobasic monohydrate, sodium phosphate diba sic anhydrous, and sodium chloride, each A.C.S. certified, acetone, 70m Spectra /Mesh Nylon filters, 16x125mm and 13x100mm borosilicate glass test tubes, Ultrafree-CL with Durapore Memb rane Centrifugal Filter tubes (0.1m pore, 2mL), Corning sterile 0.45m pore filter bottles, Fisherbrand Semimicro

PAGE 200

200 methacrylate 1.5mL disposable cuvettes, and 15mL and 50mL polypropy lene centrifuge tubes were purchased from Fisher Scientific Internat ional. Type I and Type II deionized ultrapure water was prepared with a re sistivity of at least 16 M -cm-1 using the Barnstead NANOpure Ultrapure Water System in the lab. Methods Solution preparation MTX and 5-FU in situ loaded DNA and BSA. MTX and 5-FU were in situ loaded into 5% (w/v) aqueous solutions of DNA or BSA at concentrations of 15% (w/w) and 30% (w/w). Aqueous DNA and BSA solutions were prepared at room temperature by adding 0.5g of DNA or 0.5g of BSA to 5mL of ultrapure water in a 50m L polypropylene centrifuge tube. The solutions were then mixed on a vortex for 30 seconds. Thr ee milliliters of ultrapure water were then added to the DNA or BSA solutions and placed on the rotary shaker for at least two hours until the DNA or BSA had completely dissolved. The vol ume was then brought up to 10mL and vortexed on high for 30 seconds. The DNA or BSA solution wa s then placed in the re frigerator over night to ensure the complete collapse of bubbles gene rated during vortex and rotary mixing. The percent solid concentration of the DNA or BS A solution was quantified using a Metler LJ16 Moisture Analyzer at 130C for 60 minutes. On ce the concentration was determined, 3.5mL of the DNA or BSA solution was placed into a 15mL pol ypropylene centrifuge tube. Using the true solid weight of DNA or BSA in solution obtaine d during moisture analysis, the weight of the MTX or 5-FU, XgDrug, needed to obtain 15% (w/w) or 30% (w/w) loadings was calculated by multiplying the true solid DNA or BS A weight in 3.5mL of solution (WTSOLID) by 15 or 30. The product was then divided by 100, Equations 6-1 and 6-2. DNAorBSA 100g W Drug 15g Drug XgTSOLID (6-1)

PAGE 201

201 DNAorBSA 100g W Drug 30g Drug XgTSOLID (6-2) The desired mass of MTX or 5-FU was weighe d and added to the 3.5mL of the DNA or BSA solution. The solutions were then vorte xed on high for 30 seconds, placed on the rotary mixer for 2 hours and then placed in the refrigerator until further use. MXN in situ loaded DNA. Based on data obtained in Chapter 5, 15% (w/w) MXN in situ loaded DNA solutions were prepared to determin e if MXN serves as a crosslinking agent for DNA-MXN-MS. Aqueous DNA solutions were in situ loaded as outlined in the Methods Section of Chapter 5. Aqueous DNA solutions were also in situ loaded with 10% (w/w) and 25% (w/w) MXN to determine the maximum drug loading ability of DNA. Aqueous DNA solutions were prepared and the percent solid concentra tions were determined as men tioned above. Three and a half milliliters of the DNA solution was then placed into a 15mL polypropylene centrifuge tube and the weight of the MXN, XgMXN, needed to obtain 10% (w/w) or 25% (w/w) loadings was calculated by multip lying the true solid DNA weight, WTDNA, in 3.5mL of solution by 10 or 25, Equations 6-3 and 6-4. DNA 100g W MXN 10g MXN XgTDNA (6-3) DNA 100g W MXN 25g MXN XgTDNA (6-4) Enzymatic digestion buffers. Enzymatic digestion buffers were used to digest the drug loaded DNA-MS and BSA-MS to determine their drug loadings and loading efficiencies. The DNA enzymatic digestion buffer was prepared by adding 1,800mg of EDTA, 200mg of Lcysteine hydrochloride hydrate, 125mg of papain from papaya latex, and 125mg of

PAGE 202

202 deoxyribonuclease I from bovine pa ncreas to 250mL of the 0.05M PBS solution (pH = 7.4). The BSA enzymatic digestion buffer was prepared th e same way; however, 125mg of proteinase from Bacillus licheniformis was added to the buffer in place of the deoxyribonuclease. The enzymatic digestion buffer components were s tirred at room temperature until completely dissolved and used immedi ately after preparation. Additional synthesis and ch aracterization solutions. Solutions of cellulose acetate butyrate in 1,2-dichloroethane (CAB) were us ed as the water-immiscible continuous phase during DNA-MXN-MS synthesis. The CAB solutions we re used at a concentration of 5% (w/v). Aqueous solutions of 0.1M gadolinium (III) chloride were used as the ionic crosslinking agent for all DNA-MS syntheses. All BSA-MS synthe ses were conducted with covalent crosslinking using aqueous 4% (w/v) glutaralde hyde solutions. Phosphate bu ffered saline (PBS) solution at a pH of 7.4 and a concentrati on of 0.05M was used for all in vitro drug release analysis. Solutions mentioned here were prepared using the same methods as outlined in Chapter 5. Synthesis procedure Drug loaded DNAMS and BSA-MS were prepared using optimized synthesis conditions from Chapter 4 to obtain particles with particle diameters in the range of 50nm to 10m and normal and narrow size distributions. A modified emulsion stabilization technique was used that sterically stabilizes the DNA or BSA molecule into spherical conformations and crosslinks them while in suspension. The s ynthesis process involved dispersing 3mL of the aqueous drug loaded DNA or BSA solutions (i.e. the aqueous phase) into 47mL of the organic CAB solution (i.e. the continuous phase) in a 300mL Labconco lyophilization flask. A DNAMS or BSA-MS microemulsion was then creat ed by vigorously mixing the two solutions at 1550rpm for 20 minutes at room temperature using a paddle mixe r with a two inch, two blade propeller. The DNA-MS microemulsions were then ionically crosslinked while in suspension by

PAGE 203

203 reducing the speed of the paddle mixer to 600r pm and adding 2mL of the 0.1M gadolinium (III) solution to obtain a 120%MEQ crosslink concentration. The BS A-MS were similarly crosslinked, however, 2mL of the 4% (w/v) aqueous glut araldehyde solution was added. The MS microemulsions then underwent crosslinking for 1 hour and 40 minutes at which time 50mL of acetone was added and any further reactions were allowed to reach completion for another hour. After synthesis was complete, the drug loaded DNAMS or BSA-MS underwent four rinses in acetone to remove any residual or ganic phase, crosslinking agent, or non-loaded drug. The MS were rinsed by separating the resultant DNAMS or BSA-MS suspension into four separate 50mL polypropylene centrifuge tube s. Acetone was added to 40mL and the tubes were capped and vortexed on high for 30 seconds. The MS were then collected by cent rifuging the tubes at 2600rpm for 10 minutes and decanting the acetone Fresh acetone was added again to 40mL mark and the acetone rinse was repeated once more as mentioned above and then twice more by consolidating the contents of 4 tubes to 2 tubes and then 2 tubes to one tube. After the final acetone rinse, the drug loaded DNA-MS were resusp ended in 30mL of acetone, vortexed on high for 30 seconds, and filtered using a stainless steel vacuum filtrati on device with a 70m Spectra/Mesh Nylon filter. A 70m Spectra/Mes h Nylon filter was used in place of the 20m filter to not completely eliminate the possible si ze effects of drug loadings on the DNA-MS. The device was then rinsed with 10mL of acetone to further filter any remaining DNA-MS under 70m. The centrifuge tube was then capped, vor texed on high for 30 seconds, and centrifuged at 2600rpm for 5 minutes to collect the drug load ed DNA-MS. The acetone was then decanted and a Kimwipe was secured over the mouth of tube using a rubber band to allow the drug loaded DNA-MS to dry overnight at r oom temperature. DNA-MXN-MS prepared without crosslinking agent and drug loaded BSA-MS were not filtered. After the final acetone rinse, the acetone was

PAGE 204

204 decanted and a Kimwipe was secured over the mout h of tubes using a rubber band, and the MS were allowed to dry overni ght at room temperature. Particle characterization Yield analysis. The yields of the resulting drug loaded DNA-MS and BSA-MS were calculated and expressed as percent yield values. The percent yields were calculated by dividing the final weight of the drug loaded DNAMS and BSA-MS by the amount of weight used to synthesize them, Equation 6-5. The weight of the in situ loaded drugs was also added to the equation and expressed as the pe rcent theoretical yield, Equati on 6-6. The equations for the percent yield and percent theoretica l yield are expressed below where WF is the final weight of the MS, V, and C are the volume, density, and con centration of the aqueous DNA or BSA solutions used respectively, WX is the weight of the crosslinking agent added during synthesis, and WY is the theoretical weight of the in situ loaded drug. 100 ) ) W C ((V W ( Yield %X F (6-5) 100 ) ) W W C ((V W ( Yield l Theoretica %Y X F (6-6) Dry particle size analysis. Dry drug loaded DNA-MS and BSA-MS particle diameters and size distributions were obtaine d using a Coulter LS 13 320 partic le size analyzer. MS were sonicated for 10 to 30 seconds to aerate and sepa rate the particles prior to analysis and then approximately 2mg of the MS were suspended in 2mL of methanol. The suspension was then sonicated for 30 seconds to further break up any MS aggregates and tested. The Coulter LS 13 320 particle size analyzer was se t to run at a pump speed of 73% using a protein/DNA particle diffraction model. Standards were tested in methanol before th e first run to ensure that the instrument was performing adequately. Each condition was sampled three times in which each

PAGE 205

205 sample’s particle size measurements consisted of two runs. This method of testing produced six independent particle diameters and size distributions. Data collected from these experiments were statistically analyzed using SigmaStat 3.0 software. Scanning electron microscopy. The morphology and surface topography of the drug loaded DNA-MS and BSA-MS were observed us ing scanning electron microscopy (SEM). Approximately 1mg of dry MS were mounted onto a small piece of silicon wafer which in turn was mounted onto an aluminum SEM stub using doubl e sided tape. The MS were then coated with gold-palladium for 2 minutes using a Technix Hummer V sputter coater. Images were taken on a JEOL 6335F Field Emission SEM at an accelerating voltage of 5KeV and a working distance of 15mm. In vitro drug evaluation procedures MTX, 5-FU, and MXN loading efficiency. The loading efficiencies of the drug loaded DNA-MS and BSA-MS were determined by en zymatic digestion followed by photometric analysis of the recovered entrapped drug. A pproximately 5mg of the drug loaded MS were weighed out into labeled 16x125mm glass test tubes and record ed. Ten milliliters of the enzymatic digestion buffer were then added to the MS and the test tubes were capped, parafilmed, and incubated at 37 C under stirred conditions for 48 hour s. The test tubes were then taken out of the incubator and the solutions were allowed to cool to room temperature. While the solutions were cooling, a 10% (w/v) solution of trichloroacetic acid (TCA) in ultrapure water was prepared. Two milliliters of each drug lo aded DNA-MS or BSA-MS solution were then placed into labeled 13x100mm gla ss test tubes. Two milliliters of the 10% (w/v) TCA solution was then added to each sample. The test tubes were capped and the two solutions were allowed to react for 30 minutes at room temperature. Th e samples were then centrifuged at 2,600rpm for five minutes and the supernatants from each sa mple were collected into 1.5mL methacrylate

PAGE 206

206 disposable cuvettes for UV-Vis analysis. Each sample was analyzed on a Perkin Elmer Lambda 3 spectrophotometer to determine the concentrat ion of the entrapped drug. MTX samples were analyzed at a wavelength of 308nm against a MTX in 5% (w/v ) TCA standard curve with a correlation coefficient of 0.99955. 5-FU sample s were analyzed at a wavelength of 269nm against a 5-FU in 5% (w/v) TC A standard curve with a correlation coefficient of 0.99979 and MXN samples were analyzed at a wavelengt h of 610nm against a MXN in 5% (w/v) TCA standard curve with a correlation coefficient of 0.99966. Concentrations for each of the standard curves ranged from 1g/mL to 50g/mL. The drug concentrations were then used to de termine the percent drug loadings in each of the DNAMS and BSA-MS using Equation 6-7.137, 138 The loading efficiencies were also calculated using Equation 6-8.137-139 Each condition was tested in triplicate and all collected data was statistically analyzed us ing SigmaStat 3.0 software. 100 ) MS BSA or MS DN A o f Mass MS in Drug of Mass ( Loading Drug % (6-7) 100 ) Loading l Theoretica Loading al Experiment ( Efficiency %Loading (6-8) In vitro MTX, 5-FU, and MXN release. The in vitro release of MTX, 5-FU, and MXN from DNA-MS and BSA-MS under minimum sink conditions was test ed in sterile filtered 0.05M PBS at a pH of 7.4. Approximately 2mg of each drug loaded DNA-MS and BSA-MS condition were weighed into labeled 2mL filter centrifuge tubes (0.1m pore size) to which 1.25mL of the sterile filtered 0.05M PBS solution was added. The tubes were then capped, parafilmed, and incubated at 37 C under stirred conditions. At specified time points, the filter centrifuge tubes were removed from the incubato r and centrifuged for 10 minutes at 3,000rpm. Aliquots were then removed and placed into 1.5mL methacrylate disposable cuvettes for UV-Vis analysis. The

PAGE 207

207 cuvettes were capped, labeled, parafilmed, and stored in the refrigerator until analysis. Each centrifuge filter tube was then replenished with 1.25mL of fresh 0.05M PBS, capped, parafilmed, and placed back into the incubator at which tim e the incubation process was repeated until the next time point. The aliquots were collected at hours 1-8, days 1-7, and days 10, 15, 20, 30, 40, 50, and 75 for the 10% (w/w), 15% (w/w), and 25% (w/w) DNA-MXN-MS and the 15% (w/w) DNA-MXN-MS prepared without gadolinium cross linking. Aliquots were also collected at hours 1-8 and days 1, 2, 7, 14, 21, and 35 for the MTX and 5-FU loaded DNA-MS and BSA-MS. The aliquots for each condition were analyzed on a Perkin Elmer Lambda 3 spectrophotometer to determine the drug concentrations at each time point. MTX samples were analyzed at a wavelength of 306nm against a MT X in 0.05M PBS standard curve with a correlation coefficient of 0.9948. 5-FU samples were analyzed at a wavelength of 260nm agai nst a 5-FU in 0.05M PBS standard curve with a correlation coefficient of 0.99110 and MXN samples were analyzed at a wavelength of 609nm agains t a MXN in 0.05M PBS standard curv e with a correla tion coefficient of 0.99669. Concentrations for each standard cu rve ranged from 1g/mL to 50g/mL. Each condition was tested in triplicate and all colle cted data was statisti cally analyzed using SigmaStat. Results and Discussion MTX and 5-FU In Situ Loaded DNA-MS and BSA-MS Particle analysis Percent yield. The percent yield and theoretical yi eld values for all MTX and 5-FU in situ loaded BSA-MS and DNA-MS were calculated using Equations 6-5 and 6-6. BSA-MS that were in situ loaded with 15% (w/w) MTX we re initially prepared with an emulsion speed of 1250rpm and a 8% (w/w) glutaraldehyde concentration based on the excellent MXN release properties obtained in past studies with BSAMS prepared with those conditions.11 MTX loaded BSA-MS

PAGE 208

208 produced over an 80% yield and ove r a 70% theoretical yield. MTX in situ loaded DNA-MS were prepared with optimized synthesis conditio ns from Chapter 4 and produced excellent yields of over 80%. Based on these results, BSA-MS we re re-synthesized with the same conditions used to prepare the DNA-MS to determine if increasing the mixer speed would increase the yields. Re-synthesized BSA-MS produced yields similar to thos e obtained at the 1250rpm mixer speed. These results were consistent with data obtained in Chapter 4 which illustrated that mixer speeds past 1250rpm do not influence MS yields. BSA-MS were also in situ loaded with 30% (w/w) MTX to determine the maximum loading capa bilities of BSA. BSA-MS with 30% (w/w) MTX loadings produced lower yield and theore tical yield values th an the 15% (w/w) MTX loaded BSA-MS prepared at both the 1250rpm and 1550rpm mixer speeds, Table 6-1. Table 6-1. Percent yield and theoretical yield values for drug loaded DNA-MS and BSA-MS. Condition Percent yield (%) Percent theoretical yield (%) 15% (w/w) BSA-MTX-MS (1250rpm) 84 75 15% (w/w) BSA-MTX-MS (1550rpm) 81 71 15% (w/w) DNA-MTX-MS 93 83 30% (w/w) BSA-MTX-MS 71 58 30% (w/w) BSA-5FU-MS 50 41 30% (w/w) DNA-5FU-MS 59 50 BSA-MS and DNA-MS were in situ loaded with 30% (w/w) 5-FU to further analyze the drug loading capabilities of BSA and DNA. BS A-MS produced yield an d theoretical yield values of over 40% whereas the DNA-MS produced yield and theoretical yield values of over 50%, Table 6-1. It was noted that the DNA-MS produced highe r yield and theoretical yield values than the BSA-MS for both in situ loaded drug conditions. The percent yield values for the DNA-MS were approximately 10% higher than the BSA-MS, Figures 6-1 and 6-2.

PAGE 209

209 0 20 40 60 80 100 15%(w/w)MTX BSAMS (1250rpm) 15%(w/w)MTX BSAMS (1550rpm) 15%(w/w)MTX DNAMS (1550rpm)%Yield Figure 6-1. Percent yield values for 15% (w/w) MTX in situ loaded BSA-MS and DNA-MS. 0 20 40 60 80 10030%(w/w)5-FU BSAMS30%(w/w)5-FU DNAMS%Yield Figure 6-2. Percent yield values for 30% (w/w) 5-FU in situ loaded BSA-MS and DNA-MS. Dry particle size. Each MTX in situ loaded condition produced over 60% of all particles with diameters within the 1m to 10m size range, Table 6-2. The BSA-MS produced smaller particles than the DNA-MS. The MTX in situ loaded DNA-MS displayed evidence of the production of aggregates with 8% of all particles produced displa ying diameters of greater than 10m.

PAGE 210

210 Table 6-2. Dry mean particle diam eter and size range values for MTX in situ loaded BSA-MS and DNA-MS. Condition Dry mean particle diameter (m) MS in 1m to 10m size range (%) MS larger than 10m (%) 15% (w/w) MTX BSA-MS (1250rpm) 2.9 2.7 67 1 15% (w/w) MTX BSA-MS (1550rpm) 2.8 1.4 95 0 15% (w/w) MTX DNA-MS 4.8 5.9 72 8 30% (w/w) MTX BSA-MS 3.0 1.9 96 0 MTX in situ loaded BSA-MS prepared at 1250rpm produced particles with a slightly bimodal distribution and a mean diameter of 2.9 m, Table 6-2. It was noted that the size distributions were narrow with particle diameter s that ranged from 55nm to 14m, Figure 6-3. This is important to note since DNA-MS require filtering at the end of synthesis to obtain particles with narrow diameter ranges. 0 1 2 3 4 0.010.1110100100010000Particle Size (m)Volume % Figure 6-3. Dry particle size distribution for 15% (w/w) MTX in situ loaded BSA-MS prepared at 1250rpm with 8% (w/w) GTA. BSA-MS in situ loaded 15% (w/w) MTX that were re -synthesized at the 1550rpm mixer speed produced a more normalized size distribution with particles ranging from 470nm to 8.5m,

PAGE 211

211 Figure 6-4. This was consistent with data obtained in Chapter 4 which illustrated that particle size distributions normalized as mixer speed increa sed. The normalization of the particle size distribution can be attributed to an increase in finer particles created due to the larger shear forces exerted on the MS during synthesis.10, 57, 107, 130, 131, 134-136 0 2 4 6 8 0.010.1110100100010000Particle Size (m)Volume % 1250rpm 1550rpm Figure 6-4. Dry particle size distributions of 15% (w/w) MTX in situ loaded BSA-MS prepared at 1250rpm and 1550rpm. MTX in situ loaded DNA-MS produced particles with fairly normalized size distributions and mean dry diameters of 4.8m, Figure 6-5. Particle diamet ers ranged from 60nm to 60m; however, it is assumed that if a 20m filter would have been us ed in place of the 70m filter, particles ranging past 20m in diameter would ha ve been eliminated from the yield and the size distribution would have become more normalized as seen in results obtained in Chapter 4. BSA-MS prepared at the 30% (w/w) MTX loading also produced a normalized size distribution with a mean diameter of 3.0m. The size distribution obt ained for the 30% (w/w) MTX loaded BSA-MS was almost identical to that obtained for the 15% (w/w) MTX conditions, Figure 6-6. A one way analysis of varian ce (ANOVA) further supported these results and

PAGE 212

212 indicated that there were no si gnificant size differences between the 15% (w/w) and 30% (w/w) in situ loaded BSA-MS. These findings are consistent with current literature which state that resulting particle size va lues are not affected by drug loading at high concentrations as those seen in these studies.11, 57, 143, 145, 158 The one way ANOVA also illustrated no significant particle size differences among all MTX in situ loaded BSA-MS and DNA-MS conditions. 0 1 2 3 4 0.010.1110100100010000Particle Size ( m ) Volume % Figure 6-5. Dry particle si ze distribution of 15% (w/w) in situ loaded DNA-MS. 0 2 4 6 8 0.010.1110100100010000Particle Size (m)Volume % 15%MTX 30%MTX Figure 6-6. Dry particle size distribution for BSA-MS prepared at 1550rpm and in situ loaded with 15% (w/w) or 30% (w/w) MTX.

PAGE 213

213 BSA-MS in situ loaded with 30% (w/w) 5FU produced normalized si ze distributions with particle diameters ranging from 200nm to 8.5m, Figure 6-7. There was some evidence of slight aggregate formation past the 8.5m size; however, this may be due to the formation of van der Waals interparticle at tractions which occur during drying or upon dispersion in methanol.90, 107, 125, 128, 129, 159 0 2 4 6 8 0.010.1110100100010000Particle Size (m)Volume % Figure 6-7. Dry partic le size distribution for 30% (w/w) 5-FU in situ loaded BSA-MS. 5-FU loaded BSA-MS and DNA-MS produced mean diameters that were very similar in size to one another with greater than 60% of all particles produced falling within the mesosphere size range and less than 5% of a ll particles produced larger than 10m, Table 6-3. A t-test conducted on the 5-FU data collected found no si gnificant size differences between the BSA-MS and DNA-MS. Table 6-3. Dry mean particle diameter values for 5-FU in situ loaded BSA-MS and DNA-MS. Condition Dry mean particle diameter (m) MS in 1m to 10m size range (%) MS larger than 10m in size (%) 30% (w/w) 5-FU BSA-MS 2.9 2.1 95 1 30% (w/w) 5-FU DNA-MS 3.3 4.7 68 4

PAGE 214

214 DNA-MS prepared with in situ loadings of 30% (w/w) 5-FU also produced normalized size distributions with particles ranging from 60nm to 50m, Fi gure 6-8. As noted with the DNA-MTX-MS conditions, it is assumed that pa rticles past 20m in diameter would be eliminated from the yield if a 20m filter is used in place of a 70m filter. 0 1 2 3 4 0.010.1110100100010000Particle Size (m)Volume % Figure 6-8. Dry particle size distribution for 30% (w/w) 5-FU in situ loaded DNA-MS. Scanning electron microscopy. SEM micrographs of the BSA-MS that were in situ loaded with 15% (w/w) MTX depi cted particles with spherical morphologies. The BSA-MS that were synthesized at the 1550rpm mixer speed produ ced particles that were more discreet and uniform than those obtained at the 1250rpm mixer speed, Figure 6-9. SEM micrographs also illustrated that both BSA-MS conditions produced particles with wrinkled or textured topographies. It was further noted that the BSA-MS prepared at the 1550rpm mixer speed, which were also prepared at a higher crossl ink concentration (2mL of GTA vs. 0.382mL of GTA) produced particles with more surface text ure than those prepared at the 1250rpm, Figure 6-10. The wrinkled topographies of the BSAMS prepared at the 1550r pm mixer speed may be attributed to the GTA crosslink agent used dur ing synthesis. Since the early 1980s, BSA-MS

PAGE 215

215 have been prepared in this lab using GTA cr osslinking introduced through the organic phase.4, 11, 57-59, 94 This method of crosslinking yields BSAMS with smooth surface topographies due to GTA crosslinking the BSA-MS at the stabilizing agent/BSA interface.59 The BSA-MS prepared in this study were crosslinked with GTA through the aqueous phase. This form of crosslinking results in GTA interacting with the BSA from w ithin the MS core to the outer layer yielding BSA-MS that are crosslinked more uniformly w ithin. The BSA-MS become wrinkled due to dehydration from within the MS core that results from acet one washing and MS drying.160 Thus it would be expected for BSA-MS with higher GT A concentrations to have more surface texture. Figure 6-9. SEM micrograph of 15% (w/w) MTX in situ loaded BSA-MS prepared at the A) 1250rpm mixer speed and B) 1550rpm mi xer speed (Magnifications: 3,000x). Figure 6-10. SEM micrograph of 15% (w/w) MTX in situ loaded BSA-MS prepared at the 1550rpm mixer speed with 2mL of GTA (Magnification: 11,000x).

PAGE 216

216 SEM micrographs of the 30% (w/w) MTX in situ loaded BSA-MS depi cted discreet and uniform particles that were consistent with data obtained during dry particle size analysis, Figure 6-11. The micrographs also displayed that the 30% (w/w) MTX loaded BSA-MS produced particles with wrinkled surface t opographies that were similar to those seen with the 15% (w/w) MTX BSA-MS. The formation of the wrinkled topographies observed on the BSA-MS was a result of the GTA crosslink agent used during synthesis. Figure 6-11. SEM microgra phs of the 30% (w/w) MTX in situ loaded BSA-MS at magnifications of A) 3,000x and B) 10,000x. SEM micrographs taken of the 15% (w/w) MTX in situ loaded DNA-MS displayed aggregated particles with irre gular morphologies, Figure 6-12. These images were consistent with data obtained during particle size analysis which illustrated the presence of aggregates past the 10m size. The SEM micrographs also illust rated that the MTX loaded DNA-MS were fused together which lead to particle aggregation. The amount of part icle aggregation may have been a result of unloaded MTX since unloaded DNAMS prepared with these conditions produce discreet and uniform particles w ith smooth topographies and spherical morphologi es (Chapter 4).

PAGE 217

217 Figure 6-12. SEM micrographs of 15% (w/w) MTX in situ loaded DNA-MS at magnifications of A) 2,000x and B) 6,000x. BSA-MS and DNA-MS that were in situ loaded with 30% (w/w) 5-FU produced discreet particles with spherical morphol ogies, Figure 6-13. SEM microgra phs taken of the 5-FU loaded BSA-MS depicted particles with wrinkled surf ace topographies, Figure 6-14. The increase in surface texture noted with the BSA-MS is attrib uted to the large concentration of GTA used during synthesis. Further comparisons of the 30% (w/w) 5-FU loaded MS also suggested that the BSA-MS produced larger particles than the DNA-MS. Figure 6-13. SEM micrographs of 30% 5-FU in situ loaded A) BSA-MS and B) DNA-MS (Magnifications: 3,000x).

PAGE 218

218 Figure 6-14. SEM micrograph of 30% (w/w) 5FU loaded BSA-MS (Magnification: 20,000x). SEM micrographs of the 5-FU loaded DNA-MS depicted pa rticles with smooth surface topographies. The 5-FU loaded DNA-MS were al so very uniform in size suggesting that the particle diameters obtained pa st the 2m size during particle size analysis were more representative of MS aggregat es rather than the individu al DNA-MS, Figure 6-15. The formation of DNA-MS aggregates may have been a result of attractive electrostatic or van der Waals interactions between th e particles during the drying pr ocess or upon dispersion into methanol since these interactions play a dominan t role in systems where the particles are less than 10m in size. 124, 125 Figure 6-15. SEM micrograph of 30% 5-FU in situ loaded DNA-MS (Magnification: 10,000x).

PAGE 219

219 In vitro MTX and 5-FU loading efficiency The MTX and 5-FU percent loadings and loading efficiencies of the BSA-MS and DNAMS were measured in triplicate by enzymatic di gestion followed by photometric analysis. MTX and 5-FU percent loadings were calculated usi ng Equation 6-7 to in order to obtain the loading efficiencies for the conditions te sted which were calculated usi ng Equation 6-8. The loading efficiencies for the conditions loaded with 15% (w/w) MTX ranged from 61% to 94%, Table 64. The 30% (w/w) MTX BSA-MS condition produ ced a loading efficiency of 87%. The 30% (w/w) 5-FU conditions produced poor loading effi ciencies of approximately 20% for both BSAMS and DNA-MS conditions. Table 6-4. Loading efficien cy values for MTX and 5-FU in situ loaded BSA-MS and DNA-MS. Condition Experimental loading (%) Theoretical loading (%) Loading efficiency (%) 15% (w/w) MTX BSA-MS (1250rpm) 10.6 2.2 13.9 76.1 15.8 15% (w/w) MTX BSA-MS (1550rpm) 16.0 0.6 17.0 94.3 3.8 15% (w/w) MTX DNA-MS 9.2 1.1 15.0 61.2 7.4 30% (w/w) MTX BSA-MS 24.4 5.0 28.0 87.2 17.8 30% (w/w) 5-FU BSA-MS 5.5 0.6 27.0 20.3 2.1 30% (w/w) 5-FU DNA-MS 5.0 0.7 23.2 21.3 3.0 T-tests were conducted on the MTX conditions to determine if mixer speed, amount of drug loaded, or drug loading material effected MTX loading. The t-tests illustrated no significant loading efficiency differences as a result of mixer speed or percent drug loaded; however the t-tests did find that the BSA-MS entrapped signifi cantly more MTX than the DNAMS (p = 0.002), Figures 6-16 to 6-18. These findi ngs are consistent with current literature which cite that loading efficiencies are independent of mixer speed and amount of drug loaded.134, 139 However, it has also been cited in the literatu re that increases in GTA concentration increase MTX loading which was somewhat observed with these data.161

PAGE 220

220 0 20 40 60 80 1001250rpm1550rpmLoading Efficiency (%) Figure 6-16. Loading efficiency comparison chart for 15% (w/w) MTX in situ loaded BSA-MS prepared at different mixer speeds. 0 20 40 60 80 100 12015%MTX 30%MTX Loading Efficiency (%) Figure 6-17. Loading efficiency comparison ch art for BSA-MS loaded with different MTX concentrations prepared at the 1550rpm mixer speed.

PAGE 221

221 0 20 40 60 80 100DNA BS A Loading Efficiency (%) Figure 6-18. Loading efficiency comparison chart of the 15% (w/w) MTX in situ loaded DNAMS and BSA-MS prepared at the 1550rpm mixer speed. A t-test was also conducted on the 5-FU c onditions to determine if the drug loading material effected 5-FU loading. The t-test illustrated no significant loading efficiency differences between the BSA-MS a nd DNA-MS conditions, Figure 6-19. 0 20 40 60 80 100DNABSALoading Efficiency (%) Figure 6-19. Loading efficiency comparison chart of the 30% 5-FU in situ loaded DNA-MS and BSA-MS.

PAGE 222

222 In vitro MTX and 5-FU release The in vitro MTX and 5-FU release properties of BSA-MS and DNA-MS were measured in 0.05M PBS at a pH of 7.4. Each drug loaded MS condition was test ed in triplicate using minimum sink conditions and incuba ted in 1.250mL of PBS at 37 C to simulate the tumor environment. Each MTX condition tested displayed an initial “burst” release within the first three hours, Figure 6-20 and had ceased re leasing by Day 2, Figure 6-21. 0 20 40 60 80 100 051015202530Time ( hrs ) %MTX Released 15%MTX DNA 15%MTX BSA 1250rpm 30%MTX BSA Figure 6-20. MTX release profiles for in situ loaded DNA-MS and BSA-MS for the first 24 hours. 0 20 40 60 80 100 0510152025303540Time (Days)%MTX Released 15%MTX DNA 15%MTX BSA 1250rpm 30%MTX BSA Figure 6-21. MTX release profiles for in situ loaded DNA-MS and BSA-MS for duration of study.

PAGE 223

223 These in vitro release results are consistent with studies that have been presented in the literature that site that MTX produces an initial 30% to 50% “burst” release from nanoand microparticles and then finish es releasing within 24 hours.58, 162 MTX release analyzed from both 15% (w/w) in situ loaded conditions were diffusion c ontrolled and followed first-order release kinetics where the amount of drug released from the MS decreased over time after it’s initial burst release.67, 70 The MTX release from the 30% (w/w) in situ loaded condition was also diffusion controlled and exhibite d a biphasic release pattern in which the first eight hours followed the Higuchi square root time kinetics model and the MTX rel ease after hour 8 followed first-order release ki netics, Figure 6-22.158 r2 = 0.910.0 0.3 0.6 0.9 1.2 1.5 1.8 0.00.51.01.52.02.53.0Time^0.5 (hrs)Drug relelased (mg) Figure 6-22. Higuchi squa re root time kinetics MTX in vitro release for the 30% (w/w) MTX in situ loaded BSA-MS conditi on (Hours 1 through 8). The 5-FU in situ loaded DNA-MS released a larger con centration of drug at a quicker rate than the BSA-MS condition, Figure 6-23. The in vitro release of 5-FU from DNA-MS appeared to be diffusion controlled and followed the Higuchi square root time kinetics model for release during the first eight hours and fi rst-order kinetics for the durati on of release, Figure 6-24. The 5-FU in situ loaded BSA-MS however, exhibited erosion controlled release and followed the

PAGE 224

224 Higuchi square root time kinetics model for re lease throughout the durati on of the study, Figure 6-25. 0 20 40 60 80 100 0510152025303540Time (Days)%5FU Releas e 30%5FU DNA 30%5FU BSA Figure 6-23. 5-FU release profiles for in situ loaded DNA-MS and BSA-MS. r2 = 0.950.0 0.2 0.4 0.6 0.8 0.00.51.01.52.02.53.0Time^0.5 (hrs)Drug relelased (mg) Figure 6-24. Higuchi square root time kinetics 5-FU in vitro release for the 23% (w/w) 5-FU in situ loaded DNA-MS condition (Hours 1 through 8).

PAGE 225

225 r2 = 0.920.00 0.02 0.04 0.06 05101520253035Time^0.5 (hrs)Drug relelased (mg) Figure 6-25. Higuchi square root time kinetics 5-FU in vitro release for the 27% (w/w) 5-FU in situ loaded BSA-MS condition (Hour 1 through Day 35). The data collected from the 5-FU in situ loaded DNA-MS is consistent to research presented in current literature which site that 5-FU tends to exhibit a biphasic release profile in vitro from various biomaterials such as gelatin albumin, chitosan, and poly(lactic acid)163.158, 161, 164 Release trends as those seen with the 5-FU in situ loaded BSA-MS have also been presented in the literature where the 5-FU is initially released slowly and then exhibits an increase in release due to the degradation of the highly crosslinked MS.165 DNA-MXN-MS Studies Particle analysis Percent yield. The percent yield and theore tical yield values for all MXN in situ loaded DNA-MS were calculated using Equations 65 and 6-6. Each DNA-MXN-MS condition produced yields and theoretical yi elds of over 65%, Table 6-5. Th ere were no clear yield trends observed with regard to MXN c oncentration, Figure 6-26. The MXN in situ loaded DNA-MS produced larger per cent yields values than the nonloaded (i.e. blank) DNA-MS, Figur e 6-27. In addition, the cr osslinked blank DNA-MS produced

PAGE 226

226 larger yields than the blank non-crosslinked DNA-MS, Figure 6-27. The larger yields observed for the MXN in situ loaded DNA-MS may have resulted from the formation of a tighter DNAMS network due to the intercalation inte ractions between MXN and the DNA matrix.166-168 MXN intercalates DNA through inte ractions between MXN’s aminoa lkylamino side chains and the DNA base pairs.166, 168 MXN is also capable of bi nding with DNA through electrostatic interactions between the aminoalkylamino side chains of MXN and the phosphate groups of DNA.166, 167 The electrostatic interactions begin to arise at pH’s 7.4 due to the deprotonation of the NH3 + groups in the aminoalkylamino chains.167 In addition, the larger yields observed with the crosslinked blank DNA-MS may have been attrib uted to formation of crosslinks in the DNAMS which resulted in a larger re tention of DNA in the yield. Table 6-5. Percent yield and theoretical yield values for DNA-MXN-MS. Condition Percent yield (%) Percent theoretical yield (%) 10% (w/w) MXN 7569 15% (w/w) MXN 8375 25% (w/w) MXN 8067 15% (w/w) MXN (no Gd crosslinking) 9680 Blank DNA-MS 76NA Blank DNA-MS (no Gd crosslinking) 63NA 0 20 40 60 80 10010%MXN15%MXN25%MXN%Yield Figure 6-26. Theoretical yield values for DNA-MXN-MS prepared at varying MXN concentrations.

PAGE 227

227 0 20 40 60 80 10015%MXN15%MXN NoGdBlank DNAMSBlank NoGd%Yield Figure 6-27. Percent yield values for MXN in situ loaded and non-loaded DNA-MS. Dry particle size. Each DNA-MXN-MS and blank DNA-MS produced particles with mean dry diameters of less than 10m in diameter, Table 6-6. DNA-MNX-MS prepared at the 10% (w/w), 15% (w/w), and 25% (w/w) MXN concentrations produced narrow and normalized particle size distributions, Figure 6-28. The 10% (w/w) MXN condition produced a small volume percent of aggregates between the 15 m and 30m diameter range which may have arisen due to interpar ticle bonding. The presen ce of aggregates was reflected in standard deviation of the 10% (w/w) MXN condition, however, a one way ANOVA illustrated no significant size differences among the DNA-MX N-MS prepared at the varying MXN concentrations. Table 6-6. Dry mean particle diameter values for cross linked and non-crosslinked DNA-MXNMS and DNA-MS. Condition Dry mean particle diameter (m) 10% (w/w) DNA-MXN-MS 2.7 4.5 15% (w/w) DNA-MXN-MS 2.1 2.8 25% (w/w) DNA-MXN-MS 1.8 1.9 15% (w/w) DNA-MXN-MS (no Gd crosslinking) 2.8 4.1 DNA-MS 2.2 2.1 DNA-MS (no Gd crosslinking) 5.5 8.7

PAGE 228

228 0 1 2 3 4 5 0.010.1110100100010000Size (m)%Yield 10%MXN 15%MXN 25%MXN Figure 6-28. Dry particle size distribution for DNA-MXN-MS prepared at varying MXN concentrations. Particle size distributions for non-crosslinke d conditions were normalized; however, they exhibited aggregate formation within the 13m to 55m diameter range, Figure 6-29. The aggregates present may have been a result of non-crosslinked DNA material in the yield. A larger degree of aggregates were presen t in the non-crosslinked blank DNA-MS size distributions as compared to the non-crossl inked DNA-MXN-MS, Figure 6-30. This data suggests that MXN may help to bind the DNA-MS together a nd thus reduce the amount of aggregates formed in the yield.

PAGE 229

229 0 1 2 3 4 50.010.1110100100010000Particle Size ( m ) Volume% NoGd Gd Figure 6-29. Dry particle size distribution comparison of DNA-MXN-MS prepared with and without gadolinium crosslinking. 0 1 2 3 4 50.010.1110100100010000Particle Size ( m ) Volume% MXN Blank Figure 6-30. Dry particle si ze distribution comparison of non-crosslinked DNA-MXN-MS and blank DNA-MS. Scanning electron microscopy. SEM micrographs of the DNA-MXN-MS prepared at varying MXN concentrations depicted partic les with spherical morphologies and smooth topographies, Figure 6-31. These results are cons istent with data obtain in Chapter 5 for DNA-

PAGE 230

230 MXN-MS prepared at the 1550r pm mixer speed with 120%MEQ gadolinium crosslinking. SEM micrographs of the DNA-MXN-MS also illust rated that the DNA-MXNMS produced at the 25% (w/w) MXN concentration pro duced smaller particles than those obtained at the 10% (w/w) and 15% (w/w) MXN concentrations confirming re sults obtained during part icle size analysis. In addition, particles produced at the 25% (w/w) MXN concentrati on appear to be more uniform in size. These findings are inconsistent with current literature which cite that increasing the drug concentration increases the MS particle diameters.144, 169 AB C AB C Figure 6-31. SEM micrographs of DNA-MXN-MS prepared at the A) 10% (w/w), B) 15% (w/w), and C) 25% (w/w) MXN concen trations (Magnifications: 2,000x).

PAGE 231

231 DNA-MXN-MS prepared with no gadolinium cr osslinking produced particles with smooth surface topographies; however, some particles di splayed irregular morphologies as compared to DNA-MXN-MS prepared with gadolinium crosslinking, Figure 6-32. In addition, DNA-MXNMS prepared with no crosslinking displayed partic les that were less uniform in size. The same trends were seen for DNA-MS prepared w ithout gadolinium crosslinking, Figure 6-33. Figure 6-32. SEM micrographs of DNA-MXN-MS prepared with A) gadolinium crosslinking and B) no gadolinium crosslinking (Magnifications: 2,000x). Figure 6-33. SEM micrographs of DNA-MS prepared with no gadolinium crosslinking, A) blank DNA-MS and B) MXN in situ loaded DNA-MS (Magnifications: 3,000x).

PAGE 232

232 In vitro MXN loading efficiency The MXN percent loadings and loading efficiencies of the DNA-MXN-MS were measured in triplicate by enzymatic digestion followed by photometric analysis. MXN percent loadings were calculated using Equation 67 and the loading efficiencies were calculated using Equation 6-8. The 10% (w/w) and 15% (w/w) MXN cond itions produced loading efficiencies of over 80% whereas the 25% (w/w) MXN condition only produced loading efficiencies of over 30%, Figure 6-34. It was noted, however, that the 25% (w/w) MXN samples were still blue on the final day of the digestion study suggesting that the entrapped M XN may not have fully released due to incomplete digestion of th e MS. This may explain the low loading efficiency of the 25% (w/w) MXN condition, however it has been cited in the lit erature that the drug loading efficiencies are decreased when the drug payloads are increased.135, 144, 170 0 20 40 60 80 10010%MXN15%MXN25%MXN%Entrapment Efficiency Figure 6-34. Loading efficiency comparison chart for DNA-MXN-MS prepared at varying MXN concentrations.

PAGE 233

233 DNA-MXN-MS prepared with no gadolinium cr osslinking produced loading efficiencies of over 95% which were slightly larger than those obtained for DNA-MXN-MS prepared with gadolinium crosslinking, Figure 6-35. Non-cr osslinked DNA-MXN-MS were fully digested by the end of the digestion study suggesting that incomplete digestion of the crosslinked DNAMXN-MS may have resulted in lower loading effici ency values. Table 6-7 lists the percent drug loading and loading efficiency values for each DN-MXN-MS condition tested. 0 20 40 60 80 100 NoGd Gd Entrapment Efficiency (% ) Figure 6-35. Loading efficiency comparison chart for DNA-MXN-MS prepared at with and without gadolinium crosslinking. Table 6-7. Percent loading and load ing efficiency values for DNA-MXN-MS. Condition Experimental loading (%) Theoretical loading (%) Loading efficiency (%) 10% (w/w) MXN 9.3 1.311.1 83.3 11.9 15% (w/w) MXN 12.4 1.714.8 84.3 11.2 25% (w/w) MXN 8.2 1.4 25.0 32.8 5.6 15% (w/w) MXN (no gadolinium crosslinking) 17.9 0.618.7 96.0 3.3 A one way ANOVA followed by a Tukey’s test for multiple comparisons was conducted on all DNA-MXN-MS collected data and illu strated that the 25 % (w/w) MXN condition

PAGE 234

234 produced significantly lower loading efficiencies than the 10% (w/w) and the 15% (w/w) MXN conditions (p = 0.002). There we re no significant loading effici ency differences between the 10% (w/w) and 15% (w/w) MXN c onditions. A t-test also illu strated no significant loading efficiency differences between the noncrosslinked and crosslinked DNA-MXN-MS. In vitro MXN release The in vitro MXN release properties of the DNA-M XN-MS were measured in 0.05M PBS at a pH of 7.4. Each DNA-MXN-MS conditi on was tested in triplic ate using minimum sink conditions (i.e. low volume) and in cubated in 1.250mL of PBS at 37 C to simulate the tumor environment. DNA-MXN-MS prepared with va rying MXN concentrations each released an initial “burst” of MXN within the first two hours with the 10% (w/w) MXN condition producing the largest and quickest MXN release, Figure 6-36. The 25% (w/w) MXN condition displayed the slowest sustained release within the first 24 hours followed by the 15% (w/w) and 10% (w/w) MXN conditions. The slower release of MXN fr om the 25% (w/w) condi tion suggests that the intercalation of MXN within the DNA matrix helps to control the release of MXN in vitro. MXN ceased to release for the 10% (w/w) a nd 15% (w/w) MXN conditions after 24 hours; however, the 25% (w/w) MXN condition exhibited an increase in release between days 3 and 4 and ceased releasing after day 4, Figure 6-37. This increase in release may have been a result of the degradation of the DNA-MXN-MS. These findi ngs are inconsistent with current literature which site that release rates are increased when drug payload is in creased due to a decrease in the integrity of the MS due to the larger drug payload.170 This citing furt her validates our assumptions that the MXN is helping to hold the DNA-MS matrix together in vitro. The in vitro release of MXN from the 10% (w/w) and 15% (w/w) MXN conditi ons appeared to be diffusion controlled and follow first-or der release kinetics. The in vitro release of MXN from the 25%

PAGE 235

235 (w/w) MXN condition appeared to be controlle d by both, diffusion and erosion, and followed first-order release kinetics. 0 20 40 60 80 100 0510152025Time (hrs)% Released 10%MXN 15%MXN 25%MXN Figure 6-36. The first 24 hour MXN release profiles for DNA-MXN-MS prepared at varying MXN concentrations. 0 20 40 60 80 100 01020304050607080Time (Days)%MXN Released 10%MXN 15%MXN 25%MXN Figure 6-37. MXN release profiles for DNA-MXN-MS prepared at varying MXN concentrations. DNA-MXN-MS prepared with out gadolinium cro sslinked also produced an initial “burst” release within the first three hours and ceased releasing after 24 hours, Figures 6-38 and 6-39.

PAGE 236

236 As expected, the non-crosslinked DNA-MXN-MS released MXN quicker than the crosslinked DNA-MXN-MS which exhibited a more sustained release. The in vitro release of MXN from the non-crosslinked and crossl inked DNA-MXN-MS appeared to be diffusion controlled and follow first-order release kinetics. 0 20 40 60 80 100 051015202530Time (hrs)%MXN Released No Gd Gd Figure 6-38. The first 24 hour MXN release profiles for DNA-MXN-MS prepared with and without gadolinium crosslinking. 0 20 40 60 80 100 0510152025303540Time (days)%MXN Released NoGd Gd Figure 6-39. MXN release profiles for DNA-MXNMS prepared with a nd without gadolinium crosslinking.

PAGE 237

237 Conclusions MTX and 5-FU In Situ Loaded DNA-MS and BSA-MS The objective of these studies was to further an alyze the drug loading capabilities of DNA and BSA. MTX and 5-FU in situ loaded BSA-MS and DNA-MS were prepared to produce particles with controlled size dist ributions where at least 60% of all particles prepared were within the mesosphere size range of 1m to 10m and < 5% of a ll particles were greater than 10m in size. Particles less than 1m in diam eter were also acceptabl e and hydrated particle diameters were to be less than 25m. In a ddition, DNA-MS and BSA-MS were sought to obtain drug loadings of 5% (w/w) MTX or 5-FU and releas e drug for more than 24 hours in phosphate buffered saline under minimum sink c onditions. Drug loaded BSA-MS and DNA-MS were compared with respect to particle diamet er, size distribution, mor phology, topography, drug loading, and percent drug release. Particle analysis MTX and 5-FU in situ loaded BSA-MS and DNA-MS produ ced discreet particles with yields that ranged from 40% to 81%. Drug loaded particles di splayed narrow and normal size distributions with mean diameter s of less than 10m. Each MTX and 5-FU loaded MS condition produced particles where greater than 60% of all the particles prepared fell within the 1m to 10m size range and less than 5% of all the particles were larger than 10m with the exception of the MTX loaded DNA-MS. Drug load concentration did not affect the particle diameters and size distributions of MTX loaded BSA-MS, how ever, increasing the mixer speed from 1250rpm to 1550rpm did narrow and normalize size distri butions. There were no significant size differences between MTX loaded BSA-MS and DNA-MS; however, BSA was the only material to produce MTX loaded MS with spherical morphol ogies. MTX did not readily load into DNA. In addition, 5-FU loaded BSA-MS and DNA-MS produced discreet part icles with spherical

PAGE 238

238 morphologies with no signifi cant size differences between them. 5-FU loaded DNA-MS produced particles with smooth topographies, how ever, BSA-MS loaded with MTX and 5-FU exhibited wrinkled surface topographies. The yielding surface topography was a result of the dehydration of the BSA-MS during acetone washing and drying due to the large conc entration of aqueous glutaraldehyde used during synthesis. It is expected that using a lower concentration of glutaraldehyde would result in smoother BSA-MS topographies. In vitro MTX and 5-FU loading efficiency and release The loading efficiencies for the MTX load ed conditions ranged from 61% to 94%. A maximum loading of 24% (w/w) MTX was obtai ned for the BSA-MS and 10% (w/w) for the DNA-MS. The 5-FU loaded conditions produced poor loading efficiencies of approximately 20% for both BSA-MS and DNA-MS. A maximum loading of 6% (w/w) 5-FU was obtained for BSA-MS and 5% (w/w) for DNA-MS. T-tests were conducted on BSA-MS to determin e if mixer speed or drug payload effected MTX loading. These t-tests illustrated no significan t loading efficiency differences as a result of mixer speed or percent drug loaded; however the t-tests did find that the BSA-MS entrapped significantly more MTX than the DNA-MS (p = 0.002) A t-test was also conducted on the 5-FU conditions to determine if the drug loading material effected 5-FU loading. The t-test illustrated no significant loading efficien cy differences between the BSA-MS and DNA-MS conditions. MTX loaded BSA-MS released higher concentra tions of MTX at a quick er rate than those obtained with the MTX loaded DNA-MS. Each MTX condition tested displayed an initial “burst” release within the first three hours, a nd had ceased releasing by Day 2. MTX release analyzed from both BSA-MS and DNA-MS 15% (w/w) in situ loaded conditions were found to be diffusion controlled and follow first-order rele ase kinetics. MTX con centrations released from the BSA-MS and DNA-MS were found to decrease over time after the in itial burst release.

PAGE 239

239 The MTX release from the 30% (w/w) in situ loaded BSA-MS condition was also diffusion controlled and exhibited a biphasic release pattern. The first ei ght hours of MTX release were governed by the Higuchi square root time releas e kinetics model and th e MTX release after hour 8 followed first-order release kinetics.158 The 5-FU in situ loaded DNA-MS released a larger con centration of drug at a quicker rate than the BSA-MS condition. The in vitro release of 5-FU from DNA-MS appeared to be diffusion controlled and followed th e Higuchi square root time ki netics model for release during the first eight hours and firstorder kinetics for the durati on of release. The 5-FU in situ loaded BSA-MS however; exhibited erosion controlled release and followed the Higuchi square root time kinetics model for release th roughout the duration of the study. The sustained release of 5FU achieved with the BSA-MS may have been attr ibuted to the slow degradation of the highly crosslinked BSA-MS. Overall conclusions Overall, BSA-MS produced optimal particle pr operties for intratumoral chemotherapy and were able to load MTX more efficiently than DNA-MS. BSA-MS entrapped 14% (w/w) more MTX than DNA-MS and released MTX at larger c oncentrations at a faster rate. MTX release analyzed from both BSA-MS and DNA-MS were found to be diffusion controlled and follow first-order release kinetics. MTX release from BSA-MS was f ound to be diffusion controlled and exhibit a biphasic release pa ttern by increasing the MTX re lease from 15% (w/w) to 30% (w/w). Release from 30% (w/w) MTX loaded BSA-MS were governed by the Higuchi square root time release kinetics mode l during the first eight hours and fi rst-order release kinetics after hour 8. DNA-MS and BSA-MS produced equivalent 5-FU loadings with optimal MS properties; however, DNA-MS released 5-FU at larger concentr ations and at a faster rate than BSA-MS. 5-

PAGE 240

240 FU release from DNA-MS was diffusion controlle d and followed Higuchi and first-order release kinetics, whereas 5-FU release from BSA-MS wa s erosion controlled and followed the Higuchi model throughout the release study duration. DNA-MXN-MS Studies MXN in situ loaded DNA-MS were prepared to determine the maximum drug loading ability of DNA. DNA-MXN-MS were prepared with a 120%MEQ gadolinium crosslink concentration and were loaded with 10% (w/w), 15% (w/w), and 25% (w/w) MXN. The particle diameter, size distribution, mo rphology, topography, drug loading, and percent drug release of the DNA-MXN-MS were evaluated with respect to MXN concentration. DNA-MXN-MS were also prepared with no gadolinium crosslinking to determine if MXN serves as a crosslinking agent to DNA-MS. DNA-MXN-MS were prepared with 15% (w/w) in situ loaded MXN and were compared with respect to crosslinking. Particle analysis DNA-MXN-MS produced discreet particles with yields of over 65%. Each DNA-MXNMS and blank DNA-MS produced particles with mean dry diameters of less than 10m in diameter. A one way ANOVA conducted on the DNA-MXN-MS prepared at the 10% (w/w), 15% (w/w), and 25% (w/w) MXN concentrations illustrated no significant size differences among the DNA-MXN-MS with regard to MXN c oncentration. A t-test conducted on the crosslinked and non-crosslinke d DNA-MXN-MS illustrated that size was not dependent on gadolinium crosslinking ; however, particle morphologies we re less spherical for the noncrosslinked DNA-MXN-MS than for the crosslinked DNA-MXN-MS. DNA-MXN-MS prepared at varying MXN concentrations produced particles with narrow and normalized particle size di stributions, spherical morphol ogies, and smooth topographies. Particle size distributions for non-crosslinke d conditions were normalized; however, they

PAGE 241

241 exhibited aggregate formation within the 13m to 55m diameter range. A larger degree of aggregates were present in the non-crosslinked blank DNA-MS size distributions as compared to the non-crosslinked DNA-MXN-MS which may ha ve been a result of non-crosslinked DNA material in the yield. This data suggests that MXN may help to bind the DNA-MS together through intercalation thus reducing the amount of aggregates formed in the yield. In vitro MXN loading efficiency and release A maximum MXN loading concentration of 18% (w/w) was obtained for DNA-MS and MXN loading efficiencies ranged from 33% to 96%. The 10% (w/w) and 15% (w/w) MXN conditions produced loading effi ciencies of over 80% wherea s the 25% (w/w) MXN condition only produced a significantly lowe r loading efficiency of just over 30%. However, DNA-MXNMS prepared at the 25% (w/w) MXN condition we re not fully digested at the end of the digestion study and the low loadi ng efficiency values obtained at this condition may have been a result of their incomplete digestion sugges ting that MXN may further stabilize the DNA-MS matrix through intercalation. DNA-MXN-MS prepared with no gadoliniu m ion crosslinking produced loading efficiencies of over 95% which were slightly larger than those obtained for DNA-MXN-MS prepared with gadolinium cross linking; however, a t-test found no significant loading efficiency differences between the non-crosslin ked and crosslinked DNA-MXN-MS. The 10% (w/w), 15% (w/w), and 25% (w/w ) DNA-MXN-MS conditions exhibited firstorder release kinetics and each released an initi al “burst” of MXN of ove r 30% within the first two hours with the 10% (w /w) MXN condition producing the larges t and quickest MXN release. The 25% (w/w) MXN condition displayed the slowes t sustained release within the first 24 hours followed by the 15% (w/w) and 10% (w/w) MXN c onditions. The slower release of MXN from the 25% (w/w) condition further suggests that the in tercalation of MXN w ithin the DNA matrix

PAGE 242

242 helps to control the release of MXN in vitro. MXN ceased to release for the 10% (w/w) and 15% (w/w) MXN conditions after 24 hours; however, the 25% (w/w) MXN condition exhibited an increase in release between Days 3 and 4 and ceased releasing after Da y 4 suggesting that MXN release from the 25% (w/w) c ondition is erosion c ontrolled whereas MXN release for the 10% (w/w) and the 15% (w/w) MXN conditi ons are diffusion controlled. Overall conclusions Overall, DNA-MS produced a maximum MXN loading of 18% (w/w). Resulting DNAMXN-MS mean diameters were found to not be dependent on gadolinium crosslinking nor MXN payload; however, gadolinium ion crossl inked DNA-MXN-MS produced MS with more spherical morphologies and more narrow and normal size distributions. The in vitro release of MXN from DNA-MS was diffusion controlled and fo llowed first-order release kinetics. The 10% (w/w) MXN condition produced the largest and quickest MXN release followed by the 15% (w/w) and 25% (w/w) MXN conditions. The incomplete digestion and the slow release of MXN from the 25% (w/w) MXN condition suggest th at MXN may stabilize the DNA-MS matrix through intercalation and help control th e release of MXN in vitro.

PAGE 243

243 CHAPTER 7 CONCLUSIONS Overview DNA nano-mesospheres (DNA-MS) were prepared for the first time using a modified steric stabilization method that was developed in this lab for th e preparation of albumin mesomicrospheres. The steric stabilization method was modi fied for DNA-MS synthesis by increasing the initial mixer speed to 1550rpm and by adding a filtration step to the end of the process. These two modifications resulted in discreet DNA-MS with yields of over 70% and optimal particle diameters that ranged fr om 50nm to 10m with normal and narrow size distributions. DNA-MS were prepared with gadolinium cro sslinking through ionic interactions with DNA phosphate groups. Gadolinium crosslinking was confirmed by DNA-MS stability in Grey’s balanced salt solution (BSS) and phos phate buffered saline (PBS). Evidence of gadolinium bonding with DNA was determined via energy dispersive x-ray spectroscopy (EDS) which suggest that these mesospheres may have imaging capabilities. The chemotherapy drug mitoxantrone (MXN) wa s readily incorporated into the DNA-MS during synthesis with MXN entrap ments of over 70%. 5-fluorour acil (5-FU) was also loaded into DNA-MS suggesting the significant po tential for the use of these DNA-MS as biodegradable carriers for intratumoral chem otherapy. MXN and 5-FU release profiles extending over 30 days demonstrated a controlle d and prolonged releas e of the chemotherapy drugs from the DNA-MS. The efficacy of MXN loaded DNA-MS was demonstrated in vitro using a murine Lewis lung carci noma (mLLC) cell line. The in vitro cell study illustrated that MXN loaded DNA-MS produce a cytotoxic respons e on the mLLC cells at doses as low as 1ppm. The controlled and sust ained release properties and in vitro cytotoxicity of drug loaded

PAGE 244

244 DNA-MS suggests that their clinical use may provide less frequent chemotherapy visits, which may substantially improve the quality of life fo r patients suffering from various solid tumor cancers. Overall, these DNA nano-mesospheres may offe r a possible alterna tive to conventional oral or intravenous therapies by providing local ized, controlled, and pr olonged chemotherapy. DNA Nano-Meso-Microspheres Synthesis 1. A stabilizing agent concentra tion of 5% (w/v) CAB (cellulo se acetate butyrate in 1,2dichloroethane) was found to produce sphe rical and discreet DNA-MS the most efficiently. 2. Chromium, gadolinium, and iron trivalent cat ionic crosslinking agents were found to react instantaneously with DNA, whereas reac tions with glutaraldehyde take 2 hours and reactions with genipin take over 72 hours. 3. DNA-MS synthesized with ionic and covalent crosslinking agents produced dry mean diameters of less than 20m; however, geni pin was unable to crosslink DNA (i.e. DNAMS did not turn blue) and chromium, iron, and glutaraldehyde crosslinked DNA-MS displayed multimodal dry particle size distri butions suggesting that they may not be optimal crosslinking agents for DNA-MS. 4. Gadolinium crosslinked DNA-MS produced th e smallest dry (2.6m) and hydrated (12.1m) particles, with the largest zeta pot ential values (-45.3mV), and the most narrow and normal size distributions of all the cros slinking agents tested. In addition, the gadolinium crosslinked DNA-MS displayed the mo st stable dispersability in Grey’s BSS with dispersion times exceeding 48 hours. 5. EDS analysis of DNA-MS confirmed the pres ence of the chromium, gadolinium, and iron cationic crosslinking agents st rongly suggesting that the tr ivalent cations are indeed chemically bonding with the DNA. 6. Filtration removed aggregates and particles over 20m in diameter, normalized dry and hydrated particle size distri butions, and produced only an18% decrease in yield, as compared to the non-filtered yield, for gadolinium crosslinked DNA-MS. 7. Increasing the stir speed from 950rpm to 1550rpm produced particles with mean diameters of less than 10m and normalized particle size distribut ions for all non-drug loaded and drug loaded DNA-MS and BSA-MS (i.e. MXN, MTX, 5-FU). 8. Drug pay load did not affect the particle di ameters and size distributions of MXN, MTX, and 5-FU loaded DNA-MS and BSA-MS.

PAGE 245

245 9. Increasing the gadolinium cross link concentration from 20%MEQ to 120%MEQ did not affect the mean dry particle diameters or the particle size distributions for DNA-MXNMS. 10. Gadolinium crosslinked DNA-MS and DNA-M XN-MS produce less aggregates than non-crosslinked DNA-MS and DNA-MXN-MS. 11. DNA-MXN-MS prepared at the 1550rpm mixe r speed with a gadolinium concentration of 120%MEQ produced the most optimal MS results with yields over 85%, dry particle diameters less than 5m, hydrated particle diameters less than 20m, normalized and narrow size distributions, and ex cellent stability in PBS. 12. EDS analysis confirmed the presence of gadolinium in DNA-MXN-MS prepared with gadolinium concentrations in the range of 20%MEQ to 120%MEQ with the 120%MEQ condition producing the largest counts of both gadolinium and chlorine. 13. Zeta potential analysis confirmed the in situ loading of the cationic drug MXN into the DNA-MXN-MS through a significant increase in DNA-MS surface charge (i.e. -45.3mV to -22.4mV). 14. BSA-MS in situ loaded with MTX produced discr eet particles with spherical morphologies and wrinkl ed topographies. DNA-MS in situ loaded with MTX produced discreet particles with irregular morphologies and topographies. 15. BSA-MS in situ loaded with 5-FU produced disc reet particles with spherical morphologies and wrinkl ed topographies. DNA-MS in situ loaded with 5-FU produced discreet particles with spherical morphologies an d smooth topographies. Drug Loading and Release 1. Increasing the gadolinium concentration from 20%MEQ to 120%MEQ for DNA-MXN-MS decreased MXN loading efficiencies from 91% to 84% and decreased MXN loadings from 13.5% to 12.4%; however, conditions were statistically comparable. Increasing the gadolinium concentration also decreased the in vitro release of MXN into phosphate buffered saline. 2. MXN loading and loading efficiencies for gadolinium crosslinked DNA-MXN-MS were found to be statistically comparab le to those obtained with MXN in situ loaded albumin and gelatin microspheres. 3. Non-crosslinked DNA-MXN-MS loaded 6% (w/w) more MXN than gadolinium crosslinked DNA-MXN-MS (i.e. Non-cross linked max loading was 18% (w/w) MXN where gadolinium crosslinked max loading was 12.4% (w/w) MXN). 4. DNA-MXN-MS prepared with gadolinium concentrations that ranged from 20%MEQ to 120%MEQ each produced a “drug burst” release within the first 24 hours; however, a sustained MXN release was measured up to 75 days.

PAGE 246

246 5. DNA-MXN-MS prepared at the 120%MEQ gadolinium crosslink concentration were found to release MXN most efficiently (i.e. a significant increase in MXN release was found between Day 1 and Day 75). 6. Increasing the MXN payload from 10% (w/w ) to 25% (w/w) decreased MXN loading efficiencies from 83% to 33% with a maximu m MXN loading of 18% (w/w) obtained. In addition, increasing the MXN payload from 10% (w/w) to 25% (w/w) decreased the in vitro release of MXN into phosphate buffered saline; however, low MXN loading efficiency and in vitro release values obtained for th e 25% (w/w) MXN condition may have been a result of incomplete diges tion which was observed during analysis. 7. The sustained release of MXN was measured for up to 35 days for the 10% (w/w), 15% (w/w), and 25% (w/w) DNA-MXN-MS condi tions. Each MXN condition exhibited diffusion controlled first-order release kinetics after releasi ng an initial “burst” of over 30% MXN within the first two hours. 8. BSA-MS entrapped 14% (w/w) more MTX than DNA-MS and released larger concentrations of MTX at a faster rate (i.e. BSA max loading was 24% (w/w) MTX and DNA max loading was 10% (w/w) MTX). 9. MTX release from both BSA-MS and DNA-MS were measured up to 35 days and were found to be diffusion controlled and follow firs t-order release kineti cs after the initial burst exhibited during the first three hours. 10. Increasing MTX loading from 15% (w/w) to 30 % (w/w) resulted in a diffusion controlled biphasic release pattern wher e MTX release was governed by the Higuchi square root time release kinetics model during the first eight hours and first-or der release kinetics after hour 8. 11. 5-FU loaded conditions produced poor loading efficiencies of approximately 20% for both BSA-MS and DNA-MS. A maximum loading of 6% (w/ w) 5-FU was obtained for BSA-MS and 5% (w/w) for DNA-MS. 12. DNA-MS released 5-FU at larger concentrations and at a faster rate than BSA-MS. 5-FU release from DNA-MS was diffusion controlled and followed Higuchi and first-order release kinetics, whereas 5-FU release from BSA-MS was erosion controlled and followed the Higuchi model throughout the release study duration. In vitro Cell Growth and Cytotoxicity 1. Human dermal fibroblast growth data and opt ical microscopy images illustrated that DNA derived from herring testes does not e licit an anti-proliferative response or a cytotoxic response in vitro. 2. DNA-MS crosslinked with chromium and gadolinium produced the best in vitro fibroblast growth results with proliferation rates th at either exceeded or were comparable

PAGE 247

247 to the media with cells control condi tion at both the 25g and 100g DNA-MS concentrations. 3. Iron crosslinked DNA-MS produced significantly lower prolifer ation rates than the media with cells control group for both 25g a nd 100g DNA-MS concentrations indicating that iron at DNA-MS concentrations as low as 25g elicit negative fibroblast growth responses in vitro. 4. Glutaraldehyde crosslinked DNA-MS elicited a negative fibr oblast growth response at the 100g DNA-MS concentration and not at the 25g DNA-MS concentration indicating that glutaraldehyde crosslinked DNA-MS elic it negative fibroblast growth responses at DNA-MS concentrations greater than 25g. 5. The effect of MXN on the in vitro viability of murine Lewis lung carcinoma cells was found to be dose dependent. The 1g/mL DNA-MXN-MS dose did not elicit a cytotoxic response until Day 2; however, the 10g/mL and 25g/mL DNA-MXN-MS dose conditions exhibited cytotoxic ity responses by Day 1. 6. DNA-MXN-MS conditions elicit ed significantly higher mu rine Lewis lung carcinoma cytotoxicities than the free MXN conditions at each dose suggesting that the gadolinium in the DNA-MXN-MS contributes to the in vitro cytotoxicity or that DNA-MXN-MS are more readily taken up by the murine Le wis lung carcinoma cells than free MXN.

PAGE 248

248 CHAPTER 8 FUTURE STUDIES The research presented in this dissertation wa s devoted to the synthe sis and properties of drug loaded DNA nano-meso-microspheres (DNA-MS) for intratumoral chemotherapy applications. This section highlights research topics that may be of interest to pursue for future studies. 1. Synthesis and characterization of DNA-MS crosslinked with genipin. Crosslink reaction studies presented in Chapter 3 illustrated that DNA did interact with genipin after 24 hours as indicated from the change of color of the DNA solution from white to a purplish-yellow. DNA-MS s ynthesis studies should be repeated to include longer reaction times for genipin crosslinking. DNAMS studies should look at 2, 12, 24, and 48 hour time points to determine if genipin can be used as a crosslinking agent for DNAMS. Crosslinking should be confirmed in phos phate buffered saline at a pH of 7.4 using hydrated particle sizing methods ou tlined in Chapters 3 through 6. 2. In vivo efficacy and imaging studies with mitoxantrone loaded DNA-MS crosslinked with gadolinium. In vivo studies should be conducted us ing our 16/C murine mammary adenocarcinoma cell model to determine the in vivo efficacy of mitoxantrone loaded DNA-MS. Studies should be c onducted using similar methods as those presented in Dr. Brett Almond’s dissertation.4 Data obtained with DNA-MS should be compared to BSAMS data. In vivo studies should also be conducted to determine the imaging potential of gadolinium crosslinked DNA-MS. Studies should be carried out with the assistance of faculty at the Brain Institute. It would also be interest ing to determine if gadolinium crosslinked DNA-MS could be used for magnetically guided therapy. 3. In vivo distribution testing of mito xantrone loaded DNA-MS. In vivo studies should be conducted to determine the distribution of DNA-MS within the tu mor interstitium and throughout the body using the 16/C murine mammary adenocarcinoma cell model. 4. Synthesis, characterization, and in vitro evaluations of mitoxantrone loaded DNAMS crosslinked with glutaraldehyde and gadolinium. Glutaraldehyde is assumed to crosslink DNA through the base pairs whereas gadolinium crosslinks DNA through the phosphate groups. DNA-MS synthe sis should be repeated us ing both crosslinking agents simultaneously to determine their effect on MS morphology, topogra phy, particle size, and mitoxantrone in vitro entrapment and release. 5. In vitro evaluations of folic acid delivery fr om folic acid and mitoxantrone loaded DNA-MS. Many human tumor cells are known to have overexpressed folic acid receptors on their surfaces such as breas t, ovarian, lung, renal, and colon cancers.171 Folic acid has been conjugated to various th erapeutic particles to enhance their uptake into tumor cells.171, 172 Folic acid conjugation has shown to increase the therapeutic efficacy of chemotherapy agents while al so decreasing their systemic toxicity in vivo

PAGE 249

249 using mice bearing human cancer cell lines due to an increase in tumor uptake.172 In vitro studies should be conducted using folic acid loaded DNA-MS and BSA-MS to determine if the same outcome could be ach ieved through delivery of folic acid from a MS instead of through surface conjugation. DNA-MS and BSA-MS loaded with 15% (w/w) mitoxantrone and 0.5% (w/w) and 1% (w/w) folic aci d were prepared and have been partially characterized w ith the assistance of Karly Jacobsen. Preliminary results are given in Appendix B. In vitro studies should be conducted to further characterize folic acid and mitoxantrone loaded DNA-MS and BSA-MS using a murine Lewis Lung Carcinoma cell line to determine if the delivery of folic acid increases mitoxantrone uptake. The cell study should look at the 1, 2, 4, 8, 12, 24, and 48 hour time points. The cell study should include free mitoxantrone, mitoxantrone loaded DNA-MS and BSA-MS (DNA-MXN-MS and BSA-MXN-MS), and folic acid loaded DNA-MXN-MS and BSAMXN-MS. In vitro entrapment and release studies should also be conducted to determine if folic acid effects the entrapment and release of mitoxantrone. 6. In vitro evaluations of mitoxantrone, methotre xate, or 5-fluorour acil loaded DNABSA blended MS. Studies should be conducted to look at the in vitro entrapment and release properties of mitoxantrone, met hotrexate, or 5-fluorour acil loaded DNA-BSA blended MS. MS should be pr epared with 30% (w/w) drug lo adings and compared at the 80:20 and 50:50 DNA to BSA ratios. Drug entrapment and loadings should be compared to DNA-MS and BSA-MS controls. MS should also be characterized by particle size, morphology, and topography. 7. Delivery of Tumor Killing Bacteria. Non-toxic systemic bact eria cocktails containing bacterium such as Clostridium novyi-NT have shown great potential in eradicating nonoperable solid tumors in mice.173 Their tumor killing potential may be maximized by delivering them intratumorally via MS prepared with biodegradable materials such as DNA or BSA. Bacteria loaded DNA-MS and BSA-MS should be prepared and tested in vitro for efficacy. DNA-MS and BSA-MS should be characterized by particle size, morphology, and surface topography. The tumor k illing properties of the bacteria should be analyzed in vitro with murine Lewis Lung Carci noma cells and compared to mitoxantrone releasing DNA-MS and BSA-MS. 8. Synthesis of fruit or vegetable DNA-MS. Now that DNA-MS synthesis and characterization protocols have been established, studies should be conducted on the synthesis of DNA-MS from fruit or vegeta ble DNA sources such as banana, onion, and tomato. DNA extraction protocols have been modified to produce optimal extraction yields and are presented in Appendix C. Th e next steps for these studies should include DNA purification followed by DNA-MS synthesis, DNA-MS characterization, and drug entrapment and release studies. 9. Synthesis, characterization, and in vitro evaluations of 2-methacryloyloxyethyl phosphorylcholine (MPC), phosphorylcholine, or phosphatidylcholine modified DNA-MS and BSA-MS. MPC phospholipid polymer surfaces have shown excellent in vivo blood compatibility due to hi gh fractions of free water th at can be bound to their surface.174 Surfaces containing MPC or phosphoryl choline groups help to suppress blood

PAGE 250

250 cell adhesion even in the absence of anticoagulants.174 Furthermore, liposomes that have been sterically stabilized or modified with MPC, phosphor ylcholine, phosphatidylcholine phospholipid polymers have prolonged the halflife, enhanced the ce llular permeability of solid tumor cells, provided an “anti-burst” sust ained release, and increased the uptake of chemotherapy agents in vivo. 175-179 These phospholipid modified liposomes have also shown multiple imaging or therap eutic agent loading capabilities.175-178 Therefore, DNAMS and BSA-MS should be prepared followi ng general MS synthesis procedures and modified with the aforementioned phospholip id polymers. DNA-MS may be modified through electrostatic interactions found with the phosphate groups found in the phospholipid polymers. BSA-MS may be modifi ed with covalent in teractions between the amino groups of the phospholipid polymer s and the free aldehyde groups of the BSAMS. For BSA-MS modifications, BSA-MS should be prepared by crosslinking the organic phase, rather than the aqueous phase since organi c phase crosslinking yields many free aldehyde groups at the surface of the BSA-MS.11 DNA-MS and BSA-MS should be characterized by particle size, morphology, and surface topography. The effects of phospholipid modification on drug en trapment, release, and uptake by tumor cells should also be analyzed. 10. Synthesis, Characterization, and In Vitro/Vivo Evaluations of DNA-MS and BSAMS Loaded with Antiangiogenic Drugs. Angiogenesis, which is the formation of new vascular networks in tumors, is thought to be the lead contributor to tumor metastasis.35, 180, 181 Angiogenic vessels develop rapidly in ve ssels due to vascular endothelial growth factor (VEGF).181 Bevacizumab is an antiangiogeni c drug recently approved by the FDA for its use against renal and colon cancers. 180 Bevacizumab is a monoclonal antibody that functions against VEGF and helps to inhibit capillar y and vessel growth.180 Bevacizumab is typically given in combinati on with other chemotherapy drugs such as 5Fluorouracil (5-FU) intravenously and has shown statistically significant survival improvements in human colorect al cancer clinical trials.182 Studies should be conducted to evaluate the in vitro and in vivo efficacy of bevacizumab loaded DNA-MS and BSAMS. Study should compare the in vitro efficacy of bevacizumab, bevacizumab and 5-FU, and 5-FU loaded DNA-MS and BSA-MS agai nst blank MS using murine Lewis Lung Carcinoma cells. The study should then be repeated in vivo using the murine mammary adenocarcinoma model. Loading with re spect to particle size and morphology should also be evaluated.

PAGE 251

251 APPENDIX A EDS CHARACTERISTIC X-RAY DESIGNAT IONS AND ENERGIES FOR DNA NANOMESO-MICROSPHERES Table A-1. Characteristic x-ra y designations and energies for elements analyzed via energy dispersive x-ray spectroscopy.106 Element Designation Energy (keV) Silicon K 1,2 1.739 Carbon K 0.277 Sodium K 1,2 1.041 K 1.071 Phosphorous K 2 2.012 K 1 2.013 Gadolinium M 0.914 M 1.185 M 1.209 L 5.362 L 2 6.025 L 1 6.057 L 6 6.867 L 2,15 7.102 L 7 7.207 L 1 7.785 Chlorine K 2 2.620 K 1 2.622 K 2.815 Chromium L 0.500 L 1 0.582 K 2 5.405 K 1 5.414 K 1,3 5.946 Potassium K 1 3.313 K 1,3 3.589 Iron L 3,4 0.792 K 2 6.390 K 1 6.403

PAGE 252

252 APPENDIX B FOLIC ACID MODIFICATION OF MITOXANTRONE LOADED DNA-MS AND BSA-MS PRELIMINARY RESULTS Introduction Many human tumor cells are known to have ove rexpressed folic acid receptors on their surfaces such as breast, ovarian, lung, renal, and colon cancers.171, 183 Folic acid has been conjugated to various therapeutic particles to enhance their uptake into tumor cells.171, 172 Folic acid conjugation has shown to increase the therap eutic efficacy of chemotherapy agents while also decreasing their systemic toxicity in vivo using mice bearing human cancer cell lines due to an increase in tumor uptake.172 Therefore, folic acid modi fication of DNA-MS and BSA-MS was prepared as a possible enhancement fo r tumor cell uptake. DNA-MS and BSA-MS in situ loaded with 15% (w/w) mitoxantrone (DNA-M XN-MS or BSA-MXN-MS) were modified with 0.5% (w/w) and 1% (w/w) folic acid (FA) to initiate these studies. The particle diameter, size distribution, and morphology of the folic ac id modified DNA-MXN-MS and BSA-MXN-MS were evaluated with respect to the folic acid incorporation. The particle diameters and size distributions were obtained us ing an LS Coulter 13 320 par ticle size analyzer and the morphology was assessed using a field emission s canning electron microscope. The preliminary data presented here was conducted with the as sistance of Karly Jacobsen and summarizes the particle and morphological anal ysis of the FA modified DNA-MXN-MS and BSA-MXN-MS. Materials and Methods Materials DNA sodium salt derived from herring testes Type XIV (DNA), albumin from bovine serum (BSA), cellulose acetate butyrate, HPLC grade 1,2-dichloroethane methanol, gadolinium (III) chloride hexahydrate, 25% (w/w) Grade II aqueous glutaral dehyde solution, mitoxantrone dihydrochloride, and tric hloroacetic acid were purchased fr om the Sigma-Aldrich Company.

PAGE 253

253 Acetone, 70m Spectra/Mesh Nylon filters, and 15mL a nd 50mL polypropylene centrifuge tubes were purchased from Fisher Scientific International. Type I and Type II deionized ultrapure water was prepared with a resistivity of at least 16 M -cm-1 using the Barnstead NANOpure Ultrapure Water System in the lab. Methods All solutions, synthesis, and characterization methods were performed as outlined in Chapters 4 through 6. Folic acid was loaded into MXN loaded aqueous DNA and BSA solutions using the same protocol for MXN loading. Results Preliminary particle size analysis illustrated th at FA did not affect th e particle size of the DNA-MS or DNA-MXN-MS. Each FA modi fied DNA-MS and DNA-MXN-MS condition produced an average dry mean particle size of 2. 7m 3.2m, Figure B-1. FA modification did not affect BSA-MS particle size as well; however, a 2m increase in mean dry particle sizes was observed with respect to MXN loading, Figure B2. Table B-1 gives the dry mean particle diameters for each MS condition tested. 0 1 2 3 4 5 0.010.1110100100010000Particle Size ( m ) Volume % 5%DNA Control 5%DNA 0.5%FA 5%DNA 1.0%FA 5%DNA 15%MXN 5%DNA 15%MXN 0.5%FA 5%DNA 15%MXN 1.0%FA Figure B-1. Particle size di stribution comparison of FA a nd MXN loaded DNA-MS conditions.

PAGE 254

254 0 2 4 6 8 0.010.1110100100010000Particle Size ( m ) Volume % 5% BSA Control 5%BSA 0.5%FA 5%BSA 1.0%FA 5%BSA 15%MXN 5%BSA 15%MXN 0.5%FA 5%BSA 15%MXN 1.0%FA Figure B-2. Particle size distribution comparison of FA and MXN loaded BSA-MS conditions. Table B-1. Dry mean particle diameters for FA and MXN loaded DNA-MS and BSA-MS. Condition Dry mean particle size (m) 5%DNA Control 2.9 2.7 5%DNA 15%MXN 2.7 4.6 5%DNA 0.5%FA 2.8 3.0 5%DNA 0.5%FA 15%MXN 2.9 3.3 5%DNA 1%FA 2.6 3.3 5%DNA 1%FA 15%MXN 2.5 2.7 5%BSA Control 2.8 2.2 5%BSA 15%MXN 6.1 2.6 5%BSA 0.5%FA 2.9 2.6 5%BSA 0.5%FA 15%MXN 5.3 3.4 5%BSA 1%FA 3.2 3.5 5%BSA 1%FA 15%MXN 5.3 2.9 FA modified DNA-MS and DNAMXN-MS produced discreet spherical particles with smooth surface topographies, Figures B-3 and B-4. FA modified BSA-MS also produced spherical particles with smooth topographies, Figure B-5, howeve r, FA modified BSA-MXN-MS produced smooth and discreet particles with dish shaped morphologies, Figure B-6.

PAGE 255

255 Figure B-3. SEM micrographs of DNA-MS with A) 0.5% FA, B) 1% FA, and C) no FA (Magnifications: 3,000x). Figure B-4. SEM micrographs of DNA-MXN-MS with A) 0.5% FA, B) 1% FA, and C) no FA (Magnifications: 2,000x). Figure B-5. SEM micrographs of BSA-MS with A) 0.5% FA, B) 1% FA, and C) no FA (Magnifications: 3,000x).

PAGE 256

256 Figure B-6. SEM micrographs of BSA-MXN-MS with A) 0.5% FA, B) 1% FA, and C) no FA (Magnifications: 3,000x). Preliminary Conclusions The data presented in this appe ndix illustrates that FA may be in situ loaded into DNAMS, BSA-MS, DNA-MXN-MS, and BSA-MXN-MS. The in situ loading of FA into DNA-MS and BSA-MS was visually noted by the light oran ge color of the resulting MS yields. FA modified MS produced optimal part icle size properties with norma l and narrow size distributions and dry diameters in the range of 50nm to15m. Theoretical FA loadings were attempted up to 1.0% (w/w) and should be quantified in the future using a combination of FTIR, to confirm FA binding or loading, and a microbi ological assay, to quantify FA uptake into DNA-MS and BSAMS.184 In addition, in vitro cell uptake tests using murine Lewis lung carcinoma cells were inconclusive (not presented here) in the FA theo retical range tested and should be retested to determine if FA modification increases DNA-MXN-MS and BSA-MXN-MS uptake.

PAGE 257

257 APPENDIX C FRUIT AND VEGETABLE D NA EXTRACTION PROTOCOLS Overview Based on the success obtained with DNA nano-meso-microsphere (DNA-MS) synthesis using DNA derived from herring testes, experiments were set forth to prepare DNA-MS from DNA extracted from fruit and vegetables. DNA ex traction protocols were acquired and modified to produce optimal extraction yields. DNA-MS syntheses were attempted; however, discreet particles were not obtained due to the contam ination of polysaccharides in the extracted DNA yields. DNA extraction protocols are described be low for the extraction of DNA from bananas, onions, and tomatoes. The most efficient extract ion process with the highest DNA yields is obtained using the banana DNA extraction prot ocol; however, each individual protocol will produce good extracted DNA yields. These protocols may also be modified to extract DNA from other living sources. Future studies shoul d include the purificati on of the extracted DNA using DNA purification kits such as those that may be purchased from Promega (i.e. the solution-based Wizard Genomic DNA purification kit). DNA-MS should then be prepared using methods described in Chap ters 4 through 6 using DNA extrac ted from bananas, onions, or tomatoes. Materials 6 Bananas, or bag of yellow Spanish onions, or 8 Plum Tomatoes or 4 Regular Sized Tomatoes Ethanol squirt bottle Refrigerator Test tube rack 12 – 15 50mL polypropylene centrifuge tubes 3 – 6 spatulas 200mL glass bottle with lid 1L plastic bottle with lid 2 magnetic stir bars Sodium lauryl sulfate (o r sodium dodecyl sulfate) Sodium citrate Deionized salt (sodium chloride)

PAGE 258

258 Adolf’s meat tenderizer Ethylenediamine tetraacetic acid (EDTA) 1L glass beaker 2 #6 Coffee filters Metal strainer Parafilm strips Sharpie marker Blender Thermometer 1 Round 2L bowl dish Cutting board Knife Tape 1 stir plate 1 heat plate 1 25mL pipette 1 10mL pipette 2 pipette bulbs Protocols Homogenization Medium 1. Place large magnetic stir bar into the 1L plastic bottle. 2. Weigh out 50g of the sodium lauryl sulfate and add to the bottle. 3. Weigh out 8.770g of the deionized sodi um chloride and add to the bottle. 4. Weigh out 4.410g of the sodium citrate and add to the bottle. 5. Weigh out 0.292g of the ethylenediamine tetraac etic acid (EDTA) and add to the bottle. 6. Add 1L of ultrapure water to the bottle. 7. Cap the bottle and parafilm. 8. Place bottle on stir plate and allow to stir for an hour. Enzyme Solution 1. Place a magnetic stir bar into the 250mL glass bottle. 2. Weigh out 10g of the Adolf’s meat tenderizer and add to the bottle. 3. Add 190mL of ultrapure water to the bottle. 4. Cap the bottle and parafilm. 5. Place bottle on stir plate and allow to stir at low speed throughout the extraction experiment. Banana DNA Extraction 1. Place ethanol squirt bottle in the refrigerator and allow to chill during experiment (best if placed in the refrigerator the night before – keep ethanol in refrigerator at all times). 2. Prepare 1L of the homogeniza tion medium and 200mL of th e enzyme solution (Note: Homogenization medium may alr eady be prepared from previous extraction. Medium

PAGE 259

259 should not be used if it has been more than 7 days since last prepared. Enzyme solution will need to be prepared fresh for each extraction).185, 186 3. Cut up two bananas into inch pieces and place pieces into the 500mL beaker. 4. Pour the homogenization medium over the bananas until all the bananas are covered in medium (~200mL to 300mL). 5. Using one of the spatulas, mash the banana s in the medium. Continue to mash the bananas until the bananas/me dium mixture looks syrupy. 6. Place the strainer over the 2L glass bowl di sh and place the 2 #6 co ffee filters into the strainer making sure not to leave any open sections. 7. Pour the banana/medium mixture into the st rainer and allow the mixture to filter. 8. During this time, repeat steps 3 to 7 for the remaining 4 bananas. 9. While the banana/medium solutions are filtering, label your centrifuge tubes with the following: Banana DNA Your Name Date Place a piece of invisible tape over the la bel to prevent the label from erasing. 10. Once straining is complete, pipette 20mL of the banana DNA solution into a 50mL PP centrifuge tube (Not e: can do up to 6 tubes at a time). 11. Pipette 10mL of the enzyme solution into the centrifuge tube. Cap the tube and shake vigorously for 10 to 20 seconds in orde r to get a nice foam in the tube. 12. Set the tube down and slowly add very cold etha nol to the side of the tube in order to get an ethanol layer on top of the banana DNA so lution. (Note: You should immediately see the banana DNA precipitate into the ethanol layer. Extracted DNA resembles the appearance and consistency of clear mucus.) 13. Cap and parafilm the tube. 14. Place into the refrigerator. 15. Repeat steps 10 through 14 until all the banana DNA solution has been used. 16. Once all the banana DNA solution has been used, place dishes in the sink. 17. Label an unused 50mL PP centrifuge tube using format from step 9. This will be your banana DNA collection tube. 18. Add 5mL of cold ethanol to the bottom of the centrifuge tube. 19. Using a new spatula, collect the extracte d banana DNA from all the tubes in the refrigerator by scooping out the banana DNA and placing it into the collection tube. 20. After all the banana DNA has been collected, add 5mL of cold etha nol to the tube and cap. 21. Centrifuge the tube fo r 5 minutes at ~2600rpm. 22. After centrifuging, dump out etha nol into waste cont ainer being careful to not dump out extracted DNA. 23. Once all the ethanol has been removed from th e extracted banana DNA, cap and parafilm the tube and place in freezer over night. 24. The next day, remove the frozen banana DNA from the freezer and place in the lyophilizer over night. (Not e: Settings should be -45 C>T>-41 C and 10x103Mbar>P>11x10-3Mbar)

PAGE 260

260 Onion DNA Extraction 1. Place top part of blender in freezer and allow to chill through experiment (best if place blender in freezer night before). 2. Place ethanol squirt bottle in the refrigerator and allow to chill during experiment (also best if placed in the refrigerator the night before – keep ethanol in refrigerator at all times). 3. Prepare 1L of the homogeniza tion medium and 200mL of th e enzyme solution (Note: Medium may already be prepared from previ ous extraction. Enzyme solution will need to be prepared fresh for each extraction).185, 186 4. Cut the ends and the sides off of an onion. Dice the remaining onion into half inch squares. 5. Place the diced onions into the round 2L bowl dish. 6. Repeat steps 4 and 5 for four to six onions. 7. Pour approximately 200mL to 300 mL of the homogenization medium185 over the onion in the bowl. 8. Tape the thermometer to the inside of the bow l and turn the heat pl ate on to setting 6. 9. Allow onion mixture to homogenize for 12 minutes at a temperature between 55 C and 60 C1. (Note: Do not allow mixture to get warmer than 60 C) 10. After 12 minutes, remove the thermometer a nd place the onion mixture into the freezer and allow to cool for 8 minutes. 11. During this time, label your centr ifuge tubes with the following: Onion DNA Your Name Date Place a piece of invisible tape over to prevent the label from erasing. 12. After 8 minutes, remove the onion DNA mixture and the top part of the blender from the freezer. 13. Assemble the blender and plug in. 14. Pour the onion DNA mixture into the blender and blend on low (Low button pressed with the Chop button at the same time) for 30 s econds and on high (High button pressed with the Chop button at the same time) for an a dditional 30 seconds. Allow to settle. 15. Place the strainer over the 1L glass beaker and place the 2 #6 co ffee filters into the strainer making sure not to leave any open sections. 16. Pour the blended onion DNA solution into the strainer. Allow complete straining. 17. Once straining is complete, pipette 20mL of the onion DNA solu tion into a 50mL PP centrifuge tube (Not e: can do up to 6 tubes at a time). 18. Pipette 10mL of the enzyme solution into the centrifuge tube. Cap the tube and shake vigorously for 10 to 20 seconds in orde r to get a nice foam in the tube. 19. Set the tube down and slowly add very cold etha nol to the side of the tube in order to get an ethanol layer on top of the onion DNA solution. 20. Cap the tube and parafilm. 21. The onion DNA will precipitat e into the ethanol. 22. Place into the refrigerator. 23. Repeat steps 17 through 22 until all th e onion DNA solution has been used.

PAGE 261

261 Tomato DNA Extraction 1. Place ethanol squirt bottle in the refrigerator and allow to chill during experiment (best if placed in the refrigerator the night before – keep ethanol in refrigerator at all times). 2. Prepare 1L of the homogeniza tion medium and 200mL of th e enzyme solution (Note: Homogenization medium185 may already be prepared from previous extraction. Medium should not be used if it has been more than 7 days since last prepared. Enzyme solution will need to be prepared fresh for each extraction).185, 186 3. Dice up four plum tomatoes (or two regular sized tomatoes) and place pieces into the blender. 4. Pour the homogenization medium over the tomatoes until all the tomatoes are covered in medium (~200mL to 400mL). 5. Blend the tomatoes and the homogenization medium for 20 seconds on the lowest setting. 6. Allow the foam to settle for 5 minutes. 7. Place the strainer over the 2L glass bowl di sh and place the 2 #6 co ffee filters into the strainer making sure not to leave any open sections. 8. Pour the tomato/medium mixture into the st rainer and allow the solution to filter. 9. During this time, repeat steps 3 to 8 for the remaining 4 plum tomatoes (or 2 regular sized tomatoes). 10. While the tomato/medium solutions are filtering, label your centrifuge tubes with the following: Tomato DNA Your Name Date Place a piece of invisible tape over the la bel to prevent the label from erasing. 11. Once straining is complete, pipette 20mL of the tomato DNA solution into a 50mL PP centrifuge tube (Not e: can do up to 6 tubes at a time). 12. Pipette 10mL of the enzyme solution into the centrifuge tube. Cap the tube and shake vigorously for 10 to 20 seconds in orde r to get a nice foam in the tube. 13. Set the tube down and slowly add very cold et hanol down the side of the tube in order to get an ethanol layer on top of the tomato D NA solution. (Note: You should immediately see the tomato DNA precipitate into the ethanol layer. Extracted DNA resembles the appearance and consistency of clear slime. Extracted tomato DNA will be reddish in color.) 14. Cap and parafilm the tube. 15. Place into the refrigerator. 16. Repeat steps 11 through 15 until all of th e tomato DNA solution has been used. 17. Once all the tomato DNA solution has been used, place dishes in the sink. 18. Label an unused 50mL PP centrifuge tube using format from step 10. This will be your tomato DNA collection tube. 19. Add 5mL of cold ethanol to the bottom of the centrifuge tube. 20. Using a new spatula, collect the extracted tomato DNA from all the tubes in the refrigerator by scooping out the tomato DNA and placing it into the collection tube. 21. After all of the tomato DNA has been collected add 5mL of cold ethanol to the tube and cap. 22. Centrifuge the tube fo r 5 minutes at ~2600rpm.

PAGE 262

262 23. After centrifuging, dump out etha nol into waste cont ainer being careful not to dump out the extracted DNA. 24. Once all the ethanol has been removed from th e extracted tomato DNA, cap and parafilm the tube and place in freezer over night. 25. The next day, remove the frozen tomato DNA from the freezer and place in the lyophilizer over night. (Not e: Settings should be -45 C>T>-41 C and 10x103Mbar>P>11x10-3Mbar)

PAGE 263

263 LIST OF REFERENCES 1. Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Ghafoor A, Feuer EJ, Thun MJ. Cancer statistics, 2005. CA Ca ncer J Clin 2005;55(1):10-30. 2. American Cancer Society. Cancer fact s and figures 2006. Atlanta: American Cancer Society; 2006. 3. Cukier D, Gingerelli F, Makari-Juds on G, McCullough VE. Coping with chemotherapy and radiation. New York, NY: The McGraw-Hill Companies, Inc.; 2005. 4. Goldberg EP, Hadba AR, Almond BA, Maro tta JS. Intratumoral cancer chemotherapy and immunotherapy: Opportunities for nonsyste mic preoperative drug delivery. J Pharm Pharmacol 2002;54:159-80. 5. Cheung RY, Rauth AM, Wu XY. In vivo effi cacy and toxicity of intratumorally delivered mitomycin c and its combination with doxorubi cin using microsphere formulations. AntiCancer Drug 2005;16(4):423-33. 6. Jain RK. Delivery of novel therapeutic agents in tumors: Phys iological barriers and strategies. J Natl Cancer I 1989;81:570-6. 7. Emerich DF, Snodgrass P, Lafreniere D, Dean RL, Salzberg H, Marsh J, Perdomo B, Arastu M, Winn SR, Bartus RT. Sustained re lease chemotherapeutic microspheres provide superior efficacy over systemic thera py and local bolus infusions. Pharm Res 2002;19(7):1052-60. 8. CreechJr. O, Krementz ET, Ryan RF, Wi nblad JN. Chemotherapy of cancer: Regional perfusion utilizing an extracorporeal circuit. Ann Surg 1958;148(4):616-32. 9. Anderson JH, McArdle CS, Cooke TG. Microparticulate carrier s as a therapeu tic option in regional cancer therapy: Clinical consider ation. In: N Willmott, JM Daly, eds. Microspheres and regional cancer therapy. Boca Raton, FL: CRC Press, Inc.; 1994. 10. Chen Y, Burton MA, Gray BN. Phar maceutical and methodological aspects of microparticles. In: N Willmott, J Daly, eds. Microspheres and regi onal cancer therapy. Boca Raton, FL: CRC Press, Inc.; 1994. 11. Hadba AR. Synthesis, properties, and in vivo evaluation of sustained release albuminmitoxantrone microsphere formulations fo r nonsystemic treatment of breast cancer and other high mortality cancers [Doctoral Dissertation]. Gainesville, FL: University of Florida; 2001. 12. Almond BA, Hadba AR, Freeman ST, Cuevas BJ, York AM, Detrisac CJ, Goldberg EP. Efficacy of mitoxantrone-loaded albumin micr ospheres for intratumoral chemotherapy of breast cancer. J Contro l Release 2003;91:147-55.

PAGE 264

264 13. York AM. Synthesis, properties, and in vivo evaluations of protein meso/microsphere compositions for intratumoral chemotherapy [Doctoral Dissertation]. Gainesville, FL: University of Florida; 2006. 14. Segura T, Shea LD. Surface-tethered DNA complexes for enhanced gene delivery. Bioconjugate Chem 2002;13:621-9. 15. Breaker RR. Natural and engi neered nucleic acids as tool s to explore biology. Nature 2004;432:838-45. 16. Luo D, Saltzman WM. Synthetic DNA deliver y systems. Nat Biot echnol 2000;18(1):33-7. 17. Iwata K, Sawadaishi T, Nishimura S-I, Tokura S, Nishi N. Utilization of DNA as functional materials: Preparation of filters containing DNA insolubilized with alginic acid gel. Int J Biol Macromol 1996;18:149-50. 18. Zhao C, Liu X, Nomizu M, Nishi N. Blood compatible aspects of DNA-modified polysulfone membrane-protein adsorption and platelet adhesion. Biomaterials 2003;24:3747-55. 19. Tanaka K, Okahata Y. A DNA-lipid complex in organic media and formation of an aligned cast film. J Am Chem Soc 1996;118(44):10679-83. 20. Yamada M, Kato K, Shindo K, Nomizu M, Sakairi N, Yamamoto H, Nishi N. Immobilization of DNA by uv irradiation and its utilization as functional materials. Nucleic Acids Symp Ser 1999;42:103-4. 21. Stopek JB. Biopolymer-microglia cell compos itions for neural tissue repair [Doctoral Dissertation]. Gainesville, FL: University of Florida; 2003. 22. Jaenke RS, Campeneere DD-D, Trouet A. Cadiotoxicity and comparative pharmacokinetics of six anth racyclines in the rabbit. Cancer Res 1980;40:3530-6. 23. Campeneere DD-D, Trouet A. DNA-anthracycl ine complexes. I. Toxicity in mice and chemotherapeutic activity against l1210 le ukemia of daunorubicin-DNA and adriamycinDNA. Eur J Cancer 1980;16(8):981-6. 24. York AM, Stopek JB, Enriquez IV, Martin PJ, Goldberg EP. Albu min-DNA mesosphere synthesis and characterization for local delivery of chemot herapy agents. In: 7th World Biomaterials Congress Transac tions; 2004; Sydney, Australia: Society for Biomaterials; 2004. 25. Hanahan D, Weinberg RA. The hall marks of cancer. Cell 2000;100:57–70. 26. Kaufman D, Chabner BA. Clin ical strategies for cancer trea tment: The role of drugs. In: BA Chabner, DL Longo, eds. Cancer chem otherapy and biothera py: Principles and practice. 2nd ed. Philadelphia, PA : Lippincott-Raven Publishers; 1996.

PAGE 265

265 27. Society AC. Cancer facts and figures 2007. A tlanta, GA: American Cancer Society, Inc.; 2007. 28. Hartwell LH, Lander ES. Report to nationa l cancer advisory board: Ncab working group on biomedical technology. In: National Can cer Advisory Board, ed.: National Cancer Advisory Board; 2005. 29. World Cancer Research Fund, AICR. Food, nutrition a nd the prevention of cancer: A global perspective. In. London, UK: WCRF, AICR; 1997. 30. Cancer trends progress report 2005 update National Cancer Institute, NIH, DHHS, December 2005. (Accessed October 16, 2006, 2006, at http://progressreport.cancer.gov. ) 31. Trouet A, Baurain R, Campeneere DD-D, La yton D, Masquelier M. DNA, liposomes, and proteins as carriers for antitumoral drugs. Recent Results Canc Res 1980;75:229-35. 32. Campeneere DD-D, Jaenke R, Trouet A, Baudon H, Maldague P. DNA-anthracycline complexes. II. Comparative study of the acute lesions induced in mice after intravenous administration of free and DNA bound adriam ycin. Eur J Cancer 1980;16(8):987-98. 33. Trouet A, Sokal G. Clinical studies with daunorubicin-DNA and adriamycin-DNA complexes: A review. Canc Treat Rep 1979;63(5895-898). 34. Weinberg RA. The biology of cancer. New Yo rk, NY: Garland Science, Taylor & Francis Group, LLC; 2007. 35. Harrington L, Bristow RG, Hill RP, Tannock IF. Introduction to cancer biology. In: IF Tannock, RP Hill, RG Bristow, L Harrington, eds. The basic science of oncology. Fourth ed. Toronto, Canada: McGraw-Hill Medical Publishing Division; 2005. 36. Scott RE, Wille JJJ, Wier ML, Hoerl BJ, Maer cklein PB. Carcinogenesis and defects in the integrated control of cell prol iferation and differen tiation. In: CM Veneziale, ed. Control of cell growth and proliferation. Rochester, MN: Van Nostrand Reinhold Company Inc.; 1985. 37. Bertram JS. The molecular biology of cancer. Mol Aspect Med 2001;21:167-223. 38. Pledger WJ. Regulation of cell proliferation: Serum growth factor s control an ordered series of g1 events. In: CM Veneziale, ed. Control of cell growth and proliferation. New York, NY: Van Nostrand Re inhold Company Inc.; 1985. 39. Potten C, Wilson J. Apoptosis: The life and death of cells. New York, NY: Cambridge University Press; 2004. 40. Rode H-J, Eisel D, Frost I, eds. Apoptosis, ce ll death, and cell proliferation. 3rd ed: Roche Diagnostics Corporation; 2004-2005

PAGE 266

266 41. Heath JK. Principles of cell prolifer ation. Birmingham, United Kingdom: Blackwell Science Ltd.; 2001. 42. Bowen ID, Bowen SM, Jones AH. Mitosis a nd apoptosis: Matters of life and death. London, United Kingdom: Chapman & Hall; 1998. 43. Nurse PM. Cyclin dependent kinases and cell cycle control. Bi oscience Rep 2002;22:48799. 44. Hartwell LH. Yeast and cancer Bioscience Rep 2002;22:373-94. 45. Pratt WB, Ruddon RW, Ensminger WD, Maybaum J. Principles of cancer chemotherapy. In: The anticancer drugs. 2nd ed. New York, NY : Oxford University Press, Inc.; 1994. 46. Corrie PG. Cytotoxic chemotherapy: C linical aspects. Medicine 2004;32:25-9 47. Ingwersen KC. Cell cycle kinetics and an tineoplastic agents. In: M Barton-Burke, GM Wilkes, KC Ingwersen, eds. Cancer chemot herapy: A nursing pro cess approach 3rd ed. Sudbury, Massachusetts: Jones and Bartlett Publishers, Inc.; 2001. 48. Srgel F. The return of ehrlich's 'therapia magna sterilisans' and ot her ehrlich concepts?: Series of papers honoring paul ehrlic h on the occasion of his 150th birthday. Chemotherapy 2004;50:6-10. 49. Wilkes GM. Potential toxicities and nur sing management. In: M Barton-Burke, GM Wilkes, KC Ingwersen, eds. Cancer chemot herapy: A nursing pro cess approach. 3rd ed. Sudbury, MA: Jones and Bar tlett Publishers, Inc.; 2001. 50. Rabussay D, Dev NB, Fewell J, Smith LC, Widera G, Zhang L. Enhancement of therapeutic drug and DNA delivery into cells by electroporation. J Phys D 2003;36:348-63. 51. Markman M. Principles of regional antin eoplastic drug delivery. In: M Markman, ed. Regional chemotherapy: Clinical research and practice. Totowa, New Jersey: Humana Press Inc.; 2000. 52. Sinko P, Kohn J. Polymeric drug delivey sy stems. In: MA El-Nokaly, DM Piatt, BA Charpentier, eds. Polymeric delivery system s: Properties and app lications. Washington, D.C.: American Chemical Society; 1993. 53. Kerr DJ. Introduction: Pharmacokinetic basis of regional chemotherapy. In: DJ Kerr, CS McArdle, eds. Regional chemotherapy: Th eory and practice. Amersterdam, The Netherlands: Overseas Publishers Association; 2000. 54. Heller J, Hoffman AS. Drug delivery systems. In: BD Ratner, AS Hoffman, FJ Schoen, JE Lemmons, eds. Biomaterials science: An in toduction to materials in medicine. London, UK: Elsevier Academic Press; 2004.

PAGE 267

267 55. Denkbas EB, Ottenbrite RM. Perspectives on: Chitosan drug delivery systems based on their geometries. J Bioact Compat Pol 2006;21:351-68. 56. Almond BA. Mitoxantrone-loaded albumin microspheres for localized intratumoral chemotherapy of breast cancer [D octoral Dissertation]. Gainesville, FL: University of Florida; 2002. 57. Cuevas BJ. Synthesis and properties of protein micro/mesosphe re-drug compositions designed for intratumoral cancer therapy [D octoral Dissertation]. Gainesville, FL: University of Florida; 2003. 58. Kirk JF. Protein microspheres for contro lled drug delivery and related analysis of biopolymers [Doctoral Dissertation]. Gaines ville, FL: University of Florida; 1997. 59. Longo WE. Albumin microspheres for the co ntrolled release of therapeutic agents [Doctoral Dissertation]. Gainesville, FL: University of Florida; 1983. 60. McArdle CS. Regional chemotherapy for colo rectal liver metastases. In: DJ Kerr, CS McArdle, eds. Regional chemotherapy: Theory and practice. Amsterdam, The Netherlands: Overseas Publishers Association; 2000. 61. Walter KA, Tamargo RJ, Olivi A, Burger PC, Brem H. Intratumoral chemotherapy. Neurosurgery 1995;37:1129-45. 62. Cavalheiro S, Sparapani FVDC, Franco JOB, Silva MCD, Braga FM. Use of bleomycin in intratumoral chemotherapy for cystic cran iopharyngioma. J Ne urosurg 1996;84:124-6. 63. Nakamura Y, Hamabe Y, Ikuta H, Hyon S-H, Kuroda Y. An experimental study of regional chemotherapy using CDDP-loaded microspheres for esophageal cancer. Surg Today 2002;32:335-42. 64. Boiardi A, Eoli M, Salmaggi A, Zappacosta B, Fariselli L, Milanesi I, Broggi G, Silvani A. Efficacy of intratumoral delivery of mitoxantr one in recurrent maligna nt glial tumours. J Neuro-Oncol 2001;54:39-47. 65. Fournier E, Passirani C, Montero-Menei C, Colin N, Br eton P, Sagodira S, Menei P, Benoit J-P. Therapeutic effectiveness of novel 5-fluorouracil-loaded poly(methylidene malonate 2.1.2)-based microspheres on f98 glioma-bearing rats. Cancer 2003;97:2822-9. 66. Yoo GH, Subramanian G, Boinpally RR, Iska nder A, Shehadeh N, Oliver J, Ezzat W, Piechocki MP, Ensley JF, Li n H-S, Shibuya TY, Polin L, Parchment RE. An in vivo evaluation of docetaxel delivered intratum orally in head and neck squamous cell carcinoma. Arch Ot olaryngol 2005;131:418-29. 67. Heller J, Hoffman AS. Drug delivery systems. In: BD Ratner, AS Hoffman, FJ Schoen, JE Lemons, eds. Biomaterials science: An intr oduction to materials in medicine. 2nd ed. San Diego, CL: Elsevier Academic Press; 2004.

PAGE 268

268 68. Matsumura Y, Maeda H. A new concept for macromolecular therapeutics in cancer chemotherapy: Mechanism of tumoritropic ac cumulation of proteins and the antitumor agent smancs. Cancer Res 1986;46:6387-92. 69. Hirst DG. Blood flow and its modulation in malignant tumors. In: N Willmott, JM Daly, eds. Microspheres and regional cancer therapy. Boca Raton, FL: CRC Press, Inc.; 1994. 70. Murty SB, Wei Q, Thanoo BC, DeLuca PP. In vivo release kinetics of octreotide acetate from experimental polymeric microsphere formulations using oil/water and oil/oil processes. AAPS Pharm Sci Tech 2004;5:Article 49. 71. Jacobs IC, Mason NS. Polymer delivery syst ems concepts. In: MA El-Nokaly, DM Piatt, BA Charpentier, eds. Polymeric delivery sy stems: Properties and a pplications. Washington D.C.: American Chemical Society; 1993. 72. Schwartz JB, Simonelli AP, Higuchi WI. Drug release from wax matrices I. J Pharm SciUS 1968;57:274-7. 73. Benita S, Hoffman A, Donbrow M. Microe ncapsulation of paracetamol using polyacrylate resins (eudragit retard), kineti cs of drug release and evaluatio n of kinetic model. J Pharm Pharmacol 1985;37:391-5. 74. Ehtezazi T, Washington C, Melia CD. Fi rst order release rate from porous PLA microspheres with limited exit holes on the exterior surface. J Control Release 2000;66:2738. 75. Higuchi T. Mechanism of su stained-action medication: Theoretical analysis of rate of release of solid drugs dispersed in so lid matrices. J Pharm Sci-US 1963;52:1145-9. 76. Gohel MC, Panchal MK, Jogani VV. N ovel mathematical method for quantitative expression of deviation from the Higuchi m odel. AAPS Pharm Sci T ech 2000;1:Article 31. 77. Higuchi T. Rate of release of medicamen ts from ointment bases containing drugs in suspension. J Pharm Sci-US 1961;50:874-5. 78. Pratt WB, Ruddon RW, Ensminger WD, Ma ybaum J. The anticancer drugs. In: The anticancer drugs. 2nd ed. New York, NY: Ox ford University Press, Inc.; 1994. 79. Doroshow JH. Anthracyclines. In: BA Chabner, DL Longo, eds. Cancer chemotherapy and biotherapy: Principles and pr actice. 2nd ed. Philadelphia, PA : Lipincott-Raven Publishers; 1996. 80. Wilkes GM. Chemotherapeutic agents: Standardized nursing care plans to minimize toxicity. In: M Barton-Burke, GM Wilkes, KC Ingwersen, eds. Cancer chemotherapy: A nursing approach. 3rd ed. Sudbury, MA: J ones and Bartlett Publishers, Inc.; 2001. 81. Dorr RT, Hoff DDV. Drug monographs. In : RT Dorr, DDV Hoff, eds. Cancer chemotherapy handbook. 2nd ed. Norwal k, CT: Appleton & Lange; 1994.

PAGE 269

269 82. Chu E, Allegra CJ. Antifola tes. In: BA Chabner, DL Longo, eds. Cancer chemotherapy and biotherapy: Principles and practice. 2nd ed. Philadelphia, PA: Lippincott-Raven Publishers; 1996. 83. Grem JL. 5-fluoropyrimidines. In: BA Chabne r, DL Longo, eds. Cancer chemotherapy and biotherapy: Principles a nd practice. 2nd ed. Philadel phia, PA: Lippincott-Raven Publishers; 1996. 84. Campbell NA. Biology. Fourth ed. Menlo Park, CA: The Benjamin/Cummings Publishing Company, Inc.; 1996. 85. Bonadio J. Tissue engineering via local gene delivery: Update and future prospects for enhancing the technology. Adv Drug Deliver Rev 2000;44:185-94. 86. Lehninger AL, Nelson DL, Cox MM. Principl es of biochemistry. Second ed. New York, NY: Worth Publishers, Inc.; 1993. 87. Watson JD, Crick FHC. Molecular structure of nucleic acids: A st ructure for deoxyribose nucleic acid. Nature 1953;171:737-8. 88. Trouet A. Carriers for bioactive materials. In: RJ Kostelnik, ed. Polymeric delivery systems. New York, NY: Gordon and Breach Science Publishers, Inc.; 1978. 89. von der Beucken JJJP, Vos MRJ, Thne PC, Hayakawa T, Fukushima T, Okahata Y, Walboomers XF, Sommerdijk NAJM, Nolte RJM, Jansen JA. Fabrication, characterization, and biological assessment of multilayered DNA-coatings for biomaterial purposes. Biomaterials 2006;27:691–701. 90. Morrison ID, Ross S. Colloidal dispersions : Suspensions, emulsions, and foams. New York, NY: John Wiley and Sons, Inc.; 2002. 91. Eichhorn GL, Shin YA. Interaction of me tal ions with polynuc leotides and related compounds. Xii. The relative e ffect of various metal ions on DNA helicity. J Am Chem Soc 1968;90:7323-8. 92. Kazakov SA. Nucleic acid binding and catal ysis by metal ions. In: SM Hecht, ed. Bioorganic chemistry: Nucleic acids. New York, NY: Oxford University Press, Inc.; 1996:244-87. 93. Labiuk SL, Delbaere LTJ, Lee JS. Gamma and ultraviolet radiation cause DNA crosslinking in the presence of metal ions at high ph. Photochem Photobiol 2001;73(6):579-84. 94. McCluskey RA. Synthesis and properties of submicron albumin spheres [Doctoral Dissertation]. Gainesville, FL: University of Florida; 1987. 95. Sung H-W, Huang R-N, Huang LLH, Tsai C-C, C-T. Feasibility study of a natural crosslinking reagent for bi ological tissue fixation. J Bi omed Mater Res 1998;42:560-7.

PAGE 270

270 96. Sung H-W, Chang Y, Liang I-L, Chang W-H, Chen Y-C. Fixation of biological tissues with a naturally occurring crosslinking agent: Fi xation rate and effects of pH, temperature, and initial fixative concentration. J Biomed Mater Res 2000;52(1):77-87. 97. Owen TC, inventor University of South Florida, assignee. Tetr azolium compounds for cell viability assays. US patent 5,185,450. February 9, 1993. 98. Berridge MV, Tan AS. Char acterization of the cellular reduc tion of 3-(4,5-dimethylthiazol2-yl)-2,3-diphenyltetrazolium bromide (MTT) : Subcellular localization, substrate dependence, and involvement of mitochondrial electron trans port in MTT reduction. Arch Biochem Biophys 1993;303:474-82. 99. Lehninger AJ. Biochemistry. 2nd ed. Ne w York, NY: Worth Publishers, Inc.; 1975. 100. Raspaud E, Cruz MO, Sikorav JL, Livol ant F. Precipitation of DNA by polyamines: A polyelectrolyte behavior Biophys J 1998;74(1):381-93. 101. Touyama R, K. InoueK, Y. Takeda Y, M. Yatsuzuka M, T. Ikumoto T, N. Moritome N, Shingu T, Yokoi T, Inouye H. Studies on th e blue pigments produced from genipin and methylamine. II: On the formation mechanisms of brownish-red intermediates leading to the blue pigment formation. Chem Pharm Bull 1994;42:1571-8. 102. Long EC. Fundamentals of nucleic acids. In : SM Hecht, ed. Bioorga nic chemistry: Nucleic acids. New York, NY: Oxford University Press, Inc.; 1996:3-35. 103. Barton JK, Lippard SJ. Heavy metal inter actions with nucleic aci ds. In: TG Spiro, ed. Nucleic acid-metal ion interactions. Ne w York, NY: John Wiley & Sons, Inc.; 1980. 104. Marzilli LG, Kistenmacher TJ, Eichhorn GL Structural principles of metal ion-nucleotide and metal ion-nucleic acid interactions. In: TG Spiro, ed. Nucleic acid-metal ion interactions. New York, NY: John Wiley & Sons, Inc.; 1980. 105. Bloomfield VA, Crothers DM, TinocoJr. I. Nucleic acids: Structures, properties, and functions. Sausalito, CA: Univ ersity Science Books; 2000. 106. Weast RC, Astle MJ, eds. CRC handbook of chemisty and physics. 63rd ed. Boca Raton, FL: CRC Press, Inc.; 1982-1983. 107. Meyers D. Surfaces, interfaces, and collo ids: Principles and applications. 2nd ed. New York, NY: John Wiley & Sons; 1999. 108. Arakawa H, Ahmad R, Naoui M, Tajmir-R iahi H-A. A comparativ e study of calf thymus DNA binding to Cr(III) and Cr(VI) ions: Ev idence for the guanine N-7-chromiumphosphate chelate formation. J Biol Chem 2000;275:10150–3. 109. Ouameur AA, Arakawa H, Ahmad R, Naoui M, Tajmir-Riahi HA. A comparative study of Fe(II) and Fe(III) interacti ons with DNA duplex: Major a nd minor grooves bindings. DNA Cell Biol 2005;24:394–401.

PAGE 271

271 110. Shindo D, Oikawa T. Anal ytical electron microscopy for ma terials science. Tokyo, Japan: Springer-Verlag; 2002. 111. Hu G-F. Copper stimulates proliferati on of human endothelial cells under culture. J Cell Biochem 1998;69:326-35. 112. Liu R, Liu W, Doctrow SR, Baudry M. Iron toxicity in or ganotypic cultures of hippocampal slices: Role of reactive oxygen species. J Neurochem 2003;85:492–502. 113. Kehrer JP. The Haber–Weiss reactio n and mechanisms of toxicity. Toxicology 2000;149:43–50. 114. Cornett CR, Markesbery WR, Ehmann WD Imbalances of trace elements related to oxidative damage in Alzheimer's dis ease brain. Neurotoxi cology 1998;19:339-45. 115. Jenner P. Oxidative stress in Park inson’s disease. Ann Neurol 2003;53:S26–S38. 116. Shrivastava HY, Ravikumar T, Shanmuga sundaram N, Babu M, Na ir BU. Cytotoxicity studies of chromium(III) complexes on human dermal fibroblasts. Free Radical Bio Med 2005;38:58– 69. 117. Heckl S, Debus J, Jenne J, Pipkorn R, Wa ldeck W, Spring H, Rastert R, von der Lieth CW, Braun K. CNN-Gd3+ enables cell nucleus molecular imag ing of prostate cancer cells: The last 600 nm. Cancer Res 2002;62:7018–24. 118. Pathak AP. Magnetic resonance imaging of tumor physiology. In: PV Prasad, ed. Magnetic resonance imaging: Methods and biologic ap plications. Totowa, New Jersey: Humana Press, Inc.; 2006. 119. Zhitkovich A, Voitkun V, Costa M. Fo rmation of the amino acid-DNA complexes by hexavalent and trivalent chromium in vitro: Importance of trivalent chromium and the phosphate group. Biochemistry 1996;35:7275-82. 120. Meade VM, Burton MA, Gray BN, Self GW Distribution of differe nt sized microspheres in experimental hepatic tumors. Eu r J Cancer Clin Oncol 1987;23:37-41. 121. Teipel U, Bremser JK. Particle characte rization. In: U Teipel, ed Energetic materials: Particle processing and characterization. Weinheim, Germany: WILEY-VCH Verlag GmbH & Co. KGaA; 2005. 122. Barbosa-Canovas GV, Ortega-Rivas E, Juliano P, Yan H. Food powders: Physical properties, processing, and functionality. New York, NY: Kluwer Academic/Plenum Publishers 2005. 123. Cazabat AM, Chatenay D, Guering P, Urbach W, Langevin D, Meunier J. Dynamic processes in water/oil microemulsions. In: HL Rosano, M Clausse, eds. Microemulsion systems. New York, NY: Marcel Dekker, Inc.; 1987.

PAGE 272

272 124. Feng JQ, Hays DA. Relative importance of electrostatic forces on powder particles. Powder Technol 2003;135-136:65– 75. 125. Saltiel C, Chen Q, Manickavasagam S, Schadler LS, Siegel RW, Menguc MP. Identification of the dispersion behavior of surface treated nanoscale powders. J Nanopart Res 2004;6:35–46. 126. Klokova TP, Volodin YA, Glagoleva OF. Ef fect of ultrasound on the colloidal-disperse properties of petroleum systems. Chem Tech Fuel Oil 2006;42:43-6. 127. Yong V, Hahn HT. Processing and prope rties of sic/vinyl ester nanocomposites. Nanotechnology 2004;15:1338–43. 128. Martinez CJ, Liu J, Rhode SK, Luijten E, Weeks ER, Lewis JA. Interparticle interactions and direct imaging of colloidal phases assemb led from microsphere-nanoparticle mixtures. Langmuir 2005;21:9978-89. 129. Zhu X, Jiang D, Tan S, Zhang Z. Dispersi on properties of alumina powders in silica sol. J Eur Ceram Soc 2001;21:2879–85. 130. Mateovic T, Kriznar B, Bogataj M, Mrhar A. The influence of stirring rate on biopharmaceutical properties of Eudragit RS microspheres. J Microencapsul 2002;19:29 36. 131. Benoit JP, Puisieux F. Microcapsu les and microspheres for embolization and chemoembolization. In: P Guiot, P Couvr eur, eds. Polymeric nanoparticles and microspheres. Boca Raton, FL: CRC Press, Inc.; 1986. 132. Kawaguchi Y, Oishi T. Synthesis an d properties of thermoplastic expandable microspheres: The relation betw een crosslinking density and expandable property. J Appl Polym Sci 2004;93:505–12. 133. Jameela SR, Jayakrishnan A. Glutaralde hyde cross-linked chitosan microspheres as a long acting biodegradable drug deliver y vehicle: Studies on the in v itro release of mitoxantrone and in vivo degradation of microspheres in rat muscle. Biomat erials 1995;15:769-75. 134. Yang Q, Owusu-Ababio G. Biodegradable progesterone microsphere delivery system for osteoporosis therapy. Drug Dev Ind Pharm 2000;26:61–70. 135. Gabor F, Ertl B, Wirth M, Mallinger R. Ketoprofen-poly(d,l-l actic-co-glycolic acid) microspheres: Influence of ma nufacturing parameters and ty pe of polymer on the release characteristics. J Microencapsul 1999;16:1-12. 136. Yang Q, Williams D, Owusu-Ababio G, Ebube NK. Controlled rel ease tacrine delivery system for the treatment of alzhei mer’s disease. Drug Deliv 2001;8:93–8. 137. Muvaffak A, Gurhan I, Hasirci N. Cytot oxicity of 5-fluorouracil entrapped in gelatin microspheres. J Microen capsul 2004;21(3):293-306.

PAGE 273

273 138. Tong W, Wang L, D’Souza MJ. Evaluation of PLGA microspheres as delivery system for antitumor agent-camptothecin. Dr ug Dev Ind Pharm 2003;29(7):745-56. 139. Yang Q, Williams D, Owusu-Ababio G, Ebube NK. Controlled rel ease tacrine delivery system for the treatment of Alzhei mer’s disease. Dr ug Deliv 2001;8:93-8. 140. Dobson J. Gene therapy progress and prospects: Magnetic na noparticle-based gene delivery. Gene Therapy 2006;13:283-7. 141. Holden SA, Lan Y, Pardo AM, Wesolows ki JS, Gillies SD. Au gmentation of antitumor activity of an antibody -interleukin 2 immunocytokine with chemotherapeutic agents. Clin Cancer Res 2001;7:2862-9. 142. Lee SJ, Sakurai H, Koizumi K, Song GY, Bae YS, Kim H-M, Kang K-S, Surh Y-J, Saiki I, Kim SH. MAPK regulation and caspase activ ation are required in DMNQ S-52 induced apoptosis in lewis lung carcinom a cells. Cancer Lett 2006;233:57-67. 143. Bozda S, apan Y, Vural I, Dalkara T, Dogan AL, Guc D, Hincal AA, Deluca PP. In vitro cytotoxicity of mitoxantrone-inc orporated albumin microspheres on acute promyelocytic leukaemia cells. J Microencapsul 2004;21:751-60. 144. Luftensteiner CP, Viernstein H. Statistic al experimental design based studies on placebo and mitoxantrone-loaded albumin micr ospheres. Int J Pharm 1998;171:87-99. 145. Jameela SR, Jayakrishnan A. Glutaralde hyde cross-linked chitosan microspheres as a long acting biodegradable drug deliv ery vehicle: Studies on the in vitro release of mitoxantrone and in vivo degradation of microspheres in ra t muscle. Biomaterials 1995;16:769-75. 146. Muvaffak A, Gurhan I, Hasirci N. Prolonged cytotoxic effect of colchicine released from biodegradable microspheres. J Bi omed Mater Res B 2004;71B:295-304. 147. Memisoglu-Bilensoy E, Vural I, Bochot A, Renoir JM, Duchene D, Hincal AA. Tamoxifen citrate loaded amphiphilic -cyclodextrin nanoparticles: In vitro characterization and cytotoxicity. J Control Release 2005;104:489–96. 148. Han HD, Lee A, Song CK, Hwang T, Seong H, Lee CO, Shin BC. In vivo distribution and antitumor activity of hepari n-stabilized doxorubi cin-loaded liposomes. Int J Pharm 2006;313:181–8. 149. Morrison ID, Ross S. Colloidal disper sions: Suspensions, emulsions, and foams. New York, NY: John Wiley and Sons, Inc.; 2002. 150. Bayomi MA. Aqueous preparation and evaluation of albumin-chitosan microspheres containing indomethacin. Dr ug Dev Ind Pharm 2004;30:329–39. 151. Oppenheim RC. Nanoparticulate drug deliv ery systems based on gelatin and albumin. In: P Guiot, P Couvreur, eds. Polymeric nanopartic les and microspheres. Boca Raton, FL: CRC Press, Inc.; 1986.

PAGE 274

274 152. Jin J, Song M. Chitosan and chitosan–pe o blend membranes crosslinked by genipin for drug release. J Appl Polym Sci 2006;102:436–44. 153. Mathiowitz E, Kreitz MR, Brannon-Peppa s L. Microencapsulation. Encyc Control Drug Deliv 1999;2:493-546. 154. S. Bozda Y. apan, I. Vural, T. Dalkara, A.L. Dogan, D. Guc, A.A. Hincal, Deluca aPP. In vitro cytotoxicity of mitoxantrone-i ncorporated albumin microspheres on acute promyelocytic leukaemia cells. J Microencapsul 2004;21:751-60. 155. Bellosillo B, Colomer D, Pons G, Gil J. Mitoxantrone, a topoisomerase II inhibitor, induces apoptosis of B-chronic lymphoc ytic leukaemia cells. Brit J Haematol 1998;100:142-6. 156. Bhalla K, Ibrado AM, Tourkina E, Tang C, Grant S, Bullock G, Huang Y, Ponnathpur V, Mahoney ME. High-dose mitoxantrone induces programmed cell death or apoptosis in human myeloid leukemia cel ls. Blood 1993;82:3133-40. 157. Vibet S, Maho K, Gor J, Dubois P, Bougnoux P, Chourpa I. Differential subcellular distribution of mitoxantrone in relation to chemosensitizatio n in two human breast cancer cell lines. Drug Metab Dispos 2007;35:822-8. 158. Dubey RR, Parikh RH. Two-stage optimi zation process for formulation of chitosan microspheres. AAPS Pharm Sci Tech 2004;5:Article 5. 159. von Oss CJ. Interfacial forces in aqueous media. 2nd ed. Boca Raton, FL: Taylor & Francis Group, LLC; 2006. 160. Bahukudumbi P, Carson KH, Rice-Ficht AC, Andrews MJ. On th e diameter and size distributions of bovine serum albumin (B SA)-based microspberes. J Microencapsul 2004;21:787-803. 161. Muvaffak A, Gurhan I, Gunduz U, Hasirc i N. Preparation and characterization of a biodegradable drug targeting system for anticancer drug de livery: Microsphere-antibody conjugate. J Drug Target 2005;13:151–9. 162. Santhi K, Dhanaraj SA, Koshy M, Ponnusa nkar S, Suresh B. Study of biodistribution of methotrexate-loaded bovine serum albumin nanospheres in mice. Drug Dev Ind Pharm 2000;26:1293–6. 163. Chandy T, Das GS, Rao GHR. 5-fluorourac il-loaded chitosan co ated polylactic acid microspheres as biodegradable drug carriers for cerebral tumours. J Microencapsul 2000;17:625 38. 164. Jia WJ, Liu JG, Zhang YD, Wang JW, Wang J, Sun CY, Pan YF, Yu C, Li GZ, Zhang H, Guo ML, Tu MJ, Qian ZY. Preparation, char acterization, and optim ization of pancreastargeted 5-FU loaded magnetic bovine seru m albumin microspheres. J Drug Target 2007;15:14-145.

PAGE 275

275 165. Muvaffak A, Grhan I, Hasirci N. Cytot oxicity of 5-fluorouracil entrapped in gelatin microspheres. J Microe ncapsul 2004;21:293-306. 166. Wang S, Peng T, Yang CF. Investigation on the interaction of DNA and electroactive ligands using a rapid electrochemical me thod. J Biochem Bioph Meth 2003;55:191–204. 167. Feofanov A, Sharonov S, Kudelina I, Fleu ry F, Nabiev I. Localization and molecular interactions of mitoxantrone within living K562 cells as probed by confocal spectral imaging analysis. Biophys J 1997;73:3317-27. 168. Panousis C, R.Phillips D. DNA sequence spec ificity of mitoxantrone. Nucleic Acid Res 1994;22:1342-5. 169. Bhardwaj SB, Shukla AJ, Collins CC. Effect of varying drug loading on particle size distribution and drug release ki netics of verapamil hydrochl oride microspheres prepared with cellulose esters. J Microencapsul 1995;12:71-81. 170. Yang Q, Williams D, Owusu-Ababio G, E bube NK. Controlled release of tacrine delivery system for the treatment of Alzhei mer's disease. Dr ug Deliv 2001;8:93-8. 171. Sun C, Sze R, Zhang M. Folic acid-PEG conjugated superparamagnetic nanoparticles for targeted cellular uptake and detection by MRI. J Biomed Mater Res A 2006;78A:550-7. 172. Kukowska-Latallo JF, Candido KA, Cao Z, Nigavekar SS, Majoros IJ, Thomas TP, Balogh LP, Khan MK, BakerJr. JR. Nanoparticle targ eting of anticancer dr ug improves therapeutic response in animal model of human epithe lial cancer. Cancer Res 2005;65(12):5317-24. 173. Forbes NS. Profile of a bacteria l tumor killer. Nat Biotechnol 2006;24:1484-5. 174. Ishihara K, Nomura H, Mihara T, Kurita K, Iwasaki Y, Nakabayashi N. Why do phospholipid polymers reduce protein adsorp tion? J Biomed Mater Res 1998;39:323-30. 175. Yuan F, Leunig M, Huang SK, Berk DA, Papahadjopoulos D, Jam RK. Microvascular permeability and interstitial penetration of ster ically stabilized (stealth) liposomes in a human tumor xenograft. Ca ncer Res 1994;54:3352-6. 176. Huang SK, Lee K-D, Hong K, Friend DS, Papahadjopoulos D. Micros copic localization of sterically stabilized liposomes in co lon carcinoma-bearing mice. Cancer Res 1992;52:5135-43. 177. Sengupta S, Sasisekharan R. Exploiting nanotechnology to target cancer. Brit J Cancer 2007;96:1315-9. 178. Sengupta S, Eavarone D, Capila I, Zhao G, Watson N, Kiziltepe T, Sasisekharan R. Temporal targeting of tumour cells and ne ovasculature with a nanoscale delivery system. Nature 2005;436:568-72.

PAGE 276

276 179. Mier W, Hoffend J, Haberkorn U, Eisenhut M. Current strategies in tumor-targeting. In: M Los, SB Gibson, eds. Apoptotic pathways as targets for novel therapie s in cancer and other diseases. New York, NY: Springer Sc ience+Business Media, Inc.; 2005. 180. Magnon C, Galaup A, Rouffiac V, Opolon P, Connault E, Rose M, Perricaudet M, Roche A, Germain S, Griscelli F, Lassau N. Dyna mic assessment of antiangiogenic therapy by monitoring both tumoral vascularization and tissue degeneration. Gene Ther 2007;14:10817. 181. Bazan-Peregrino M, Seymour LW, Harri s AL. Gene therapy targeting to tumor endothelium. Cancer Ge ne Ther 2007;14:117-27. 182. Hurwitz H, Fehrenbacher L, Novotny W, Cart wright T, Hainsworth J, Heim W, Berlin J, Baron A, Griffing S, Holmgren E, Ferrara N, Fyfe G, Rogers B, Ross R, Kabbinavar F. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. New Engl J Med 2004;350:2335-42. 183. Zhang Y, Sun C, Kohler N, Zhang M. Se lf-assembled coatings on individual monodisperse magnetite nanoparticles for efficient intracellu lar uptake. Biomedical Devices 2004;6:3340. 184. Tamura T. Microbiological assay of folate s. In: MF Picciano, ELR Stokstad, JF Gregory, eds. Folic acid metabolism in health and di sease: Contemporary issues in clinical nutrition. New York, NY: Wiley-Liss; 1990. 185. Protocol: DNA extraction from onion. (Accessed January 27, 2006, at http://dwb.unl.edu/Teacher/NSF/C08/C08Li nks/www.framingham.k12.ma.us/fhs/depts/sci ence/Biotechnology/prot ocols.html#oniondna. ) 186. Introduction to DNA extractions. Access Excellence @ the Nati onal Museum. (Accessed August 4, 2005, at http://www.accessexcellence.org/AE /AEC/CC/DNA_extractions.html. )

PAGE 277

277 BIOGRAPHICAL SKETCH Iris Vanessa Enriquez Cartagena was born on May 16, 1979 in Ponce, Puerto Rico to parents Iris and Pedro Enriquez. She and her pa rents moved to El Paso, Texas when she turned one. Her younger sister, Laura Susana was born in El Paso, Texas a l ittle after she turned four at which point her family moved to Spangdahlem, Germa ny. Iris and her family spent five exciting and very fun years in Germany before moving down to Melbourne, Florida where Iris attended Holland Elementary, De Laura Junior High School and Satellite High School. While at De Laura, Iris became very interested in math a nd was nominated to interview for a special NASA program entitled NURTURE, which stood for NASA’ s Unique Residential Tutoring program for Up-and-coming Replacement Engineers. Iris wa s one of the 50 students chosen out of the 500 from Brevard County that applied for the prog ram. Iris visited NASA at the Kennedy Space Center four times a year from grades 9 th rough 12 and shadowed engineers in various disciplines. On her last two vi sits, Iris shadowed materials sc ience engineers in the Failure Analysis Department and fell in love with the discipline. Iris was accepted to the Departme nt of Materials Science and E ngineering at the University of Florida in January of 1997. She graduate d from Satellite High School in May of 1997 and started school in June of that year. At the begi nning of her junior year, Iris began a three co-op rotation with the Kimberly-Clark Corporation in Neenah, Wisconsin. While at Kimberly-Clark, her interests in materials scienc e and polymers science grew and developed. Once Iris returned from her co-op rotation, she finished up her senior year. Iris graduated cum laude from the Department of Materials Science and Engineering with a specialty in polymers science in August of 2002 and decided to purs ue graduate school. Iris was accepted to the graduate program in the Department of Materials Science and Engineering in August of 2002 and was awarded the NSF Alliance for Graduate Education and

PAGE 278

278 the Professoriate Fellowship. She began pursuin g her doctorate degree in materials science and engineering and joined Dr. Goldbe rg’s research group in January 2003. Since then, Iris has been working on the development of DNA mesospheres for intratumoral chemot herapy applications. In May of 2004, she was awarded her master’s of science degree and in October of 2005 she married James Schumacher. In August 2007, Ir is was awarded a Doctorate of Philosophy in Materials Science and Engineering from the University of Florida, specializing in biomaterials. After graduation, Iris plans to move to Atlant a with her husband James, to pursue a career in materials science and engineering/biomaterials science at the Kimberly-Clark Corporation.